Kinetic profile of the transcriptome changes induced in the choroid ...

Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932 & 2009 ISCBFM All rights reserved 0271-678X/09 $32.00 www.jcbfm.com

Kinetic profile of the transcriptome changes induced in the choroid plexus by peripheral inflammation Fernanda Marques1, Joaõ C Sousa1, Giovanni Coppola2, Ana M Falcao1, Ana Joaõ Rodrigues1, Daniel H Geschwind2, Nuno Sousa1, Margarida Correia-Neves1 and Joana A Palha1 1

Life and Health Sciences Research Institute (ICVS), School of Health Sciences, University of Minho, Campus Gualtar, Braga, Portugal; 2Department of Neurology, David Geffen School of Medicine-UCLA, Los Angeles, California, USA

The choroid plexus, being part of the blood–brain barriers and responsible for the production of cerebrospinal fluid, is ideally positioned to transmit signals into and out of the brain. This study, using microarray analysis, shows that the mouse choroid plexus displays an acute-phase response after an inflammatory stimulus induced in the periphery by lipopolysaccharide (LPS). Remarkably, the response is specific to a restricted number of genes (out of a total of 24,000 genes analyzed, 252 are up-regulated and 173 are down-regulated) and transient, as it returns to basal conditions within 72 h. The up-regulated genes cluster into families implicated in immune-mediated cascades and in extracellular matrix remodeling, whereas those down-regulated participate in maintenance of the barrier function. Importantly, several acute-phase proteins, whose blood concentrations rise in response to inflammation, may contribute to the effects observed in vivo after LPS injection, as suggested by the differential response of primary choroid plexus epithelial cell cultures to LPS alone or to serum collected from animals exposed to LPS. By modulating the composition of the cerebrospinal fluid, which will ultimately influence the brain parenchyma, the choroid plexus response to inflammation may be of relevance in brain homeostasis in health and disease. Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932; doi:10.1038/jcbfm.2009.15; published online 25 February 2009 Keywords: choroid plexus; cerebrospinal fluid; lipopolysaccharide; blood–brain barrier; inflammation; transcriptome

Introduction The precise mechanisms through which peripheral inflammatory stimuli trigger brain inflammation are poorly understood. Interest on the subject is increasing in light of the importance of inflammation in the central nervous system (CNS) in multiple sclerosis (MS), and the more recent implications of CNS inflammation in neurodegenerative diseases such as Parkinson and Alzheimer diseases. Although most studies addressing communication between the periphery and the CNS focus on the blood–brain barrier (BBB) formed by endothelial cells of the brain capillaries, considerably less

Correspondence: Dr JA Palha, Life and Health Sciences Research Institute (ICVS), School of Health Sciences, University of Minho, Campus Gualtar, Braga 4710-057, Portugal. E-mail: [email protected] Received 31 October 2008; revised 28 January 2009; accepted 29 January 2009; published online 25 February 2009

attention has been paid to the involvement of the blood–cerebrospinal fluid (CSF) barrier (BCSFB), formed by the choroid plexus (CP) epithelial cells. The CP, located within the cerebral ventricles, is composed of a vascularized stroma surrounded by a tight layer of epithelial cells that restrict cellular and molecular traffic between the blood and the CSF (Redzic and Segal, 2004). The CP is best known for its role in the production of the CSF that fills the brain ventricles and the subarachnoid space, and it is ideally located to transmit signals into and out of the brain (Emerich et al, 2005). To date, studies published on the CP response to peripheral inflammatory stimuli have mostly focused on single proteins or groups of proteins. Among these are immune mediators such as interleukin IL-1b and TNF (Nadeau and Rivest, 1999; Quan et al, 1999), enzymes such as prostaglandin D2 synthase (Marques et al, 2007) and bacteriostatic proteins such as lipocalin 2 (LCN2) (Marques et al, 2008). Accessory molecules important for leukocyte

Choroid plexus acute inflammatory response F Marques et al 922

adhesion such as L-selectin, intercellular adhesion molecule 1 (ICAM-1), and vascular cell adhesion molecule 1 have also been described as up-regulated on CP epithelial cells during inflammation (Engelhardt et al, 2001), eventually facilitating access of immune cells into the brain. Notably, it has been recently suggested that the CP may be the primary route of immune cells entry into the CSF in the experimental allergic encephalomyelitis model of MS (Brown and Sawchenko, 2007). Similarly, the CP may facilitate entry of bacteria, as indicated in a mouse model of infection with Streptococcus suis (Dominguez-Punaro et al, 2007). These observations, suggesting that the contribution of the CP to the brain’s innate and adaptative immune responses is still only poorly understood, prompted the present detailed investigation of the CP’s response to acute peripheral inflammation. Specifically, we evaluated the overall kinetic response of the mouse CP to peripheral administration of the Gram-negative bacteria cell wall lipopolysaccharide (LPS) and found that the CP, and particularly its epithelial cells, activates a rapid and transient acute-phase response to LPS that is ultimately reflected in the composition of the CSF. On the basis of complementary studies in CP epithelial cells in culture, we propose a signaling cascade that may underlie this response.

Materials and methods Animals and Lipopolysaccharide Injection All experiments were carried out using 8 to 9 weeks old C57BL/6 male mice and 8 to 9 weeks old Wistar rats (Charles River, Barcelona, Spain), in accordance with the European Community Council Directive 86/09/EEC guidelines for the care and handling of laboratory animals. Animals were maintained under 12 h light/dark cycle at 22.51C and 55% humidity and fed with regular rodent’s chow and tap water ad libitum. To reduce the stressinduced changes in the hypothalamus-pituitary axis, animals were handled for 1 week before the beginning of the experiment. Animals were injected intraperitoneally with LPS (5 mg/g of body weight; Escherichia coli, serotype O26:B6; Sigma, St Louis, USA) or vehicle alone (0.9% NaCl). Mice were anesthetized with ketamine hydrochloride (150 mg/kg) plus medetomidine (0.3 mg/ kg), transcardially perfused with cold saline and killed 1, 3, 6, 12, 24, or 72 h after LPS injection. For mRNA studies, the CP was rapidly removed from all ventricles, stored in RNA later (Ambion, Austin, TX, USA) and kept at 801C. The CP isolation was made under conventional light microscopy (SZX7, Olympus, Hamburg, Germany). At least two pools of CP (from three mice each) were prepared for each time point. This experiment was performed twice: CP samples from one experiment were used for microarray analysis; CPs collected in an independent experiment were used to confirm, by quantitative real-time–polymerase chain reaction (qRT–PCR), the results obtained in the array study. In this second experiment, five pools of CP were prepared for each time point. Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932

The CSF was collected from the cisterna magna and pooled from several mice. An aliquot of each pool was used to verify the absence of blood contamination and the remainder immediately frozen until use. Rats were similarly anesthetized and killed 3 and 6 h after LPS injection. Blood was collected and used to stimulate primary cultures of rat CP epithelial cells.

