Notes CSBE Symposium Day 2: From Systematic to Synthetic Biology
September 5, 2007
The last day of the symposium was also very good. My notes aren’t as long this time, which may or may not be a good thing, depending on your point of view.
Today was a half-day with the last two talks containing information not really in my discipline. This meant that I didn’t follow them as well as the others, and therefore my notes aren’t much use. However, I have included the names of the authors and titles of the talks as indicators of what was discussed.
Jussi Taipale, University of Helsinki
“Systems Biology of Cancer”
How do growth factors and oncogenes regulate cell proliferation? Questions include:
+ Multicellularity: how is cell cycle regulation integrated with signals and transcriptional networks controlling differentiation?
+ organ-specific growth control
+ specificity of oncogenes to particular tissues/tumor types
Many oncogenes regulate the same processes. Cancer is a highly multigenic disease. There are only a few pheontypes to cancer. The main ones are unrestricted growth, invasion of other organs, metastasis. ~350 genes controlling essentially 3 phenotypes. They use computational (prediction of targets of oncogenic TFs) and experimental (expression profiling of cancers with known mutations) methods to identify transcriptional targets of oncogenic signalling pathways. They needed to determine the affinity of all single-base mismatch oligos for all three GLI TFs. Very often the highest-affinity is known, but not the lower-affinity sites.
Regulatory SNPS (rSNPS): placed all known SNPs into human genome and aligned against mouse to discover the impact of SNPs on binding sites and regulatory areas. rSNPs are thought to explain much of individual variation in the human population, and thus are likely to contribute to predisposition to diseases such as cancer. Application of EEL to prediction of regulatory SNPs. Initial analysis against HAPMAP data looks promising, however other data sets need to be done to confirm results.
Also, they look at transcriptional circuits regulating TFs. For screening, they initially started with flow cytometry analysis looking at Drosophila S2 cells as they have similar cell cylces to human. They found that DNA-content phenotypes are detectable with flow cytometry. They also did genome-wide pooling to analyze functional redundancy: the closest homologues for all drosophila proteins were identified using BLASTP. It doesn’t look like there’s much redundancy.
+ Systems biology of the metazoan cell cycle
They have id’ed approx 600 genes which affect the cell cycle in S2 cells. They get an 80% hit rate of known strong effectors based on alaysis of 19 different protein complexes and pathways. Approx 650 genes have been cloned to Gateway vectors for the analysis of overexpression phenotype, enzyme-substrate relationships (half-life etc), PPIs (TAP-tag, fragment2hybrid), and subcellular localizations. They also did an analysis of the transcriptional network. The transcriptional analysis includes: identification of target genes of all TFs affecting the cell cycle (whole-genome profiling after RNAi of all TFs affecting cell cycle or cell size, and determination of binding specificities of the TFs followed by EEL analysis in Drosophila species), ID of pathways affecting the activities of the TFs (whole-genome profiling of all strong hits, and clustering), ID’ing of signalling inputs to cell cycle machinery and unstable proteins that are transcriptionally regulated.
Mark Bradley, University of Edinburgh
“High-throughput chemical biology”
+ Encoded Libraries
A way to interrogate 10000 molecules on a DNA microarray: 10000 peptide compounds and 10000 tags, attached to each other via a linker. The tags allow us to ID the compound its attached to, and makes it possible to deliver the compound to a specific location on a 2D DNA microarray. peptide attached to a linker, which is attached to a tag, which is attached to PNA, which can attach to the DNA on the microarray. It is better to have a PNA/DNA than DNA/DNA.
The peptides all contain a quencher and a fluorescein donor. When a protease comes along it will cleave the peptide and liberate the quencher and give us fluorescence.
They have a 10000-member FRET-based library. Then treat with protease (3d) and put onto a 2d microarray. This is a transformation of 10000 solution assays into a 2d microarray. These are high-density, clean, arrays made with an OGT custome DNA microarray. Every PNA has a preferential “home” to go to in the array. There are 22,500 oligos on the array for replicates plus 2,500 controls. DNA is printed in random locations by OGT (Agilent) and use BlueGnome software analysis. All binding duplicates are compared.
They display the data using 40 cube plots with 1000 peptides per cube with one position defined. xyz are three different amino acids with the 4th amino acid being fixed.
Peptide Arrays and Cell Binding: Have also started using this method to identify ligands for cells.
+ Cellular Chips and Polymer Manipulation
A polymer coating provides specificity for white blood cells when removing them using filters (Sepacell) from whole blood. They have a program to identify new bio-compatible polymers for topics like prevention of binding. One approach they like is ink-jet printing. They want to do the same thing but rather than 3 colors, they want to do it with polymers or monomers.
+ Microwell Array Technology: single-cell loading and transfection
You can get 4000 wells on a microscope slide. If you seed with about 10000 cells per mL, you get >85% of wells with one cell per well. You can then propogate within the wells.
+ Future Directions
encoded proteomes for arraying all proteins; peptide arrays via inkjet printing; and more.
Jamal Tazi, CNRS, Montpellier
“Small molecule screens for splicing inhibitors”
Paul Ko Ferrigno, Leeds Institute of Molecular Medicine
“Label-free protein microarrays for systems biology”