Figure 4. Monitoring the interaction between CaM and Cx50p^141–166 by (¹H,¹⁵N)-HSQC spectroscopy.

(A) An overlay of HSQC spectra of holo-CaM (black) with the spectrum of the holo-CaM–Cx50p^141–166 (grey). (B) An overlay of the HSQC spectra of holo-CaM (black) with the spectrum of the holo-CaM–SCx50p (grey). (C) Chemical-shift perturbation in CaM induced by addition of 2-fold molar excess of Cx50p^141–166. The weight-average chemical-shift change (Δδ) was calculated using eqn (5). Residues with Δδ> 0.1 p.p.m. were mapped to the three-dimensional structure of holo-CaM (PDB code 3CLN). (D) The chemical-shift change of Gly³³ during titration of holo-CaM with Cx50p^141–166. The disappearance of the peak (free form) was accompanied by the appearance of the corresponding peak (bound form) at a new position. (E) Structural basis of the difference in directionality of chemical shift change. The structure of the CaM–CaMKII complex (PDB code 1CDM) is shown with the peptide (inside) and the CaM residues (outside). Leu¹¹⁶ and Met¹⁴⁵ of CaM are within 5 Å of Lys²⁹⁸p and Lys³⁰⁰p of the peptide respectively. These peptide positions correspond to Lys¹⁴⁶p and Arg¹⁴⁸p for Cx43, and Arg¹⁴⁹p and Glu¹⁵¹p for Cx50, as can be seen in the superposition of their sequences. In Cx50, Glu¹⁵¹p may be pulled away from Met¹⁴⁵ by Lys¹⁴⁷p. In contrast, in Cx43, Arg¹⁴⁸p may be pushed towards Met¹⁴⁵ by Lys¹⁴⁴p. These differences in peptide sequences can cause changes of chemical shifts in different directions.