Figure 5.

ERK2 interacts with the PNT domain and phosphoacceptor region of PntP2^142–252. (A) Superimposed ¹⁵N-HSQC spectra of PntP2^142–252 at pH 7.5 in the absence (red) and presence (blue) of a 0.25 molar equivalent of active ERK2. (B) Addition of EKR2 leads to a significant reduction in the relative signal intensities of specific amides in the phosphoacceptor region, as well as in helices H0 and H5 of PntP2^142–252 (blue histogram bars). There is also an overall reduction in ⁵N-HSQC signal intensities to an average value of ∼0.4 (horizontal solid line) attributed to sample dilution and increased global relaxation rates owing to the formation of a high-molecular-mass complex. Missing data correspond to prolines, amides with overlapping signals, or residues not observed owing to rapid HX at the elevated sample pH of 7.5 that was required to prevent ERK2 aggregation. (C) The homology model of PntP2^142–252 generated with SwissModel³⁷ using Ets1^29–138 (2JV3.pdb) as a template. Amides showing the largest change in signal intensity upon ERK2 binding (below the horizontal dashed line in (B)) are highlighted in blue. Also shown in stick format (carbon, green; oxygen, red) are the side chains of residues identified by mutagenesis to be important for kinase docking,²⁶ as well as the three phosphoacceptors (Thr144, Thr151, and Thr154) and Phe166 at the H0/H1 bend. (D) X-ray crystallographic structure of the heterodimeric complex formed by the SAM/PNT domains of Mae (green) and Yan (cyan) (1SV0.pdb).⁷ Binding of Mae to the corresponding region of PntP2^142–252 to prevent Rolled ERK2 docking would require displacement of the dynamic helix H0. The structural figures were rendered with PyMol.³⁸