Mapping cancer signaling networks by an integrated multiplexed tissue imaging platform

Abstract

Aberrant cellular signaling networks are implicated in major diseases including cancer, but are difficult to reliably quantitate, as many signaling proteins are expressed at low abundance and further reduced following specimen collection. MTIP is an integrated tissue-imaging platform that leverages bright, fluorescent reporters and sensitive spectral instrumentation, along with automated staining, image acquisition and non-parametric image analysis, to attain reproducible, multiplexed quantitation of signaling proteins in tissue. MTIP captured the phosphoactivity of six key PI3K/MAPK proteins (pAKTS473, pAKT308, pPRAS40, pS6, peIF4G and pERK1/2) at high precision (coefficient of variation, CV <10%), four-log dynamic range and subcellular resolution. We demonstrated the MTIP platform's capability to capture a diversity of PI3K/MAPK networks present in breast tumors. These protein networks are heterogeneously distributed across the tumor tissue and associated with subgroups of cells and underscore the importance of accessing information about signaling networks in spatially intact tissue. Analysis of PI3K/MAPK networks by hierarchical clustering showed that PI3K/MAPK networks do not strictly correlate with PI3K pathway mutations, also pointing to the value of functional signaling network profiling together with genomic information. MTIP's reliable quantitative capability can be applied to guide therapeutic development and selection in precision medicine.