UC Irvine

Breast cancer is a heterogeneous malignancy. Profiling its molecular information helps physicians tailor treatment for individual patients and also aids scientists to further explore its underlying mechanisms. However, existing techniques, like immunohistochemistry or immunofluorescence, only allow detection of one or few targets at a time within a sample section, making them time and labor-extensive approaches and hindering comprehensive characterization of the tumor. Fluorescence lifetime is the duration before a population of excited fluorescent molecules returns to the ground state. It is an inherent property of the molecule and can be measured using fluorescence lifetime imaging microscope (FLIM). The phasor approach significantly simplified FLIM data analysis and broadened its applications in research and clinical fields. Although Phasor FLIM has been extensively employed in detecting endogenous fluorescent molecules, its application in detecting exogenous species has rarely been recorded. The objective of this work is to quantitatively identify multiple targets of interest simultaneously with the help of FLIM and phasor analysis.We first implemented FLIM and phasor analysis to distinguish three biomarkers that were restricted in different cell compartments (plasma membrane, cytoplasm and nucleus). Then, we quantitatively recovered three biomarkers that were co-expressed on the cell membrane. We observed that lifetime-based quantification accuracy was decreased if one or more probes were significantly brighter than the others. To improve quantification accuracy, we reduced the intensity of brighter probes by mixing unlabeled and dye-labeled antibodies. Consequently, resolving accuracy reached more than 80% for all three biomarkers in most cases. We concluded that FLIM enabled lifetime-based multiplexing of three biomarkers, and matching the intensity of probes was crucial for accurate biomarker quantification (Chapter 3). Next, we employed Phasor FLIM to discriminate and quantify five molecules that have clinical relevance to breast cancer diagnostics. As this biomarker panel consisted of two surface biomarkers and three nuclear biomarkers, we performed quantification for whole cells as well as specific subcellular compartments (inside or outside nuclei). The results indicate that FLIM can provide both quantitative and spatial molecular information. In addition, we developed a new algorithm that performed phasor transformation in two harmonics to unmix up to 5 components (Chapter 4). Finally, we created a new detection panel consisting of five nuclear biomarkers and exploited Phasor FLIM to identify potential drug-able targets for triple-negative breast cancer (TNBC). Moreover, we performed FLIM in a separate color channel to define the cell plasma membrane and nuclei. Lifetime-based segmentation was beneficial for tracking biomarker translocation (Chapter 5). In summary, FLIM and phasor analysis, combined with spectral colors, hold enormous potential to be leveraged as a powerful high-content molecular analysis tool for cancer diagnostics and research.