Conformations of the central transforming region (Ile 55‐Met 67) of the p21 protein and their relationship to activation of the protein

Abstract

The GTP-binding p21 protein, encoded by the ras-oncogene, becomes transforming if amino acid substitutions are made at critical positions in the polypeptide chain, e.g., at Gly 12, Gly 13, Ala 59, Gln 61 and Glu 63. Most of these substitutions occur in two phosphate-binding loop regions, Tyr 4-Thr 20, herein designated as segment 1, and Ile 55-Met 67, herein designated as segment 2. These two segments are homologous to two corresponding regions in the two purine nucleotide binding proteins, bacterial elongation factor (EF-tu) (Val 12-Thr 28 corresponds to segment 1: his 78-Ile 92 corresponds to segment 2) and adenylate kinase (ADK) (Lys 9-Cys 25 corresponds to segment 1 and Tyr 95-Arg 107 corresponds to segment 2). We find that the conformations of the segment 1 region in the p21 protein, EF-tu and ADK are similar to one another and that the conformation of the segment 2 region of EF-tu is superimposable on that of segment 2 of ADK. Furthermore, the relative position of the two segments in EF-tu is strikingly similar to that of the two segments in ADK. In the originally proposed X-ray structure for the p21 protein, the conformation of segment 2 in the p21 protein is not similar to that found for the other two proteins, and its disposition relative to segment 1 and the remainder of the protein is also different from that observed for the other two proteins. We previously computed the low-energy conformations for segment 2 in the p21 protein using conformational energy calcuations based on ECEPP (Empirical Conformational Energy Program for Peptides) with Gln, Pro, Leu, Lys, and Arg at positions 61. We also predicted that the position of segment 2 in the p21 protein adopts a position and conformation identical to the ones observed in EF-tu and ADK in the activated protein. We now find that the computed global minimum lowest-energy structure is superimposable on segment 2 of EF-tu and on the 95-107 sequence of DK and that our predicted position for this sgment in the activated protein coincides with that found in a more recent X-ray crystallographic structure for the p21 protein.