The role of PI3K-p110α in breast cancer

Jean J. Zhao,Zhenning Liu,Alexander Miron,Xin Lu,Li Wang,J. Dirk Iglehart,Thomas M. Roberts
IF: 11.2
2005-01-01
Cancer Research
Abstract:Proc Amer Assoc Cancer Res, Volume 46, 2005 2589 The PI3K signaling pathway has been known to be up-regulated in many tumors via activation of receptor tyrosine kinases (RTKs) or inactivation of PTEN. It was only realized recently that the PIK3CA gene encoding the catalytic subunit p110α, a member of the PI3K family, is mutated at high frequency in human cancers. Mutations in PIK3CA were found in 40% of breast tumors, making this the most commonly mutated gene in breast cancer. Mutations in PIK3CA cluster into two major “hot spots” located in the helical and kinase domains. To evaluate the oncogenic potential of these mutant forms of p110α and investigate their mechanism(s) of action, we have developed a transformation system utilizing telomerase immortalized primary human mammary epithelial cells (HMECs). The study of kinases in HMECs demanded creation of a system devoid of interfering viral oncogenes. In the absence of viral oncogenes HMECs transformation is dependent on the activation of PI3K for growth in soft agar and as xenografts in mice. Interestingly, the kinase activities and transforming potencies of the two “hot spot” mutant alleles of p110α (E545K and H1047R) derived from human cancers are much higher than those of constitutively activated p110α by myristoylation (Myr-p110α). Unlike activating EGFR mutations, the activation of both E545K and H1047R is independent of growth factor stimulation. Thus HMEC cells transformed with activated PI3K-p110α are relatively insensitive to RTK inhibitors, including specific IGF-1R and EGFR inhibitors. Moreover, tumors derived from transformed HMEC cells expressing constitutively activated PI3K-p110α are highly vascularized particularly in the peripheral zone. We have also constructed a corresponding helical “hot spot” mutant allele in p110α that was roughly 10 fold less effective in activating Akt but still showed residual transforming ability. We have been able to analyze transcriptional profiles of our PI3K transformed HMEC to identify unique signatures of PIK3CA mutants in human breast tumor samples. Finally we have constructed a conditional knockout of the PIK3CA gene in mice to study the role of PI3K-p110α in breast development and breast cancer formation in vivo. Taken together, using our HMEC derived functional transformation system, we have demonstrated the oncogenic potential of tumor derived mutant alleles of PIK3CA in breast cancer formation and established a valuable model for preclinical studies of new specific inhibitors.
What problem does this paper attempt to address?