Up-regulation of the Vascular Cell Adhesion Molecule-1 (VCAM-1) Induced By Theiler's Murine Encephalomyelitis Virus Infection of Murine Brain Astrocytes

Abstract

The present article reports the up-regulation of the expression of the vascular cell adhesion molecule-1 (VCAM-1) by SJL/J mouse brain astrocytes infected with Theiler's murine encephalomyelitis virus (TMEV). Complementary RNA (cRNA) from mock- and TMEV-infected cells was hybridized to the Affymetrix whole murine genome U74v2 DNA microarray. Hybridization data analysis revealed background expression in untreated cells and the up-regulation of three sequences coding for VCAM-1, as described by the SCOP (Structural Classification Of Proteins) database. The authors further studied its regulation, confirming and validating their mRNA increase by reverse transcriptase–polymerase chain reaction (RT-PCR) and quantitative real-time RT-PCR. The presence of the 100-kDa VCAM-1 protein in mock- and TMEV-infected cells was demonstrated in the cell membrane by a specific cell-based enzyme-linked immunosorbent assay (ELISA), in addition to flow cytometry and confocal immunohistochemistry. Further, Western blots were used to quantify the amount of VCAM-1 molecules in cell extracts. All these data demonstrated a mean 75% increase in the expression of VCAM-1 on the surface of TMEV-infected cells. Three inflammatory cytokines, interleukin-1α (IL-1α), interferon gamma (IFNγ), and specially tumor necrosis factor alpha (TNF-α), some of which are also induced by TMEV in astrocytes (IL-1α and TNF-α), were potent inducers of VCAM-1 expression. To demonstrate whether the VCAM-1 molecules were biologically active, mediating adhesion to other cells as the integrin α4-expressing CD4+ T lymphocytes, the authors used a cell adhesion test. It was also demonstrated by immunohistochemistry that in vivo VCAM-1 expression is enhanced after TMEV intracraneal infection. The present data show a small but statistically significant overexpression of VCAM-1 after astrocyte infection with TMEV that could play a significant role in vivo.

Keywords::

• astrocytes
• cell adhesion
• VCAM-1
• viral infection

INTRODUCTION

Several families of related proteins contain the molecules involved in cell adhesion processes. These include cell adhesion molecules (CAMs) and integrins, which recognize a large number of surface-bound and soluble ligand proteins (Leahy 1997). The recognition of some proteins on the cell surface results in effects on the cell contractile apparatus, resulting in assembly and adhesion (Arnaout et al. 2005). Very recently, an adhesion molecule, named GlialCAM, has been described. This protein is predominantly expressed in central nervous system (CNS) glial cells, mainly by oligodendrocytes and astrocytes (Fabre-Kontula et al. 2008).

The vascular cell adhesion molecule-1 (VCAM-1 or CD106) consists of seven extracellular immunoglobulin G (Ig) domains, a single membrane-spanning region, and a cytoplasmic domain (Osborn 1990). VCAM-1 is expressed in endothelium cell activated by cytokines and in macrophage cell types during some immune responses (Rice et al. 1989; Rice et al. 1990; Springer 1990). This protein interacts with integrin VLA-4 family members, mainly with integrin α4 (CD 49d) or the heterodimer α4β7 present on the surface of leukocytes, basophils, eosinophils, and other immune cells (Elices et al. 1990; Walsh et al. 1991; Weller et al. 1991). Therefore, the interaction between VCAM-1 and integrin α4 plays an important role in innate and acquired immunological responses and in the migration of immune cells to sites of inflammation.

Theiler's murine encephalomyelitis virus (TMEV) is a picornavirus that persistently infects murine CNS (Theiler 1937). After intracerebral inoculation of the BeAn low-neurovirulence strain, genetically susceptible strains of mice develop a chronic demyelinating disease currently used as an experimental model for multiple sclerosis (MS) (Clatch et al. 1986; Gerety et al. 1994). Myelin breakdown is mediated by the immune system rather than being a direct consequence of viral infection of oligodendrocytes, the myelin-forming cells (Clatch et al. 1986; Lipton and Dal Canto 1976; Peña Rossi et al. 1991; Roos et al. 1982). Focal inflammation occurs in the white matter of the brain and spinal cord and mononuclear immune cell infiltrates were observed in demyelinating areas (Clatch et al. 1986). Chemokines that participate in the recruitment and influx of immune cells are involved in this inflammatory process, being synthesized in large amounts by astrocytes, both in vitro and in vivo, after infection with TMEV (Palma and Kim 2001, 2004; Rubio et al. 2006; Rubio and Sanz-Rodriguez 2007). Furthermore, VCAM-1 and chemokines could also be involved in the migration of CD4+ T lymphocytes (Clatch et al. 1986), macrophages (Jelachich et al. 1995; Levy et al. 1992; Schlitt et al. 2003), neutrophils (Rubio et al. 2006), and monocytes (Rubio and Sanz-Rodriguez 2007) into the inflammatory areas in the brain of SJL/J mice susceptible to TMEV-induced demyelination.

In this article, we have studied the expression of genes coding for VCAM-1 by DNA hybridization using arrays from Affymetrix and the expression of such molecules in the cell membranes of uninfected or TMEV-infected astrocytes. We also discussed their possible role in demyelination mediated by immune processes triggered by TMEV infection. We propose that the up-regulation of VCAM-1 participates in the influx of integrin α4-expressing immune cells to sites of inflammation within the CNS.

