Mar 21, 2017

Decoupling time and location from single-cell RNA sequencing analysis

Gcmdern 10x

A key challenge across all single-cell RNA-sequencing (scRNA-seq) techniques is the preservation of each cell’s transcriptional profile throughout the entire sample handling process. A distinct advantage to the 10x Genomics Chromium™ Single Cell 3’ Solution is that hundreds to tens of thousands of cells are processed in under 7 minutes, with cell lysis beginning immediately after encapsulation into a droplet environment. However, like all scRNA-seq approaches, fresh starting material in the form of a single cell suspension is still required, potentially placing limitations on the experimental design. Development of cryopreservation storage and successful resuscitation methods across a diverse number of cell types is therefore essential for disconnecting time and location of sampling from subsequent single-cell sequencing experiments. To that end, two recent publications have developed similar cryopreservation techniques that can be readily implemented into scRNA-seq workflows.

Most recently, in "Single-cell transcriptome conservation in cryopreserved cells and tissues," a publication that appeared in Genome Biology, researchers assessed the impact of sample cryopreservation on scRNA-seq data. They reported using the cryoprotectant dimethyl sulfoxide (DMSO) formulated into common cell media preparations as a freezing solution. Cells from human (HEK293, K652), mouse (NIH3T3) and canine (MDCK) cell lines along with primary tissues (human peripheral blood mononuclear cells and mouse colon cells) were cryopreserved in this freezing solution by gradually cooling (-1 °C/min) to -80 °C followed by storage at either -80 °C or in liquid nitrogen. The authors observed that the freezing process resulted in an elevated proportion of damaged cells. Nevertheless, they reported that across all sample types the 670 fresh and 816 cryopreserved cells sampled produced libraries of comparable complexity. A linear relationship between the number of sequencing reads and unique transcript counts or the number of detected genes, was observed between fresh and frozen cells, suggesting equal transcriptome capture efficiencies for both conditions.

Our Nature Communications manuscript with collaborator Jason Beilas, "Massively parallel digital transcriptional profiling of single cells," which demonstrates the massively parallel digital transcriptional profiling of single cells, also describes a cryopreservation technique for human peripheral blood mononuclear cells (PBMC’s). To examine the impact of freezing, scRNA-seq data from 90,431 fresh and 14,046 frozen PBMCs was compared. The two data sets (fresh and frozen) showed a high similarity between their average gene expression. In addition, the number of genes and unique molecular identifier (molecule) counts detected was very similar, suggesting that the conversion efficiency of the system is not compromised when profiling frozen cells. Furthermore, subpopulations were detected from frozen PBMCs at a similar proportion to that of fresh PBMCs.

Subsequently, 10x Genomics further refined the freezing procedure and has released two demonstrated protocols for the cryopreservation of cell lines (HEK293 and NIH3T3) and Human peripheral blood mononuclear cells. The cryoprotectant DMSO is formulated into common cell media preparations as a freezing solution. Cells are then washed into this solution and gradually cooled (-1 °C/min) to -80 °C followed by storage in liquid nitrogen. There are also detailed instructions on how to correctly thaw, wash and quantify cells once they are ready for processing using the 10x Genomics Chromium™ Single Cell 3’ Solution. Example datasets generated using the cryopreservation procedures are also available on the 10x Genomics website.

Useful Links:

  • Guillaument-Adkins et al. Genome Biology "Single Cell Transcriptome Conservation in Cryopreserved Cells and Tissues" (2017) 18:45
  • Zheng et al. Nature Communications "Massively parallel digital transcriptional profiling of single cells" (2016)
  • Demonstrated Protocols for Single Cell Sample Prep
  • Single Cell Datasets