Introducing dysfunctional Protein-Protein Interactome (dfPPI) - A platform for systems-level protein-protein interaction (PPI) dysfunction investigation in disease

Abstract

Protein-protein interactions (PPIs) play a crucial role in cellular function and disease manifestation, with dysfunctions in PPI networks providing a direct link between stressors and phenotype. The dysfunctional Protein-Protein Interactome (dfPPI) platform, formerly known as epichaperomics, is a newly developed chemoproteomic method aimed at detecting dynamic changes at the systems level in PPI networks under stressor-induced cellular perturbations within disease states. This review provides an overview of dfPPIs, emphasizing the novel methodology, data analytics, and applications in disease research. dfPPI has applications in cancer research, where it identifies dysfunctions integral to maintaining malignant phenotypes and discovers strategies to enhance the efficacy of current therapies. In neurodegenerative disorders, dfPPI uncovers critical dysfunctions in cellular processes and stressor-specific vulnerabilities. Challenges, including data complexity and the potential for integration with other omics datasets are discussed. The dfPPI platform is a potent tool for dissecting disease systems biology by directly informing on dysfunctions in PPI networks and holds promise for advancing disease identification and therapeutics.

Figure 1:. A snapshot of dysfunctional Protein-Protein Interactome (dfPPI)

a Interactome capture and identification. Native Biological Baits: Unlike traditional affinity purification that uses genetically tagged proteins introduced into cells, dfPPI utilizes the naturally occurring epichaperomes as baits. This method captures the native state of protein interactions, making it particularly useful for studying disease mechanisms in authentic biological contexts. Because epichaperomes’ interactors are the proteins being rewired by stressors, the dfPPI elucidates the broader network of dysfunctional PPIs associated with disease states. Chemical Probes: The use of chemical probes to trap and isolate epichaperome-interactor assemblies is a critical component of dfPPI as it enables efficient identification of both direct and indirect interactors, including those of low abundance and transient. Trapping direct and indirect interactors (e.g., as part of protein assemblies) is an important feature of dfPPI as it may enable an accurate determination of the context-specific function of a specific protein despite a potentially ‘noisy’ dataset (i.e. with many contaminants). Interactome Identification: Captured interactors in dfPPI are identified using mass spectrometry with label-free strategies, such as spectral counting and ion intensity-based quantification. These methods accurately identify proteins interacting with epichaperomes, enabling assessment of protein interactions and network dynamics in disease states. b A bioinformatics pipeline has been created to analyze dfPPI datasets. Its basic functions are to identify proteins with altered connectivity in specific biological contexts, construct context-specific PPI network maps, functionally map these changes, and derive biological insights from these dysfunctions. Disease-Specific PPI Networks: By focusing on the rewiring of PPIs under disease-specific stressors, dfPPI helps to map out how these stressors alter proteome connectivity. This is crucial for understanding the pathways and mechanisms through which diseases impact cellular functions. Functional Insights: dfPPI enables accurate pathway predictions by mapping the functional roles of proteins across varying cellular contexts. The ability to analyze context-dependent alterations in PPIs provides insights not just into the protein interactions themselves but also into their functional consequences. This can lead to a better understanding of how diseases manifest at the molecular level and how they might be treated by targeting these altered networks.

Figure 2.. Application of dfPPI to MDA-MB-468 triple negative breast cancer cells

a The YK5-B probe (and a control probe) were utilized to explore protein-protein interactions (PPI) underlying the malignant phenotype in the MDA-MB-468 cell line. Comparison of dfPPI identified interactors with protein expression levels, as detected by a proteomic-based approach, revealed that the incorporation of proteins into PPIs was independent of their overall expression within the specific cellular context of MDA-MB-468 cells. Reactome pathway enrichment analysis of the 2,481 Grade A epichaperome interactors mapped these proteins to biological processes known to be altered in the context-specific of MDA-MB-468 cancer cells. b Focusing on mitotic processes, particularly those crucial from the G₂/M transition through anaphase, the analysis identified proteins and mitotic PPI networks essential for the proper division of MDA-MB-468 cells, providing insight into the highly proliferative nature of these cancer cells. This study thus highlights a mechanism of altering mitosis by rewiring the assembly of mitotic regulator proteins in such a way that the fitness of mitotic processes is increased. Adapted from Rodina et al. Nature Communications 2023 [ref. 11].

Figure 3.. Application of dfPPI to Alzheimer’s disease.

The PU-beads probe was utilized to explore protein-protein interactions (PPI) underlying pathologic changes between normal aging and Alzheimer’s disease (AD) in brain tissue homogenates obtained from post-mortem human brain specimens. PPI mapping revealed significant changes, including both loss and gain of PPIs, with 942 proteins losing connection partners and 1,191 proteins forming new interactions. Reactome mapping demonstrated the functional impact of these PPI alterations, highlighting affected pathways. This study thus underscores the proteins, their interactions, and the impacted processes contributing to the altered brain networks observed in AD, including dysfunctions in synaptic plasticity, metabolic processes, and inflammatory phenotypes. Adapted from Inda et al., Nature Communications, 2020 [ref. 15].

Figure 4.. Applications of dfPPI to the development of treatment paradigms.

a PU-beads were applied to MiaPaca2 pancreatic cancer cells to investigate protein-protein interactions (PPIs) involved in evading drug mechanisms. PPI maps were generated at baseline and after drug addition, alongside constructing epichaperome-interactome functional connectivity maps. MDA-MB-468 cells were used as a positive control for hyperconnected PPI networks, where dfPPI identified 3,046 PPIs initiated by epichaperomes with protein pathways through one or more interactors within those pathways. In MiaPaCa2 cells, dfPPI identified 930 baseline PPIs (no drug added), with 1,502 PPIs deployed upon drug addition. The combined effect of inhibiting baseline PPIs and deploying new bypass PPIs created a temporary state of network hyperconnectivity and maximal PPI network capacity, approaching that observed in MDA-MB-468 cells at baseline. b Analysis of the deployed PPIs led to the hypothesis that cellular rebound occurs through the reactivation of pre-existing protein pathways, constitutively active at baseline in MiaPaCa2 cells but utilized during recovery through alternative paths. c Applying these dfPPI-derived hypotheses, the study demonstrated that pancreatic tumors engineered into a hyperconnected state became highly vulnerable to trametinib, a MEK inhibitor targeting the MAPK pathway known to be dysfunctional in this context. Conversely, trametinib alone, without priming cells into the hyperconnectivity state, was ineffective in these tumors due to compensatory increases in alternate signaling activity, including the AKT/mTOR pathway. This study underscores how dfPPI facilitates real-time comprehension of the intricate PPIs within tumors. It led to the discovery of a treatment approach that induces tumors into a state of PPI network hyperconnectivity and maximal capacity, suppressing rebound pathways. This strategy primes PPI networks to heighten vulnerability, and enhances treatment efficacy, thereby converting traditionally ineffective therapeutics into potent agents. Adapted from Joshi et al. Communications Biology 2021 [ref. 16].