MATERIALS AND METHODS

Astrocyte Cultures

Astrocyte cultures were prepared by mechanical dissociation of the cerebral cortex from newborn SJL/J mice purchased from Harlan Laboratories (Indianapolis, IN) and maintained at the Instituto Cajal. The cortex was isolated under a dissecting microscope and cleaned of choroid plexus and meninges. Cell suspensions were filtered through 135-mm pore size nylon cell strainer (Falcon-Becton Dickinson, Le Pont De Claix, France) into Dulbecco`s modified Eagle medium (DMEM) containing 10% fetal calf serum (FCS) and penicillin-streptomycin. (Gibco BRL, Paisley, Scotland). After centrifugation, cells were filtered again through a 40-mm cell strainer and cultured in 75-cm² tissue culture flasks (Costar, Cambridge, MA) at 37°C. The medium was changed after 4 days in culture and subsequently 2 times a week for the entire culture period. Cultures were enriched for astrocytes by the removal of less adherent microglia and oligodendrocytes by shaking overnight at 37°C at 250 rpm in a table top shaker (Thermo Forma, Marietta, OH). Cellular confluence was observed 10 days after plating, producing around 1 × 10⁷ cells per flask and showing a polygonal flat morphology. The cultures were endotoxin-free (Rubio and Sierra 1993). A mean of 98% astrocytes was confirmed by indirect immunofluorescence staining of methanol-fixed cultures using rabbit anti-glial fibrillar acidic protein (GFAP) antiserum (Dakopatts, Glostrup, Denmark). The lack of noticeable mature oligodendrocytes or microglial/macrophage cells was determined using a guinea pig anti-myelin basic protein (MBP) antiserum and monoclonal anti-Mac-1 antibody (Serotec, Oxford, UK). Secondary fluorescein-labeled antibodies were purchased from Sigma Chemical (St. Louis, MO) (Rubio et al. 2003).

Viruses and Infection

For these studies, a strain of TMEV isolated in 1957 from a feral mouse in Belem, Brazil, called BeAn 8386, was used. Baby hamster kidney cells (BHK-21) were grown at 37°C in DMEM containing 10% FCS and penicillin-streptomycin. BHK-21 cell cultures were infected for 48 h at 33°C, sonicated, and centrifuged at 4°C to remove cells debris. Virus titers were determined by a standard plaque assay on BHK-21 cells with 1% Noble agar (Difco Laboratories, Detroit, MI) and staining with 0.2% crystal violet in 20% methanol. Purified astrocytes in 75-cm² tissue culture flasks were infected with BeAn virus at several multiplicities of infection (m.o.i) in a volume of 10 ml of DMEM containing 0.1% BSA, at room temperature for 1 h. After infection, cells were washed, 10 ml of DMEM plus 10% FCS was added and flasks incubated at 37°C for different periods of time. Cells used for mock infections were incubated with a virus-free BHK-21 cell lysate.

cRNA Target Preparation, Hybridization, and Data Analysis

Three independent cultures of SJL/J astrocytes infected at a m.o.i. of 10 were harvested 24 h postinfection, washed with phosphate-buffered saline, and total RNA isolated by using the RNeasy Mini Kit (Qiagen, Valencia, CA). Ten micrograms of RNA were converted to cDNA by using the kit SuperScript Choice System (Gibco BRL). Second strand synthesis was performed using T4 DNA polymerase, and cDNA was isolated by phenol-chloroform extraction. Isolated cDNA was transcribed using the BioArray High Yield RNA transcript Labeling Kit (Enzo Biochem, New York, NY) with biotin-labeled UTP and CTP to produce biotin-labeled cRNA. Labeled cRNA was isolated using the RNeasy Mini Kit and fragmented in 100 mM potassium acetate, 30 mM magnesium acetate, 40 mM Tris-acetate (pH 8.1) for 30 min at 94°C. Hybridization performance was analyzed using Test 2 arrays (Affymetrix, Santa Clara, CA) and spike and housekeeping controls. Target cRNA was hybridized to the murine genome U74v2 microarray (Affymetrix) according to manufacturer's protocols. Briefly, 15 mg of fragmented cRNA was hybridized for 16 h at 45°C with constant rotation (60 rpm). The microarrays were washed and stained with streptavidin-conjugated phycoerythrin (SAPE) using the Affymetrix GeneChip Fluidic Station 400. All hybridization steps were performed by Progenika Biopharma S.A. Derio, Spain.

Each gene on the U74v2 array is represented by 20 different 25-base cDNA oligonucleotides complementary to a cRNA target transcript (perfect match). As a hybridization specificity control, an oligonucleotide containing a single-base substitution corresponding to each perfect-match cDNA oligonucleotide (mismatch) is represented on the array. By using Affymetrix- defined absolute mathematical algorithms describing perfect-match and mismatch intensities, each gene was defined as absent or present and assigned a binding intensity value. The virus-induced changes were quantified by the signal log ratio corresponding to the log₂ of the fold change for each gene. Standard errors in mean transcriptional expression of a gene were also calculated.