Microarray Experimental Design and Data Analysis Total RNA was isolated with Trizol (Invitrogen, Carlsbad, CA, USA) following manufacturer’s instructions. After quality assessment using the Agilent Bioanalyzer (Agilent Technologies, CA, USA), 100 ng of total RNA from three pooled controls and two pooled samples from each time point were amplified and labeled with Illumina TotalPrep RNA Amplification Kit (Illumina Inc, San Diego, CA, USA). The labeled cRNA was then hybridized using the recommended protocol in a total of two Illumina Wholegenome Mouseref-8 expression Beadchips (Illumina Inc). This mouse beadchip contains eight arrays, each comprising a total of 24,000 well-annotated RefSeq transcripts. After scanning, raw data from BeadStudio software (Illumina Inc) was read into R/Bioconductor and normalized using quantile normalization. A linear model was applied to the normalized data using Limma package in R/ Bioconductor (Gentleman et al, 2004). A contrast analysis was applied and differentially expressed genes were selected using a Bayesian approach with a false discovery rate of 5%. The differentially expressed genes were categorized using Gene Ontology from Biomart (http://www.biomart. org/) or Ingenuity tools (Redwood City, CA, USA). Enrichment analysis was performed using the DAVID (http:// david.niaid.nih.gov/david/ease.htm) and the Ingenuity softwares.

Gene Expression Measurements by Quantitative Real-Time–Polymerase Chain Reaction As described earlier, 500 ng of total RNA isolated were amplified using a SuperScript RNA Amplification System (Invitrogen) according to the manufacturer’s instructions. After amplification, RNA was reverse transcribed into first strand cDNA using random hexamers of the Superscript First-strand Synthesis System for RT–PCR (Invitrogen). The qRT–PCR analysis was used to measure the expression levels of selected mRNA transcripts. Primers were designed using the Primer3 software (Rozen and Skaletsky, 2000) on the basis of the respective GenBank sequences. The expression level of the reference gene hypoxanthine guanine phosphoribosyl transferase (Hprt) (accession number from GenBank: NM_013556) was used as internal standard for normalization, as we have first confirmed that its expression is not influenced by the experimental conditions (Marques et al, 2007). All other accession numbers and primer sequences are available on request. Reactions using equal amounts of total RNA from each sample were performed on a LightCycler instrument (Roche Diagnostics, Basel, Switzerland)


with the QuantiTect SYBR Green RT–PCR reagent kit (Qiagen, Hamburg, Germany) according to the manufacturer’s instructions. Product fluorescence was detected at the end of the elongation cycle. All melting curves exhibited a single sharp peak at the expected temperature.

Cerebrospinal Fluid Analysis Immune mediators (IL-6, IL-10, IL-12p70, TNF, IFNg, and CCL2) were measured using BD cytometric bead array (BD Biosciences, CA, USA) according to the manufacturer’s instructions in 10 mL of pooled CSF samples obtained from animals injected with LPS at different time points. Three to four pools of CSF from up to five animals each were used for the various time points studied. The detection limit for all proteins was 20 pg/mL.

Primary Cultures of Rat Choroid Plexus Epithelial Cells The CP is composed of a vascularized stroma surrounded by a tight layer of epithelial cells. Therefore, to study to what extent the response observed in vivo was mediated by epithelial cells, we extended our study to the in vitro response of primary cultures of rat CP epithelial cells. We chose to perform the primary cultures with rats for two reasons: (1) the amount of tissue is considerable greater when compared with that of the mouse, (2) as a second animal model for determining the CP response. Epithelial cells from rat CP were prepared as described previously by Strazielle and Ghersi-Egea (1999) with minor modifications. Briefly, neonates (postnatal day 3 or 4) were killed and CP were dissected under conventional light microscopy (SZX7, Olympus). The tissue was rinsed twice in phosphate buffered saline (without Ca2 + and Mg2 + ) followed by a 25-min digestion with 0.1 mg/mL pronase (Sigma) at 371C. Predigested tissue was recovered by sedimentation and briefly shaken in 0.025% of trypsin (Invitrogen) containing 12.5 mg/ml DNAseI (Roche). The supernant was then withdrawn and kept on ice with 10% fetal bovine serum (Invitrogen). This step was repeated five times. Cells were pelleted by centrifugation and ressupended in culture media consisting of Ham’s F-12 and DMEM (1:1) (Invitrogen) supplemented with 10% fetal bovine serum, 2 mmol/L glutamine (Invitrogen), 50 mg/mL gentamycin (Sigma), 5 mg/mL insulin, 5 mg/mL transferrin, 5 ng/mL sodium selenite (ITS, Sigma), 10 ng/mL epidermal growth factor (Sigma), 2 mg/mL hydrocortisone (Sigma), 5 ng/mL basic fibroblast growth factor (Invitrogen). For further enrichment, cells were incubated on plastic dishes for 2 h at 371C. A differential attachment on plastic dish occurred, and supernant containing mostly epithelial cells was collected and placed for seeding on laminin (Boehringer Ingelheim, GmbH, Germany) coated transwells (Corning, Lowell, MA, USA). To assess purity, the cell monolayers were immunostained with an anti-transthyretin antibody (specific for CP epithelial cells) (kindly provided by Maria Joaõ Saraiva). Cell counting under the microscope revealed that at least 95% of the cells stained positive for transthyretin, thus confirming the purity of the cultures.

Stimulations were performed after the formation of confluent cell monolayers, approximately after 7 days in culture. The CP epithelial cells were stimulated with LPS (200 ng/mL) in the basal side (which corresponds to the membrane facing the blood in vivo) for 6, 12, and 24 h. In another set of experiments, CP epithelial cells were similarly stimulated for 6 h with serum collected from rats 3 and 6 h after LPS or saline injection. Total RNA from CP epithelial cells cultures was extracted using a Micro Scale RNA Isolation Kit (Ambion, Austin, TX, USA). A total of 1,000 ng RNA were reverse-transcribed into first strand cDNA using Oligo-dt from the Superscript First-strand Synthesis System for RT–PCR (Invitrogen). The qRT–PCR analysis was used to measure the expression levels of selected mRNA transcripts.

