A crosslinking analysis of GAP-43 interactions with other proteins in differentiated N1E-115 cells

Abstract

It has been suggested that GAP-43 (growth-associated protein) binds to various proteins in growing neurons as part of its mechanism of action. To test this hypothesis in vivo, differentiated N1E-115 neuroblastoma cells were labeled with [(35)S]-amino acids and were treated with a cleavable crosslinking reagent. The cells were lysed in detergent and the lysates were centrifuged at 100,000 x g to isolate crosslinked complexes. Following cleavage of the crosslinks and analysis by two-dimensional gel electrophoresis, it was found that the crosslinker increased the level of various proteins, and particularly actin, in this pellet fraction. However, GAP-43 was not present, suggesting that GAP-43 was not extensively crosslinked to proteins of the cytoskeleton and membrane skeleton and did not sediment with them. GAP-43 also did not sediment with the membrane skeleton following nonionic detergent lysis. Calmodulin, but not actin or other proposed interaction partners, co-immunoprecipitated with GAP-43 from the 100,000 x g supernatant following crosslinker addition to cells or cell lysates. Faint spots at 34 kDa and 60 kDa were also present. Additional GAP-43 was recovered from GAP-43 immunoprecipitation supernatants with anti-calmodulin but not with anti-actin. The results suggest that GAP-43 is not present in complexes with actin or other membrane skeletal or cytoskeletal proteins in these cells, but it is nevertheless possible that a small fraction of the total GAP-43 may interact with other proteins.

Keywords: CaM, calmodulin; DMSO, dimethyl sulfoxide; DSP, dithiobis (succinimidyl propionate); DTT, dithiothreitol; GAP-43, growth-associated protein of 43 kDa; Neuromodulin; SDS, sodium dodecyl sulfate; cytoskeleton; filopodia; lipid rafts; palmitoylation.

Figure 1.

DSP treatment of N1E-115 cells yields crosslinked complexes that will not enter a one-dimensional gel. Molecular weight standards are shown at the right of each figure. Exposure time = 1 h.

Figure 2.

Calmodulin, but not actin, co-immunoprecipitates with GAP-43 following DSP treatment of cells. (a, b): DMSO. (c, d): DSP. Anti-β-galactosidase (a, c). Monoclonal anti-GAP-43 (b, d). (e) The immunoprecipitation supernatant from (d) was treated with monoclonal anti-calmodulin. Exposure time = 2 days (a–d) and 4 days (e).

Figure 3.

GAP-43 is not present in the 100,000 × g pellet fractions. (a) DMSO-treated cells. (b) DSP-treated cells. Exposure time = 4 h.

Figure 4.

Polyclonal anti-GAP-43 yields co-immunoprecipitated calmodulin but not actin. (a) Cells treated with DMSO. (b) Cells treated with DSP. Exposure time = 6 days.

Figure 5.

Anti-actin does not yield co-immunoprecipitated GAP-43 following DSP treatment of cells. (a) normal rabbit serum. (b) rabbit anti-actin. Exposure time = 2 days.

Figure 6.

GAP-43 does not sediment with the membrane skeleton. Cells were treated with DMSO (a, c) or DSP (b, d). Anti-GAP-43 immunoprecipitates (a, b) and membrane skeleton pellets (c, d). Exposure times = 2 days (a, b) and 4 h (c, d).

Figure 7.

Addition of DSP to cell lysates. (a, c): DMSO. (b, d): DSP. Anti-GAP-43 immunoprecipitates (a, b) and pellet fractions (c, d). Exposure times = 2 days (a, b) and 4 h (c, d).