QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stallcup, W. B. Articles by Dahlin-Huppe, K. Search for Related Content PubMed PubMed Citation Articles by Stallcup, W. B. Articles by Dahlin-Huppe, K.

Chondroitin sulfate and cytoplasmic domain-dependent membrane targeting of the NG2 proteoglycan promotes retraction fiber formation and cell polarization

The Burnham Institute, La Jolla Cancer Research Center, 10901 North Torrey Pines Road, La Jolla, CA 92037, USA

View larger version (12K): [in a new window]   Fig. 1. Evaluation of chondroitin sulfate content by immunoblotting. Immunoblot analysis with rabbit anti-NG2 antibody was performed on detergent extracts of B28 cells transfected with wild-type and mutant NG2. Half of each sample was treated with chondroitinase ABC to remove chondroitin sulfate chains (+), while the other half was left untreated (-). In the case of B28/NG2.6 cells, which express wild-type NG2 (c6/wt), this chondroitinase treatment converts the high-molecular-weight smear representing the mature proteoglycan (asterisk) into the 300 kDa core glycoprotein (arrowhead). A similar increase in the quantity of the core is seen with extracts of cells expressing the NG2/S1342A mutant (clones 42 and 44), indicating that this form of NG2 still contains chondroitin sulfate. By contrast, chondroitinase treatment has no apparent effect on extracts of cells expressing the NG2/S999A mutant (clones 51 and 65), suggesting that this form of NG2 is not modified with chondroitin sulfate.

View larger version (129K): [in a new window]   Fig. 2. Chondroitin sulfate affects targeting of NG2 to retraction fibers. B28 cells expressing wild-type NG2 (clone 6, a,b,g,h), NG2/S1342A (clone 42, c,d), NG2/S999A (clone 51, e,f), NG2/L1 (clone 5, i,j) and NG2/CNTN (clone M, k,l) were grown on poly L-lysine-coated dishes overnight. In panels g and h, cells expressing the wild-type NG2 were chondroitinase treated for 1 hour at 37°C. All sets of cells were then colchicine treated (10-5 M) for 30 minutes at 37°C and stained with monoclonal anti-NG2 antibody (b,d,f,h,j,l) and poly-specific rabbit anti-B28 antibody (a,c,e,g,i,k). After colchicine treatment, all cells exhibit dense arrays of retraction fibers as shown by the staining with poly-specific antibody. However, NG2 staining in the NG2/S999A cells is much weaker than in the wild-type or NG2/S1342A cells. Removal of chondroitin sulfate from cells expressing wild-type NG2 also results in diminished localization of NG2 to the retraction fibers (g,h). Diminished, punctate NG2 staining in retraction fibers is also seen in the NG2/CNTN transfectants, whereas no retraction fiber staining for NG2 is observed for NG2/L1. Bar, 10 µm (j).

View larger version (103K): [in a new window]   Fig. 3. Localization of NG2 to cellular retraction fibers. Parental B28 cells (g,h) and B28 cells expressing wild-type NG2 (clone 6, a,b), NG2/S1342A (clone 42, c,d), and NG2/S999A (clone 51, e,f) were grown overnight on poly L-lysine-coated plates, fixed with 2% paraformaldehyde, and then double-labeled with monoclonal anti-NG2 antibody (b,d,f,,h) and rabbit antibody against whole B28 cells (a,c,e,g). The paraformaldehyde fixation is important because antibody incubations and repeated washing of living cells tend to induce cell ‘rounding’, accompanied by retraction fiber formation. Thus, the fixed cells give a more accurate picture of cell morphology under actual culture conditions. In the case of cells expressing wild-type NG2 and NG2/S1342A, abundant retraction fibers visible by staining with the poly-specific B28 antiserum are also heavily labeled by the anti-NG2 antibody. B28 cells and transfected cells expressing NG2/S999A have many fewer retraction fibers. In the NG2/S999A transfectants these fibers are not well stained by the anti-NG2 antibody, as seen previously in Fig. 2. B28 cells are negative for NG2. Bar, 10 µm (h).

