Single Cell Genomics

Single-cell analysis technologies have been instrumental in facilitating significant developments in cellular biology. Rapid progress in this growing field is necessary for the continued transformation of how we learn about and approach the clinical research, diagnosis, prognosis, and treatment of a wide range of disease types. However, the more we learn about the immense heterogeneity of cell populations, the more we realize that bulk population analysis and current tools for preparing samples for single-cell analysis damage or negatively affect cells during sample preparation, and do not provide adequate representation of the true cell population within the sample.

Conventional single-cell sample preparation methods and technologies tend to damage cells, have low capture efficiency, and generally result in low cell viability – resulting in inaccurate population representation of the biological sample or insufficient cells for a complete experiment. Often, scientists discover these issues late in the experimental pipeline, which contributes to their increasing frustration and costs in irreplaceable time, resources, and funds.

The urgency for technological platforms that are gentle (to minimize cell damage) and efficient (to maximize the capture of all live-cell types in the population) is imperative to realize the scientific and clinical potential of single-cell analysis. The LeviCell™ demonstrates such a promise.

The 10X single cell sequencing platforms have helped to revolutionize our understanding of biology and how the measurement of individual cells’ gene expression can uncover the previously hidden messages commonly unseen in bulk mRNA experiments.

Unfortunately while the single cell sequencing workflow has proven to be fast, simple and reliable, obtaining the most complete and accurate single cell data depends on up-front sample preparation steps. Traditional methods currently in use tend to damage cells, have low capture efficiency, and often completely fail to process sensitive and difficult to work with samples such as dissociated tissue samples. The most common sample enrichment methods, Flow Cytometry and Bead-based enrichment methods, both often lead to significant cell stress and gene expression changes due to the high pressures used (Xiong et al., 2002; Romero-santacreu et al., 2009; Van Den Brink et al., 2017) or cell signaling effects due to the cell surface binding required by these methods (Kornbluth and Hoover, 1989; Christaki et al., 2011). Due to these shortcomings many promising single cell experiments delivered biased results or fail to delivery results entirely.

The urgency for technological platforms that are gentle (to minimize cell damage), and efficient (to maximize the capture of all live cells in the population while maintaining the original cellular representation) is imperative to realize the scientific and clinical potential of single-cell analysis.

The LeviCellTM demonstrates such a promise.

To demonstrate the benefits of the LeviCellTM in removing dead cells and debris, each sample of 400k cells was split into fractions prior to the each enrichment process. One half of each sample was washed into a 10X library preparation buffer, while the other half was enriched for live cells using the LeviCellTM. After quantitation, the output from the LeviCellTM was put directly into the Chromium Next GEM single Cell 3’ kit step without additional washing. Approximately 15k cells were used for each method. Final cDNA concentration was estimated by BioAnalyzer. While all samples succeeded at this step, there was a clear improvement in quality of the cDNA and the results library quality from the material generated by the LeviCellTM.

All samples were normalized before library preparation to account for differences yields, after normalization 15K cells were loaded from each process into the Chromium system.

All samples were sequenced on the Illumina 6000 S2 sequencer to a depth of ~500 million read per sample. Sequencing metrics, including Q30 bases in reads, barcodes, and UMI’s were nearly identical across all samples. The fraction of reads mapped overall were very similar across all samples, and only small differences were observed in the fraction of reads mapping to intergenic, intronic, and exonic regions of the genome.

The most striking differences were seen when comparing the detected genes in the non-enriched sample to its corresponding live-enriched sample. This figure highlights the fact that the early stage bladder cancer had a nearly three-fold increase in the median number of genes detected per cell, while the late stage cancer sample showed a 50% increase.

The distribution of UMI counts per cell in the standard method shows a small cluster of cells with the majority of the UMI counts. This same cluster of cells is represented by a single cluster when grouped by gene expression changes. In contrast, in the LeviCellTM enrichment sample there is a larger number of cells with significant numbers of UMI counts. This same group of cells is represented by four gene expression clusters (lower right), indicating a larger range of cellular diversity captured in the sequencing results.

A similar trend is observed with the late stage cancer sample, although the trends are less pronounced. Here, the sample processing with the standard method shows a large number of cells with high diversity of UMI counts, and correspondingly five clusters of gene expression. The sample after LeviCell^TM enrichment has a larger number of cells with high diversity of UMI counts, and six clusters of gene expression clusters, indicating a superior view of the expression profiles.

Related Content

10X single cell sequencing