Tenascin supports lymphocyte rolling

Abstract

Tenascin is a large extracellular matrix molecule expressed at specific sites in the adult, including immune system tissues such as the bone marrow, thymus, spleen, and T cell areas of lymph nodes. Tenascin has been reported to have both adhesive and anti-adhesive effects in static assays. We report here that tenascin supports the tethering and rolling of lymphocytes and lymphoblastic cell lines under flow conditions. Binding was calcium dependent and was not inhibited by treatment of lymphocytes with O-glycoprotease or a panel of glycosidases including neuraminidase and heparitinase but was inhibited by treatment of cells with proteinase K. Binding was to the fibrinogen-like terminal domain of tenascin as determined by antibody blocking studies and binding to recombinant tenascin proteins. When compared to rolling of the same cell type on E-selectin, rolling on tenascin was found to be smoother at all shear stresses tested, suggesting that cells formed a larger number of bonds on the tenascin substrate than on the E-selectin substrate. When protein plating densities were adjusted to give similar profiles of cell detachment under increasing shears, the density of tenascin was 8.5-fold greater than that of E-selectin. Binding to tenascin was not dependent on any molecules previously identified as tenascin receptors and is likely to involve a novel tenascin receptor on lymphocytes. We postulate that the ability of tenascin to support lymphocyte rolling may reflect its ability to support cell migration and that this interaction may be used by lymphocytes migrating through secondary lymphoid organs.

Figure 3

The domain structure of the hexabrachion arm of human tenascin (above) and the bacterial expression proteins used in this study (below), modified from reference .

Figure 1

Cell accumulation on plastic-adsorbed tenascin. (a) Ability of various cell types to accumulate on tenascin under flow. Cells were accumulated for 40 s each at 0.27 and 0.53 dynes/cm². Cells were counted after the 0.53 dynes/cm² accumulation step; all adherent cells were rolled. One of two representative experiments is shown with the range indicated by bars. (b) Effects of inhibitors and enzyme treatments on SKW3 cell accumulation on tenascin. SKW3 cells were treated for 10 min at 25°C with either 0.1% azide and 50 mM 2-deoxyglucose (Azide+DOG), inclusion of 5 mM EDTA or 10 mM EGTA and 1 mM Mg²⁺ in the assay, or treatment for 40 min at 37° with chondroitinase ABC, heparitinase, hyaluronidase, neuraminidase, neuraminidase and exo-β-galactosidase, O-glycoprotease, or proteinase K. Control binding represents binding of control-treated cells to the same tenascin substrate immediately before binding of treated cells. Mean values for two experiments are shown with the range indicated by bars.

Figure 1

Cell accumulation on plastic-adsorbed tenascin. (a) Ability of various cell types to accumulate on tenascin under flow. Cells were accumulated for 40 s each at 0.27 and 0.53 dynes/cm². Cells were counted after the 0.53 dynes/cm² accumulation step; all adherent cells were rolled. One of two representative experiments is shown with the range indicated by bars. (b) Effects of inhibitors and enzyme treatments on SKW3 cell accumulation on tenascin. SKW3 cells were treated for 10 min at 25°C with either 0.1% azide and 50 mM 2-deoxyglucose (Azide+DOG), inclusion of 5 mM EDTA or 10 mM EGTA and 1 mM Mg²⁺ in the assay, or treatment for 40 min at 37° with chondroitinase ABC, heparitinase, hyaluronidase, neuraminidase, neuraminidase and exo-β-galactosidase, O-glycoprotease, or proteinase K. Control binding represents binding of control-treated cells to the same tenascin substrate immediately before binding of treated cells. Mean values for two experiments are shown with the range indicated by bars.

Figure 2

Accumulation, rolling velocity, and resistance to detachment of SKW3 cells on tenascin. (a) Accumulation of SKW3 cells on plastic-adsorbed tenascin at various shears. Cells introduced at a concentration of 5 × 10⁶ cells/ml were accumulated for 40 s at the indicated shear stresses in separate experiments. The mean values of two experiments are shown with the range indicated by bars. (b) Rolling velocities of SKW3 cells initially accumulated on plastic-adsorbed tenascin for 40 s at 0.27 dynes/cm². Velocity measurements were done on all cells in the field; bars represent SEM for each measurement. (c) Detachment of SKW3 cells bound to plastic-immobilized tenascin under increasing shear. Cells were initially accumulated for 40 s at 0.27 dynes/cm² and then perfused with cell-free medium for 10 s at each of the indicated shears. Cell accumulation at the end of the 0.27 dynes/cm² accumulation step was considered 100% binding; the numbers of cells remaining at the end of subsequent steps were compared to this value to determine percent detachment. The mean values of two experiments are shown; bars representing the range of measurements are present but are too small to be seen.

