October 15, 2017 (Vol. 37, No. 18)
Like snowflakes, fingerprints, and retinas, every cell has a unique personality. Whether wandering through the bloodstream as solitary nomads or gregariously congregating in tissue, each cell has a unique story to tell.
Single-cell genomics helps obtain and write those complex biographies, yielding insights into normal cell dynamics, and provides information about when cells go rogue and turn into tumors.
Single-cell genomics, a field still in its infancy, captures genome-wide data at the resolution of individual cells. With its ability to depict thousands to millions of individual cells at once, the potential is enormous for shedding new light on fields such as immunology, neurology, and cancer. However, a common challenge of heterogeneous systems is that they often contain massive numbers of cells. For example, a centimeter-sized tumor can consist of hundreds of millions of cancer cells. Additionally, each cell has only a tiny bounty of DNA, making it a challenge to accurately amplify and sequence individual cells.
Despite these limitations, innovative technologies and platforms are being developed to not only enhance throughput, but also to quickly evaluate the vast datasets generated. Recent meetings, such as the Precision Medicine World Conference and SelectBio’s RNA-Seq, Single Cell Analysis & Single Molecule Analysis conference, present new forums to discuss this young and exciting field. Some approaches are employing microfluidics in the form of nanodrops to encase cells or microfluidics chips with gel beads containing bar-coded oligonucleotides. Another emerging field is the use of cell nuclei for single nuclei RNA-seq. The latter can avoid the harsh treatment to separate or lyse whole cells that have the potential to alter gene expression. Also on the horizon are computational pipelines and visualization tools for efficiently processing million-cell datasets.
Earlier methods examining gene expression usually averaged all the transcriptomes obtained from bulk tissue. This global snapshot necessarily lost vital information at the single-cell level. While recent advances enabled mRNA-seq analysis of individual cells, technologies could only examine a few hundred cells at a time. Additionally, cells often first needed separating by flow cytometry or microfluidics.
However, the emergence of high-throughput single-cell RNA-seq (scRNA-seq) has begun revolutionizing the field. “This technology has offered an exciting, cutting-edge technique that provides unprecedented insight into the expression patterns of thousands of single cells,” reports Muriel Breteau, Ph.D., technical applications specialist, Dolomite Bio. She adds, “Individually analyzing thousands of single cells from a tissue can vastly improve the [understanding of] complexity of biological systems to derive vital information [about] diseases and immunity.”
Dolomite Bio has developed a microfluidics platform that enables the Drop-seq protocol (Figure 1). Dr. Breteau explains, “Our technology allows highly parallel genome-expression profiling of cells using nanoliter droplets. Thousands of cells can be individually captured in these droplets. Inside the droplet, and within only a few minutes, the individual cell is lysed and its mRNA captured on uniquely barcoded oligonucleotides attached to beads. After recovering the beads, they are subjected to reverse transcription, library preparation, and finally sequencing” (Figure 2).
As an example, a cell suspension can be loaded into an agitated remote reservoir and beads injected. Droplets are then collected in an output reservoir at ~2,800 droplets per second. Within a 15 minute run, more than 6,000 single-cell libraries can be generated.
Dr. Breteau points out, “There are many applications for this novel technology ranging from fundamental research identifying new cell types, to defining tumor heterogeneity, clonal evolution, and effects of drug treatments in oncology applications. One of the key features of the system is that it is open and flexible. Researchers can innovate by utilizing their own reagents and protocols instead of being required to use just [Dolomite Bio’s products].”
In the future, Dr. Breteau envisions the day when the technology will be applied to personalized medicine. “Because this is essentially a lab-on-a-chip microfluidics device, it could be possible to obtain a small blood sample and get quick answers, for example, as to the very early stages of cancer detection.”
Profiling Large and Complex Cell Populations
The ability to profile a large number of cells is becoming increasingly important for rare-cell detection and for the comprehensive classification of biological systems, according to Grace Zheng, Ph.D., senior scientist, 10x Genomics. “Because of the limited throughput with most scRNA-seq methods, our company devised a comprehensive, scalable solution to quickly characterize and profile the transcriptome of hundreds to millions of cells.”
Single cells, reverse transcription reagents, “gel beads” containing barcoded oligonucleotides, and oil are combined on a microfluidic chip to form reaction vesicles called “Gel Beads in Emulsion,” or GEMs, which are formed in parallel within the microfluidic channels of the chip, allowing the user to process hundreds to tens of thousands of single cells in a single seven-minute Chromium™ Controller run.
Dr. Zheng reports, “Cells are loaded at a limiting dilution to maximize the number of GEMs containing a single cell, while ensuring a low doublet rate and maintaining a high cell recovery rate of up to ~65%. The speed, reproducibility, and high cell capture rates of the Chromium Solution facilitates profiling and discovery of precious and rare-cell populations.”
A challenge associated with the accumulation of such massive scRNA-seq datasets is their efficient processing. According to Dr. Zheng, such computational analysis has not been scalable beyond tens of thousands of cells, and has been a limiting factor to enabling large-scale cell atlas efforts. “Our computational pipelines and visualization tools can now efficiently process million-cell datasets, and can be easily used by a noncomputational biologist. These are the types of tools needed to enable large-scale cell atlas studies.”
