No abstract

In this study, a sensitive, yet robust, biosensing system with real-time electrochemical readout was developed. The biosensor system was applied to the detection of carcinoembryonic antigen (CEA), which is a common marker for many cancers such as pancreatic, breast, and colon cancer. Real time detection of CEA during a medical procedure can be used to make critical decisions regarding further surgical intervention. CEA was templated on gold surface (RMS roughness ∼3-4 nm) coated with a hydrophilic self-assembled monolayer (SAM) on the working electrode of an open circuit potentiometric network. The subsequent removal of template CEA makes the biosensor capable of CEA detection based on its specific structure and conformation. The molecular imprinting (MI) biosensor was further calibrated using the potentiometric responses in solutions with known CEA concentrations and a detection limit of 0.5 ng ml(-1) was achieved. Potentiometric sensing was then applied to pancreatic cyst fluid samples obtained from 18 patients when the cyst fluid was also evaluated using ELISA in a certified pathology laboratory. Excellent agreement was obtained between the quantitation of CEA obtained by both the ELISA and MI biosensor detection for CEA. A 3-D MI model, using the natural rms roughness of PVD gold layers, is presented to explain the high degree of sensitivity and linearity observed in those experiments.

The molecular imprinting technique has tremendous applications in artificial enzymes, bioseparation, and sensor devices. In this study, a novel molecular imprinting (MI) biosensor platform was developed for the detection of a broad range of biomolecules with different sizes. Previously this method has been applied to 2D molecular imprinting, where the height of the self-assembled monolayer (SAM) of around 2 nm limited the maximum dimensions of the molecule that can be imprinted to create template-shaped cavities. In order to match the size of the imprinted molecules with the height of the SAM, we propose a model for 3D molecular imprinting where the analyte is sequestered within a niche created by the surface roughness. The SAM is assembled on the walls of the niche, forming a 3D pattern of the analyte uniquely molded to its contour. Surfaces with multi-scale roughness were prepared by evaporation of gold onto electropolished (smooth) and unpolished (rough) Si wafers, where the native roughness was found to have a normal distribution centered around 5 and 90 nm respectively. Our studies using molecules with size ranging on a nanometer scale, from proteins of a few nanometers to bacteria of hundreds of nanometers, showed that when the size of the analyte matched the roughness range of the gold surface, the molecular imprinting process was optimized for the best biosensing performance. After optimization, the MI biosensor platform enabled the identification and quantification of a broad range of biomolecules with great discrimination abilities. Hemoglobin under different pH values and several mutated fibrinogen molecules can also be well differentiated through the test.