Single-cell studies of the genome and transcriptome are all the rage as sequencing technologies are reaching maturity. No two cells are exactly alike; the only way to understand this biological heterogeneity, or to capture information from rare cells, is to profile them individually.

Proteomes of single cells may soon be routinely profiled. Credit: Marina Corral Spence/Springer Nature

Proteins represent the main functional machinery of cells, so how the expressed proteome differs from cell to cell is a question of high interest. However, capturing proteomic information from individual cells has proven to be a substantial technical challenge. Mass spectrometry—the workhorse technique of proteomics—enables near-complete proteomes to be detected and quantified, but such experiments are typically carried out with tens of thousands of cells or more. In a typical single mammalian cell, reliable analysis by mass spectrometry has been possible for only the most abundant proteins.

Newer approaches may soon bring change. Researchers in the mass spectrometry field are devising advanced sample preparation and separation approaches to reduce sample losses during processing, and therefore facilitate deeper, quantitative analysis of single-cell proteomes. Nanopore sensors, now an established technology for DNA and RNA sequencing on the single-cell level, are also being explored for protein sequencing—a bigger challenge, given the much greater complexity of amino acid chemistries in comparison to nucleic acids.

Targeted, antibody-based approaches to protein profiling in single cells are also being increasingly applied on a broader scale. Mass cytometry, which relies on antibodies conjugated with rare-earth mass tags, now allows profiling of upward of 100 different protein targets in single cells in high throughput. In the Cell Atlas project, antibody-based immunofluorescence confocal microscopy was used to map the subcellular distributions of more than 12,000 human proteins across multiple cell lines (Science 356, eaal3321; 2017).

Even an old method for protein sequencing—Edman degradation—is poised to make a comeback. By fluorescently labeling selected amino acids and imaging the decrease in fluorescence during consecutive rounds of Edman degradation, researchers recently showed that it is possible to obtain sparse peptide sequences that could be identified via matching to a reference protein database (Nat. Biotechnol. 36, 1076; 2018).

We look forward to seeing intrepid methods developers turn these and other nascent ideas for single-cell proteomics into reality.