Individual cells vary widely, whether in a healing organ or a cancerous growth. Most of the time, however, cells must be studied as populations, masking differences between individual cells. A new method, published this month by Hideki Kambara and his team at the Hitachi Central Research Laboratory in Tokyo, will allow scientists to look at gene expression in single cells1.

Quantitative real-time PCR (qRT-PCR) along with DNA chip technology is widely used for gene-expression analysis, but because the technique requires minute amounts of DNA to be copied and analyzed, it can introduce noise and bias into the data. The standard procedure uses a cDNA library, and the mRNA transcripts in all of the cells are converted into DNA, which can then be analyzed. This method works well for groups of cells, but for individual cells, it is not so easy to use. mRNA transcripts from a single cell must be amplified to produce enough DNA to analyze, and if there is a low copy number, the amplification can make it hard to distinguish between real gene expression and noise in the PCR data. To avoid amplification, Kambara's team has developed a method in which, for the first time, the cDNA library generated from each cell is reusable.

Uniquely, the team immobilized the cDNA molecules on magnetic beads and added the beads to the qRT-PCR reaction. Juggling various parameters, including the temperature of the PCR reaction and the number of beads, Kambara optimized the experiment and was able to protect the integrity of the cDNA library so that as many as 20 genes could be analyzed. After each qRT-PCR reaction, the beads could be cleaned of any PCR products and used again. “We are planning to improve the method further — gene expressions for more than 100 genes will be carried out with a single cDNA library [in the] near future,” Kambara says.

The field of single-cell analysis is taking off: South San Francisco, California–based Fluidigm has developed a 96-well plate assay looking at one gene per cell. But Kambara can accurately examine up to 20 genes from the unamplified template cDNA of one single cell.

“I think technically this reusable library is very intriguing,” says Kai Lao, principal scientist at Applied Biosystems in Foster City, California. “I never would have thought this could be done.” The practical ramifications are less clear, however. Lao recently published a paper with Azim Surani at the University of Cambridge, UK, that describes how they amplified a single mouse blastomere to look at expression of thousands of genes, and Lao says that Applied Biosystems is already marketing tests that accurately look at around 350 genes using amplified cDNA from a single cell.

Can looking at cells individually tell you more than looking at cells in a group? “I thought all the cells in a tissue might have similar gene-expression patterns,” Kambara recalls. But the experiments showed that these patterns fluctuated. “At first I thought it must be an artefact and repeated the measurements, using many single cells as well as model mRNA samples. The results were very reproducible. The gene-expression patterns were fluctuating for single-cell samples.”

The fluctuations in gene expression might play a role in cell responses to certain environmental cues, and these nonuniform expression patterns can be clarified by singe-cell analysis. This is extremely pertinent in the field of stem cell science, in which the availability of cells is limited. Single-cell analysis may be able to uncover the subtle genetic differences between pluripotent and multipotent progenitor cells and could help explain how stem cells respond to environmental cues that trigger self-renewal or differentiation.

For the moment, gene-expression analysis for a single cell is not very easy and is rather time consuming, “but imagine the future, where gene expressions for thousands of cells can be easily carried out in a short period of time. Which would you prefer to have, the averaged gene-expression data or individual ones”? Kambara asks.

Related articles

A closer look at the single cell