Statistical Analysis The values are reported as mean±s.e. Statistical significance was determined using the nonparametric Mann–Whitney test, with differences considered significant at P < 0.05.

Results Reproducibility of the Gene Array Data

Two pooled CP samples from animals injected with LPS and killed at 1, 3, 6, 12, 24, or 72 h were compared with three pooled CP samples from salineinjected animals. The data were analyzed in the R-Bioconductor. Quality control using inter-array Pearson correlation and clustering based on variance allowed us to ensure reproducibility between the replicates (data not shown). After data normalization, differentially expressed genes were selected based on a false discovery rate of 5%. This analysis yielded a list of 252 up-regulated and 173 downregulated genes at one or more of the experimental time points. Kinetic Profile of the Choroid Plexus Transcriptome After a Peripheral Inflammatory Stimulus

Figure 1 depicts the number of genes whose expression was up-regulated or down-regulated throughout the experimental period; it shows that the CP displays a rapid (acute phase) response to LPS that peaked at 3 to 6 h after LPS administration and that gradually returned to baseline by 72 h. When genes were grouped with respect to the fold changes (Table 1), it became evident that the fold change was of higher magnitude for up-regulated genes (complete list of genes is available as Supplementary data Table S1). Identification of Altered Gene Pathways

Gene ontology and biological pathway analyses of differentially expressed genes, performed using the Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932


Figure 1 Kinetic profile of the CP response to LPS. The number of genes whose expression was found altered at 1, 3, 6, 12, 24, and 72 h after the peripheral injection of LPS as compared with saline-injected mice. The genes whose expression was up-regulated are represented in black, and those that were down-regulated are represented in grey.

Ingenuity software and the DAVID program, showed that the biological pathways mostly altered are associated with the innate immune response (Table 2). Confirmation of Array Results by Quantitative Real-Time–Polymerase Chain Reaction on a Set of Relevant Genes

Within each pathway, and using RNA extracted from CP pools of an independent experiment, a number of genes (Lcn2, Il6, Stat3, Saa3, Cxcl1, Timp1, Tlr2, Icam1, Cd14, Socs3, Stat1, Irf1, and Irf7 as upregulated genes and Cldn5, Cldn11, Lama2, and Gjb6 as down-regulated genes) were chosen for qPCR analysis; this analysis confirmed the array data. Figure 2 exemplifies the kinetic expression profile of some of these genes. Cerebrospinal Fluid Concentration of Cytokines

A significant correlation was found between the CP mRNA and the protein CSF concentration of the chemokine CCL2 (Figure 3A) and the cytokine IL-6 (Figure 3B). The CSF levels of CCL2 (n = 4) and IL-6 (n = 4) increased from 124.7±13.6 to 332.3±51.7 and from below detection to 29.4±7.7 pg/mL, respectively, within 1 h of LPS administration; 12 h later, the levels of both proteins returned to those of basal conditions and remained as such for all time points measured until 72 h. TNF levels also increased by LPS, as reported previously (Marques et al, 2007). All other measured cytokines (IL-10, IL-12p70, and IFNg) were below the detection limit in the CSF from control and from LPS-treated animals; of relevance, no differences were observed in their gene expression profile in the array. Response of Primary Cultures of Rat Choroid Plexus Epithelial Cells to Various Stimuli

To study whether the observed response in vivo was mediated by the CP epithelial cells, which is of Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932

relevance given their role in producing most of the CSF, we extended our study to the in vitro response of primary cultures of rat CP epithelial cells to various stimuli: LPS alone or serum collected from rats treated with LPS. Figure 4 shows that, in response to both stimuli (LPS and serum from LPStreated rats), the genes encoding for IL-1b, IL-6, and CXCL1 were up-regulated within 6 h. However, the observed changes in the expression of Irf1 is unlikely to result from LPS per se, but rather from other mediator(s) present in the serum of LPS-treated rats, as the up-regulation was confined to cells treated with the latter preparation. Figure 4 presents data from one of the three independent experiments.

Characterization of the Choroid Plexus Acute Response

Proteins Involved in Barrier Function and in Cell Adhesion: Most of the genes differentially expressed after the LPS injection were up-regulated but, interestingly, those encoding for proteins involved in the formation of tight junctions were downregulated (Table 2). Among the genes whose expression was down-modulated were those encoding for claudins 3, 5, and 11 (Cldn3, Cldn11, Cldn5) and for proteins of the extracellular matrix that participate in cell-to-cell interactions such as endothelial cellspecific molecule 1 (Esm1), nidogens 1 and 2 (Nid1, Nid2), protocadherin 7 (Pcdh7), laminin alpha 2 (Lama2), and decorin (Dcn). On the contrary, the expression of genes encoding for proteins that may facilitate leukocyte migration, including Icam1, mucosal vascular addressin cell adhesion molecule 1 (Madcam1), selectin platelet (Selp), and selectin endothelial cell (Sele), was up-regulated. Degradation of the extracellular matrix is achieved through the action of proteases and the role of matrix metalloproteinases has been extensively investigated in this context (Flannery, 2006). We observed an up-regulation in the expression of genes encoding for proteases such as collagenase 13 (Mmp13),


Table 1 Genes whose expression was most altered at each time point after LPS Up-regulated Accession 1h NM_008176

Gene

F. change Accession

14.9

NM_029928

5.4

NM_023612

NM_013652 NM_007913

Suppressor of cytokine signaling 3 (Socs3) Chemokine (C-C motif) ligand 4 (Ccl4) Early growth response 1 (Egr1)

5.0 4.4

NM_009987 NM_172411

NM_013654 NM_008361 NM_009841 NM_008416 NM_133662 NM_010234

Chemokine (C-C motif) ligand 7 (Ccl7) Interleukin 1 beta (Il1b) CD14 antigen (Cd14) Jun-B oncogene (Junb) Immediate early response 3 (Ier3) FBJ osteosarcoma oncogene (Fos)

4.1 4.0 3.2 3.1 3.0 2.9

NM_007707

3h NM_008176

Chemokine C-X-C motif) ligand 1 (Cxcl1)

Down-regulated Gene

F. change

Protein tyrosine phosphatase, receptor type, B (Ptprb) Endothelial cell-specific molecule 1 (Esm1)

