NEW YORK (GenomeWeb News) – Using a tiny microfluidics chip, a team of California researchers has developed a new method for assessing gene expression in single cells, according to research that’s scheduled to appear online this week in the Proceedings of the National Academy of Sciences.
The group developed the integrated microfluidic bioprocessor by combining an affinity capture system with tiny pumps and miniscule chambers in which RT-PCR reactions occur in just a couple hundred nanoliters of liquid. Using the device, the researchers demonstrated that cells exposed to small interfering RNA targeting the GAPDH gene fall into two categories: cells with complete GAPDH silencing and cells containing about half the normal level of GAPDH transcripts.
The device has the potential to help those bent on answering certain types of research questions. But, researchers say, it is not ideal for evaluating gene expression in all research contexts, leaving room for other gene expression technologies as well.
Along with microarrays, which provide an analog readout of gene expression, researchers have come up with several schemes for assessing gene expression digitally, including real-time PCR, mRNA-seq, serial analysis of gene expression, and so. But many of the existing methods provide information about the average gene expression in a group of cells, rather than the expression in each individual cell.
Richard Mathies, a chemist affiliated with the University of California at San Francisco and the University of California at Berkeley and director of UC Berkeley’s Center for Analytical Biotechnology, was senior author on the new paper. He told GenomeWeb Daily News that ensemble approaches — in which researchers take a clump of cells and “smoosh ‘em all up and mix up all that [genetic] information” — miss outliers and information about distinct groups of cells in a sample.
In contrast, he said, his team’s approach provides a way to quantitatively assess gene expression and compare one cell to the next using a microfluidic device that allows researchers to do RT-PCR on RNA from a single cell in one step.
To begin with, researchers trick the cells into taking up a synthetic sugar molecule. Cells metabolize the sugar and then present an azido group on the cell surface. They then react that group with modified single-stranded DNA, which binds complementary DNA, anchoring cells on the capture pad one at a time.
The cell is lysed and subjected to RT-PCR in a “super small volume” (the reactor contains just one or two hundred nanoliters). The device uses and relies on the same basic technology that’s used in some sequencing platforms, forensic identification, and infectious disease detection, Mathies explained. By using such a small reaction volume, the effective concentration of DNA increases, diminishing the signal-to-noise ratio.
“We actually harvest every product made by the PCR reaction,” he said.
For instance, when Mathies and his colleagues used the integrated chip to study gene expression in eight Jurkat cells that had been exposed to siRNA targeting GAPDH, they found two completely different groups of cells: two cells in which about half the normal amount of GAPDH mRNA was expressed and six cells expressing little to no GAPDH. In contrast, when the expression of an assembly of these cells is measured, the siRNA appears to curb roughly 80 percent of GAPDH expression.
Although they only tested the expression of one gene, Mathies said it is possible to do multiplexed gene expression assays on the integrated chip. The number of genes that can be tested simultaneously depends on the desired resolution, he said, but the chips can easily test the expression of at least 25 genes in parallel. He also noted that it’s possible to clean and re-use the chips.
Dublin, Calif.-based Microchip Biotechnologies is developing aspects of the technology for commercial purposes. Mathies said the company holds the rights to the technology described in the paper, but is currently focusing on the forensics and genetic sample preparation applications of similar technology rather than rushing to commercialize the integrated gene expression system.
Mathies is a founder of Microchip Biotechnologies.
For his part, Mathies plans to use the device for studying stem cell differentiation. He is currently trying to drum up support for a project following gene expression changes in single stem cells. He noted that the technique may also prove useful for cancer research — providing a window into the gene expression changes associated with cancer progression.
The researchers are continuing to improve the microfluidic system and eventually, Mathies said, he hopes to be able to detect a single transcript from a single cell — a goal that he called “perfectly feasible” with this platform. Mathies argued that the new approach offers increased sensitivity and precision over existing methods aimed at assessing gene expression in individual cells.
But the approach has drawbacks for those interested in assessing gene expression across the genome, Mathies conceded, since it relies on primers targeting particular candidate genes. He noted that other approaches, including microarrays, are more appropriate for getting a genome-wide look at gene expression.
Paola Capodieci, a senior scientist at Yonkers, New York-based life sciences company Aureon Laboratories, called Mathies’ approach “quite interesting.” But she told GenomeWeb Daily News that the most appropriate method for assessing gene expression depends largely on the research questions being asked.
Capodieci has been helping develop a fluorescent-based method for measuring gene expression in single cells without destroying tissues — an approach that is expected to particularly appeal to pathologists.
In 2005, Capodieci was lead author on a paper in Nature Methods describing peT-FISH, a multiplex fluorescent in situ hybridization assay for measuring gene expression in single cells from paraffin-embedded human tissue.
To measure gene expression using FISH, researchers prepare labeled oligonucleotides targeting DNA in the gene or genes of interest to internally label nascent, pre-mRNA transcripts fluorescently via in situ hybridization.
By detecting signals from multiple nascent RNAs in intact tissue, the method allows investigators to compare gene expression differences between cells in the same tissue. Like Mathies’ approach, that allows researchers to look not just at the average gene expression, but at cell-to-cell variation.
So far the peT-FISH method is semi-quantitative, Capodieci explained. For instance, she said, it has been used to accurately distinguish between gene expression profiles in cells treated with drugs and untreated cells. Capodieci said she and her colleagues are working towards increasing the sensitivity of their approach and developing an automated, quantitative system for assessing gene expression.
Another firm, Seattle-based NanoString, which spun out of the Institute for Systems Biology, has developed a system for assessing gene expression digitally using color-coded fluorescent reporters that target genes of interest. By scanning these barcodes, the company’s system can reportedly profile hundreds of genes in a single reaction in an automated way.
Last February, NanoString announced that the Baylor College of Medicine and the University of Miami had signed on as early access users of the company’s nCounter digital gene expression platform. In late 2007, the company also placed instruments at the University of Washington and the California Institute of Technology.
Fluidigm is also targeting the market for single-cell gene expression. The company claims that its BioMark Dynamic Arrays can study up to 96 genes in a single cell.