Investigation of biomarker-targeted SERS nanoparticles for multiplexed molecular imaging of fresh tissue specimens

Abstract

Biomarker-targeted surface-enhanced Raman scattering (SERS) nanoparticles (NPs) have been explored as a viable option for targeting and imaging multiple cell-surface protein biomarkers of cancer. A previously-developed Raman-encoded molecular imaging (REMI) technique utilizes such targeted NPs for topical application onto excised tissues to enable rapid visualization of a multiplexed panel of cell surface biomarkers at surgical margin surfaces. While it has been demonstrated that REMI may potentially be used to guide tumor-resection procedures, the strategy would benefit from improvements described in this thesis. First, an investigation into channel-compressed spectrometry revealed that up to 64 times fewer spectral channels may be used to accurately demultiplex up to five SERS NP flavors (compared to our previous methods). This strategy offers the potential for improved imaging speed and/or detection sensitivity with a low-channel count detector in future REMI systems. Next, the complexities in nonspecific accumulation, diffusion, and chemical binding of targeted NPs in fresh tissues were explored in a microscopic investigation quantifying the specific vs. nonspecific accumulation of topically-applied NPs as they diffuse into fresh tissue. The findings from this study led us to hypothesize and later demonstrate that by reducing NP diffusion, nonspecific accumulations of NPs in tissue is reduced, thereby allowing for molecular imaging of fresh tissue surfaces with higher NP ratios (targeted vs. untargeted), and that the staining can be achieved more rapidly than before (6-min topical application). A third, and final, study is presented, to help establish optimized protocols for the staining and rinsing of fresh tissue specimens for REMI using a mathematical model that incorporates multi-layer diffusion in addition to binding and nonspecific retention compartments. The goal of this final study is for this forward model to ultimately be used to enable quantitative methods of evaluating molecular expression, which could enable improved assessments of tumor margins (e.g., the use of multi-stage staining/rinsing processes to allow kinetic model fitting of data). In summary, the first two studies enable the design of more rapid molecular imaging systems, and NP agents with improved sensitivity and contrast, respectively, for rapid molecular imaging of fresh tissues suitable for intraoperative clinical settings. The mathematical model study is valuable for accurately quantifying biomarker expression levels to potentially increase the sensitivity and specificity of tumor detection at surgical margins.