Model Systems: Yeast Peroxisome Biogenesis

The Aitchison group has therefore focused on the early aspects of signaling and gene regulatory networks that govern information flow leading to organelle biogenesis. Over the past year, the Aitchison group has focused on the following approaches to study the early events of peroxisomal biogenesis:

Establish phosphoproteomics as a quantitative tool for understanding information flow through signaling networks
To understand information flow in signaling networks, phosphorylation states throughout the peroxisome proteome were identified and the temporal changes in phosphorylation states in response to stimuli were quantified. A procedure to maintain protein phosphorylation during analysis has been developed. PMCID: PMC2812858 [Available on 2010/7/11]. Journal of Cell biology

Extend the quantitative imaging analysis of signaling molecules to the entire genome (with the Microfluidics and Imaging Core) Genomically integrated, GFP-tagged peroxisomal matrix enzymes were used to systematically analyze the effect of gene deletions on peroxisome biogenesis. In this approach, three dimensional confocal data were captured of oleate-induced yeast cells to identify deletion strains displaying aberrant peroxisomal morphology. The resulting image data captured and curated. Peroxisome biogenesis processes are presumably distributed to a number of sites throughout the cell including the nucleus. PMID: 20395639 Free Article

Model transcription factor and cell signaling networks, and interpret peroxisome images (with Computational Biology Research group and the Informatics Core)

Modeling TF networks. Modeling and simulations of the TF network regulating the oleate response has advanced to depict the trade-offs between sensitivity of the system, and its ability to filter noise. In this case, a time-frequency analysis was used.

Cell signaling networks. Thousands of peptides that are differentially phosphorylated in response to oleic acid exposure were used to identify groups of genes responding coordinately with one or several phosphorylation events. The Aitchison group is working with the Baliga group to expand approaches developed for transcriptional regulatory analysis.

Capturing and interpreting peroxisome images. Automated image analysis was applied to obtain the number of peroxisomes per cell, as well as the size and the intensity of the fluorescent signal for each peroxisome. To enable the determination of these cell-by-cell statistics, cell segmentation and K-means clustering was performed. This enabled the association of each peroxisome to a cell. An image analysis method that determines the volumes of the peroxisomes from the image stacks was also developed. Thresholding is then performed to define peroxisome areas in each slice. Combining the thresholded slices back into an image stack allows us to readily obtain the peroxisome volumes.

Analysis Scheme. Using both hand annotation and automated image analysis data was analyzed in combination with matrix decomposition. Each peroxisome was mapped to an individual cell, and the number of peroxisomes in each cell, cross-sectional area of each peroxisome, and mean intensity of the fluorescence signal for each peroxisome were determined. Using these a feature matrix was obtained that consisted of the mean of the number of peroxisomes per cell, the mean intensity, and the mean area for each knock out.

Analyze single cell responses (with the Microfluidics and Imaging Core, and the Informatics Core) In order to study the variation and noise of the peroxisomal induction response, flow cytometry was used as a method to reveal single cell heterogeneity in a population of cells exposed to oleic acid. Flow cytometry produces a multiparametric output which includes the fluorescence intensity of the reporter, forward scattered light, which correlates with cell volume and side scattered light, which correlates with cell granularity. In the oleate response, the cell size and granularity varies dramatically as part of the normal response. A regression model was developed that removes the effect of the morphological variability introduced by different cell sizes and granularities. Thus the method is resilient to heterogeneous populations, and the entire population can be observed.