1.7

1.6 1.5

NM_010171 NM_026524 NM_010496 NM_001029934 NM_152804 NM_020278

Chemokine (C-X3-C) receptor 1 (Cx3cr1) RIKEN cDNA 2310007B03 gene (2310007B03Rik) Coagulation factor III (F3) Mid1 interacting protein 1 (Mid1ip1) Inhibitor of DNA binding 2 (Idb2) Ubiquitin specific peptidase 32 (Usp32) Polo-like kinase 2 (Drosophila) (Plk2) Leucine-rich repeat LGI family, member 1 (Lgi1)

1.7

1.5 1.5 1.5 1.5 1.4 1.4

50.9

NM_013723

Podocalyxin-like (Podxl)

3.9

NM_011315

Chemokine (C-X-C motif) ligand 1 (Cxcl1) Serum amyloid A 3 (Saa3)

28.8

NM_023055

3.0

NM_008491

Lipocalin 2 (Lcn2)

19.8

NM_029928

NM_013652 NM_013654 NM_007707

Chemokine (C-C motif) ligand 4 (Ccl4) Chemokine (C-C motif) ligand 7 (Ccl7) Suppressor of cytokine signaling 3 (Socs3) Intercellular adhesion molecule (Icam1) Interferon, alpha-inducible protein (G1p2) Tissue inhibitor of metalloproteinase 1 (Timp1) Serine proteinase inhibitor, clade A, member 3N (Serpina3n)

18.4 15.1 14.9

NM_023612 NM_010171 NM_007409

Solute carrier family 9, isoform 3 regulator 2 (Slc9a3r2) Protein tyrosine phosphatase, receptor type, B (Ptprb) Endothelial cell-specific molecule 1 (Esm1) Coagulation factor III (F3) Alcohol dehydrogenase 1 (class I) (Adh1)

13.5 11.3

NM_009987 NM_133249

2.2 2.1

11.1

NM_010686

10.6

NM_013805

Chemokine (C-X3-C) receptor 1 (Cx3cr1) Peroxisome proliferative activated receptor gamma coactivator 1 b (Ppargc1b) Lysosomal-associated protein transmembrane 5 (Laptm5) Claudin 5 (Cldn5)

Serum amyloid A 3 (Saa3) Lipocalin 2 (Lcn2)

31.1 26.7

NM_013723 NM_010686

Chemokine (C-X-C motif) ligand 1 (Cxcl1) Serine proteinase inhibitor, clade A, member 3N (Serpina3n) Tissue inhibitor of metalloproteinase 1 (Timp1) Interferon, alpha-inducible protein (G1p2) Intercellular adhesion molecule (Icam1) Glutathione peroxidase 3 (Gpx3) Guanylate binding protein 4 (Gbp4) Chemokine (C-C motif) ligand 5 (Ccl5)

11.6

NM_029928

9.8

NM_023612

7.3

NM_008128

7.3 6.8 5.8 5.5 4.9

Lipocalin 2 (Lcn2) Serum amyloid A 3 (Saa3) Serine proteinase inhibitor, clade A, member 3N (Serpina3n) Chemokine (C-X-C motif) ligand 1 (Cxcl1) Glutathione peroxidase 3 (Gpx3)

NM_010493 NM_015783 NM_011593 NM_009252 6h NM_011315 NM_008491 NM_008176 NM_009252 NM_011593 NM_015783 NM_010493 NM_008161 NM_008620 NM_013653 12 h NM_008491 NM_011315 NM_009252 NM_008176 NM_008161 NM_009778 NM_008620 NM_013653 NM_011593 XM_355243.1 24 h NM_008491 NM_011315 NM_009252

2.8 2.4 2.4 2.3

2.1 2.1

Podocalyxin-like (Podxl) Lysosomal-associated protein transmembrane 5 (Laptm5) Protein tyrosine phosphatase, receptor type, B (Ptprb) Endothelial cell-specific molecule 1 (Esm1)

2.6 2.6

2.2

NM_009320

Gap junction membrane channel protein beta 6 (Gjb6) Solute carrier family 6, member 6 (Slc6a6)

2.1

NM_032398 NM_001001309 NM_009349 NM_020278

Plasmalemma vesicle associated protein (Plvap) Integrin alpha 8 (Itga8) Thioether S-methyltransferase (Temt) Leucine-rich repeat LGI family, member 1 (Lgi1)

2.0 1.9 1.9 1.8

52.7 46.9 18.1

NM_023612 NM_008973 NM_177470

Endothelial cell-specific molecule 1 (Esm1) Pleiotrophin (Ptn) Acetyl-Coenzyme A acyltransferase 2 (Acaa2)

2.6 2.5 2.3

11.8

NM_010686

1.9

10.5

NM_145434

Complement component 3 (C3) Guanylate nucleotide binding protein 4 (Gbp4) Chemokine (C-C motif) ligand 5 (Ccl5)

8.6 7.5

NM_007472 NM_008481

Lysosomal-associated protein transmembrane 5 (Laptm5) Nuclear receptor subfamily 1, group D, member 1 (Nr1d1) Aquaporin 1 (Aqp1) Laminin, alpha 2 (Lama2)

6.5

NM_146257

Tissue inhibitor of metalloproteinase 1 (Timp1) Proteoglycan 4 (Prg4)

6.1

NM_007833

5.9

2.5 2.3

1.9 1.9 1.9

Solute carrier family 29 (nucleoside transporters), member 4 (Slc29a4) Decorin (Dcn)

1.8

NM_025953.1

Potassium channel, subfamily K, member 4 (Kcnk4)

1.8

Histocompatibility 2, class II antigen E beta (H2Eb1) Response to metastatic cancers 1 (Rmcs1) Solute carrier family 29 (nucleoside transporters), member 4 (Slc29a4)

2.0

Lipocalin 2 (Lcn2)

19.2

NM_010382

Serum amyloid A 3 (Saa3) Serine proteinase inhibitor, clade A, member 3N (Serpina3n)

14.8 7.6

NM_207105 NM_146257

1.8

1.8 1.6

Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932


Table 1 Continued Up-regulated Accession

Gene

Down-regulated F. change Accession

NM_009778

Complement component 3 (C3)

6.8

NM_172411

NM_008161

Glutathione peroxidase 3 (Gpx3)

6.8

NM_008128

NM_029803

Interferon, alpha-inducible protein 27 (Ifi27) Chemokine (C-C motif) ligand 7 (Ccl7)

