Tenascin-R Is Antiadhesive for Activated Microglia that Induce Downregulation of the Protein after Peripheral Nerve Injury: a New Role in Neuronal Protection

MATERIALS AND METHODS

Materials. Mouse monoclonal Abs tn-R1 and tn-R2 (clones 597 and 596) to TN-R have been characterized (Pesheva et al., 1989). Monoclonal Ab tn-R6 was raised in BALB/c mice using an equimolar mixture of adult brain-derived chick and human TN-R immunoaffinity purified on a tn-R2 Ab column as antigen and for the screening of positive hybrid clones. The Ab belongs to the IgG1 subclass and recognizes a protein epitope of yet unknown location present on the 160 and 180 kDa isoforms of TN-R derived from chick, mouse, rat, bovine, and human brain. In Western blot analyses of brain extracts from a number of vertebrates, tn-R6 reacts with two single bands of 160 and 180 kDa corresponding to TN-R (P. Pesheva, E. Spiess, K. Winterhalter, and T. Peshev, unpublished observations). Polyclonal Abs to mouse brain-derived TN-R 160 (pTN-R), which react with TN-R 160 and TN-R 180 from rat brain but not with TN-C, were produced in rabbits. Polyclonal Abs to mouse brain-derived TN-C (recognizing TN-C in the rat) and laminin (LN) from Engelbreth–Holm–Swarm (EHS) mouse sarcoma were produced in rabbits (Pesheva et al., 1994). Rat monoclonal Ab J1/tn2 recognizes the fibronectin type III domain fnD of TN-C (Götz et al., 1996). Mouse monoclonal Abs O4 and O1 recognize glycosphingolipids expressed at the cell surface of OLs of different maturation stages (Sommer and Schachner, 1981; Bansal et al., 1989). Monoclonal Ab Ox-42 recognizing the rat complement receptor type 3 (CR 3) complex was purchased from Serotec (Oxford, England). Biotin-labeled Vicia villosa agglutinin was purchased from Sigma (Deisenhofen, Germany). TN-R was purified from brains of adult Wistar rats by immunoaffinity chromatography on a monoclonal tn-R2 Ab column (Pesheva et al., 1989). These TN-R preparations consisted of an approximately equimolar mixture of the 160 and 180 kDa isoforms and were recognized by all monoclonal Abs to TN-R used in this study when analyzed by Western blot and ELISA.

Surgery. For all surgical operations, 3-month-old female Wistar rats (strain HsdCpb:Wu) were used. Sixty-two animals underwent surgery, and two animals served as normal controls. The surgical operations on facial and hypoglossal nerves were performed under a microscope by trained microsurgeons. The animals were anesthetized with ether, and after an intraperitoneal injection of 1.4 ml of Avertin (2 gm of tribromethanol, 1 ml of 3-pentanol, 8 ml of absolute ethanol, and 90 ml of 0.9% NaCl), the main trunk of the facial or hypoglossal nerve was unilaterally exposed and immobilized. Nerves were axotomized under conditions allowing (transection and suture) or not allowing (resection) subsequent regeneration.

Transection and suture. Forty rats underwent an unilateral transection and immediate end-to-end suture of the facial (20 rats) or the hypoglossal (20 rats) nerve. The main trunk of the facial nerve was exposed and transected at its emergence from the foramen stylomastoideum distally to the posterior auricular branch. The proximal stump was then microsurgically sutured to the distal stump with two 11–0 atraumatic sutures (Ethicon), an operation known as facial–facial anastomosis (FFA) (Guntinas-Lichius et al., 1994). Animals undergoing FFA were divided into six groups (of 2 or 4 rats each) that were allowed to survive for 3, 5, 7, 14 (4 rats each), 56, and 112 postoperative days (PODs) (2 rats each). For unilateral transection of the hypoglossal nerve, the nerve trunk was exposed and transected between its medial and lateral branches, followed by an immediate end-to-end suture [(hypoglossal–hypoglossal anastomosis (HHA)]. Animals undergoing HHA were also divided into six groups (of 2 or 4 rats each, as above) and allowed to survive for the same time periods as described above.

