Visualization of Small Molecules in Multicellular Tumor Spheroids by Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging

Multicellular tumor spheroids are a 3D cellular model system that replicate the chemical microenvironments of tumors. This cellular model has played an important role in drug evaluation and clinical chemotherapy. Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) has emerged as a powerful analytical technique to visualize molecular distributions. Work presented in this dissertation focuses on the application of MALDI-MSI for both targeted and untargeted analyses in tumor spheroids. In Chapter 2, we optimized a quantitative MSI (qMSI) method to monitor a time-dependent penetration process of a tumor therapeutic. Irinotecan (IR) was selected as a proof of concept drug. By use of a deuterium internal standard, a calibration curve was generated to quantify IR throughout a 72-hour penetration to the spheroids. Moreover, quantification was performed in the inner and outer regions of the spheroids separately. This study is the first time to obtain MALDI-qMSI of a therapeutic within the scale of a single spheroid. In Chapter 3, we optimized the conditions of spheroid fast photochemical oxidation of proteins (FPOP) experiments by performing MSI on spheroids treated with hydrogen peroxide at different concentrations and time lengths. Hydrogen peroxide is too small to be easily detected directly by MALDI but it oxidizes lipids. This lipid oxidation can be used as a surrogate to determine the penetration of hydrogen peroxide. Four lipids were selected as biomarkers according to a discriminative analysis. Putative identifications were perfomed based on database searching and tandom MS was used for more accurate identification. In collaboration with the Jones Lab, proteomic analyses were performed to study protein modifications and residue-level modification after FPOP to interrogate protein interactions within a native tumor microenvironment.Chapter 4 focuses on data analysis for MSI in collabration with the Desaire Lab. Previously, a customized machine learning (ML) method was utilized for classification. These studies demonstrated that metabolic changes in colon spheroids induced by the monoclonal antibody, Cetuximab, are detectable, but modest, at 24 hr, and substantial at 72-hour treatment. Moreover, we designed and applied a user-friendly feature selection workflow based on R for MSI data of spheroids. After developing the code in R, the workflow was tested on spheroids made from four tumor cell lines. Comparisons were performed between each two classes of spheroids with Top 100 features. The results were summarized to find 4 to 26 representative features, respectively, for each cell line. Chapter 5 includes several collaborations using MALDI-MSI to visualize the penetration of various cancer therapeutics in spheroids. Regorafenib, a multikinase inhibitor, treated SILAC spheroids were imaged to ensure penetration for a proteomic study. MSI was performed on cerulenin and TVB-2640, FASN inhibitors, treated spheroids to ensure drug penetration for lipidomic studies. In collaboration with the Hu Lab, we attempted to visualize vincristine in mouse tissues to study the influence of OATP1B2 knockout. Lastly, collaborating with the Xue Lab, MALDI-MSI was utilized to select drug candidates from a series of pyrazole-based drugs functioning dually as Wnt inhibitors and AMPK activators by evaluation of the penetration process of these novel compounds.