ST. LOUIS (GenomeWeb) – Single-cell genomic approaches are offering researchers ever deeper looks into what's going on in heterogeneous and seemingly homogeneous cells and tissues, as underscored by numerous presentations on Sunday here at the Association of Biomolecular Resource Facilities annual meeting.
Cancer researchers, for instance, are interested in examining single cells from tumors to study how the disease progresses from a small group of cells with driver mutations to subclones that spin off circulating tumor cells that metastasize and expand further, now harboring a mix of old and new mutations, McGill University's Jiannis Ragoussis noted at the meeting.
"Single cell is a big, hot area these days," added Sridar Chittur from the State University of New York at Albany at a separate ABRF session.
With the ABRF's Genomics Research Group, Chittur and his colleagues have been applying single-cell genomics approaches to profile cells from the SUM149PT breast cancer cell line that both have and have not been treated with the HDAC inhibitor TSA.
In particular, they've turned to Fluidigm's C1 microfluidics-based platform, which allows for the capture, lysis, and amplification of single cells on its chip, and is, according to Chittur, fairly straightforward and hands-off to use.
He noted, though that users still have to know the size of their cells and have to be studying spherical cells — and the cells have to be fairly concentrated as well as viable.
The C1 system only has one port, so Chittur said that users have to be cognizant of the time the run will take and the possible influence that batch effects might have on their experiments if they have to keep a set of samples in a holding pattern.
After capture and amplification, Chittur and his colleagues used Clontech's SMART kit to generate full-length cDNAs followed by the Illumina Nextera XT kit to generate libraries for RNA sequencing.
Working with single cells is challenging, especially because of the many rounds of amplification that take place. Indeed, the sequencing results from Chittur and his colleagues highlighted just that — while data from their treated breast cells looked good, their controls failed as the data from those were all from a contaminating organism. Intriguingly, that organism isn't studied in any of the labs involved in the work.
Even with good results, he noted that single-cell RNA-seq is inherently noisy as compared to bulk cell RNA-seq.
McGill's Ragoussis, meanwhile, in a session dedicated to single-cell genomics, described the single-cell exome workflow that he and his team used to analyze treatment-resistant breast cancer.
To understand the heterogeneity of tumors and their progression, Ragoussis said that the disease has to be studied at various stages with high-resolution tools, and single cell analysis then becomes "critical."
He and his colleagues collected breast cancer biopsy samples as well as patient-derived mouse xenografts for analysis. They, too, used the Fluidigm C1 platform to separate cells and lyse and amplify their contents for sequencing. They were able capture the whole exomes of 81 single cells, with more than three-quarters of the reads mapping uniquely on target.
After sequencing, Ragoussis and his team had some 60,000 SNPs, approximately half of which were thought to be damaging. Using variant calling tools like GATK and Samtools, they identified mutations in BRCA1 and TP53 that were shared among all the cells and loss of heterozygosity mutations that affected 16 TCGA cancer driver genes.
Overall, the single-cell sequencing results were similar to what the researchers observed in their xenograft models.
Single-cell analyses are likewise of interest to immune researchers. During the same session, the Mayo Clinic's Timothy Niewold told attendees about how he studied two kinds of immune monocytes, defined by their cell surface markers — CD14 and CD16 — simultaneously using single-cell approaches by fluorescently labeling them prior to isolation.
RT-PCR analysis of the cells indicated that transcription, even of housekeeping genes, appears to occur in burst-like patterns. It may be, he said, like a thermostat in which transcription kicks back on after having turned off for a bit.
He and his colleagues also noted variations between cells from the same person as well as a greater diversity of expressed genes than expected. Bulk cells, he added, came out with the average expression, but there wasn't necessarily one cell that expressed that average.
Their gene expression data could broadly classify the cells as classical monocytes, which had high levels of expressed CD14, or non-classical monocytes expressing CD16.