Figure 4

Localization of antibody binding sites with immunoblotting and inhibition of cell binding to tenascin. (a) Recognition of tenascin and recombinant tenascin proteins by antibodies. Proteins were spotted on nitrocellulose and detected with the indicated antibodies and alkaline phosphatase–anti-Ig. Antibodies included normal rabbit serum (NRS); rabbit antibodies against intact tenascin (α-TN), recombinant TNfnA-D (α-TNfnA-D), recombinant TNfn1-5 (α-TNfn1-5), and a mixture of recombinant TNfn6-8 and TNfbg (TNfn68fbg); a negative IgG control mAb; and mAb M112, M139, M168, and M171 against tenascin. (b) Inhibition of SKW3 cell binding to plastic-adsorbed tenascin by tenascin antibodies. Binding of SKW3 cells was measured before and after treatment of the spots of plasticimmobilized tenascin with polyclonal antisera (1:50), ascites (1:100), or purified mAb (50 μg/ml) for 10 min. Binding to control NRStreated spot was 250 cells and binding to IgG control-treated spot was 224 cells. Mean values of two experiments are shown, with the range indicated by bars.

Figure 4

Localization of antibody binding sites with immunoblotting and inhibition of cell binding to tenascin. (a) Recognition of tenascin and recombinant tenascin proteins by antibodies. Proteins were spotted on nitrocellulose and detected with the indicated antibodies and alkaline phosphatase–anti-Ig. Antibodies included normal rabbit serum (NRS); rabbit antibodies against intact tenascin (α-TN), recombinant TNfnA-D (α-TNfnA-D), recombinant TNfn1-5 (α-TNfn1-5), and a mixture of recombinant TNfn6-8 and TNfbg (TNfn68fbg); a negative IgG control mAb; and mAb M112, M139, M168, and M171 against tenascin. (b) Inhibition of SKW3 cell binding to plastic-adsorbed tenascin by tenascin antibodies. Binding of SKW3 cells was measured before and after treatment of the spots of plasticimmobilized tenascin with polyclonal antisera (1:50), ascites (1:100), or purified mAb (50 μg/ml) for 10 min. Binding to control NRStreated spot was 250 cells and binding to IgG control-treated spot was 224 cells. Mean values of two experiments are shown, with the range indicated by bars.

Figure 5

Ability of recombinant tenascin proteins to support SKW3 cell rolling. Cells were accumulated for 40 s at 0.27 dynes/ cm² shear stress; adhesion events were counted throughout this 40 s period. “Accumulation” represents cells that rolled a distance of at least four cell diameters. “Tethering events” were defined as cells that tethered under flow and remained bound to the substrate for at least one second. Mean values of two experiments are shown, and the range is indicated by bars.

Figure 6

Comparison of KG1a cells rolling on tenascin and E-selectin. (a) Detachment profiles of KG1a cells on tenascin and two plating dilutions of E-selectin. (b) Rolling velocities of KG1a cells on tenascin and E-selectin. The mean values of two experiments are shown; ranges are indicated by bars.

Figure 6

Comparison of KG1a cells rolling on tenascin and E-selectin. (a) Detachment profiles of KG1a cells on tenascin and two plating dilutions of E-selectin. (b) Rolling velocities of KG1a cells on tenascin and E-selectin. The mean values of two experiments are shown; ranges are indicated by bars.

Figure 7

Rolling on tenascin is smoother than rolling on E-selectin at all shear stresses examined. (a–c) Displacement and velocity profiles of cells rolling on tenascin and E-selectin at 2.7 dynes/cm² shear stress. (c–e) Change in velocity profiles of cells rolling on tenascin and E-selectin at 10 μm/s. The variance of velocity measurements is as shown.

Figure 8

Binding of SKW3 cells to tenascin is not inhibited by mAbs to known tenascin ligands or RGD peptide. SKW3 cells were treated with the indicated mAbs or GRGDSP peptide before flow assays; M168 and GRGDSP were also added to the flow chamber before the addition of cells. For GRGDSP peptide experiments, cells were maintained in the continued presence of the peptide. Observed binding was compared to binding of cells treated with control IgG CBRp150/2E1 (mAb experiments; control binding was 162 cells), normal rabbit serum (rabbit Ab; control binding was 171 cells), or untreated cells (GRGDSP peptide experiments; control binding was 293 cells) to obtain percent control binding. The mean values of two experiments are shown; ranges are indicated by bars.

Figure 9

Immunofluorescent staining of human tonsil with tenascin mAb M168. Tenascin is heavily expressed by HEV and the reticular fiber network of the interfollicular T cell areas, while B cell follicles show sparse expression. “F” denotes a B cell follicle; three prominent HEV are visible below the follicle.