Skip to main content
Premium Trial:

Request an Annual Quote

CRISPR-Based Wearable Diagnostics Enable SARS-CoV-2 Detection Through Mask

MIT-Sensor-Mask-01_Felice Frankel.jpeg

NEW YORK – As the COVID-19 pandemic progressed through 2020 and into 2021, it became increasingly apparent that mask wearing was essential to slowing the spread of SARS-CoV-2. Now, researchers at Harvard University's Wyss Institute for Biologically Inspired Engineering are looking to turn masks into wearable devices that could help diagnose viral infections in individual users long after the current pandemic has run its course.

In a paper published on Monday in Nature Biotechnology, the researchers described their development of a synthetic biology CRISPR-based wearable sensor that can be integrated into a face mask for the noninvasive detection of SARS-CoV-2 through a person's breath at room temperature within 90 minutes. Other than the push of a button, the sensor requires no user intervention, and it indicates the presence of virus by changing color.

Specifically, the researchers used freeze-dried, cell-free, or FDCF, genetic circuits in combination with specifically designed flexible and textile substrates to create practical wearable biosensors for small molecule, nucleic acid, and toxin detection. They then integrated these sensors into flexible multi-material substrates such as silicone elastomers and textiles, using genetically engineered components including CRISPR-Cas12a complexes.

"Most diagnostics are very invasive. They swab far up the nasopharyngeal cavity or they'll do saliva tests. This is one of the first that we've seen that reliably looks at nucleic acid from just breath, which is very noninvasive. And it does so at sensitivities rivaling what we do in laboratories for normal PCR-based sampling," co-first author Peter Nguyen said. "So, we think it'd be highly useful for anything that is aerosol or anything that's respiratory transmitted."

The Cas12-based sensor was paired with recombinase polymerase amplification, or RPA, and freeze-dried into a one-pot reaction. In the presence of a target double-strand DNA sequence, isothermally generated RPA amplicons activate Cas12a guide RNA complexes, which leads the active Cas12a to engage in trans-single-strand DNase activity. This cleaves quenched single-strand DNA fluorophore probes, resulting in a fluorescence output.

The researchers integrated the sensor into a variety of wearable items capable of sensing different viral and bacterial targets. In the study, they demonstrated the efficacy of the device on the aerosol-enabled mask for the detection of SARS-CoV-2. Importantly, when the researchers compared the sensitivity of their mask sensor against that of World Health Organization-endorsed standard laboratory-based RT-PCR assays for SARS-CoV-2, they observed a limit of detection for their sensors of 500 copies of SARS-CoV-2 in vitro transcribed RNA, matching that of the PCR assays. The mask sensors also didn't cross-react to RNA from other commonly circulating human coronavirus strains.

However, they also showed that the sensor could be used in real-time monitoring of environmental exposure and biohazard detection, and tested a jacket they'd created that contained a distributed arrangement of wearable FDCF multi-sensor arrays designed to detect species of bacteria such as those a doctor might encounter while seeing patients.

According to Nguyen, one of the sensor's most important features was that it was optimized to work at room temperature. He also emphasized the completely independent nature of this diagnostic sensor from the lab, noting that although diagnostics development has boomed during the pandemic, the common denominator for most of the tests has been the need for laboratory equipment, a technician, a doctor to order the test, or someone to collect the sample during the testing procedure.

"What we want to do is see if we can short-circuit that or replace the laboratory completely. Now, we've shrunk the entire laboratory down into the wearable. You can wear it on your suit, or you can wear it inside a mask," Nguyen said.

Combined with the ability to freeze-dry all the components, the ability to work at room temperature was the biggest step in uncoupling the sensor from the lab, he added. And although taking the test out of the lab setting may have lowered the sensitivity, isothermal amplified-CRISPR is already two to three orders of magnitude more sensitive as a testing modality than RT-PCR in the lab.

"So, we took a hit [on sensitivity], but that hit that we took brought us back to the same level as the RT-PCR diagnostics," Nguyen said.

The face mask sensor also doesn't have any electronic components. Because face masks are inexpensive and disposable, for the most part, the sensors also have to be inexpensive. To that end, the signal that indicates whether someone is infected or not is colorimetric, said co-first author Luis Soenksen. The CRISPR reaction also cleaves reporters that are attached to the complex, leading to a colored line in the sensor, much as would develop in a pregnancy test to indicate a positive result.

Nguyen and Soenksen did note, however, that the other wearables they developed would likely be more sophisticated in their design. For example, the jacket would be connected wirelessly to a phone and could ping the user if exposed to a certain toxin or biohazard such as Ebola.

"For that one, you're not focused on a single thing where you're asking, 'Am I COVID-19 positive?' You're wearing it as protection and you're not able to constantly check yourself. So, that one's nice to have something where it'll alert you," Nguyen said.

Soenksen added that the jacket diagnostic also demonstrated the technology's ability to detect more than one target at once. In that experiment, the researchers developed a multiplexed system for the detection of three orthogonal drug-resistant bacterial species, envisioning that it could be worn on the lab coat of a doctor doing rounds in a hospital.

"Maybe this is integrated into a lab coat or some sort of clinical garment so that if they are walking around and they get exposed to sputum or blood from a patient and that's infected, they don't go around spreading those bugs out," Soenksen said.

How far the sensor could be multiplexed is a question the researchers have yet to answer. Soenksen estimated one sensor could probably be configured to detect at least a dozen targets at once. Nguyen also noted that multiplexing strategies would depend on the logic of what's being detected, adding that face masks could be configured to detect not only SARS-CoV-2 but also multiple variants at once.

"There's no physical reason why we can't do that," he said.

In addition to the question of multiplexing, there are still a couple of parameters left to work out. For example, the sensors are currently a one-use only product, Soenksen said, though he and Nguyen are working on a way that the fabric could be washed and reloaded with reagents for multiple uses. And the prototypes that were used in the paper were all made by hand, so the 90-minute time for a result on the mask sensor was a conservative estimate based on the highest result they got from all their tests, Nguyen said. If the sensors were optimized through manufacturing, it might be possible to get that time down to an hour or less for SARS-CoV-2. Testing times varied for the other targets, depending on the amount of sample splashed onto the sensor and the targets themselves.

Those challenges aside, the researchers are already considering their options for commercializing the sensors and getting them into patients' hands. They've submitted a patent application and are looking for commercial partners who could help them figure out a mass manufacturing process. Manufacturing would help optimize the device itself and bring pricing down to minimal levels. Right now, Nguyen estimated that the price per sensor is about $5, based on the prototype. But he added that with the minimal levels of reagents and enzymes needed to make it work, that could probably come down to $1 or $2 per sensor.

They're also considering which route they might take for regulatory approval. The most likely option might be to have the sensor platform approved as a medical device, Nguyen said, though Soenksen noted that the best strategy might be to go the route of a CLIA-waived in vitro diagnostic determination. It's likely that because the sensor requires a button-push activation that there might be an intermediate step of seeking regulatory approval for use in a doctor's office before seeking ultimate approval for use in the home, they added.