Down to a Single Cell

Biological system functionality is defined by the intricate interplay of its diverse cell types. Yet, cells are often studied in bulk, obscuring intracellular variability and single-cell contributions. This can hinder our ability to study complex diseases like cancer, in which cellular heterogeneity is a significant driver of prognosis. The scientific community needs to step away from averaged protein readouts and towards examining single-cell behavior.

The field of single-cell proteomics is still in its infancy, but is already delivering on its promise to capture transient cellular changes in dynamic systems by capitalizing on key advances in MS/MS-based approaches – particularly those affording improved sensitivity. This is key to make the most of biological samples’ limited protein content (because, although present in quantities an order of magnitude greater than mRNA, cellular protein cannot be amplified!). This is particularly tricky when working with samples which are themselves limited, such as patient biopsies.

So how do we boost our proteomic outputs in light of limited sample? There are two main ways: i) by minimizing sample loss with meticulous preparatory steps, and ii) by adopting the most appropriate instrumentation to enhance experimental accuracy.

When it comes to minimizing sample loss, proteins’ “sticky” nature is an issue; they cling to the sides of pipette tips and tubes at every stage of sample preparation, making conservation tricky. Robotic liquid handlers with non-contact pipetting can help – as can minimizing sample volumes to an absolute minimum (less than 1 μl). Single cell samples can be isolated through fluorescence-activated cell sorting or laser capture microdissection, which allows for the study of primary tissues in situ.

Experimental accuracy can be improved by including “booster” samples in the multiplexing mix. These increase the number of ions available, allowing us to enhance the lower limit of detection of MS analyses. In our lab, we use tandem mass tags (TMTs; isobaric labels for the accurate quantification of peptides and proteins in MS/MS-based analysis) for this. We dedicate a single TMT channel to tag a group of up to 200 booster cells, leaving other channels available for single cells. This boosts the protein identification rate by providing optimal ion levels for identifying and quantifying peptides that represent a significant portion of the cellular proteome.

New-generation MS systems are also equipped with real-time search features in which MS³ scans are only triggered for peptide precursors identified at the MS² level. Real-time data acquisition vastly improves proteome coverage, with up to 95 percent of peptides quantified at MS³ level. Matching MS² fragments with the correct peptide sequence provides improved accuracy. By not wasting runtime on unnecessary MS³-level scans, real-time searching also significantly improves the productivity of single-cell proteomics workflows as a whole. Field asymmetric ion mobility spectrometry (FAIMS) can also filter out background ions and provide deeper sample coverage.

“How effective are these interventions?” I hear you ask. According to a proof-of-concept study we conducted in a heterogeneous acute myeloid leukemia model system (1), very! In fact, we were able to apply TMT multiplexing with a FAIMS-optimized workflow and precision robotics to conduct an entire sample preparation workflow on more than 3,000 cells. Our in-house data processing workflow, SCeptre (single-cell proteomics readout of expression), allowed us to visualize the data and pinpoint protein expression levels within each individual cell type from the sample.

Though it may seem overwhelming to transition from the safe familiarity of bulk protein analyses into the novel and technically demanding world of single-cell proteomics, I would say the time is now. The ability to examine a biological system down to a single cell can pave new paths in research – and clinics. Laboratories can set themselves up for future successes now by venturing into simple proof-of-concept experiments or collaborative efforts that eventually develop into a tailored single-cell workflow.