Chemical mapping of the PPIase interactome in C. elegans - development of a systems biology toolbox

Caenorhabditis elegans (C. elegans) is a free living round worm comprising only 959 cells. Its genome produces 19,000 proteins (compared with some 25,000 from the human genome). We will introduce chemicals into the worms and look to see how they respond, particularly in terms of changes in their proteins. Small molecules which might be hormones, drugs or toxins can have profound effects on biological systems. Changes are usually caused by the small molecules (ligands) binding directly to a particular 'target protein' and stopping it from performing its normal function. These types of molecules are known as inhibitors. Currently we have a very poor understanding of the 'knock-on' effects in the cell when a given protein is bound to a ligand. We want to design molecules that bind to particular families of proteins in C.elegans and then modify their behaviour. We will monitor the effect of the chemical on the whole worm and identify which of the other proteins are affected. In this project we will only look at one family of proteins, the immunophilins, for which we have considerable background knowledge. Over 20 different proteins belong to this family. We know that they can act as enzymes and also help protein folding. We have synthesised new classes of molecules that inhibit these immunophilins and also show biological activity in the C.elegans. We have already put fluorescent-labels on some of these inhibitor molecules so that we see (using an optical microscope) where the molecules are taken up by the worm. We will see how proteins in worm change when they are exposed to these inhibitor ligands. By testing the many hundreds of different inhibitors that will be made in this project, we will identify molecules that have specific biological effects. Many of the proteins in C. elegans are similar to those in mammals, and so we will also learn about the effect of these chemicals in animal systems. Our data from worms could have a major impact on human and veterinary biology in areas like gene regulation, developmental biology, and drug discovery.

Our current understanding of the complex effects of biologically active small molecules in multicellular organisms is primitive and the work outlined here will provide a framework to develop a systematic study linking protein chemo-regulation and cellular effects in C. elegans. The long-term goal will be to generate data that maps out the effect of representative compounds (currently some 3 million available compounds) on the biological activity of the 19,000 proteins in C. elegans. In this project we will focus on a smaller group of some 25 PPIase proteins (Peptidyl Prolyl Isomersaes) present in C. elegans. The biological hypothesis for this initial program is that PPIases play a central role in cellular regulation and that chemical inhibition of different members of the PPIase family will enable us to manipulate biological function in C. elegans. There is an intriguingly disparate body of published data showing the importance of PPIases in protein folding, immunosuppression, protein trafficking, chaperone activity, hormone regulation, HIV infectivity and collagen formation. The approach outlined below will use C. elegans as a model organism. Designer molecular tags will be used as probes to modulate and monitor the biological activity of PPIases in the different cells in C. elegans. Fluctuations in the mRNA levels and expression of other genes affected by changes in PPIase activity will be monitored using DNA microarrays and quantitative mass spectrometry afforded by stable isotope amino acid labelling. The resulting map of chemically induced biological effects and PPIase-protein interactions will determine the main roles of PPIases in the cell and also their interaction with other regulatory pathways. Proteins identified in the PPIase interactome will be mapped both temporally and spatially throughout the life cycle of the C. elegans to provide information on the composition of the animal at different stages of development. We already have an excellent starting point for this project. Database mining methods have been used to identify dimedone as a template for the synthesis of 3 families of compounds that inhibit cyclophilin in the low micromolar range. Binding data using mass spectrometry, enzymatic assay and spectroscopic binding along with X-ray structural data provide strong Structure Activity Relationships for the design of further compounds with improved solubility and enhanced binding. These compounds already show significant biological activity against C. elegans. We were the first to describe the gut deformities and cuticle shedding defects caused by cyclosporine A 1, and, interestingly, the new families of inhibitors show similar phenotypes. Proteins that are up or down regulated by the various chemotherapies will be identified using microarray and mass spectrometric techniques. We will also make use of our database mining software (Lidaeus) to identify potential new and novel PPIase interactome inhibitors suggested by virtual screening (eg steroids as FKBP inhibitors 2) will also be tested The large amounts of biological and chemical data thus generated, will be linked with the available chemical information in relational databases. We will build on our existing relational database (EDULISS) containing data on 2.5 million available compounds and develop a web-based resource that will offer general public access. A graphical representation of the PPIase interactome will be developed to provide information showing which proteins talk to each other and, most importantly, how the individual chemicals attenuate communication between pairs of proteins in the (PPIase) interactome.