Representation of the asymmetrical dimer of the EGFR kinase domain that has been shown by Zhang et al. to be involved in the activation of EGFR. Adapted from figure 6 of the highlighted paper.

Epidermal growth factor (EGF) causes its receptor, EGFR, to dimerize and — through a previously unknown mechanism — become activated to effect downstream signalling, cell growth, proliferation and differentiation. Although EGFR is crucial for healthy cell growth, its mutant variants are linked to several forms of cancer. Zhang, Kuriyan and colleagues examined the structural mechanisms of EGFR activation and report their findings in Cell, providing insights for cancer treatments.

Following crystallography screens, these researchers studied two forms of active-conformation EGFR kinase-domain dimers. The first dimer was symmetrical, with the two kinase domains interacting in a mutual head-to-tail fashion. By contrast, the second dimer was asymmetrical, with the C-lobe of one monomer (B) nestled against the N-lobe of the other monomer (A) (see figure). Kuriyan and colleagues proposed that, in the asymmetrical dimer, the C-lobe of monomer B forces the allosteric activation of monomer A, in a manner that is similar to the activation of cyclin-dependent kinases by cyclins.

To test this idea, specific residues that are central to the symmetrical dimer interface were mutated. These mutations had no effect on the phosphorylation of residues that are indicative of EGFR activation. The fact that mutations of the dimer interface do not affect activation indicates that the symmetrical dimer is not involved in the activation process. But, when a similar mutational study was carried out on residues of the asymmetrical dimer interface, the phosphorylation of the same 'EGFR-activity-indicative' residues was almost completely abrogated, indicating that this dimer interface is crucial for EGFR activation.

To check that the abrogation of EGFR activity was not due to misfolding, two models were tested. A monomer with a mutation in its C-lobe interface region should be able to be activated, but not function as an activator. Conversely, a monomer with a mutation in its N-lobe interface region should be able to function as an activator, but be activation resistant. Co-transfectional studies with mutated EGFR showed that these predictions were correct.

Furthermore, the researchers solved the crystal structure of a form of the EGFR kinase-domain dimer that was mutated in the C-lobe face of the dimer interface. As expected, this mutant displayed an inactive conformation, confirming the important role of the dimer interface in triggering the active conformation of the kinase. This inactive structure also explains the activating effects of two sets of EGFR mutations that are commonly found in human lung cancers.

The EGFR signalling pathway is crucial in normal development, and its malfunction, as shown by cancer-causing mutations of EGFR, is disastrous. Zhang et al. describe a mechanism of allosteric activation that allows greater levels of regulation than standard activation by trans-phosphorylation. They also show that, although symmetry is thought to be beautiful, in the case of EGFR, it's asymmetry that turns things on.