Leonardo Rosa Valdo^1,2, Giovanna Lopes de Araújo^1,2, Guilherme Reis-de-Oliveira², Fabio Cesar Gozzo^1,2

Protein–protein interactions (PPIs) have been widely studied to better understand how cell signaling works and how this knowledge can be applied to disease research and the development of targeted therapies. A recent study mapped the PIK3CA interactome in breast cancer cells and identified novel interacting proteins that modulate PI3K-AKT signaling, providing insights into how alterations in these interactions may contribute to tumorigenesis. Mass spectrometry-based approaches were applied in this study to investigate these interactions, including affinity purification coupled with mass spectrometry (AP-MS), which analyzes interactions on a one-by-one basis. However, this technique has limitations due to its labor-intensive workflow requiring the expression and purification of the proteins of interest and the potential for false positives, as it may detect interactions that would not naturally occur under native cellular conditions.

To overcome these challenges, cross-linking mass spectrometry (XL-MS) offers a more robust approach, providing higher confidence in detected interactions and enabling the study of protein complexes in their native cellular environment, as well as providing structural information about the complex. In this approach, a cross-linking reaction is performed inside the cell by fixing complexes with formaldehyde, followed by membranes permeabilization with detergents, allowing the reagent to access intracellular proteins while preserving the cytoskeletal structure. While this technique has already been applied to adherent cells, its application to non-adherent cells has proven challenging, since permeabilization often results in complete cell lysis, preventing an effective intracellular cross-linking and leads to loss of spatial context. To get around this problem, we applied the protocol in neuroblastoma cells SH-SY5Y and used milder surfactants like tween-20 during the permeabilization step to improve the preservation of cellular structures, enabling successful cross-linking within intact cells. 

Furthermore, XL-MS data analysis relies heavily on specialized software tools to identify cross-linked peptides. In this context, our group developed SCOUT, a fast and optimized tool for cross-link identification, and FLASHLIGHT, a companion tool that integrates XL-MS results with interaction networks from STRING and UniProt databases to visualize the interactome, associated biological functions, and subcellular localization of the interactions. To assess the quality of the results, SCOUT was compared with other widely used tools, such as MeroX and pLink.

Agradecimentos: IQ – UNICAMP, Centro de Pesquisa Boldrini, FAPESP, CNPq e CAPES.