Lab In A Lunchbox

Portable Biosensor System Detects Toxins And Biomolecules With High Sensitivity

By Tiffany Straza

Many of our favorite local traditions are directly affected by toxins in the environment. Anyone who has visited the beach, perhaps to dig clams, and has seen "No Harvesting” signs has experienced the consequences. Shellfish harvesting is stopped during domoic acid-producing algal blooms, when shellfish may store large quantities of the neurotoxin that causes Amnesic Shellfish Poisoning, or ASP. Predicting outbreaks and monitoring environmental health requires measuring the concentrations of toxins in certain conditions.

However, most analyses of toxins are done in laboratories–after time has been spent collecting the samples and transporting them back to a study facility. It is now possible to complete those analyses in seconds to minutes using a sensor system developed at the University of Washington (UW) with support from the Washington Sea Grant Program.

Small biosensors capable of detecting chemicals in the parts-per-trillion range have been developed by Sinclair Yee, professor of electrical engineering (EE), Clement Furlong, research professor of Medical Genetics, and Tim Chinowsky, research assistant professor of EE. A self-contained, lunch-box sized system, containing several of the dime-sized sensors along with interfacing programs, can detect toxic domoic acid at 100 parts per billion, cortisol (a hormone) at 271 parts per trillion, and protein toxins as low as 2.8 parts per trillion.

The sensitivity of this technology is important, but just as important is usability. Previous sensor systems were too bulky and expensive for practical field work. Some of the main goals for these biosensors were to make them light-weight, portable, battery-operated, and user-friendly. "Basically, ‘push button here'; in other words, the user will be so busy with other things that they aren't going to worry about which button to push. Just push one button,” says Yee. "One big button, so they can do it with a glove on,” laughs Furlong.

The researchers have partnered with Texas Instruments Inc. (TI). TI now builds the sensor chip, and the researchers build the system around that main component. "This contains the sensor, [and] contains the optical system to operate the sensor itself,” says Yee. The sensors work using antibodies attached to the gold-plated surface of the sensor.

A sensor is prepared with an antibody specific to the chemical to be tested. The chemical binds to the antibody, changing the optical properties, or refractive index (RI), of the surface. Changes in the RI are converted to electrical signals, recorded by software in the system.

Because the sensors rely on the contaminant binding to an antibody specific to that contaminant, a limited number of chemicals can be tested per biosensor. A sensor system packages several specific biosensors together, enabling multiple-contaminant analysis. The portable systems now test three chemicals per chip, and have eight chips in the system, enabling the simultaneous testing of 24 analytes.

"It's very much ready for second-tier testing. We're trying to now establish some kind of collaboration with military laboratories to do some live-agent testing with the system, because within the university we cannot do anything like that. It's just not equipped,” says Yee. "The collaboration with the different government labs is a crucial portion of how to finalize the final instrumentation, because the feedback of the laboratory results in the army would allow us to do some modification [and] redesign.”

Three main labs are involved in the collaboration right now. Having several labs involved is necessary because of the variety of agents testable by the sensors. "Each agent requires different modification of the surfaces, and each one requires its own expertise,” says Yee.

Once the sensor is built, though, it has a shelf-life of at least one year with glass storage. One year is the longest time period tested so far, and the sensor worked just as well as when it was new, says Yee. While time of use depends on the agent being tested, it is possible to complete 50-60 cycles for months with the same chip.

Redesigning the system is the focus now, with special emphasis on reducing the component cost. At present, the sensor system costs $20,000 to $25,000. Within a few years, they hope to reduce that to $5,000 to $10,000. The operating costs, however, are already quite low. The cost is a few cents per assay, in part because there is no need for other chemical reagents; the sensor surface does all the work itself.

"If you're mapping a spill or a spread of an agent, you don't have to do grab samples, send them back to the lab and wait. So there's a huge value in the real-time analyses in containing both spills and releases," says Furlong, "and it's inexpensive to do it with this device.”

For more information, visit http://www.wsg.washington.edu/publications/seastar/ archive/storyarchives/sensor.html

Tiffany Straza is an undergraduate senior in oceanography at the University of Washington.