Resection. Sixteen rats underwent an unilateral resection of the motor nerve. In eight rats, a piece of 8–10 mm length of the right temporal, zygomatic, buccal, upper, and lower divisions of the marginal mandibular branch was removed surgically. In the other eight rats, a piece of 8–10 mm length of the hypoglossal nerve was removed (Neiss et al., 1992; Guntinas-Lichius et al., 1994). Animals undergoing resection of the facial or hypoglossal nerve were divided into four groups each that were allowed to survive for 7, 14, 30, and 112 PODs (2 rats each).

At the end of each postoperation period, animals were anesthetized with ether, and their vascular systems were rinsed with 0.9% NaCl. Rats were then treated with fixative by transcardial perfusion with a periodate–lysine–paraformaldehyde fixative containing 4% paraformaldehyde (PA) and 10 mm sodiummeta-periodate in 0.2 m lysine–HCl buffer, pH 7.4 (McLean and Nakane, 1974). The brainstems were subsequently removed and cut into 50-μm-thick coronal vibratome sections in 0.1m phosphate buffer, pH 7.4.

Immunocytochemistry. Immunostaining of tissue sections was performed as described previously (Angelov et al., 1995, 1996b). The immunoreaction product was visualized using a horseradish peroxidase–avidin–biotin complex and diaminobenzidine tetrahydrochloride (DAB). In control experiments, the omission of primary or biotinylated secondary Abs resulted in a complete lack of immunoreaction product.

Perineuronal nets around motoneurons were identified by biotinylated Vicia villosa agglutinin (Sigma). Tissue sections were incubated with biotinylated Vicia villosa agglutinin (1:200) for 3 hr, and lectin binding was detected by Cy2-conjugated streptavidin (1:100) (Sigma).

The expression of specific cell surface markers by cultured microglial cells and oligodendrocytes from early postnatal rat brain was analyzed by indirect immunofluorescence (Pesheva et al., 1998) using Cy3-conjugated secondary Abs (Jackson ImmunoResearch, West Grove, PA).

Electron microscopy. After HRP-DAB reaction, sections were rinsed in 0.1 m cacodylate buffer, pH 7.2, incubated for 2 hr in 1% OsO₄ and 1.5% K₃Fe^III(CN)₆ in the same buffer, and dehydrated in graded acetones. Sections were then embedded in araldite (Durcupan ACM; Fluka, Buchs, Switzerland) and further processed for electron microscopy.

Glial cell cultures. Microglial cells were obtained from cerebral glial cultures of 1- or 2-d-old rat pups (from Wistar or Lewis rats) as described previously (Pesheva et al., 1998). After 11–14 d in culture, microglial cells were shaken off the astrocytic monolayer and immediately used for cell adhesion assays (see below) or maintained for several days in culture in DMEM containing 10% or 5% fetal calf serum (FCS). This procedure yielded a homogeneous population of microglial cells expressing marker molecules for activated microglia, such as CR 3 and galectin-3 (Graeber et al., 1988; Pesheva et al., 1998).

OLs were obtained from cerebral glial cultures of neonatal rat pups (McCarthy and de Vellis, 1980). After 12 d in culture in DMEM containing 10% FCS, OLs were shaken off the astrocytic monolayer, collected by centrifugation (600 × g for 10 min at 4°C), and resuspended in the same medium. Contaminating microglial cells were removed from the cell suspensions by preadhesion to untreated culture dishes (twice for 30 min at 37°C). Nonadherent cells were then collected by centrifugation, resuspended in DMEM containing 5% FCS, plated onto poly-l-lysine (PLL)-coated (0.1 mg/ml in water) tissue culture dishes at a density of 1–2 × 10⁶ cells/ml, and maintained for several days in culture. Under these conditions, 100% of the cells were O4-positive, and the majority of them were O1-positive, defining them as mature OLs (R. Probstmeier and P. Pesheva, unpublished observations).

