The importance of distinguishing cell-cell complexes from singlets in flow cytometry and single-cell sorting: a case for an image-based cell sorter

Abstract

Abstract Our recent work highlights that care needs to be taken when interpreting single-cell data originating from flow cytometry acquisition or cell sorting: We found that doublets involving T cells bound to another immune cell are indeed very often present in the live singlet gate of human samples acquired by flow cytometry, in PBMC as well as in tissues. This hidden ‘contamination’ of cell-cell complexes generates atypical gene signatures of mixed cell lineage in what is assumed to be single cells, which can lead to data misinterpretation, such as the description of novel immune cell types with dual lineage properties. Based on the example of T cell-monocyte complexes, we investigated experimental and data analysis strategies to help distinguishing between singlets and cell-cell complexes in non-imaging flow cytometry and single-cell sorting. We found robust molecular signatures in both T cell-monocyte and T cell-B cell complexes that can distinguish them from singlets at both protein and mRNA level. Imaging flow cytometry with appropriate gating strategy (matching the one used in cell sorting) and direct microscopy imaging after cell sorting were the two preferred methods to validate the presence of cell-cell complexes in suspicious dual-expressing cells. In parallel, we also investigated the molecular profile of T cells and monocytes present in a doublet and found that their association is not random, but enriched for LFA1/ICAM1 interactions and gene expression profiles associated with disease-specific immune responses. Taken together, these data highlight the importance of systematically investigating for the presence of cell-cell complexes in single-cell data derived from non-imaging flow cytometry.