O28 Label-free single-cell proteomics identifies immune cell populations in CYLD cutaneous syndrome tumours

Abstract Introduction and aims Label-free mass spectrometry has recently advanced such that a thousand proteins can be identified from a single cell. We aimed to utilize these technological advances to resolve immune cells populations in the cutaneous tumour microenvironment. We studied CYLD cutaneous syndrome skin tumours (CCS), where corroborative single-cell datasets exist, to elucidate biological processes and cell populations otherwise masked in bulk proteomic analysis. Methods CCS tumour tissue was mechanically and enzymatically digested to produce a single-cell suspension, labelled with an anti-CD45 antibody, and sorted using fluorescence-activated cell sorting for our downstream proteomic workflow. Tryptic peptides were loaded on Evosep tips prior to liquid chromatography-mass spectrometry analysis. Our workflow coupled the Evosep liquid chromatography system running a 40 samples per day method with a Bruker timsToF-HT allowing data-independent acquisition with parallel accumulation and serial fragmentation (DIA-PASEF). Data files were searched against a neural network derived spectral library with DIA-NN 1.9 to reveal protein group intensities. We used gene ontology and human protein atlas databases to broaden protein group classifications hence minimizing missing value bias during cell clustering. Results We developed a streamlined single-cell proteomics workflow for dermatological research. Single-cell proteomic analysis of CCS tumour tissue identified 7515 peptides from 1434 proteins (n = 48 cells). We identified a mean of 718 proteins per cell with a 19% coefficient of variation in a proof-of-concept dataset. Uniform manifold approximation and projection clustering distinguished cell populations with enriched gene ontology terms including skin development and signalling by Rho GTPases. Furthermore, 89 differentially expressed proteins were identified between CD45−/+ groups, where we observed increased pattern recognition receptor abundance and immune system process ontologies. Conclusions We have leveraged the capabilities of ultrasensitive mass spectrometry to study single-cell proteomics in the cutaneous tumour microenvironment. This proof-of-concept study can distinguish between cell populations in CCS and may be applied to investigate common skin tumours.