RT-PCR Analysis of VCAM-1 Expression

Total RNA from normal or TMEV-infected astrocyte cultures was purified using the RNeasy Mini purification kit (Qiagen). Samples were reverse transcribed using Moloney murine leukemia virus reverse transcriptase (RT) (Promega, Madison, WI). Primer pairs used for mouse VCAM-1 amplification were purchased from R&D Systems, Minneapolis, MN (catalog number: RDP-194-025). A primer set for mouse β-actin, used as house keeping positive control, were synthesized by Sigma-Genosys (St. Louis, MO). The cycling conditions of the reverse transcriptase–polymerase chain reaction (RT-PCR) reactions consisted of 5 min at 94°C and 30 cycles of 40 s at 62°C and 1 min at 72°C, followed by a final 10 min extension at 72°C, using a Perkin Elmer Cetus 480 DNA thermal cycler. The resulting PCR products were purified using S-400 MicroSpin columns (Pharmacia Biotech, Uppsala, Sweden) and electrophoresed in 2% NuSieve agarose gels (FMC Bio Products, Rockland, ME) in TAE buffer, stained with ethidium bromide, and photographed in a Gelstation system (TDI, Barcelona, Spain). The DNA molecular weight markers used were the pBR322 DNA-MspI digest from New England Biolabs, Beverly, MA.

Quantitative Real-Time PCR of VCAM-1

Total RNA was extracted from mock- or TMEV-infected astrocytes cultures, with the RNeasy Mini purification kit (Qiagen) as above. cDNA was prepared from RNA using Moloney murine leukemia virus RT (Promega). Quantitative PCR was performed using the ABI PRISM 7000 Sequence Detector System instrument (Applied Biosystems, Weiterstadt, Germany). Primers for VCAM-1 and for the control house keeping gene β-actin were the same used in RT-PCR. Real-time PCRs were performed following the supplier's instructions (Applied Biosystems) in a 20-μl volume reaction using the Power SYBR Green PCR Master Mix from Applied Biosystems. All reactions were done in triplicates. The standard curves for VCAM-1 were generated using serial dilutions of cDNAs from astrocytes and its expression was normalized for β-actin expression by using the standard curve method described by the manufacturer. Data were analyzed using the 7000 System SDS software from Applied Biosystems and expressed as mean ± SEM in boxplots.

Cell-Based ELISA

The study of VCAM-1 expression was performed by using an anti-VCAM-1 affinity chromatography purified goat antibody to mouse VCAM-1, purchased by R&D Systems, and Protein A labeled with horseradish peroxidase (HRP) (Amersham Biosciences, Little Chalfont, UK; catalog number: NA 9120). Confluent astrocyte monolayers in multiwell, 6-well plates (Costar) were used in the binding experiments performed at 4°C. TMEV-infected (at a m.o.i. of 10) or mock-infected cells were incubated for 30 min with different amounts of anti-VCAM-1 antibodies or affinity purified rabbit IgG antibodies to the receptor for interferon α/β (Santa Cruz Biotechnology, Santa Cruz, CA; Catalog code sc-706) as a negative control, in 1 ml of phosphate-buffered saline (PBS)–0.1% bovine serum albumin (BSA) (Gibco). Samples were revealed with PrA-HRP linked diluted 1:5000, washed four times, and revealed with tetramethylbenzidine (TMB) as substrate for HRP during 15 min at room temperature in the dark. The yellow color produced after the addition of acid stop solution was read at 450 nm in a Titertek Multiskan plate reader.

Flow Cytometry Characterization

Sham- or TMEV-infected astrocytes were detached from 25-cm² plastic culture flasks (Costar) by using Versene (Gibco BRL). Cell viability was assessed by trypan blue exclusion and was invariably >95%. After washing, cells were incubated with either anti-VCAM-1 antibodies (at a concentration of 1 μg/ml) or an irrelevant P3X63 mAb at the same concentration, as negative control, in a total volume of 500 l of DMEM plus 0.5% BSA medium. After two washes, cells were incubated with the secondary fluorescein isothiocyanate (FITC)-conjugated rabbit anti-goat antibody (Sigma) diluted 1:200 and analyzed using a flow cytometer (Coulter EPICS-XL). The percentage of cells expressing VCAM-1 and the mean fluorescence intensity was determined. The expression indexes were calculated using the following equation:

Confocal Immunochemistry of Astrocyte Cultures

Astrocytes were trypsinized and plated in Chamber Slide culture chambers (NUNC, Naperville, IL). Confluent cultures uninfected or TMEV-infected were fixed by exposing the cells to 4% formaldehyde in PBS for 30 min at 4°C. Cell morphology was analyzed before confocal microscopy by staining with Hoechst 33258 at 2.5 μg/ml in PBS (Sigma). After washing with PBS, cell membranes were permeabilized with 0.1% Triton X-100 for 30 min at room temperature, washed, and incubated with anti-VCAM-1 antibodies (1 μg/ml) diluted in PBS containing 0.1% BSA and 0.1% Triton, or anti-GFAP goat antiserum diluted 1:500 (Dakopatts, Denmark) as a positive control, for another 30 min at 4°C. Cells were then washed and incubated for 1h with the secondary FITC-conjugated rabbit anti-goat antibody (Sigma) diluted 1:100 and exhaustively washed with PBS. Finally, the upper structures of the chambers were removed and the slides mounted in 50% glycerol. Immunofluorescence staining was examined with a Leica TCS NT confocal laser scanning microscope. FITC emission was excited using an argon laser of 488 nm and the emission signal was filtered with a 515–565-nm filter. Cells were subjected to optical serial sectioning to produce images in the X-Y plane and images were recorded digitally in a 768 × 576 pixel format. Nonspecific fluorescence was assessed by incubating cells with the secondary fluorescent antibody alone and this value was then subtracted from all specific images. Quantitative analysis was performed using an Analytical Imaging Station (Imaging Research, St. Catharines, Ontario, Canada).

