NEW YORK – Researchers from Genentech and the Technical University of Munich have published a study exploring the optimal settings for single-cell proteomic experiments.
Detailed in a paper published this week in Nature Methods, the study examined the conditions under which carrier proteome-based single-cell mass spec workflows offered the best quantitative performance and developed a software tool to help users optimize experimental conditions and evaluate the quality of single-cell proteomic data.
The researchers looked at the Single Cell Proteomics by Mass Spectrometry (SCOPE-MS) workflow developed several years ago by scientists at Northeastern and Harvard Universities. Variations of the approach have since been adopted by other researchers pursuing single-cell analysis.
The method uses isobaric labeling of peptides from both the single-cell sample of interest and another larger sample source (such as dozens or hundreds of the equivalent cell). By including the second, larger sample, the researchers are able to ensure that even analytes present only at low abundance in the single-cell samples are present in relatively high abundance in the overall sample, making them more likely to be fragmented and detected by the mass spec.
Since the Northeastern and Harvard University team published the approach in 2017, other teams, most prominently one at Pacific Northwest National Laboratory, have developed versions of the method. In a study published last year, PNNL researchers combined a sample prep workflow optimized for single-cell analysis with a version of the SCOPE-MS approach to identify up to 2,300 proteins in single cells with a throughput of around 100 cells per day.
But while the SCOPE-MS carrier proteome approach allows for identification of very low abundance proteins, it presents challenges for protein quantitation, said Christopher Rose, a senior scientist at Genentech and senior author on the Nature Methods paper. This is because the larger amount of carrier sample that is used, the less time the mass spec is able to use for measuring ions from the single-cell sample. And the fewer ions it measures from the single-cell sample, the less accurate and reproducible the protein quantitation.
"As you add more carrier proteome, you are going to be able to identify more peptides because there is more signal for the mass spectrometer," said Rose. "But while your identifications will increase, the quality of your [quantitative] data will decrease."
The question of how high researchers can push the ratio of carrier sample to single-cell sample in SCOPE-MS-style experiments has been a point of inquiry for users of the approach since it was introduced, Rose said. In the Nature Methods paper, he and his colleagues set out to systematically address the question.
The researchers used a mixture of 13 synthetic peptides along with a carrier proteome combined at levels of 5x, 50x, 250x, and 500x, finding that quantitative accuracy decreased as the amount of carrier proteome increased. Quantitative accuracy could be maintained at higher carrier proteome levels but only by samples with more ions; the authors noted that experiments using 300x of greater levels of carrier sample require sampling of significantly more ions than are typically measured in a single-cell mass spec experiment.
Rose said the researchers determined that a threshold of 20x carrier-to-single-cell sample would ensure high-quality data in most experiments. He added that users "could probably push it to 100x," but that to obtain good quantitative data at that level it would be necessary to pay close attention to instrument parameters and data analysis. He said the PNNL single-cell work demonstrated how the method's quantitative accuracy could be maintained at this higher threshold.
Rose and his coauthors noted that their work likely understated the variability of actual single-cell mass spec experiments as they used for their work with bulk proteome digests diluted to single-cell levels, which let them avoid the challenges inherent in preparing single-cell samples for mass spec analysis.
In a recent paper in the Journal of Proteome Research, Northeastern University researchers including Nikolai Slavov, one of the developers of the SCOPE-MS method, similarly looked at how carrier proteome size affected quantitative accuracy. That study arrived at a more generous threshold, determining that carrier samples up to around two hundredfold larger than the target sample could be used "without adverse effects on quantification."
Rose noted that his team's findings are applicable not only to single-cell experiments but to any mass spec experiments using a carrier sample-based workflow. Researchers have employed the technique for a variety of purposes outside single-cell proteomics — studies of protein phosphorylation in small samples, for instance. The approach was originally devised by Proteome Sciences, which owns the tandem mass tag isobaric labeling reagents typically used in such experiments.
Regarding single-cell proteomics generally, Rose said that his team at Genentech was still focused primarily on better working out the technical aspects of the approach, but he added that he saw a number of potential applications.
"I think where it could have the biggest benefit is in identifying novel cell markers … potentially ones that you might want to follow up and develop, say, an antibody for a more defined, targeted workflow," he said.
He said that as the technology improved it could become feasible to look at things like drug response and development of resistance on a cell-by-cell basis.
"Where we hope to see it is if we get down to see, say, 3,000, 5,000 proteins quantified, then we might start to actually identify some of those proteins that could be useful for looking at pathways and how they are potentially responding to therapeutics," he said. "I think at the moment we view single-cell proteomics as an evolving field."