View larger version (21K): [in a new window]   Fig. 4. Quantitation of retraction fiber formation. (A) Using cultures like those described in Fig. 3, we examined at least 200 cells to determine the percentage of cells that displayed retraction fibers, as determined by staining with the poly-specific rabbit antibody against B28 cells. The figure compares parental B28 cells (P) and B28 cells transfected with wild-type NG2 (clone 6), NG2/S1342A (clones 42 and c44), NG2/S999A (clones 51 and c65), NG2/L1 (clone 5) and NG2/CNTN (clone M). Mean and standard error values were calculated from data obtained with three replicate plates of each cell type. (B) Using the same sets of cells described in A, we counted the number of retraction fibers per cell. Values for at least 13 cells are plotted for each cell type in the figure. Horizontal bars show the mean value for each cell type. Cells with small numbers of retraction fibers are the same ones identified as low responders in A.

View larger version (144K): [in a new window]   Fig. 5. NG2-mediated cell polarization. B28 cells transfected with wild-type NG2 were serum-starved overnight, resuspended in DMEM/BSA for 1 hour, and then allowed to spread for 8 hours in DMEM/FCS on plates coated with mAb D120. Cells were then fixed and double-stained with rabbit anti-NG2 (a) and mouse monoclonal anti-ß-actin (b), or with rabbit anti-B28 (c) and mouse monoclonal anti-fascin (d). Retraction fibers at one pole of the cell are strongly NG2-positive, whereas actin and fascin are localized to lamellipodia at the opposite pole. Bar, 10 µm (a).

View larger version (13K): [in a new window]   Fig. 6. Involvement of the Rho GTPase in cell polarization. NG2-transfected B28 cells were serum-starved overnight, resuspended in DMEM/BSA for 1 hour, then allowed to spread for 8 hours on PLL- or mAb-coated plates. Cells were then fixed with paraformaldehyde and stained with poly-specific rabbit antibody against B28 cells. The percentage of morphologically polarized cells was determined under a variety of conditions. On PLL plates the control condition was 1% BSA in DMEM, to which were added either 10 ng/ml of PDGF-AA and PDGF-BB or 1 µg/ml LPA. For mAb-coated plates the control condition is mAb CD44 coating, whereas the stimulatory condition is mAb D120 coating. When used, the C3 exoenzyme (3 µg/ml) and the Y27632 (Y) rho-associated kinase inhibitor (20 µM) were added to cells during the 1 hour suspension period and maintained in the cultures throughout the spreading phase of the experiment. Mean and standard error values for each condition were calculated from data obtained from three replicate plates.

View larger version (146K): [in a new window]   Fig. 7. Effect of NG2 chondroitin sulfate chain on cell polarization. Cell polarization was examined after 8 hours on PLL plates in the presence of 1 µg/ml LPA (d-f) and on mAb D120 plates in DMEM/BSA (a-c). Under both sets of conditions, B28 cells transfected with wild-type NG2 (a,d) and with NG2/S1342A (b,e) exhibited a high tendency to polarize. By contrast, many fewer instances of polarization were observed for NG2/S999A-transfected cells (c,f). Bar, 10 µm (f).

View larger version (17K): [in a new window]   Fig. 8. Quantitation of cell polarization. B28 cells transfected with wild-type NG2 (clone 6) and with NG2/S1342A (clones 42 and 44) and NG2/S999A (clones 51 and 65) were compared with parental B28 cells (P) for their ability to polarize after 8 hours of spreading on PLL-coated (A) and mAb-coated dishes (B). After spreading, the cells were fixed with paraformaldehyde and stained with poly-specific anti-B28 antibody as shown in Fig. 7 to allow determination of the percentage of polarized cells. Mean and standard error values were calculated from data obtained with three replicate plates for each cell type. On PLL dishes the addition of LPA to the control BSA-containing medium stimulated polarization in wild-type and NG2/S1342A transfectants, but not in parental cells or NG2/S999A transfectants. On D120-coated dishes, wild-type and NG2/S1342A transfectants had much a greater tendency to polarize than on mAb CD44-coated dishes, whereas NG2/S999A transfectants did not distinguish significantly between the two surfaces.