Graphene transistor detects SARS-CoV-2 in less than a minute

A small transistor that can accurately detect tiny traces of RNA from SARS-CoV-2 could herald a new breed of rapid COVID-19 screening tests (J. Am. Chem. Soc. 2021, DOI: 10.1021/jacs.1c06325). Although the device is not yet ready for clinical use, it already combines high sensitivity with speedy results.

That could give it an advantage over some of today’s COVID-19 tests. Current nucleic acid tests, for example, seek out viral RNA using techniques such as quantitative reverse transcription polymerase chain reaction (qRT-PCR). These tests are sensitive but slow, typically taking a few hours to complete. Conversely, disposable tests for antigens—usually, the spike proteins that bristle the surface of the virus—provide faster results but are not as reliable as qRT-PCR. Alternative tests that offer both speed and accuracy could be particularly useful for screening asymptomatic people who nevertheless carry SARS-CoV-2 and may be infectious.

The new test, developed by Dacheng Wei of Fudan University and colleagues, is based on a transistor made of graphene, an atom-thin sheet of carbon atoms. This is decorated with Y-shaped fragments of DNA that target two genes encoded in the RNA genome of SARS-CoV-2. Each arm of the Y-shaped DNA carries a sequence that can bind to one of the two viral genes. As more viral RNA binds to the DNA probes, it progressively changes the current flowing through the graphene transistor beneath, enabling the researchers to measure the viral load in a sample. Crucially, the method does not require a nucleic acid amplification step, one of the main bottlenecks of qRT-PCR.

The researchers tested the device using artificial saliva samples containing viral RNA and nasopharyngeal swab samples collected from patients with confirmed COVID-19 infections. After heating the samples for 30 min and adding them directly to the chip, the team found that the device produced results in under a minute. “The detection limit of our method is also very low, detecting about three molecules of SARS-CoV-2 RNA in 100 µL of solution,” says Derong Kong, a graduate student in Wei’s lab and first author of the paper.

That limit of detection makes the test 20 times as sensitive as the standard for qRT-PCR assays set by the US Centers for Disease Control and Prevention. “From a sensor perspective, they have really good results,” says Delphine Bouilly of the University of Montreal, who develops nanobiosensors and was not involved in the work.

The device could also detect very low levels of viral RNA in pooled samples where one positive sample had been mixed with four negative ones. Pooling samples from different people is one way to speed up testing, but when this approach is used with qRT-PCR, the dilution involved generally means that longer amplification cycles are needed.

Bouilly says that similar graphene-based biosensors have been around for a decade or so, but they often rely on single-stranded DNA probes that can sometimes get tangled up, reducing the sensitivity of the devices. In contrast, the Y-shaped DNA fragments have a rigid helical structure at their base that prevents them from tangling. Using two RNA binding points on each probe makes the device even more sensitive, she adds.

Other graphene transistor–based diagnostic devices have used antibody probes to recognize viral antigens (ACS Nano 2020, DOI: 10.1021/acsnano.0c02823). But the synthetic DNA probes used in the new transistor are “much less expensive, and they’re also much more customizable” than antibodies, Bouilly says. Kong estimates that the materials in the device only cost about $1.60 and says that the biosensor could be decorated with different DNA probes to detect other viruses.

For now, the transistor is not yet optimized for mass production or clinical use, and it performs best when it is used only once. Wei’s team aims to make improvements so that the device can be cleaned and reused without losing its sensitivity.