Phosphoinositide 3-kinase-related overgrowth: cellular phenotype and future therapeutic options

Somatic activating mutations in PIK3CA, which encodes the p110α catalytic subunit of phosphoinositide-3-kinase (PI3K) are frequently found in cancers and have been identified in a spectrum of mosaic overgrowth disorders ranging from isolated digit enlargement to more extensive overgrowth of the body, brain, or vasculature. We aimed to study affected dermal fibroblasts with a view to inform therapeutic studies, and to observe cancer-associated mutations in isolation.

We measured PIP3 concentrations in dermal fibroblasts with endogenous PIK3CA mutations and in wild type fibroblasts using mass spectrometry, and we measured downstream signalling events with ELISA and immunoblotting. Cellular proliferation was evaluated with 5-bromo-2'-deoxyuridine incorporation, and cell size assessed by fluorescence-activated cell sorting (FACS). Glycolysis and mitochondrial tests were performed with an extracellular flux analyser (Seahorse Bioscience, Billerica, MA, USA), and mitochondrial potential was measured by FACS-based JC1 staining. Experiments were repeated after exposure to 5 nmol everolimus for 72 h.

Mutant fibroblasts had two times higher basal PIP3 concentrations than wild-type fibroblasts (p=0·0017), with concomitant AKT and p70S6 activation downstream. The rate of cellular proliferation was higher in mutant cells under low serum conditions, but median cell size was not statistically different. Glycolytic capacity was similar between mutant and wild type fibroblasts, but subtle differences in mitochondrial function were detected with blunted responses to uncoupling agents and reduced membrane potentials. Treatment with everolimus reversed aberrant AKT(ser473) and p70S6 signalling, slowed cellular proliferation, and reversed mitochondrial abnormalities, but was associated, paradoxically, with increases in PIP3 concentrations.

These experiments demonstrate activation of the PI3K-AKT pathway in affected fibroblasts with increased proliferation, but no hypertrophy. Moreover, we identified changes in mitochondrial function in keeping with the known propensity of AKT to modulate elements of the Warburg effect. These results suggest that inhibitors of the mammalian target of rapamycin (mTOR) might be beneficial, but these inhibitors will require formal evaluation in clinical trials. More targeted therapy with p110α inhibitors is an enticing future option.

Wellcome Trust, Sackler Fund, National Instititute for Health Research.