(786a) Improving Detection Specificity of Label-Free Optical Biosensors

Ce Shi¹, Simin Mehrabani¹ and Andrea M. Armani^1,2*

¹Mork Family Department, University of Southern California,

²Ming Hsieh Department of Electrical Engineering-Electrophysics, University of Southern California,

Los Angeles, California 90089, USA

*Corresponding author:armani@usc.edu

Whispering gallery mode optical cavities fabricated on a silicon substrate from silica and silicon have demonstrated promise in the field of biodetection, both for fundamental biological investigations and as a diagnostic tool.  For example, arrays silicon microring devices have performed detection and discrimination of multiple cancer markers in serum and silica resonators have been used to determine binding coefficients of proteins.^1-2  In both cases, the cavity surfaces were functionalized to detect the molecule of interest, and detection was performed by monitoring the change in the resonant wavelength of the cavity.  Therefore, the specificity of detection relied on the accuracy of the functionalization molecule while the sensitivity was the result of the device performance or photon lifetime in the cavity.  While there are numerous methods of achieving high performance optical cavities, often accurate discrimination between molecules is not possible, as multiple analytes can bind to the same targeting moiety, resulting in similar detection signals.  Therefore, a complementary and simultaneous analysis method would further improve the reliability of this method.

This improvement can be achieved by analyzing both the resonant wavelength shift and the binding off-rates of the materials.³  In the present work, toroidal optical microcavities are covalently functionalized with biotin, which binds robustly to streptavidin.  Using solutions containing either free streptavidin or streptavidin functionalized polystyrene beads, a pair of experiments are performed to demonstrate this approach.  First, free streptavidin in solution is detected and the off-rate is determined.  In this control experiment, a single resonant wavelength shift is observed, as expected.  Then a mixture of the two solutions is injected and detected.  Surprisingly, a pair of shifts is detected.  Based on mass transport calculations and a comparison with the initial control data, the first peak is the free streptavidin and the second peak is the polystyrene beads.  Additionally, the two complexes have different dissociation constants, allowing further verification.

This work demonstrates the feasibility of combining resonant frequency shift with a kinetic and mass transport analysis to improve the reliability of resonant cavity-based detection.³  This approach is very straightforward to implement in systems which involve fluid flow and could enable these devices to be used in solutions with high levels of contaminants.

   (1)     Washburn, A.; Gunn, L. C.; Bailey, R. C. Analytical Chemistry 2009, 81.

   (2)     Soteropulos, C.; Hunt, H.; Armani, A. M. Applied Physics Letters 2011, 99, 103703.

   (3)     Shi, C.; Mehrabani, S.; Armani, A. M. Optics Letters 2012, 37.