Western Blots of Mock- or TMEV-Infected Astrocytes Cell Membranes

Astrocytes from at least three independently prepared cultures were washed with cold PBS and detached from the plastic flasks with Versene using a disposable cell scraper (Costar). After centrifugation for 10 min at 1000 rpm, the pellet of cells were resuspended in ice-cold 50 mM Tris.HCl, 1 mM EDTA, 0.32 M sucrose, pH 7.4, containing a protease inhibitor cocktail (1:100, Sigma) and homogenized using a Polytron homogenizer (model PT 10-35) at setting 3, for 15 s. The crude nuclear pellet, generated by centrifugation at 2000 × g for 10 min at 4°C (Sorvall RC5C, Newtown, CT) using a rotor SS-34, was discarded. The supernatant was pelleted by centrifugation at 22,500 × g for 30 min. The resulting pellet of membranes was diluted in 50 mM Tris·HCl, 1 mM EDTA, pH 7.4, containing protease inhibitor cocktail, frozen into aliquots and stored at −80°C until used. Purified membranes from mock- and TMEV-infected SJL/J mouse astrocyte cultures were solubilized with 20 mM Tris·HCl, pH 7.4 buffer, 1 mM MgCl₂, 1 mM phenylmethylsulfonyl fluoride (PMSF) (Sigma), 10 µg/ml leupeptin (Sigma) containing 1% sodium dodecyl sulfate (SDS) and 1% β-mercaptoethanol. After centrifugation and electrophoresis on 12% polyacrylamide gels, separated proteins were transferred to nitocellulose membranes and blocked with 10% defatted milk in Tris buffer. Membranes were incubated with anti-VCAM-1 affinity purified antibodies (R&D Systems) or with an anti-β-actin polyclonal rabbit antibody (Abcam, Cambridge, MA) at a concentration of 0.5 μg/ml. After washing in Tris·HCl, pH 7.4, containing 0.1% Tween, nitocellulose membranes were incubated with a 1:5000 dilution of Protein A–HRP for 2 h at room temperature. After three washes with Tris buffer–0.1% Tween, bands were visualized by chemoluminescence (ECL; Amersham Biosciences). Bands were quantified by computing densitometry using a Molecular Dynamics (Sevenoaks, Kent, UK) model 300A densitometer.

Proinflammatory Cytokine Treatments of Astrocytes

Cells were treated for 48 h with 10 ng/ml of the following mouse recombinant proinflammatory cytokines: mouse recombinant interleukin-1α (IL-1α) purchased from R&D Systems; recombinant interferon gamma (rIFN-γ) of murine origin from Holland Biotechnology (Leiden, The Netherlands); recombinant murine tumor necrosis factor alpha (TNF-α) from Innogenetics (Antwerp, Belgium); and rIL-6 from Boehringer Mannheim (Germany). Purified membranes from treated astrocytes were solubilized and analyzed by Western blots as previously stated. To test antibodies against IL-1α and TNF-α, astrocytes were trypsinized from 75-cm² flasks to 6-well multiwell plates (Costar) and infected for 1 h at room temperature with TMEV at a m.o.i. of 10. After washing, the cultures were incubated for 48 h in 2 ml of complete medium per well, containing 1, 10, or 100 μg/ml of the IgG fraction of goat antiserum against mouse IL-1α or goat affinity isolated antibodies anti-TNF-α, both from Sigma. After this period of time, purified membranes were solubilized and analyzed by Western blots.

Cell Adhesion Assay

A variant of the murine CD4+ T-helper cell clone D10 (Ojeda et al. 1995) was maintained in Click's EHAA (Eagle's Ham's Amino Acids) medium (Gibco BRL) supplemented with 10% FCS in the presence of 5 IU/ml of human recombinant IL-2 (Hoffman-La Roche, Nutley, NJ) and used for the cell adhesion assay. This cell line express normal levels of integrin α4 at his cell membrane, as demonstrated by flow cytometry (Ojeda et al. 1995), and therefore is susceptible of adhesion to the VCAM-1–expressing astrocytes. Before adhesion, D10 cells were washed and resuspended in adhesion medium (DMEM + 0.5% BSA), without a detectable loss of viability, as determined by trypan blue exclusion. Cells were labeled with the fluorogenic esterase substrate BCECF-AM (Molecular Probes, The Netherlands; catalog number: B-3051). Labeled cells (1 × 10⁵ per well in 100 μl) were added in triplicate to 96-well dishes (Costar) containing a fresh culture of mock- or TMEV-infected astrocytes (5 × 10⁴ cells/well). Plates were centrifuged for 15 s at 400 rpm, placed at 37° C for 5 min and unbound cells removed by 3 washes with adhesion medium. Bound cells were quantified using a fluorescence analyzer (CytoFluor 2300, Millipore, Bedford, MA) (Teixido et al. 1992; Sanz-Rodriguez et al. 2001). For adhesion blockade experiments, astrocytes were pretreated with anti-VCAM-1 antibody at a concentration of 50 μg/ml per well for 1 h at 37° C, washed, and used in the cell adhesion assay as above.

