Digital detection of biomarkers for low-cost, high-sensitivity diagnostics
Until about a hundred years ago, clinical diagnosis relied mostly on medical history and physical examination. However, there are many diseases that exhibit similar symptoms, making it difficult to achieve direct diagnosis based on clinical presentation. Doctors have always sought complementary validation, and rudimentary clinical tests date back thousands of years, when the color and odor of urine was utilized in diagnostics. In modern medicine, in vitro tests are a cornerstone of clinical practice, with the sensitivity of standard immunoassays measuring protein biomarkers at picomolar concentrations.1 However, while this level of sensitivity is sufficient for the diagnosis of infectious diseases when clear symptoms are present, it falls short—perhaps by a factor of many thousands—for the detection of proteins that are important in cancer,2 neurological disorders,3 and the early stages of infection.4 Devastating epidemics have exposed the limitations of current technologies, and emphasized the importance of continuing innovation and refinement of in vitro diagnostics, especially at an affordable cost.
Synergistic collaboration of medicine with engineering has been crucial for advances in disease diagnostics. Perhaps one of the most exciting recent technological developments in biomarker analysis is single-molecule counting or digital detection, an approach that provides resolution and sensitivity beyond the reach of ensemble measurements.5, 6 Digital detection not only provides very high sensitivity, but also has the potential to make the most advanced disease diagnostic tools available at low cost. One can draw an analogy from the transition of the audio recording industry to digital media. Prior to digital recording, the sound quality from vinyl analog media depended on the sophistication of the device accurately reproducing the precise analog signal. Following the digital recording of audio onto compact discs, sound quality no longer depended on the reader because it is easier to measure the presence or absence of a signal than to detect the absolute amount. In a similar fashion, detection of single particles, when possible, is easier than the precise measurement of the ensemble quantities.7
We demonstrated an optical imaging technique, known as the single-particle interferometric reflectance imaging sensor (SP-IRIS) that can detect single nanoscale particles.8 The technology is based on interference of light from an optically transparent thin film—the same phenomenon that gives rainbow colors to a soap film when illuminated by white light. In SP-IRIS, the presence of particles modifies the interference of light reflected from the sensor surface, producing a distinct signal that reveals the size of the particle (which is not otherwise visible under a conventional microscope). Using this simple platform, we demonstrated label-free identification of various viruses in multiplexed format in complex samples. Size discrimination of the imaged virions allowed for rejection of non-specifically bound particles, thus we achieved a limit-of-detection competitive with state-of-the-art laboratory technologies. We showed simultaneous detection of Ebola and Marburg vesicular stomatitis virus pseudotypes in serum or whole blood9 and recently demonstrated in-liquid real-time detection of viruses in a microfluidic cartridge.10 SP-IRIS has also shown promising results for detection of protein11 and DNA molecules labeled with small gold nanoparticles, showing attomolar sensitivity and meeting the requirements for most in vitro tests.
SP-IRIS offers a robust and low-cost platform for detection of a variety of targets such as proteins, nucleic acids, and whole viruses in a simple assay format and with high sensitivity (at single pathogen or molecule resolution) from complex samples with minimal sample preparation. Figure 1 shows a summary of the concept, illustrating multiplexed detection of viruses and molecular biomarkers. Through integration with microfluidics, it may be possible to implement SP-IRIS for critical point-of-care applications (such as detection of biomarkers from unprocessed blood) for early diagnosis or sample-to-answer molecular diagnosis. This would be particularly advantageous for infectious disease control and containment during an outbreak in resource-poor rural settings.
Recently, we demonstrated the feasibility of DNA-directed antibody immobilization for detection of individual viruses on a microarray surface using the SP-IRIS platform.12 DNA microarrays are easier to prepare than protein microarrays and are highly reproducible. Furthermore, DNA chips can be stored at room temperature for an extended period of time without denaturation. When viral diagnostics are needed, especially in urgent outbreak situations, the DNA microarrays can be functionalized quickly.
The digital detection of proteins and nanobioparticles using microarray technologies is poised to impact diagnostics. The future of all analytical measurements may be based on single-molecule detection and counting, representing the highest achievable signal fidelity. With SP-IRIS, we have shown that digital detection is possible at low cost.
Our continuing development efforts will target making the detection platform and assay more robust and portable. We are exploring a disposable passive cartridge, improving the shelf-life of assay chips, and improvements in instrument automation, image acquisition, and analysis. Furthermore, we are collaborating with biosafety level 4 laboratories to test the detection platform with hot virus samples.
M. Selim Ünlü is a distinguished professor of engineering, appointed in electrical and computer engineering, biomedical engineering, physics, material science and engineering, and graduate medical sciences. His research interests are in the areas of nanophotonics and biophotonics.