Observing proteins with precision in living cells has been a challenge for researchers across various disciplines. Now, Stefan Kubicek, PhD, and his team at CEMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, have created a new approach called visual proteomics cells (vpCells) that can fluorescently label proteins without impeding their endogenous function or regulatory mechanisms. Their study, “Pooled multicolor tagging for visualizing subcellular protein dynamics,” was published in Nature Cell Biology. Kubicek emphasized the method’s potential for future applications in both fundamental research and drug discovery. “These results provide a first glimpse into the versatility of the vpCells method,” said Kubicek. “We expect many more future applications, from fundamental cell biology to applied drug discovery.”
Proteins function dynamically in cells, and tracking their location precisely in living cells has been a major technical challenge. Current methods involve individually labeling individual proteins with fluorescent markers. Labeling and observing multiple proteins in a cell typically requires cell death and fixation, resulting in a static view of protein location and function.
The vpCells approach enables researchers to fluorescently label five different proteins in living cells. Simultaneous labeling of multiple proteins using unique fluorescent colors allows for automated high-throughput screening using AI. This method also preserves the native function and regulatory mechanisms of labeled proteins, offering a significant advancement over previous techniques.
Key features of vpCells include the ability to genetically attach fluorescent proteins to proteins of interest using the CRISPR/Cas9 gene-editing tool, enabling the systematic exploration of all human proteins. Researchers developed a genome library to enable universal exploration of all human proteins using vpCells. Because this method utilizes five complementary fluorescent colors, both researchers and AI can distinguish between different proteins and cellular structures. Of the five color markers, two are used to track different proteins of interest, one identifies the nucleus, another the cell membrane, and the fifth is used to distinguish between cell clones.
The effectiveness of vpCells was demonstrated through two applications in the present study. First, over 4,500 cell lines were generated as reporters for more than 1,100 proteins to both train AI model development and to establish normal protein localization. Next, living reporter cells were used to screen the effects of over 1,000 small-molecule substances on 61 proteins relevant to cancer cells, aimed to uncover potential therapeutic targets. The team found that of those small-molecule substances, 44 changed protein function within six hours. Moreover, one of those proteins had a similar effect to an approved drug used to treat multiple myeloma, inhibiting protein transport from the nucleus.