Intracerebral Mice Inoculations and Brain Sections Confocal Immunochemistry

Five 6-week-old SJL/J mice were anesthetized with Fluothane and intracerebral injections were made with a 25-μl Hamilton syringe at a site 1 mm right lateral and 2 mm rostal of the bregma. Twenty microliters of a suspension of BeAn virus (2 × 10⁶ plaque-forming units [PFU]) were infused at a rate of 1 μl/5 s. Five control injected mice received 20 μl of DMEM. Four days after injections, brains were removed and samples were processed by immunohistochemistry. Animals were perfused through the heart with 4% paraformaldehyde in PBS. After perfusion, brains were removed, immersed in the same fixative for 3 h, and left overnight in PBS. Vibratome sections of 30–40 μm were processed free-floating for immunohistochemistry. Sections were incubated with primary antibody (rabbit anti-TMEV antiserum obtained as explained in Rubio and Capa [1993], goat anti-mouse VCAM-1 antibody, or goat anti-GFAP antiserum) all diluted 1:1000, followed by FITC-conjugated goat anti-rabbit or Cy-3–conjugated rabbit anti-goat antibody (Amersham) diluted 1:1000. After several rinses in PBS, sections were examined for confocal microscopy as previously stated, with an excitation peak of 647 nm for Cy-3.

RESULTS

TMEV Overinduction of VCAM-1 genes

The whole-genome DNA hybridization microarray analysis allows a comparison of gene expression profile of uninfected or TMEV (BeAn strain)-infected astrocytic cells. The purpose of doing the chip analysis was to study the overall up- and down-regulation of genes induced by TMEV infection. After 24 h of infection at a m.o.i. of 10, the Affymetrix GeneChip SE437 and SE438 microarrays mouse gene expression analysis revealed that three sequences with a PFAM (Protein FAMilies) database hit description of “Immunoglobulin domain” and a SCOP (Structural Classification Of Proteins) description of “VCAM-1,” were up-regulated in infected astrocytes (Table 1). One of those sequences was absent in sham-infected astrocytes (sequence 92559 at) as its phycoerythrin signal is lower than the background of the chip (56.84). Two more sequences (especially 92558 at) were present in uninfected cells and were clearly up-regulated in TMEV-infected astrocytes. The three sequences showed increases of 6 to 6.4 times in infected cells. All of them were located in mouse chromosome number 3 and had a GO (Gene Ontology) database cellular component notation indicating that they are transmembrane molecules involved in cell-to-cell adhesion. Therefore, our DNA array gene expression analysis showed an up- regulation of the SCOP vascular cell adhesion molecule-1 (VCAM-1), which was further analyzed in this study.

Table 1. PFAM and SCOP described as “Immunoglobulin domain” and “VCAM-1” up-regulated sequences in TMEV-infected astrocytes

Affy metrix sequence	Mock-infected astrocyte signal	TMEV-infected astrocyte signal	Signal log ratio	Fold change	Chromosome location	PFAM^a description	SCOP^b description
92559 at	30.3 Absent	209.6 Present	3.2	6.4	3	Immunoglobulin domain	VCAM-1
92560 at	99.2 Marginal	815.5 Present	3.2	6.4	3	Immunoglobulin domain	VCAM-1
92558 at	333.5 Present	2569.4 Present	3.0	6	3	Immunoglobulin domain	VCAM-1

^aPFAM: “Protein FAMilies” database. ^bSCOP: “Structural Classification Of Proteins” database.

mRNA from VCAM-1 in Astrocytes

Both RT-PCR and quantitative real-time RT-PCR (qPCR) techniques were chosen to validate the previous microarray hybridization data. The presence of VCAM-1 mRNA in sham-infected and TMEV-infected astrocyte cultures was first determined by reverse transcriptase–PCR. Specific mouse VCAM-1 primer pairs were used to amplify a 355–base pair fragment, faintly present in uninfected cultures (Figure 1, lane “0”) but strongly present in cultures infected at a m.o.i. of 1 or 10 (Figure 1, lanes “1” and “10”). Negative controls, in which we omitted the Moloney murine leukemia virus RT, yielded no bands (Figure 1, −). A positive-control primer pair provided by the kit and producing a band of 300 base pairs (Figure 1, +) was also included.

Figure 1. Expression of VCAM-1 mRNA in TMEV-infected astrocytes. RNA from astrocytes mock-infected (0) or infected at a m.o.i. of 0.1, 1, and 10, 24 h postinfection, were reverse transcribed and PCR amplified by using mouse VCAM-1 primer pairs. MW: DNA molecular weight markers (pBR322 DNA-MspI digest); −: negative control where the reverse transcriptase was omitted; +: positive control provided by the kit and producing a band of 300 base pairs. β-Actin primers were used as a house keeping gene positive control and compared with DNA molecular weight markers VIII (Roche Diagnostics, Manheim, Germany).

The qPCR boxplot showed that mRNA synthesis for VCAM-1 was increased 3-folds at a m.o.i. of 0.1 and 11- to 12-folds (50th percentile) at a m.o.i. of 1 and 10 respectively (p < .05) (Figure 2). All these results were consistent with the changes obtained by the microarray DNA hybridization assay.

Figure 2. Boxplot of mRNA synthesis of VCAM-1 after infection at different m.o.i., as determined by quantitative real-time RT-PCR. Total RNA was purified 24 h postinfection. The lower bound of the boxes represents the 25th percentile, the upper bound the 75th percentile, and the line inside the 50th percentile. Significant differences compared with the mock-infected cultures, which was assigned a value of 1, as determined by the Student's t test, *p < .05, **p < .01. p < .05 was considered statistically significant.

Cell-Based ELISA

The expression changes of the studied genes were further validated at the protein level. In order to study changes in the transmembrane expression of VCAM-1, we set up a specific cell-based ELISA. Figure 3A shows the titration of the anti-VCAM-1 antibody on uninfected astrocytes. At concentrations of 5 μg/ml or higher, a plateau was reached. Therefore the antibody concentration of 5 μg/ml was used in further experiments, depicted in Figure 3B and C. A negative titration control was performed with an antibody directed against the receptor for interferon α/β, a molecule that is not expressed on the uninfected mouse astrocyte cell surface. The increase in the binding of anti-VCAM-1 antibody, demonstrating a parallel increase in the expression of such antigenic adhesion molecule, is shown in Figure 3B. This increase is proportional to the TMEV multiplicities of infection used. Finally, the analysis of the kinetics of VCAM-1 expression revealed a significant and transient increase between 8 and 24 h after infection (Figure 3C).