Microglia-conditioned media (MCM) were obtained after 2 or 3 d in culture in DMEM containing 5% FCS, cleared by centrifugation (100,000 × g for 20 min at 4°C), and immediately frozen at −70°C until use or directly applied to cultured OLs. After 2 d in culture on PLL-treated multiwell plates (4- or 6-well plates; Nunc, Roskilde, Denmark), OLs were cultured for 2 d in MCM or DMEM containing 5% FCS in the absence or presence of tumor necrosis factor-α (TNF-α) (200 U/ml; PeproTech, Rocky Hill, NJ) or interleukin-1β (IL-1β) (10 U/ml; PeproTech). The resulting OL-conditioned media were cleared by centrifugation (100,000 ×g for 20 min at 4°C) and stored at −70°C until use.

Cell adhesion assays. For cell adhesion assays on purified protein substrates, tissue culture dishes were coated with PLL (concentration of 0.01%), washed three times with water, and air-dried. TN-R and LN from EHS sarcoma (20 μg/ml in PBS) were coated onto the PLL layer as 2 μl droplets for 60 min at 37°C, and dishes were then washed twice with PBS. After 60 min of incubation at 37°C with PBS containing 2% heat-inactivated bovine serum albumin (BSA), Abs to TN-R (final concentration of 100 μg/ml) were added to the solution, and dishes were incubated overnight at 4°C and subsequently washed three times with PBS. Microglial cells derived from mixed glial cultures of 1- or 2-d-old rat pups (see above, Glial cell cultures) were plated onto the substrates at a density of 1 × 10⁶ cells/ml in DMEM containing 10% FCS. After 30 min of incubation at 37°C, nonadherent cells were washed away with PBS, and dishes were incubated for 30 min at room temperature with 4% PA in PBS. Adherent cells were then stained for 20 min with 0.5% toluidine blue in 2.5% Na₂CO₃, washed once with water, and air-dried. The substrate spots were visualized by ELISA using monoclonal tn-R2 and polyclonal Abs to LN, and the respective secondary alkaline phosphatase-conjugated Abs (Promega, Madison, WI) or in Ab perturbation experiments, the secondary Abs only. For estimation of the number of cells adhering to the different substrates, cells from micrographs were counted in microscopic fields corresponding to 800 μm². Mean ± SD of number of adherent cells in five different microscopic fields are represented as percentages.

For cell adhesion assays on brainstem cryosections, two rats underwent unilateral resection of the facial nerve, and another two underwent resection of the hypoglossal nerve. After a postoperative survival period of 10 d, the brainstems were quickly removed and immediately frozen at −159°C in melting isopentane precooled in liquid nitrogen. The brainstem regions containing the facial or hypoglossal nuclei were then cut into 10-μm-thick cryosections, mounted onto sterile coverslips (11 mm in diameter), placed into wells of 24-well plates, and kept at −70°C until use. Microglial cells shaken off the astrocytic monolayer (see above) were labeled for 10 min at room temperature with bisbenzimide (20 μg/ml in Ca²⁺- and Mg²⁺-free HBSS). Cells were subsequently washed three times (600 × g for 5 min at 4°C) with cold DMEM containing 10% FCS (culture medium) and resuspended in culture medium. Frozen 24-well plates containing brainstem cryosections were kept for 10 min at room temperature and incubated for 90 min at 37°C with 200 μl/well culture medium containing or not containing monoclonal and polyclonal Abs to TN-R (100 μg of IgG/ml culture medium). For each individual Ab, five equidistant cryosections through the nuclei (with an interval between the sections of ∼200 μm) were preincubated as described above and used as a substrate for microglial adhesion. The medium was then carefully removed and single-cell suspensions of labeled microglial cells (1 × 10⁶ cells/ml culture medium; 200 μl/well) were added to the wells. After 30 min of incubation at 37°C, unbound cells were gently washed away, once with 500 μl of culture medium and twice with 500 μl of PBS containing 1% BSA. Cryosections were subsequently treated for 30–45 min at room temperature with 4% PA containing 5% sucrose prewarmed at 37°C, washed twice with PBS, and embedded. For quantitative analysis, the boundaries of the nuclei on each section were delineated in the bright-field image using a CCD video camera system (Optronics Engineering) and image analyzing software Optimas 6.1 (Optimas) and were superimposed onto the fluorescence image of bisbenzimide-labeled cells (filter set 01; Zeiss, Jena, Germany). The number of labeled cells adherent to individual nuclei of the operated and unoperated side was counted and related to the surface area of the nucleus. The data obtained are represented as mean ± SD of the number of cells per area (500 μm²) of a nucleus from the unoperated (control) and operated side, respectively. To measure statistically significant Ab effects, the Student’s t test was applied.

