Phosphorylation of Thr{sup 654} but not Thr{sup 669} within the juxtamembrane domain of the EGF receptor inhibits calmodulin binding

Calcium-calmodulin (CaM) binding to the epidermal growth factor receptor (EGFR) has been shown to both inhibit and stimulate receptor activity. CaM binds to the intracellular juxtamembrane (JM) domain (Met{sup 645}-Phe{sup 688}) of EGFR. Protein kinase C (PKC) mediated phosphorylation of Thr{sup 654} occurs within this domain. CaM binding to the JM domain inhibits PKC phosphorylation and conversely PKC mediated phosphorylation of Thr{sup 654} or Glu substitution of Thr{sup 654} inhibits CaM binding. A second threonine residue (Thr{sup 669}) within the JM domain is phosphorylated by the mitogen-activated protein kinase (MAPK). Previous results have shown that CaM interferes with EGFR-induced MAPK activation. If and how phosphorylation of Thr{sup 669} affects CaM-EGFR interaction is however not known.In the present study we have used surface plasmon resonance (BIAcore) to study the influence of Thr{sup 669} phosphorylation on real time interactions between the intracellular juxtamembrane (JM) domain of EGFR and CaM. The EGFR-JM was expressed as GST fusion proteins in Escherichia coli and phosphorylation was mimicked by generating Glu substitutions of either Thr{sup 654} or Thr{sup 669}. Purified proteins were coupled to immobilized anti-GST antibodies at the sensor surface and increasing concentration of CaM was applied. When mutating Thr{sup 654} to Glu{sup 654} no specific CaM binding could be detected. However, neither single substitutions of Thr{sup 669} (Gly{sup 669} or Glu{sup 669}) nor double mutants Gly{sup 654}/Gly{sup 669} or Gly{sup 654}/Glu{sup 669} influenced the binding of CaM to the EGFR-JM. This clearly shows that PKC may regulate EGF-mediated CaM signalling through phosphorylation of Thr{sup 654} whereas phosphorylation of Thr{sup 669} seems to play a CaM independent regulatory role. The role of both residues in the EGFR-calmodulin interaction was also studied in silico. Our modelling work supports a scenario where Thr{sup 654} from the JM domain interacts with Glu{sup 12} in the calmodulin molecule. Phosphorylation of Thr{sup 654} or Glu{sup 654} substitution creates a repulsive electrostatic force that would diminish CaM binding to the JM domain. These results are in line with the Biacore experiments showing a weak binding of the CaM to the JM domain with Thr{sup 654} mutated to Glu. Furthermore, these results provide a hypothesis to how CaM binding to EGFR might both positively and negatively interfere with EGFR-activity.