|John Sheridan, Dickinson College
Mentors: Dr. Robert Murphy
Multiple Cell Type Imaging and Analysis with Subcellular Protein Features
Fluorescence microscopy has been widely used to visually study subcellular location patterns of organelles. However, visual assessment of images by a human observer can be tedious and inaccurate, while the use of an automated system can provide a more objective and accurate depiction. A method to statistically analyze a fluorescence micrograph by extracting subcellular location features (SLF) was previously established. However, this approach has not been applied to analyzing how organelle patterns vary in cell lines. The goal of this project is to study the location patterns of specific organelles —the Golgi complex, lysosomes, mitochondria, and the nucleus— in varying cell types so that we may learn how organelle patterns vary between different cell lines. To do this, we have chosen 12 cell types from various lineages. In order to visualize these organelles we labeled the cells with different probes: Bodipy FL (Golgi), LysoTracker (lysosomes), MitoTracker (mitochondria), and Hoechst (DNA). We acquired three-dimensional images for multiple probes simultaneously using fluorescence microscopy. From these images we can computationally extract various SLFs, and then use these features to classify organelle patterns in single and multiple cell lines.
|Kalin Vasilev, Gettysburg College
Mentor: Dr. Jon Jarvik
Attacking the Secretome of Mammalian Cells Using CD-Tagging
We proposed a novel method for the identification and description of a cell's secreted proteins. The secretome can serve as a biomarker to identify and monitor different cell lines and specifically tumors (Gretzer and Partin, 2003). Secreted proteins can be effectors of metastasis (Jessani et al., 2002), and they, or their receptors, can serve as targets for therapeutic drugs (Ross and Fletcher, 1998). Using the highly sensitive CD-tagging protein trapping approach we identified numerous NIH 3T3 cell secreted protein products. In these experiments a self-inactivating CD-tagging retroviral vector was used to tag proteins with luciferase moiety (luc) and green florescence protein (GFP). The procedures applied were direct modifications of Jarvik (2002). Luciferase tags were used since they provide sensitive and rapid method for identifying and quantitating secreted proteins using a standard plate reading luminometer. The evaluation of twenty nine 96-well plates produced 15 luciferase tagged secreted proteins. Six of them were imaged using spinning disk confocal microscope allowing us to characterize some as low secretors and some as strong. The molecular analysis of the secreted proteins will define their exact nature. Some important questions related to the sensitivity of the assay and its success rate are still to be further examined. The successful implementation of this technology is allowing us to apply it to the study of cancerous cells.