4.7

NM_145434

4.4

NM_172802

Guanylate nucleotide binding protein 4 (Gbp4) Chemokine (C-X-C motif) ligand 1 (Cxcl1) Interferon induced transmembrane protein 3 (Ifitm3)

3.9

Glutathione peroxidase 3 (Gpx3) Nuclear factor, interleukin 3, regulated (Nfil3) Interferon, alpha-inducible protein 27 (Ifi27) X-box binding protein 1 (Xbp1) Scleraxis (Scx) Complement component 3 (C3)

NM_013654 NM_008620 NM_008176 NM_025378 72 h NM_008161 NM_017373 NM_029803 NM_013842.2 NM_198885.2 NM_009778.1 NM_009252.1 M69067 NM_023612.3 NM_023612.3

Gene

F. change

1.4

NM_008973

RIKEN cDNA 2310007B03 gene (2310007B03Rik) Gap junction membrane channel protein beta 6 (Gjb6) Nuclear receptor subfamily 1, group D, member 1 (Nr1d1) Fascin homolog 2, actin-bundling protein, retinal (Fscn2) Pleiotrophin (Ptn)

3.8

NM_201639

Desmuslin (Dmn), transcript variant 1

1.4

3.7

NM_080575

Acetyl-Coenzyme A synthetase 2 (AMP forming)-like (Acas2l)

1.4

3.3 1.9

NM_011066 NM_010902

1.6 1.4

1.8

NM_009427

Period homolog 2 (Drosophila) (Per2) Nuclear factor, erythroid derived 2, like 2 (Nfe2l2) Transducer of ErbB-2,1 (Tob1)

1.6 1.6 1.6

NM_177470 NM_207105 NM_023671

Serine proteinase inhibitor, clade A, member 3N (Serpina3n) Histocompatibility 2, D region (H2l)

1.6

NM_009895

1.4

NM_009948

Osteoclast inhibitory lectin (Ocil) Endothelial cell-specific molecule 1 (Esm1)

1.3 1.3

NM_013807 NM_011404

metallopeptidase domain 7 (Adam7), and ADAM metallopeptidase with thrombospondin type 1 and 4 motifs (Adamts1, Adamts4). Notably, the expression of the gene encoding for the tissue inhibitor metalloproteinase 1 (Timp1) was up-regulated as early as 3 h after LPS injection, probably indicating a response against degradation of the extracellular matrix. Taken together, these observations suggest that the barrier function of the BCSFB may be transiently compromised. Acute-Phase Proteins: One of the most investigated, but still poorly understood, characteristics of the acute-phase response is the up-regulation and downregulation of the positive and negative acute-phase proteins, respectively. Adaptive changes to the stimulus in the synthetic and secretory liver machineries lead to alterations in the levels of these proteins. Several well-described acute-phase liver proteins (Ceciliani et al, 2002; Gabay and Kushner, 1999) were also found up-regulated in the CP soon after the LPS injection. For instances, peak expression of serum amyloid A1 (Saa1), serum amyloid A3 (Saa3), lipocalin 2 (Lcn2), and Il6 were observed in the CP at 3 and 12 h after the LPS injection. The production of some of the peripheral acute-phase proteins by the CP, and despite of specific characteristics of the CP response could represent a brain defence mechanism that is similar to that mediated by the liver in the periphery. Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932

Acetyl-coenzyme A acyltransferase 2 (Acaa2) Response to metastatic cancers 1 (Rmcs1) Chloride channel, nucleotide-sensitive, 1A (Clns1a) Cytokine inducible SH2-containing protein (Cish) Carnitine palmitoyltransferase 1b, muscle (Cpt1b) Polo-like kinase 3 (Drosophila) (Plk3) Solute carrier family 7 member 5 (Slc7a5)

1.5 1.4

1.4 1.4

1.4 1.4 1.3 1.3 1.3 1.3 1.3 1.3

Signaling Transduction Pathways: The expression of genes belonging to various well-described signaling transduction pathways was up-regulated by LPS. These include several intermediate modulators and transcription factors such as nuclear factor kappa B (NfkB1), the interferon regulatory factors 1, 2, 7, and 9 (Irf1, Irf2, Irf7, Irf9), activated protein-1 (Junb and Fos), Stat1 and Stat3 (Figure 2; Supplementary data Table S1). Many of the proteins encoded by these genes have, as downstream targets, genes that encode for cytokines such as Il1b, Il6, and chemokines such as chemokines (C-C motif) ligands 3 and 4 (Ccl3, Ccl4 ), all of which were seen to be up-regulated. Interestingly, the CP is not only able to activate these pathways but also to synthesize molecules responsible for the induction of inhibitory mechanisms, as illustrated by the results for proteins belonging to the SOCS family. In our array, the expression of genes encoding for the suppressors of cytokine signaling 2 and 3 (Socs2, Socs3) was upregulated.

Discussion This study reveals that peripheral inflammatory stimuli can elicit an acute-phase response in the CP. As part of the BBBs, and because it determines the composition of the CSF, changes in the CP transcriptome can potentially play an important role


Table 2 Clustering of the genes whose expression was altered in the choroid plexus upon peripheral LPS injection Immune molecules

Antigen presentation pathway Signaling pathways

Acute phase response signaling Glucocorticoid receptor signaling Cell adhesion molecules

