There are many motivators for the optimization of single cell proteomics, including the ability to examine individual cells or cell types within mixed populations of a tissue or tumor, thus removing the diluting effect of averages in a heterogenous sample. Studying single cells using proteomics also removes the reliance on assumptions regarding the stochastic relationship between transcript and protein abundance that is inherent in other “omics” methods and has the ability to capture important post-translational modifications, such as phosphorylation, that can provide information on protein function. However, it’s primary benefits have also been its most significant hurdles, with limited amounts of starting material, and no built-in amplification strategies, the methods used for single cell proteomics must be optimized from sample capture, to peptide extraction, through LC-MS/MS and data analysis. A paper recently presented in Analytical Chemistry by Li and colleagues outlines a workflow for integrated proteome, phosphoproteome, and metabolome analysis from a single cell that addresses many of these technical challenges. The key optimizations, including surface treatments of microwells for improved extraction of metabolites and peptides, and spectral library-based data-independent acquisition (DIA) mass spectrometry powered by PEAKS software, resulted in excellent metabolite, peptide and phosphopeptide identification. Importantly, the methods show high sample-to-sample reproducibility and did not result in any bias in the physiochemical characteristics of the peptides and proteins captured in the single-cell analysis (vs. a bulk sample control). As a proof of concept, the authors tested their workflow by treating the single cells with nocodazole, an antineoplastic agent, to show that treatment could clearly separate the metabolites, proteomes and phosphoproteomes by principal component analysis (PCA), identifying this workflow as a powerful method to analyze the treatment response of single cells.
How was PEAKS used?
In projects where starting material is limited, methods to increase peptide identifications are key to robust proteome coverage. The use of data-independent-analysis (DIA)-MS, which fragments all peptides measured during MS1 instead of only select peptides, generates extensive datasets. However, it requires previous knowledge of the fragments generated from the peptides in your sample to achieve maximum identification rates. To accommodate this, custom spectral libraries can be generated with representative samples analyzed by data-dependent acquisition (DDA)-MS. In Li et al, DDA raw files were searched in PEAKS using a UniProtKB/Swiss-Prot human database and then the resulting spectra was used to create a spectral library for HeLa cells. This library was then used to analyze the data generated by DIA-MS analysis of the single cells, maximizing the peptide identifications in their workflow. Like the DDA database search workflow, the DIA workflow in PEAKS can accommodate PTM analysis that allowed the authors to asses the phosphoproteome as well.
Data exported from PEAKS was compatible with downstream data analysis in Persus Software Platform and PCA analysis with MetaboAnalyst.
Li, Yuanyuan, et al. “An Integrated Strategy for Mass Spectrometry-Based Multiomics Analysis of Single Cells.” Analytical Chemistry, no. 42, American Chemical Society (ACS), Oct. 2021, pp. 14059–67. Crossref, doi:10.1021/acs.analchem.0c05209.
Single-cell-based genomics and transcriptomics analysis have revealed substantial cellular heterogeneity among seemingly identical cells. Knowledge of the cellular heterogeneity at multiomics levels is vital for a better understanding of tumor metastasis and drug resistance, stem cell differentiation, and embryonic development. However, unlike genomics and transcriptomics studies, single-cell characterization of metabolites, proteins, and post-translational modifications at the omics level remains challenging due to the lack of amplification methods and the wide diversity of these biomolecules. Therefore, new tools that are capable of investigating these unamplifiable “omes” from the same single cells are in high demand. In this work, a microwell chip was prepared and the internal surface was modified for hydrophilic interaction liquid chromatography-based tandem extraction of metabolites and proteins and subsequent protein digestion. Next, direct electrospray ionization mass spectrometry was adopted for single-cell metabolome identification, and a data-independent acquisition-mass spectrometry approach was established for simultaneous proteome profiling and phosphoproteome analysis without phosphopeptide enrichment. This integrated strategy resulted in 132 putatively annotated compounds, more than 1200 proteins, and the first large-scale phosphorylation data set from single-cell analysis. Application of this strategy in chemical perturbation studies provides a multiomics view of cellular changes, demonstrating its capability for more comprehensive investigation of cellular heterogeneity.