Fishing Out Dilute Disease Biomarkers

Monitoring disease-related biomarkers called microRNAs (miRNAs) could be easier and more accurate with a new technique that detects directly the nucleic acids in blood samples at sub-femtomolar concentrations (Anal. Chem., DOI: 10.1021/ac201618k).

Scientists think that these short RNA segments are signposts for certain diseases. The nucleic acids don’t carry protein-assembly instructions, but can regulate gene expression. Since specific miRNAs signal different stages of cancers, diabetes, Alzheimer’s and other diseases, scientists think that finding them can help monitor and predict the course of these diseases.

Detecting miRNAs quickly and cheaply in blood samples hasn’t proved so easy: They aren’t present in large amounts, and blood’s other biomolecules complicate methods to extract the molecules. Current techniques often rely on amplifying the nucleic acids using polymerase chain reaction to produce higher concentrations that are easier to detect and measure. But this amplification can introduce sequence errors, leading to misidentification of the miRNAs. A more ideal method would detect the miRNAs directly, says chemical engineer Patrick Doyle atMassachusetts Institute of Technology.

To produce such a method, Doyle and his graduate student Stephen Chapin developed a system that amplified the detection signal, instead of the miRNAs. Their technique uses absorbent gel particles that are 70 µm wide and look like dominoes. The scientists etch sites into the particles to attach DNA molecules that selectively recognize specific miRNA sequences of interest.

After the scientists add the particles to a blood sample and allow miRNAs to hybridize with the gel-bound probes, they introduce another, longer DNA molecule that binds to the end of the probe DNA with a long, over-hanging DNA sequence. Then the researchers add a circular DNA template and DNA polymerases to the gel particles to replicate the overhanging sequences about 100 to 1,000 times. This process builds long DNA chains of repeating sequences at each bound miRNA site. The scientists can tag each repeating unit with fluorescent markers to amplify the detection signal and make it easy to detect bound miRNAs using fluorescence spectroscopy.

Doyle and Chapin tested their technique with real serum samples from a healthy donor and a patient with prostate cancer. They looked for a few miRNA sequences including miR-141, which appears in the blood of people with prostate cancer. They report detecting miRNA in serum at concentrations of 300 aM, which amounts to about 10,000 molecules in 50 µL. In blood serum, there are typically 10,000 to 100,000 molecules of miRNAs per 50 µL, says Chapin. “We’re certainly in the zone where we can detect clinically relevant miRNAs in the circulatory system,” he says.

Muneesh Tewari, who works on miRNAs at the Fred Hutchinson Cancer Research Center in Seattle, is impressed that the technique can pick out different miRNAs in one go from serum, without isolating the RNA. “It is remarkable that the sensitivity appears to approach that of PCR-based methods,” he says, adding that it is too early to tell how useful the technique will be in a clinical setting. But he says, “the results so far are promising”.