When doctors need to identify the strain of bacteria responsible for a patient’s infection, they use blood cultures, which can take days to return results. To speed up diagnosis, scientists want to develop microfluidic methods that diagnose infections by spotting microbes’ genetic material in blood. Unfortunately, sample preparation for such devices remains slow and difficult to automate. Now researchers have developed a simple and quick microfluidic method to extract and purify bacterial RNA from human blood (Anal. Chem., DOI: 10.1021/ac301021d).
“A glaring weakness common to the vast majority of microfluidics is the lack of integrated sample preparation on the device,” says Juan Santiago of Stanford University. Ideally, scientists would like a single microfluidic device to automatically extract, purify, and analyze genetic material from complex biological fluids.
In a previous study, Santiago extracted and purified human genomic DNA from blood in a microfluidic channel using isotachophoresis, a type of electrophoresis (Anal. Chem., DOI: 10.1021/ac901965v). For the new device, Santiago and his collaborators, Anita Rogacs and Yatian Qu, wanted to increase sensitivity by detecting RNA. There are often many strands of RNA for each gene, he says, increasing the signal even with only a few bacteria in a sample.
Unfortunately, while Santiago’s previous method isolated high quality DNA from blood, it produced RNA of poor quality, he says. Ubiquitous enzymes in blood called RNases degraded the RNA before the device could analyze the nucleic acids.
So Santiago and his colleagues devised a four-chemical mixture to protect RNA in blood samples. Sodium hydroxide, a strong base, rips open the cells in the sample. Then dithiothreitol, a reducing agent, disarms RNases by breaking a critical disulfide bond in them. Triton X-100, a surfactant, increases the solubility of proteins and other cell debris so as not to gunk up the microchannel, Santiago says. Finally, the mixture contains high concentrations of synthetic RNA. These nucleic acids get between any remaining RNases and the target RNA, says Santiago, “like a herd protecting its young.”
After one minute in the four-chemical mix, the solution gets neutralized when the researchers add a buffer. Then they transfer drops of the mixture to a reservoir at one end of a microfluidic channel. This reservoir contains a second buffer. When the researchers apply a current to the microfluidic chamber, the molecules migrate through the channel at different rates based on their charge and size. One buffer contains molecules that speed through the channel quickly and the other contains ones that move slowly. The two sets of buffer molecules sandwich and concentrate the nucleic acids between them.
After five minutes of applying current, the researchers collect the nucleic acids at a second reservoir at the end of the channel, leaving the slow-moving blood proteins stranded in the channel. The researchers can then mix the purified nucleic acids with reagents to amplify them.
The researchers tested the device on human blood mixed with known concentrations of the soil bacterium Pseudomonas putida. They found that they could detect P. putida RNA with as little as one bacterial cell per 30 nl of blood. Santiago says this sensitivity is enough for some clinical applications and is six orders of magnitude better than the only other microfluidic method that extracts RNA from blood.
The work represents “a very important development,” says Pak Wong of the University of Arizona, Tucson, in the move toward all-in-one microfluidic-based assays. Santiago is working on a multi-faceted chip in which nucleic acids purified by isotachophoresis are automatically delivered into another chamber for analysis.