Figure 3. Cell-based ELISA demonstrating the expression of VCAM-1 in the cellular membrane of uninfected or TMEV-infected astrocytes. (A) The titration of anti-VCAM-1 antibody and of an irrelevant antibody (anti-IFNα/β receptor) on uninfected astrocytic cells. (B) Increase in the optical density read at 450 nm obtained in the ELISA assay of cells infected at increasing m.o.i. (from 0 to 10 infecting viral particles per cell). (C) Kinetics of the increase of reactivity in the cell-based ELISA from 0 to 48 h after infection. Results represent mean values ± SD of triplicate samples. *p < .05.

Flow Cytometry Detection of VCAM-1

We further studied the expression of VCAM-1 in astrocytes using immunochemical techniques such as flow cytometry and confocal microscopy immunohistochemistry. The flow cytometer analysis (Figure 4) showed that almost 100% (96.16–97.26%) of both sham- and TMEV-infected astrocyte cultures were stained by anti-VCAM-1 antibodies. Nevertheless, the mean fluorescence intensity increased from 9.9 in uninfected to 15.6 in virus-infected cells, demonstrating again a clear up-regulation of the adhesion molecules present in the cell membrane. As a negative control, cells that were incubated with an irrelevant monoclonal antibody (P3X63 mAb) were not stained. Those findings were resumed in the bar chart at the bottom of Figure 4, where a statistically significant increase in VCAM-1 expression is illustrated.

Figure 4. Cytofluorometric analysis of astrocytic cells stained with anti-VCAM-1 antibodies before or after TMEV infection (m.o.i. of 10). Negative controls were stained with an irrelevant antibody (P3X63 mAb). The relative percent of stained cells is indicated, followed by the mean fluorescence intensity, in parentheses in the upper corners of each quadrant. The figure shows a representative result of three separate experiments. The expression index obtained from the above results is depicted as a bar chart in the bottom half of the figure. *p < .05.

Confocal Immunohistochemistry

Figure 5 shows the comparative low-magnification images obtained by confocal immunohistochemistry of astrocytes grown in Chamber Slide culture chambers, after staining with anti-astrocyte-specific marker GFAP or with anti-VCAM-1 antibodies. The immunostaining with the anti-GFAP antibody, used as a positive control, showed the expected labeling in the cytoplasm of astrocytes, with no staining in the cell nucleus (Figure 5A). A clear increase in the staining with anti-VCAM-1 was detected in TMEV-infected astrocytes (Figure 5C) in comparison with sham-infected astrocytes (Figure 5B). Quantitative analysis performed using the Analytical Imaging Station showed a mean increase of 124.71% in bright intensity when we compared infected and sham-infected astrocytes. The above results confirmed those obtained with flow cytometry and further demonstrated that the presence of the protein product increases in the cell membrane after infection with TMEV.

Figure 5. Immunofluorecence confocal analysis of SJL/J astrocytes stained with antibodies to the astrocyte-specific marker GFAP (A) or with antibodies anti-VCAM-1 (B and C). Astrocytes in A and B were sham-infected and those in C were TMEV-infected (m.o.i. of 10). A was taken at ×63 magnification and B and C at ×20 magnification to show a broad field.

Western Blot Quantification of VCAM-1 Increase

Western blots of total proteins from uninfected (0) or from astrocytes infected at different m.o.i.'s (0.1–10) are shown in Figure 6 (black bars). At a m.o.i. of 0.1, nonsignificant changes in the antigen concentration were detected. The increase become statistically significant (p < .05) when cells were infected at m.o.i.'s of 1–10. A Western blot of the housekeeping β-actin gene product is also shown as a control for equal protein loading. The percent optical density of the bands were quantified by using a Molecular Dynamics model 300A densitometer and is shown as a bar chart at the bottom of Figure 6.

Figure 6. SDS-PAGE and Western immunoblot analysis with antibodies to VCAM-1 of membrane proteins extracts from astrocytes uninfected (0) or infected at increasing m.o.i.'s (0.1–10), showed as black column bars. The regulation of the expression of VCAM-1 in astrocytes treated with 10 ng/ml each of the recombinant inflammatory cytokines IL-1α, IL-6, IFN-γ, and TNF-α for 48 h was shown as empty bars. Purified membranes from such astrocytes were extracted with SDS and tested in Western blots as detailed in Materials and Methods. A positive control with β-actin was included to check the equal loading of the samples. The percent density of the bands is plotted in the bar chart at the bottom of the figure. Data represent the mean ± SD of triplicate samples. Significant differences compared with the untreated control groups as determined by the Student's t test, *p < .05, **p < .01.