Western blot analysis. Two rats underwent unilateral resection of the facial nerve, and at POD 7, the animals were anesthetized and decapitated. The regions of the brainstem containing the facial nuclei were quickly removed and separated along the midline into an operated and unoperated portion. The tissue pieces were subsequently homogenized in 10 mm Tris-HCl, 150 mm NaCl, 5 mm EDTA, 5 mm EGTA, pH 7.4, containing 1 mm spermidine, 1 μmaprotinin, 5 μm soybean trypsin inhibitor, 1 mm phenylmethylsulfonyl fluoride, 1 mmiodoacetamide, and 1 mm type III egg-white trypsin inhibitor (extraction buffer) extracted for 90 min at 37°C, and insoluble material was cleared by centrifugation (100,000 ×g for 30 min at 4°C). TN-R in tissue extracts of equal protein concentration was selectively immunoprecipitated by using pTN-R and pansorbin cells (Calbiochem, La Jolla, CA) or monoclonal Ab tn-R2 and rabbit anti-mouse IgG (Jackson ImmunoResearch) linked to pansorbin cells as a carrier. After an end-to-end rotation overnight at 4°C, pansorbin cells were washed three times with 1 ml of extraction buffer (100,000 × g for 2 min at 4°C), resuspended in 40 μl of SDS sample buffer (Laemmli, 1970), and boiled for 4 min. Pansorbin cells were removed by centrifugation, and protein samples were subjected to SDS–PAGE using 7% polyacrylamide slab gels and further analyzed by Western blot using monoclonal TN-R Abs or pTN-R, secondary HRP-conjugated Abs (Boehringer Mannheim, Mannheim, Germany), and the ECL method for detection (Pierce, Rockford, IL).

Sandwich ELISA. Media conditioned by OLs (see above, Glial cell cultures) were analyzed by sandwich ELISA using monoclonal Abs to TN-R as a linker (tn-R1 or tn-R2 at a coating concentration of 20 μg/ml) and pTN-R immunoaffinity purified on a TN-R column for detection. After incubation of OL-conditioned media (overnight at 4°C, 100 μl/well) with the immobilized monoclonal Abs, polyclonal Ab binding (for 2 hr at 37°C) was detected by biotinylated Fab fragments of sheep anti-rabbit IgG and streptavidin-HRP conjugate (Boehringer Mannheim). Values for the TN-R content under different culture conditions were plotted onto a standard curve prepared in parallel from purified rat TN-R (2 μg/ml to 3 pg/ml) and represented as mean ± SD of triplicate measurements.

RT-PCR. Total RNA was isolated from 2–4 × 10⁶ OLs cultured under different conditions by phenolchloroform extraction (Chomczynski and Sacchi, 1987), and 3 μg of RNA was used for oligo-dT-primed single-strand cDNA synthesis with superscript reverse transcriptase using the SuperScript preamplification system (Life Technologies, Eggenstein, Germany) in 20 μl of reaction volume. Single-strand cDNA (4 μl) was amplified in 20 μl of reaction volume using the rat TN-R-specific primers 5′-GACATACAAGTCCACCGAT-3′ and 5′-CTGTGAGACGATGGATGTA-3′ in 10 mm Tris-HCl, pH 9.0, 50 mm KCl, 1.5 mm MgCl₂, 0.1% Triton X-100, 0.2 μm dNTP, and 0.26 μm primer each. This amplification leads to a 827 bp product. As a control, glycerinaldehyde-3-phosphate-dehydrogenase (GAPDH)-specific primers were used with 1 μl of template cDNA. Products were amplified through 39 cycles of 45 sec at 94°C, 45 sec at 50°C, and 90 sec at 72°C. PCR products (8 μl) were analyzed on 1% agarose gels.