The crystal structure of the p110 catalytic subunit of PI3K in complex with the p85 regulatory subunit illuminates the mechanisms by which p110 oncogenic mutations activate the protein.
Perturbation of the phosphatidylinositol 3-kinase (PI3K) pathway has long been associated with cancer development, with recent results identifying mutations that commonly occur in PIK3CA, the gene encoding the p110 catalytic subunit. Bert Vogelstein, Sandra Gabelli, Mario Amzel and colleagues have determined the structure of the complex formed between p110 and part of its regulatory subunit (p85, encoded by PIK3R1) in an attempt to further understand the nature of the common mutations in PIK3CA.
Image courtesy of Nicola McCarthy
Using a baculovirus system the authors were able to produce stable levels of p110 only when coexpressed with p85. However, this led to protein aggregation, so the authors determined whether different truncated versions of p85 would improve solubility. Expression of residues 322–600 of p85, which contain the inter-SH2 domain (iSH2) that binds p110 and the amino-terminal SH2 domain (nSH2), gave the highest protein yield. The complex was catalytically active and the resulting p110–niSH2 crystals were used to determine the structure to 3.05 Å resolution.
Mutations in PIK3CA have been found in the N-terminal adaptor binding domain (ABD), the C2 domain thought to interact with membranes, the helical domain of unknown function and the kinase domain. By examining the crystal structure the authors were able to more fully understand how these mutations disrupt the function of p110. For example, Arg 38 and Arg 88 in the ABD are often mutated to cysteine, histidine or glutamine and the new structure indicates that these mutations alter the interaction of the ABD with the kinase domain, changing the activity of the enzyme. Mutations in the C2 domain were thought to alter the interaction of p110 with the cell membrane. However, the crystal structure data indicate that mutation of Asn 345 in the C2 domain disrupts the interaction with the iSH2 domain of p85, leading to p110 activation. Moreover, although less common, a mutation in p85 that leads to truncation of the protein at residue 571 was thought to constrain the function of the p85 inhibitory domain. However, results from the structure indicate that the truncation destabilizes the coiled-coil part of iSH2, disrupting the interaction with Asn 345 in p110.
The new structure was also able to confirm some previous biochemical findings. The helical domain of p110 has two mutational hotspots at Glu 542 and Glu 545, and to a lesser degree at Gln 546, which are located on an exposed region. Biochemical analyses had indicated that these residues interact with the N-terminal SH2 domain of p85, inhibiting p110. As this part of the p85 subunit was not well-resolved in the crystal structure, the authors employed modelling techniques using previous crystal structures. Their findings indicate that mutation of these residues does prevent the inhibitory effect of the N-terminal p85 domain.
Hopefully, the insights provided by this new crystal structure will prove useful for the design of p110-specific or mutation-specific drugs.
Nicola McCarthy
References
Huang, C.-H. et al. The structure of a human p110/p85 complex elucidates the effects of oncogenic PI3K mutations. Science318, 1744–1748 (2007) | Article | PubMed |
catalytic subunit of PI3K in complex with the p85
