Preprint Club
A cross-institutional Journal Club Initiative

Keywords

● Cell surface proteome

● Proximity labelling

● Membrane biology

Main Findings

The nanoscale interactions of individual proteins on the cell surface are likely very significant but are poorly understood. Although recent developments in “hypothesis driven” and “discovery-driven” techniques to evaluate protein-protein interactions are useful for studying interactions of specific proteins, there is an unmet need for a complete surface proteome to comprehensively understand the complex interactions between large numbers of proteins on the cell surface. In this preprint, Floyd, et al., not only provide this comprehensive analysis of the complete surface proteome but also demonstrate the power of this new information for learning fundamental mechanisms driving cell function. First, the authors describe their approach for systematic proximity labelling which utilizes proximity labelling, quantitative proteomics, and unbiased machine learning network analysis to infer the spatial arrangement of 390 cell surface proteins. After confirming the association of known clusters of proteins in both Jurkat cells (T cells) and Daudi cells (B cells), the authors demonstrate the power of this unbiased approach by investigating a subset of proteins that were not expected to be found on the cell surface. Many of these  “Maverick” proteins were previously thought to be limited to mitochondria, but found to be present on the surface of a large number of cell lines. Beyond identifying new proteins on the cell surface, this approach also allows a new ability to identify novel protein-protein interactions that were conserved across cell types. In one compelling example, the authors were able to identify and confirm a novel role of IL10RB as a regulator of Type I IFN. Finally, the authors demonstrate that surface protein distribution is independent of protein abundance, suggesting active mechanisms in the spatial regulation of surface proteins. Altogether, this proof of principal study demonstrates that analyzing the surface protein distributions using large-scale PL datasets opens new avenues for addressing longstanding questions in membrane biology.

Limitations

The authors did an excellent job utilizing multiple orthogonal techniques to validate their unexpected conclusions (i.e. single cell TCR sequencing, bulk TCR sequencing, flow cytometry, etc). In addition, this work required a massive effort to robustly combine PL, mass spectrometry, and machine learning for a large number of targets (65 surface antigens, 339 mass spectrometry experiments, 25,806 fold change enrichment values in Jurkat cells alone). However, opportunities remain for future work to build on their conclusions due to several limitations of the current work. Most importantly, this work focuses solely on proteins, omitting other important biomolecules including lipids and glycans. The authors, including Dr. Bertozzi, are well-positioned to fill this gap and provide another major leap forward in future work. This work is also limited by the primary use of immortalized cell lines and the focus on average values of protein expression and protein-protein interactions across cells, which effectively focuses the results on the majority cell state. Now that the ground has been laid with immortalized cell lines, we expect future work to identify important roles for the differential interactions of proteins throughout each stage of primary lymphocyte activation.

Significance/Novelty

The development of this pipeline for complete surfaceome characterization enables deep understanding of novel protein-protein interactions. In addition, unbiased mapping allows identification of unexpected relationships on the cell surface including the presence of “Maverick” surface antigens and novel protein-protein interactions. We expect this work to be of great interest to the immunology community because it provides a new lens from which to view our field, akin to the progression from flow cytometry to spatial proteomics.

Preprint rating

· Scientific quality 5/5

· Novelty 4/5

· Significance 4/5

Credit

Reviewed by Adam Grippin as part of a cross-institutional journal club between the Icahn School of Medicine at Mount Sinai, the University of Oxford, the Karolinska Institute, the University of Toronto, and the University of Texas MD Anderson Cancer Center.

The author declares no conflict of interests in relation to their involvement in the review.