As an example of the power of the technology, the company profiled 1.3 million neurons from two embryonic murine brains. More than 100 single-cell libraries were completed in two days. Major neuronal and nonneuronal cell types were detected from different cortex layers, the hippocampus, and subventricular zones. Dr. Zheng notes, “We readily detected diverse and rare interneurons without the need to enrich by flow-cytometry sorting.”
Another important single-cell application is the characterization of paired T-cell receptor alpha and beta chains in tens of thousands of T cells. This application allows comprehensive immune repertoire profiling, for determining which functional subsets of T cells have undergone clonal expansion. Dr. Zheng believes this will be especially valuable in areas of infectious diseases and immuno-oncology.
Single-Cell Nuclei Transcriptomes
Cells of the central nervous system are particularly difficult to isolate as intact whole cells. Neurons are highly interconnected with axons and dendrites. To separate them by physical means often causes considerable damage. Further, proteolytic degradation of surface proteins to dissociate whole cells from tissue may stress the cell sufficiently enough to alter gene expression.
Roger Lasken, Ph.D., director of single-cell genomics, J. Craig Venter Institute, and colleagues have developed an alternative approach that focuses on the nucleus and its stash of mRNA while eliminating traditional harsh treatment utilizing whole cells. He reports, “RNA is made in the nucleus, processed, spliced, and exported to ribosomes in the cytoplasm. We’ve made extensive comparisons of nuclear and cellular transcriptomes and have demonstrated that nuclei can substitute for whole cells in most RNA-seq applications. For most genes, single-nuclei RNA-seq (snRNA-seq) provides expression signatures that are very similar to those obtained from whole-cell controls.”
As an example, Dr. Lasken describes a study of memory carried out by collaborators using mouse hippocampus tissue. “They characterized the transcriptome from mouse neurons using snRNA-seq. The mice were initially raised in a very plain home cage, but then placed into a cage with a more enriched environment including tunnels and other objects. As they got excited they were turning on transcription processes in specific neurons involved in making memories.”
In collaboration with Fred H. Gage, Ph.D., at the Salk Institute for Biological Studies and others, the investigators discovered major changes in the neuronal transcriptome program including upregulation of immediate early genes (IEG) such as FOS, ARC, and EGR1. Dr. Lasken summarizes, “The IEG expression of single nuclei that were derived from activated neurons was consistent with known responses to behavioral experiences. Analysis of the entire transcriptome revealed extensive upregulation of genes involved with synapse ion channels and other features suspected to play a role in memory formation. This is the first look at the entire transcriptional profile in individual neurons activated by external stimuli, and is a critical step in ultimately discovering how a memory is captured and stored.”
Dr. Lasken says nuclear transcriptomes also can be obtained from postmortem human brain tissue stored at −80 °C, making brain archives accessible for RNA-seq from individual neurons. Further, snRNA-seq can be applied to other types of tissue as well as individual cells including eggs, circulating T cells, and cultured cells. He concludes, “I believe snRNA-seq is a robust technique for examining heterogeneous transcriptional profiles from a variety of samples. This is quite a breakthrough!”
Emerging Applications and Instrumentation
With the fast-paced progress developing in single-cell genomics, it is no surprise that instrumentation is expanding in parallel with applications. BD Genomics, a division of BD (Becton Dickinson and Company), recently presented information on their current and emerging platforms and applications for single-cell genomics and how they have begun to apply these to the study of solid tumors and liquid biopsies.
John Mikszta, Ph.D., director, genomic sciences at the company explains, “There are three key challenges in the field. The first is the complex and challenging workflow of preparing libraries for next-generation sequencing (NGS). Often, library prep is difficult and expensive. To address this problem, we developed the BD CLiC system that miniaturizes library prep reactions to enable cost savings for the end-user. The system has scalable throughput and can process 24–384 samples per run in a fully automated fashion; the user simply hits ‘go’ and walks away.”
A second challenge is applying single cell genomics to clinical oncology research, such as the study of tumor heterogeneity, in order to distinguish all cells in the tumor microenvironment. “For single-cell genomic analysis of solid tumors, we provide reagents and methods to dissociate tumors into single-cell suspensions. These suspensions are then amenable to cell-surface protein analysis by BD FACS flow cytometry platforms and single-cell gene-expression analysis with a product called BD Precise. Individual cells are sorted into 96 well plates prefilled with reagents for single cell gene expression. The use of molecular indexing eliminates PCR bias and improves accuracy by >100-fold.”
The company recently developed the BD Rhapsody platform and reagents, according to Dr. Mikszta (Figure 3). “This allows evaluation of single-cell gene expression from hundreds to thousands of cells. The platform has a simple one-day workflow that allows substantial cost reductions per sample.”
Further, in collaboration with clinical researchers at leading academic medical centers, the company is developing a system for rare-cell enrichment based on magnetic depletion and acoustic focusing. This cell-enrichment system is placed upstream of a traditional BD FACS sorter to enable circulating tumor cell capture for downstream single-cell genomics.
Dr. Mikszta asserts, “While the rare-cell-enrichment technology used in these experiments is still in early development, we already see that it will offer a powerful platform for rare-cell isolation and characterization.”