Proteins that contribute to integrity of ECM Transporters

Chemokines: Ccl4, Cxcl2, Cxcl13, Ccl4, Cxcl1, Ccl5, Ccl3, Cxcl16, Ccl7, Cxcl9, Ccl9, Ccl11, Ccl2, Ccl19, Cxcl10 m and Cxcl12 k Interleukins: Il1b, Il15, Il6 m Other molecules with cytokine activity: Csf1, Csf3, Spp1 m Antigen presentation pathway: H2T23, H2K1, Psmb8, Psmb9, Tap2 m and H2Eb1 k TLR and co-estimulatory molecules: Cd14, Tlr2 m JAK/STAT signaling pathway: Socs3, Cish, Socs2, Stat1, Stat3 m and Pias3 k MAPK signaling pathway: Map3k1, Map3k6, Map3k3, Map3k8, Fos, Junb m and Mapk4 k NF-KB signaling pathway: Bcl3, Tnfrsf5, Egfr, Prkr, Il1b, Map3k3, Map3k8, Nfkb1, Nfkbia, Nfkbie, Ngfb, Relb, Ripk1, Tlr2, Tnfaip3 m and Hdac2 k Complement signaling: C2, C3, C6, Slp, H2Bf, Serping1 m Interferon signaling: Ifit3, Ifitm1, Ifngr2, Irf1, If2, If7, Isgf3 g, Mx1, Oas1 g, Psmb8, Stat1 m IL-10 signaling: Bcl3, Cd14, Fos, Il6, Il1b, Junb, Nfkbia, Nfkbie, Socs3, Stat3 m IL-6 signaling: A2 m, Bcl3, Cd14, Cebpb, Fos, Il6, Il1b, Junb, Nfkbia, Nfkbie, Stat3 m Acute phase response: A2 m, Bcl3, C2, C3, Cebpb, H2Bf, Fos, Il6, Il1b, Junb, Map3k1, Nfkbia, Nfkbie, Ripk1, Saa1, Saa3, Serpina3n, Serping1, Socs2, Socs3, Stat3 m Glucocorticoid receptor signaling: A2 m, Bcl3, Ccl3, Ccl5, Ccl11, Cxcl13, Cdkn1a, Cebpb, Cxcl2, Dusp1, Fkbp5, Fos, Icam1, Il6, Il1b, Junb, Map3k1, Nfkb1, Nfkbia, Nfkbie, Sele, Stat1, Stat3 m and Tgfbr2 k Tight junctions: Cldn3, Cldn11, Cldn5 k Gap junctions: Gjb6 k Leukocyte transendothelial migration: Icam1, Madcam1, Selp, Sele m and Pecam1 k Extracellular matrix: Pcdh7, Esm1, Ptn, Nid1, Nid2, Lama2, Dcn, Tgfbi, Emid2k Proteases/proteases inhibitors: Adamts1, Adamts4, Serping1, Serpina3n, Mmp13, Psmb8, Psmb9, Adam7, Usp18, Psme2b, Casp4, A2 m, Timp1m Solute carrier: Slc7a5, Slc43a3, Slc15a2, Slc31a2, Slc38a2, Slc7a11 m and Slc6a6, Slc23a2, Slc29a4, Slc7a10, Slc9a3r2, Slco2b1, Slc5a6, Slc9a3r2, Slco1a4, Slc22a5 k ABC transporter: Tap2 m

15m1k 3m 3m 5m1k 2m 5m1k 6m1k 15m1k 6m 11m 10m 11m 21m 23m1k

3k 1k 4m1k 9k 13m 6m10k 1m

m denotes up-regulated; k denotes down-regulated.

in mediating the brain’s response to inflammation. Notably, the kinetics of the CP response proved to be similar to that of the liver (Ceciliani et al, 2002; Gabay and Kushner, 1999) in terms of response onset and shut-off; and in some of the acute-phase proteins secreted (e.g., SAA, LCN2, IL-6, IL-1b) into the CSF. In addition, the liver and the CP acute responses share common signaling pathways, for example, NFkB, JAK/STAT, and MAPK. Despite the above commonalities, the CP epithelium expresses a number of distinct genes after being challenged with an acute inflammatory stimulus; these genes are implicated in its barrier function although, at least within the time frame of the present experiments, we did not observe any gross CP morphologic changes. Changes in the functionality and integrity of the BBB are relevant to the pathogenesis of a variety of diseases of the CNS (e.g., HIV-1 encephalitis, Alzheimer disease, MS, Parkinson disease, ischemia, and tumors) (Persidsky et al, 2006). For example, proinflammatory substances produced by microglia and specific disease-associated proteins or cells (b-amy-

loid in Alzheimer disease and reactive T cells in MS) often mediate BBB dysfunction. Although it is not known exactly how the vascular endothelial cells that form the BBB respond to an inflammatory stimulus, it is plausible that the CP response shares many of the mechanisms already described for the BBB. A number of studies support the idea that changes in endothelial cell tight junctions and cytoskeletal organization facilitate leukocyte migration, through the interaction of chemokines and adhesion molecules, such as E-selectin and vascular cell adhesion molecule 1 with leukocytes (Engelhardt, 2008). The herein presented data for the CP show decreased expression of tight junction proteins and increased expression of proteins that facilitate migration. These changes are likely to underlie the preferential migration of T lymphocyte across the BCSFB that has been observed in animal models of MS (Brown and Sawchenko, 2007) and suggest a role for the CP in cell entry into the CSF and ultimately, into the brain parenchyma. Together with studies implying particular CP proteins in Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932


Figure 2 Analysis, by qRT–PCR, of the gene expression kinetic profile of selected genes. Confirming the array results, the expression of Saa3, Cxcl1, Irf1, Irf7, Stat1, and Socs3 (A–F) was up-regulated and that of Cldn5, Cldn11, and Gjb6 (G–I) was down-regulated. As some genes were not found expressed in saline-injected animals, the mRNA expression profile is presented as the expression rate between the gene of interest and the reference gene.

Figure 3 The CSF levels of CCL2 and IL-6 after peripheral injection of LPS. The concentration of CSF CCL2 (A) and of IL-6 (B) increased 1 h after LPS administration and rapidly returned to basal levels. b.d., below detection (20 pg/mL).

diseases such as Alzheimer disease (Sousa et al, 2007), this study warrants further studies to examine the CP transcriptome in neurologic and psychiatric diseases. Apart from being a potential site of cell entry into the brain, the CP is also an active site of protein synthesis. Several inflammatory markers have been described in the CSF of individuals with brain pathology (Andreasen and Blennow, 2005; Giovannoni, 2006). The precise origin of these molecules is still equivocal, but CP epithelial cells might be important for their trafficking/entry into the CSF. The nature of the proteins secreted by the CP may change in response to disease or specific stimuli, as shown here for CCL2 and IL-6, and in previous Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932

studies for various cytokines, carrier proteins, and iron-related proteins (Marques et al, 2007, 2008; Hughes et al, 2002; Thibeault et al, 2001). In fact, several inflammatory chemokines (CCL2, CCL5, CXCL10, CXCL12, and CXCL13) whose gene expression is altered in the CP after peripheral LPS injection are detectable in the CSF of MS subjects (Trebst and Ransohoff, 2001) and CCL2 mRNA levels in the CP are increased after a single systemic bolus of LPS (Thibeault et al, 2001). The CP is known to secrete proinflammatory and antiinflammatory molecules, proteins that promote and inhibit extracellular matrix remodeling, and proteins to which neuroprotective and/or toxic properties have been ascribed to. Their beneficial or detrimental effects


Figure 4 Analysis, by qRT–PCR, of the gene expression profile from rat CP epithelial cells. Primary cultures of rat CP epithelial cells were exposed to LPS alone for 6, 12, and 24 h; or for 6 h to serum obtained from rats 3 or 6 h after injection with LPS. Although both LPS and serum from LPS-injected rats induced the expression of Il1b, Il6, and Cxcl1, only the serum from LPS-injected mice had an effect on the expression of Irf1. The mRNA expression profiles are presented as the fold change increase in relation to the respective control (exposure to cell media for the stimulation with LPS, and exposure to serum collected from saline-injected rats for the stimulation with serum obtained from LPS-treated rats).

may depend on other factors such as age and the presence of pathogenic and/or disease conditions, which should next be investigated.

