
The reliable and fast identification of different types of cells and protein detection at low concentrations are of utmost importance in many biomedical applications. The detection of cancer cells in bio-fluids such as urine and blood could be crucial to detect patient response to therapy, to monitor drug resistance effects, as well as for early identification of disease recurrence, and ultimately for a personalized medicine approach. We developed bright, spectrally rich, multiplexing SERS Biotags (SBTs), composed of a silver nanoparticle dimer core, and an affinity tag coating to bind to surface receptors on cells. The SERS spectrum of an individual SBT acts as a unique barcode that is easily differentiable in a composite SERS spectrum originating from many tags. The SERS intensities achieved are comparable to fluorescence and can all be excited with one laser. We have now developed a platform that employs SBTs for the identification of individual cells based on the ratio between surface receptors. We synthesized a cocktail of two to four SBTs, each one targeting a different surface receptor on the cell's membrane, with one SBT being a universal control. Point-by- point 2D Raman maps depicting the ratio of each receptor to the universal control were constructed with subcellular resolution from cells simultaneously incubated with the SBTs while in suspension, thus simulating the cells' capture from blood. We demonstrate the identification of individual cells by spectral unmixing of the Raman signature through a quantitative model based on classical least squares signal deconvolution that our lab developed. In summary, we demonstrate the multiplexed identification of cancer cells by SERS subcellular imaging and quantitative ratiometric analysis of surface receptor expression using up to four SBTs simultaneously. In protein assays the presence of signals when the target is not present (false-signals) is a major factor limiting an assay's detection threshold. We showed that by using two tags simultaneously and applying chemometric analysis, true binding events can be separated from false-signals, thus improving the limit of detection of the assay.

The reliable and fast identification of different types of cells and protein detection at low concentrations are of utmost importance in many biomedical applications. The detection of cancer cells in bio-fluids such as urine and blood could be crucial to detect patient response to therapy, to monitor drug resistance effects, as well as for early identification of disease recurrence, and ultimately for a personalized medicine approach. We developed bright, spectrally rich, multiplexing SERS Biotags (SBTs), composed of a silver nanoparticle dimer core, and an affinity tag coating to bind to surface receptors on cells. The SERS spectrum of an individual SBT acts as a unique barcode that is easily differentiable in a composite SERS spectrum originating from many tags. The SERS intensities achieved are comparable to fluorescence and can all be excited with one laser. We have now developed a platform that employs SBTs for the identification of individual cells based on the ratio between surface receptors. We synthesized a cocktail of two to four SBTs, each one targeting a different surface receptor on the cell's membrane, with one SBT being a universal control. Point-by- point 2D Raman maps depicting the ratio of each receptor to the universal control were constructed with subcellular resolution from cells simultaneously incubated with the SBTs while in suspension, thus simulating the cells' capture from blood. We demonstrate the identification of individual cells by spectral unmixing of the Raman signature through a quantitative model based on classical least squares signal deconvolution that our lab developed. In summary, we demonstrate the multiplexed identification of cancer cells by SERS subcellular imaging and quantitative ratiometric analysis of surface receptor expression using up to four SBTs simultaneously.

In protein assays the presence of signals when the target is not present (false-signals) is a major factor limiting an assay's detection threshold. We showed that by using two tags simultaneously and applying chemometric analysis, true binding events can be separated from false-signals, thus improving the limit of detection of the assay.