Final Report Summary - PIKBETA/CANCER (Role of the PI3Kbeta isoform in signalling and cancer invasion.)
PI 3-kinases (PI3Ks) generate lipids in cell membranes which bind a variety of intracellular protein effectors, affecting their localization and/or activity. Given that global inhibition of PI3K may be deleterious to the organism, a main research effort is under way to understand the roles of the different isoforms of PI3K, and to develop compounds that selectively target specific of PI3K isoforms, for use in different disease areas.
Mammals have three classes of PI3K. Only class I PI3Ks are known to play a role in cancer. They consist of a p110 catalytic subunit (p110alpha, p110beta, p110delta or p110gamma) bound to a regulatory subunit. p110alpha and p110beta are ubiquitously expressed whereas p110delta and p110gamma are mainly restricted to leukocytes. Recent work has revealed that PI3K isoforms have surprisingly distinct roles in biology. Compared to the other class I PI3K isoforms, the organismal role and signalling context of p110beta is still unclear.
The main objective of this proposal is to decipher a role and mechanism of action of p110β in breast cancer cell invasion and metastasis, at the cellular and organismal level. We also aimed to investigate how p110beta isoform transmits signals to downstream effectors in the cell, with a focus on endosomal biology and trafficking.
Major findings of the project:
1- p110beta activity is a suppressor of breast cancer invasion in vitro: Using in vitro invasion assays (transwell chambers and 3D collagen gel high throughput invasion assays, I discovered that inhibition of p110beta by selective pharmacological inhibitors enhances murine and human breast cancer cell invasion. Inhibition of p110alpha or p110delta did not give rise to the same biological phenomena. Moreover, murine breast cancer cells pre-treated with p110-selective inhibitors exhibited increased lung colonization upon injection into the tail vein of syngeneic mice.
2- p110beta activity is a suppressor of breast cancer metastasis in vivo
In order to assess p110beta kinase activity role in cancer cell metastasis, we injected human breast cancer cell lines into immuno-suppressed NOD-SCID mice then treated the mice with either vehicle or p110beta pharmacological inhibitor for several days. Our results clearly showed that upon p110beta inhibition, more metastasis was observed both in the liver and lungs.
p110beta plays a role in rab5-dependent endosomal trafficking and receptor signalling: It is now very well established that endosomal trafficking of receptors highly impact on cancer cell biology (proliferation, migration, metastasis…). P110 beta isoform has been reported to bind active Rab5 in vitro, a major player in early endocytosis. We used high resolution spinning disk confocal microsopy to visualize rab5 positive endosomes. We observed that under p110beta inhibition, rab5 positive endosomes were abnormally enlarged suggesting a delay in endosomal trafficking. We then hypothesized that because they are not degraded, surface receptors might be activated (phosphorylated) and signal for longer. We used phospho-receptor tyrosine kinase arrays to assess phosphorylation of many receptors over time both in murine and human breast cancer cell line and both showed that p110beta inhibition enhanced Receptor phosphorylation as compared to control. These data was consistent with en enhanced phosphorylation of AKT2 observed upon p110beta inhibition.
These studies have implications for the fundamental understanding of PI3K function and the use of PI3K inhibitors in cancer. Indeed, the development of p110 inhibitors is in progress, mainly for use in thrombosis. Based on the correlation between PTEN inactivation and p110β-sensitivity, p110β inhibitors are also explored for use in oncology. While these compounds might in the first instance induce growth inhibition in some cancers, they could, based on our data, also promote cancer progression. Our work therefore also has clinical implications. Several companies are currently trialling pan-class I PI3K inhibitors in oncology. Based on our observations, it would be advisable to remove reactivity to p110β from these compounds for more effective long-term cancer therapy.