Often an anxious event, false positives, a test result which incorrectly indicates that a particular condition or attribute is present, can take a psychological toll.

To increase the accuracy of medical screening and reduce the incidence of false positives, UC Santa Barbara researcher Tracy Chuong, along with UCSB chemistry and biochemistry professors Martin Moskovits and Galen Stucky and Stanford chemical engineering professor Tom Soh, designed a biomedical assay that eliminates the readout of these faulty results.

Not only does the assay provide greater accuracy, it also reduces the wait time for results. It is an improvement on the popular enzyme-linked immunosorbent assay (ELISA), which detects concentrations of proteins that correlate with conditions from pregnancy to allergies to infectious disease.

The mechanism of an ELISA starts when blood or other biological fluid is dropped onto plates with little wells whose surfaces bind antibodies and proteins. A second binder is added to these wells, which are tagged with “reporter molecules” that will activate and usually change color if a target protein is detected. The test can vary in the number of steps, intermediate steps and their sequences, or in the types of detection molecules or enzymes, depending on the information being sought.

The ELISA remains an indispensable tool for detecting specific target molecules in complex biological samples. ELISA’s reliability results from the use of two affinity reagents to capture and sandwich the target protein as a requisite for generating a reporter signal, an approach that reduces false-positive signals. This “sandwich” assay structure has inspired many groups to develop hybrid versions of ELISA that incorporate novel reporting techniques such as polymerase chain reaction (PCR), optical detection and surface-enhanced Raman spectroscopy (SERS) to further improve sensitivity.

“We’re not trying to be the best assay,” Chuong said, referring to tests that can get results out of ever more miniscule sample amounts. “Instead, we looked at how we can make what’s already there work better by reducing erroneous results.”

Gold nanoparticle probes functionalized with TBA-29, an enzyme, and the Raman label NBT, a chemical compound, capture the target in solutions that contain various concentrations of thrombin and a nontarget protein, albumin. These assemblies are in turn captured by a film functionalized with TBA-15 to form SERS hot spots, which generate a true positive. / courtesy of pnas.org

Chuong and her colleagues designed a sandwich-style protein detection assay that uses two Raman reporters, each conjugated with a distinct affinity reagent. This approach is less susceptible to false positives, resulting in a significantly lower limit of detection (LOD) compared with its single-reporter–based analog.

To use SERS to detect binding events, they connect one affinity reagent to gold nanoparticles and the other to a gold metal film. The gold nanoparticles and film sandwich structure formed in the presence of the target molecule becomes a SERS hot spot, generating a strong and unambiguous signal that indicates a true positive.

“What we realized was that in the process of doing this assay, everything comes down to that one binder and the reporter,” Chuong said.

If the reporter happens to bind to other surfaces or other proteins, she said, it can indicate an abnormal concentration of target proteins that correlate with a disease. Other substances in the test sample may also prompt the reporter to activate.

“Then you suddenly have a false positive,” she said.

Key to this technology are gold nanoparticles, infinitesimally tiny bits of gold whose electromagnetic properties enhance the chemical signature of whatever molecules happen to be close by.

“When we have these gold nanoparticles come together, there are these interesting little electromagnetic fields that are generated within these gold surfaces when light hits,” Chuong said. The chemical signal of the labels next to the binder gets amplified by these nanoparticles, she explained.

“You can see everything within that binding spot,” she said, and the signals can be compared from one area of binding to another. Reactions that are not the ones being sought will not producethe same signals as the true positives and can be weeded out in the analysis.

With this assay, they were able to produce a lower LOD while still resulting in a faster assay than ELISA with the capacity to discriminate between true and false binding. In this way, the researchers anticipate that future iterations of the assay will deliver both greater speed and higher sensitivity.