Proposed Model of the Choroid Plexus Response to Peripheral Inflammation

Irrespective of the final determinants of the balance between benefit and detriment, the present data provide interesting insight into the signaling transduction pathways that ultimately lead to the observed gene expression profiles in response to a

pathogenic stimulus (Figure 5). Previous work showed that serum concentrations of cytokines (e.g., IL-1b and IL-6) are increased after peripheral administration of LPS (Ramadori and Christ, 1999). Receptors for these cytokines are present in CP epithelial cells (Chodobski and Szmydynger-Chodobska, 2001) and were detected by our array analysis under basal conditions. Thus, LPS- and cytokine-mediated signaling transduction pathways may be simultaneously engaged by peripheral inflammation. In fact, the cytokine-mediated pathways may play a general role in the CP response as the secretome of LPS-treated cultured CP epithelial cells Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932


Figure 5 Suggested pathways of the CP epithelial cells response to peripheral inflammation. The genes whose expression was, at one or more time points, down-regulated are represented in green, and whose expression was up-regulated at one or more time points are represented in red.

displays a much reduced number of altered proteins (Thouvenot et al, 2006). This interpretation is consistent with that of the present study in which the up-regulation of Irf1 was not mediated by LPS alone, but by other molecules (e.g., IL-1b, IL-6, TNF) that are present in the serum on the peripheral inflammatory stimulus. With respect to LPS signaling, we confirmed that the gene encoding for the Toll-like receptor 4 (Tlr4), the LPS cognate receptor, is expressed in the CP (Laflamme and Rivest, 2001; Chakravarty and Herkenham, 2005). However, the expression of the gene encoding for TLR4 is not influenced by LPS whereas that encoding for TLR2 is induced, as has been reported previously (Laflamme et al, 2001). Of notice, the differential expression of TLRs in response to inflammation has been described in other organs (Ojaniemi et al, 2006). Interestingly, TLR2 is not liganded by LPS. However, a mediatory role for proinflammatory cytokines (e.g., IL-1b and TNF) in the up-regulation of Tlr2 has been proposed (Matsumura et al, 2003), as might as well be the case for the CP. Activation of TLR2-mediated signaling may also result from the interaction of increased serum levels of SAA, recently shown as a functional ligand for TLR2 (Cheng et al, 2008). Journal of Cerebral Blood Flow & Metabolism (2009) 29, 921–932

Activation of TLRs and of receptors for IL-1b, IL-6, and TNF results in the induction of several transcription factors, including interferon regulatory factor 3 (IRF3), activator protein-1, and NF-kB (Colonna, 2007; Honda and Taniguchi, 2006). Our analysis shows that several genes encoding for proteins belonging to the NF-kB, MAPK, STAT-JAK, and IRFs signaling pathways are induced in the CP during the acute-phase response. All of these pathways, together with others such as those mediated by cAMP (Reyes-Irisarri et al, 2008), participate in the development and regulation of innate and adaptative immune responses (Rawlings et al, 2004), and the CP, like the BBB endothelium (Laflamme et al, 1999; Laflamme and Rivest, 1999), responds to peripheral LPS by activating some of these pathways. Termination of the CP response at 72 h may be, at least in part, a consequence of the negative feedback inhibition of STAT signaling by SOCS/CIS (Naka et al, 2005), as some of the genes encoding for proteins in this pathway are up-regulated after LPS administration. A similar mechanism has been described in the BBB (Lebel et al, 2000). In summary, this study shows that the CP can mount an acute-phase response. This response includes the release of proinflammatory and antiin-

Choroid plexus acute inflammatory response F Marques et al

flammatory immune modulators into the CSF, and a relaxation of the CP barrier properties. The evidence provided for a potentially important role of the CP in the cross talk between the immune system and the CNS warrants further studies that will add to our understanding on how homeostasis is maintained in the brain under normal and pathologic states.

Acknowledgements This work was supported by grant POCTI/SAU-NEU/ 56618/2004 from the Portuguese Foundation for Science and Technology (FCT) /FEDER; and from a grant from the DANA foundation; Marques F, Falcao AM, and Rodrigues AJ are recipients of fellowships from FCT/FEDER.

References Andreasen N, Blennow K (2005) CSF biomarkers for mild cognitive impairment and early Alzheimer’s disease. Clin Neurol Neurosurg 107:165–73 Brown DA, Sawchenko PE (2007) Time course and distribution of inflammatory and neurodegenerative events suggest structural bases for the pathogenesis of experimental autoimmune encephalomyelitis. J Comp Neurol 502:236–60 Ceciliani F, Giordano A, Spagnolo V (2002) The systemic reaction during inflammation: the acute-phase proteins. Protein Pept Lett 9:211–23 Chakravarty S, Herkenham M (2005) Toll-like receptor 4 on nonhematopoietic cells sustains CNS inflammation during endotoxemia, independent of systemic cytokines. J Neurosci 25:1788–96 Cheng N, He R, Tian J, Ye PP, Ye RD (2008) Cutting edge: TLR2 is a functional receptor for acute-phase serum amyloid A. J Immunol 181:22–6 Chodobski A, Szmydynger-Chodobska J (2001) Choroid plexus: target for polypeptides and site of their synthesis. Microsc Res Tech 52:65–82 Colonna M (2007) TLR pathways and IFN-regulatory factors: to each its own. Eur J Immunol 37:306–9 Dominguez-Punaro MC, Segura M, Plante MM, Lacouture S, Rivest S, Gottschalk M (2007) Streptococcus suis serotype 2, an important swine and human pathogen, induces strong systemic and cerebral inflammatory responses in a mouse model of infection. J Immunol 179:1842–54 Emerich DF, Skinner SJ, Borlongan CV, Vasconcellos AV, Thanos CG (2005) The choroid plexus in the rise, fall and repair of the brain. Bioessays 27:262–74 Engelhardt B, Wolburg-Buchholz K, Wolburg H (2001) Involvement of the choroid plexus in central nervous system inflammation. Microsc Res Tech 52:112–29 Engelhardt B (2008) Immune cell entry into the central nervous system: involvement of adhesion molecules and chemokines. J Neurol Sci 274:23–6 Flannery CR (2006) MMPs and ADAMTSs: functional studies. Front Biosci 11:544–69 Gabay C, Kushner I (1999) Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 340:448–54

Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, Ellis B, Gautier L, Ge Y, Gentry J, Hornik K, Hothorn T, Huber W, Iacus S, Irizarry R, Leisch F, Li C, Maechler M, Rossini AJ, Sawitzki G, Smith C, Smyth G, Tierney L, Yang JY, Zhang J (2004) Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 5:R80 Giovannoni G (2006) Multiple sclerosis cerebrospinal fluid biomarkers. Dis Markers 22:187–96 Honda K, Taniguchi T (2006) IRFs: master regulators of signalling by Toll-like receptors and cytosolic patternrecognition receptors. Nat Rev Immunol 6:644–58 Hughes PM, Botham MS, Frentzel S, Mir A, Perry VH (2002) Expression of fractalkine (CX3CL1) and its receptor, CX3CR1, during acute and chronic inflammation in the rodent CNS. Glia 37:314–27 Laflamme N, Lacroix S, Rivest S (1999) An essential role of interleukin-1beta in mediating NF-kappaB activity and COX-2 transcription in cells of the blood-brain barrier in response to a systemic and localized inflammation but not during endotoxemia. J Neurosci 19:10923–30 Laflamme N, Rivest S (1999) Effects of systemic immunogenic insults and circulating proinflammatory cytokines on the transcription of the inhibitory factor kappaB alpha within specific cellular populations of the rat brain. J Neurochem 73:309–21 Laflamme N, Rivest S (2001) Toll-like receptor 4: the missing link of the cerebral innate immune response triggered by circulating gram-negative bacterial cell wall components. FASEB J 15:155–63 Laflamme N, Soucy G, Rivest S (2001) Circulating cell wall components derived from gram-negative, not gram-positive, bacteria cause a profound induction of the gene-encoding Toll-like receptor 2 in the CNS. J Neurochem 79:648–57 Lebel E, Vallieres L, Rivest S (2000) Selective involvement of interleukin-6 in the transcriptional activation of the suppressor of cytokine signaling-3 in the brain during systemic immune challenges. Endocrinology 141:3749–63 Marques F, Sousa JC, Correia-Neves M, Oliveira P, Sousa N, Palha JA (2007) The choroid plexus response to peripheral inflammatory stimulus. Neuroscience 144:424–30 Marques F, Rodrigues AJ, Sousa JC, Coppola G, Geschwind DH, Sousa N, Correia-Neves M, Palha JA (2008) Lipocalin 2 is a choroid plexus acute-phase protein. J Cereb Blood Flow Metab 28:450–5 Matsumura T, Degawa T, Takii T, Hayashi H, Okamoto T, Inoue J, Onozaki K (2003) TRAF6-NF-kappaB pathway is essential for interleukin-1-induced TLR2 expression and its functional response to TLR2 ligand in murine hepatocytes. Immunology 109:127–36 Nadeau S, Rivest S (1999) Regulation of the gene encoding tumor necrosis factor alpha (TNF-alpha) in the rat brain and pituitary in response in different models of systemic immune challenge. J Neuropathol Exp Neurol 58:61–77 Naka T, Fujimoto M, Tsutsui H, Yoshimura A (2005) Negative regulation of cytokine and TLR signalings by SOCS and others. Adv Immunol 87:61–122 Ojaniemi M, Liljeroos M, Harju K, Sormunen R, Vuolteenaho R, Hallman M (2006) TLR-2 is upregulated and mobilized to the hepatocyte plasma membrane in the space of Disse and to the Kupffer cells TLR-4 dependently during acute endotoxemia in mice. Immunol Lett 102:158–68 Quan N, Stern EL, Whiteside MB, Herkenham M (1999) Induction of pro-inflammatory cytokine mRNAs in the

931



brain after peripheral injection of subseptic doses of lipopolysaccharide in the rat. J Neuroimmunol 93:72–80 Persidsky Y, Ramirez SH, Haorah J, Kanmogne GD (2006) Blood-brain barrier: structural components and function under physiologic and pathologic conditions. J Neuroimmune Pharmacol 1:223–36 Ramadori G, Christ B (1999) Cytokines and the hepatic acute-phase response. Semin Liver Dis 19:141–55 Rawlings JS, Rosler KM, Harrison DA (2004) The JAK/ STAT signaling pathway. J Cell Sci 117:1281–3 Redzic ZB, Segal MB (2004) The structure of the choroid plexus and the physiology of the choroid plexus epithelium. Adv Drug Deliv Rev 56:1695–716 Reyes-Irisarri E, Perez-Torres S, Miro X, Martinez E, Puigdomenech P, Palacios JM, Mengod G (2008) Differential distribution of PDE4B splice variant mRNAs in rat brain and the effects of systemic administration of LPS in their expression. Synapse 62:74–9 Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132:365–86

Sousa JC, Cardoso I, Marques F, Saraiva MJ, Palha JA (2007) Transthyretin and Alzheimer’s disease: where in the brain? Neurobiol Aging 28:713–8 Strazielle N, Ghersi-Egea JF (1999) Demonstration of a coupled metabolism-efflux process at the choroid plexus as a mechanism of brain protection toward xenobiotics. J Neurosci 19:6275–89 Thibeault I, Laflamme N, Rivest S (2001) Regulation of the gene encoding the monocyte chemoattractant protein 1 (MCP-1) in the mouse and rat brain in response to circulating LPS and proinflammatory cytokines. J Comp Neurol 434:461–77 Thouvenot E, Lafon-Cazal M, Demettre E, Jouin P, Bockaert J, Marin P (2006) The proteomic analysis of mouse choroid plexus secretome reveals a high protein secretion capacity of choroidal epithelial cells. Proteomics 6:5941–52 Trebst C, Ransohoff RM (2001) Investigating chemokines and chemokine receptors in patients with multiple sclerosis: opportunities and challenges. Arch Neurol 58:1975–80

Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http:// www.nature.com/jcbfm)
