Researchers have developed a new technique that allows them to examine huge amounts of information from a single cell or zoom out and see data patterns among thousands upon thousands of cells — all in a single experiment. In work published Monday in the journal Nature Communications, scientists from the biotech company 10x genomics and Fred Hutchinson Cancer Research Center describe their method, which could help researchers dive deep into the ecosystem of cancer or other diseases.
The platform allows researchers to analyze which genes are turned on (and to what level) in tens of thousands of cells at once.
“What the technology allows you to do is to be able to identify different types of cells and how many there are, and also infer what they’re doing based on gene expression,” which could give researchers a better understanding of diseases such as leukemia, noted co-author Dr. Jerald Radich, a Fred Hutch physician-scientist who specializes in leukemia research.
Radich is working with the new technology to gain a better understanding of which cell types contribute to leukemia relapse. Once that’s understood, “you can imagine using it in the clinic as an adjunct to the ways that we look at residual disease [low levels of remaining leukemia cells that can contribute to relapse],” he said.
Fred Hutch’s Dr. Jason Bielas, the paper’s lead author, developed methods and designed the experiments needed to validate the platform, known as the Chromium Single Cell 3’ Solution. He and his team were able to analyze nearly 70,000 cells in a single experiment and use gene expression patterns to group individual cells by type.
Bielas also developed additional methods to detect subtle DNA variations and further expand the technology’s applications. Current methods to detect leukemic cells in patients often rely on surface markers. Using only gene expression information and slight differences in gene sequences, the team was able to distinguish between donor and recipient blood cells in patients who had received bone marrow transplants to treat their leukemia — an important component of patient care after transplant.
“We developed methods that we believe can be used to improve clinical care of cancer patients and save lives, in addition to addressing fundamental questions in biology,” said Bielas, who studies how errors in the genetic code contribute to disease and develops new technologies to address long-standing questions in mutation research.
Our genetic code acts like a recipe book for proteins; roughly, each gene carries the instructions for one or more proteins. To make a protein, our cells first make many copies of its gene, like photocopying a recipe and handing the copies to dozens of protein-making chefs. A cell’s transcriptome is the entire collection of the “photocopies,” or transcripts, of all the genes for which that cell is currently producing proteins. Transcriptomes vary widely, changing by cell type and even by what environment each cell has recently experienced.
Cancer is made up of millions of different types of cells, but historical methods to assess the disease have averaged the information from all these cells into one output. This type of experiment doesn’t allow for more sophisticated analyses of cancer. It’s like analyzing the reading habits of all the patrons in a library system by throwing books from all library branches into one pile. It may give a snapshot of, say, all the science fiction books in the system, but it doesn’t tell you which library branch’s patrons request these books the most.
New single-cell techniques are giving scientists the chance to examine biological systems cell by cell. They can now see whether a single cell is overproducing transcripts of a specific gene. By combining this information with information about other cells’ transcription of that gene, they can determine whether that gene is often transcribed in certain types of cells. But with this opportunity comes a new challenge: including enough cells in each experiment to give an accurate picture of, for example, a specific disease. The technology developed by 10x genomics and validated and tested using methods and experimental design from Bielas’ team provides a way to record the tens or hundreds of thousands of gene transcripts of many thousands of cells — all at once.
All told, “we looked at a quarter of a million cells in this paper,” Bielas said.
Bielas and the 10x genomics team hypothesized that comparing the transcriptomes of thousands of cells would allow them to determine different cell types’ hallmark gene expression patterns and enable them to discover a specific cell’s identity without looking at it under a microscope or knowing anything about the patterns of proteins on its surface.
Other techniques for separating cells by type rely on prior knowledge of the particular molecules that mark specific cell types. Researchers who want to read the transcriptome of every cell that group must complete more experimental steps.
The Chromium Single Cell 3’ Solution platform employs what’s known as a gel bead in emulsion, or GEM. Each bead gets encapsulated, along with no more than a single cell, in an oil droplet. Inside each droplet, a biological reaction occurs that transforms gene transcripts into a new, more stable format that can be read by sequencing machines. Every bead carries a unique barcode that tags the transcripts from its paired cell, and another tag that increases the accuracy of the transcript information.
The team found that the platform could be used to distinguish mouse cells from human cells and even between different kinds of human cells. They were able to analyze nearly 70,000 human blood cells in a single experiment and identify the gene expression patterns that distinguished each cell type.
Bielas’ primary research focus is on mutations, the typos in our genetic recipe book. Sometimes just a single letter of a gene’s sequence is altered, but Bielas suspected that these subtle variations could be used to extend the potential of high-throughput single-cell technology to determine whether two cells arose from different origins.
Such a capability has potential clinical applications. Detecting low levels of cancerous cells or new donor cells after transplant is key to monitoring patient health and detecting disease relapse as early as possible.
Bielas and his Fred Hutch team developed methods that allowed them to examine blood cells from a leukemia patient who had received a bone marrow transplant and distinguish donor cells from the recipient’s own cells. They were also able to group donor and host cells into different cell types, including leukemic cells from a particular lineage which would have been difficult to detect using standard clinical methods — an advance with the potential to change clinical practice and improve patient care, said Bielas.
Typically, researchers hoping to peer at the inner workings of cancer cells would need to be able to sort them from healthy cells prior to analyzing their gene transcripts. This requires them to know more about the cells — such as their hallmark surface molecules — and also requires more steps that can damage sensitive cells and result in lost data.
“In this case, you get all the information in one snapshot,” said Bielas.
Radich is excited to begin using the technology to learn more about the cellular interactions that drive cancer development and relapse after treatment. Cancer is essentially an ecosystem, he said.
“This is one of the ways you can understand more about what’s really going on [in the cancer ecosystem] by deconstructing the bulk population and seeing what cells are really there, and how do they change over time, and how do their functions change. And does that make any difference?” he said.
“Once you start understanding what the ecosystem does, and what it takes to keep the cancer at bay, then you can start manipulating that, potentially therapeutically,” Radich said.
Radich focuses much of his research on understanding who relapses and why. “A lot of the stuff we want to do has been waiting on new technologies. And this one has some promise,” he said.
Bielas is resurrecting an old project — looking at all the different types of immune cells that slip into tumors and which may affect progression and patient outcomes. “We’d like to look at the association between [those cells] and [cancer] mutations to see if they play a role,” he said. But he expects that other researchers will find many new ways to use the Chromium technology.
There are many other possible applications for the technique, Bielas said. “It’s kind of endless.”
The datasets generated in this work can be found here.
Sabrina Richards, a staff writer at Fred Hutchinson Cancer Research Center, has written about scientific research and the environment for The Scientist and OnEarth Magazine. She has a Ph.D. in immunology from the University of Washington, an M.A. in journalism and an advanced certificate from the Science, Health and Environmental Reporting Program at New York University. Reach her at email@example.com.