VCAM-1 Induction by Inflammatory Cytokines

Cytokines have been reported to be strong inducers of VCAM-1 in macrophages and endothelial cells (Rice et al. 1989; Springer 1990). Here, their possible role in the up-regulation of VCAM-1 expression in astrocytes of SJL/J mice was tested by Western blotting. The effects of four recombinant inflammatory cytokines, IL-1α , IL-6, IFN-γ, and TNF-α, were assessed. As shown in Figure 6 (empty bars), the treatment with IL-6 dowregulated VCAM-1 expression. Conversely, IL-1α, IFN-γ, and in particular TNF-α, all at a concentration of 10 ng/ml, significantly increased VCAM-1 production. To determine the minimal dose of cytokines required for such an induction, astrocytes were previously incubated with 1, 10, and 100 ng/ml. VCAM-1 up-regulation was already detected after stimulation with 1 ng/ml of IL-1α, IFN-γ, or TNF-α (not shown). Human astroglioma cell lines as well as primary fetal astrocytes showed also a marked increase in VCAM-1 expression when treated with proinflammatory cytokines, particularly with TNF-α (Winkler and Benveniste 1998). Due to the fact that cytokine induction levels can vary from experiment to experiment, we compared the levels of VCAM-1 induction by TMEV and cytokines side by side, at the same time in similar cultures (Figure 6).

As both IL-1α and TNF-α (but not IFN-γ) were produced by astrocytes infected with TMEV (Rubio and Capa 1993; Sierra and Rubio 1993) and because these molecules may constitute an important stimulus for autocrine expression, we tested neutralizing antibodies against both cytokines on TMEV-induced up-regulation of VCAM-1 (Figure 7). No reduction of the levels of VCAM-1 expression was detected, even at concentrations of 100 μg/ml of both neutralizing antibodies in the culture medium. We obtained the same negative results when both antibodies were added together to the cultures (not shown), as was demonstrated previously in the case of the induction of the chemokine CXCL1 (KC) (Rubio and Sanz-Rodriguez 2007). Therefore, the above results suggest that VCAM-1 induction by either cytokines or TMEV follows independent pathways, despite of the fact that both inductions peaked 24 h after infection.

Figure 7. Lack of effect of increasing amounts of antibodies against IL-1α and TNF-α on the overinduction of VCAM-1 by TMEV infection. Astrocytes in 6-well multiwell plates were infected with TMEV at a m.o.i. of 10 for 8 h at room temperature, washed, and then incubated for 48 h in medium containing different amounts of anti-IL-1α and anti-TNF-α purified neutralizing antibodies. Treated astrocytes were tested in the cell-based ELISA test. The percent optical densities obtained are plotted. Data represent the mean ± SD of quadruplicate samples.

Cell Adhesion of a CD4+ T-Cell Line to Astrocytes

Finally, we investigated whether monolayers of astrocytes adhered to an integrin-expressing fluorescent T-cell line using VCAM-1 interactions. We choose the murine CD4+ T-helper cell clone D10 because it expresses high levels of integrin α4 at its cell membrane (Ojeda et al. 1995), and it is therefore susceptible of adhesion to cells carrying VCAM-1. The increase in adhesion by TMEV-infected astrocytes was of 122.22% compared with mock-infected cells (Figure 8, empty bars). This increase was in good agreement with the increase in VCAM-1 expression previously demonstrated by immunochemical techniques: quantitative analysis of the confocal bright intensity (124.71%) and mean fluorescent increase by flow cytometry (157.57%). The percentage of adhesion obtained, of around 50%, with an input of 10 × 10⁴ D10 lymphocytic cells seeded on 5 × 10⁴ astrocytes per well, means that each astrocyte specifically adhered to one CD4+ T lymphocyte. This 1 to 1 ratio was an expected logical value for such heterogeneous in vitro system. The use of anti-VCAM-1 antibody at a very high concentration (50 μl/ml) produced a 29% decrease of adhesion in mock-infected astrocytes (Figure 8, left black bar) and around 12% in TMEV-infected cells (Figure 8, right black bar). This fact suggests that VCAM-1–mediated adhesion was not the unique way of adhesion in this system.

Figure 8. Cell adhesion assay of the D10 clone CD4+ T lymphocytes, labeled with the fluorogenic esterase substrate BCECP-AM, to mock- or TMEV-infected astrocytes (m.o.i. of 10). The percent of adhesions are shown in the absence (white bars) or in the presence of 50 μg/ml of anti-VCAM-1 purified antibody (black bars). Background binding of D10 clone cells to the plastic of the wells (no astrocytes) represents a mean of 3% adhesion. Data represent the mean ± SD of triplicate samples of three independent experiments. *p < .05.

Increase in VCAM-1 Expression by In Vivo Infection

The cortex, septum, and nucleus accumbens of uninfected or TMEV-infected mouse brains were examined 4 days after intracerebral inoculations to determine if an increase in VCAM-1 expression was detected. Profuse green staining specific for TMEV viral particles was founded as far as 1.5 mm from the needle track (Figure 9A).

Figure 9. Confocal images of the sections of mouse brains 4 days after intracerebral inoculations of control medium (B) or TMEV virus (C), stained for VCAM-1, at the level of the nucleus accumbens. (A) Sections stained for TMEV after viral infection. The rectangle shows the area stained in B and C at higher magnification. Scale bars: 200 μm (A), 30 μm (B and C). (D) The quantitative analysis of VCAM-1 immunoreactivity in graphical form, presented as mean optical density ± SD. Significant differences of the TMEV-infected group (C) compared with the control group (B) as determined by the Student's t-test, *p < .05.

VCAM-1 staining on serial sections through the nucleus accumbens revealed VCAM-1 cell membrane and cytoplasmic red staining in both mock- (Figure 9B) and TMEV- (Figure 9C) infected brains. Colocalization of VCAM-1 and GFAP stains were founded in those areas, showing that astrocytes were the main population involved (not shown).

Using the Analytical Imaging Station software we performed a quantitative immunoreactivity analysis of areas of 152.000 μm² captured from serial brain sections (10 per brain from five brains of control medium-injected animals and five brains of TMEV-injected animals). This analysis demonstrated a significant optical density increase of 141.50% in TMEV-infected astrocytes, similar to that founded in cultured cells (Figure 9D).

DISCUSSION

As stated in the Protein FAMilies (PFAM) database description in Table 1 and as it has been previously demonstrated by other authors, vascular cell adhesion molecule-1 (VCAM-1) is a cellular adhesion protein belonging to the immunoglobulin superfamily (Osborn et al. 1989). These molecules mediate interactions between cells and between cells with the extracellular matrix. They also maintain tissue structure and are involved in cell differentiation, migration, and motility (Barclay 2003; Rojas and Ahmed 1999).

Several genes were found to be up-regulated in mouse astrocytes after infection with TMEV. We have recently demonstrated, using microarray DNA hybridization technology, up-regulation of the chemokines KC (CXCL1) and MIP-2 (CXCL2), in the main cellular component of glial population (Rubio et al. 2006; Rubio and Sanz-Rodriguez 2007). In this article we have focused on VCAM-1 because its possible implication in inflammatory processes within the CNS.

Here, we have confirmed the expression of VCAM-1 in naïve astrocytes and its overexpression in TMEV infected cells by hybridizing cRNA from both types of cells to the U74v2 total genome DNA microarray from Affymetrix. Three different sequences coding for the VCAM-1 domains, located in chromosome number 3, were up-regulated with fold change increases of 6–6.4 times (Table 1). The hybridization results were further validated by RT-PCR and qRT-PCR. Both techniques demonstrated an increase in mRNA when probed with VCAM-1–specific primers (Figures 1 and 2).

The protein product of VCAM-1 gene/s was located in the cell membrane of mock or TMEV-infected astrocytes as confirmed by using three different immunochemical methods: a cell-based ELISA (Figure 3), flow cytometry (Figure 4), and confocal microscopy (Figure 5). The results obtained demonstrated the availability of VCAM-1 molecules to antibodies on the cell membrane and an increase in VCAM-1 expression after TMEV infection. These results were quantitatively lower than that obtained after mRNA analysis, but it is well know that change increases in mRNA often misrepresent the real differences in protein synthesis.

It has been demonstrated that VCAM-1 was expressed in endothelial cells activated by inflammatory cytokines (Osborn 1990; Rice et al. 1989). Several of those cytokines (IL-1α, TNF-α) have been previously demonstrated by ourselves to be produced by astrocytes upon TMEV infection (Rubio and Capa 1993; Sierra and Rubio 1993). In addition, recombinant IL-1α and TNF-α were also potent inducers of VCAM-1 synthesis in this system, as demonstrated by Western blots in Figure 6. Maximal VCAM-1 expression was simultaneous to the production of both cytokines by TMEV, suggesting some overlapping. Nevertheless, neutralizing antibodies against IL-1a and TNF-a fail to reduce TMEV-induced VCAM-1 production (Figure 7), indicating that both inductions follows distinct pathways.

In normal brain development stages VCAM-1 plays a role in migration, guidance and recognition in astrocytes and other glial cells (Fabre-Kontula et al. 2008). In pathological situations, as MS and other forms of demyelination, VCAM-1 expressed in the cell membrane of astrocytes could mediate the adhesion to integrin α4-containing CD4+ T lymphocytes, as it was shown in vitro by us (Figure 8). To verify the in vivo pathological relevance of our findings, we demonstrated a significant increase in VCAM-1 expression (mean 141.71%) in brain astrocytes infected by TMEV after intracerebral injections (Figure 9). In summary, we have shown that infection with TMEV increases expression of VCAM-1 on astrocytes in vitro and in vivo and increases adhesion of T cells to infected astrocytes in vitro. The results suggest that VCAM-1 overexpression on infected astrocytes might be relevant and influence the course of disease by playing a role in the adhesion of immune cells after crossing the blood-brain barrier in an endothelial cell-mediated process. Those cells, expressing its counter-receptor integrin α4, were previously recruited by chemokines that were found to be present in vivo in the serum and the brain of demyelinated sick SJL/J mice (Rubio et al. 2006; Rubio and Sanz-Rodriguez 2007).

Astrocytes play an important role in the immune response that triggers demyelination after intracerebral infection by the TMEV low-neurovirulence BeAn viral strain. Demyelinating lesions induced in this model contains intensive infiltrates of mononuclear immune cells and therefore a cellular immune response is strongly indicated in myelin destruction (Lipton and DalCanto 1976). It has been proposed that cytokines and chemokines released in the site of TMEV infection leads to recruitment and homing into the CNS of virus-specific CD4+ T cells as those used by us in the cell adhesion assay, in addition to monocyte/macrophages, neutrophils, and other immune cells. Those cells would cause demyelination by a nonspecific bystander response resulting in stripping of myelin lamellae (Clatch et al. 1986). The results presented here points to a role of VCAM-1 in the TMEV-induced experimental model of MS by facilitating and increasing CD4+ T cells infiltration inside CNS demyelinating areas. Our results extend previous findings showing that VCAM-1 expression in astrocytes is necessary for T-cell entry into the CNS parenchyma in experimental autoimmune encephalomyelitis (EAE), another experimental model for MS (Gimenez et al. 2004). Finally, an increase of similar magnitude in both human astrocyte VCAM-1 expression and adhesion to monocytes, was induced by human immunodeficiency virus type 1 (HIV-1) (Woodman et al. 1999).

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.