Fragment based discovery is a fairly new but also fairly well established technique. There is a difference between fragment screening and targeted library screening, however notes below generally apply to both:
Libraries depending on what you want to discover you should try to obtain a targeted (fragment?) library of useful size both in terms of each molecule and in terms of number of molecules. In the best case scenario you should be able to select a set of compounds that *you* (and your chemi-informatics specialist) find attractive. This may not always be possible. GRI (Cincinnati, Ohio) has a fairly good collection of compounds that was inherited from P&G Pharmaceuticals and since then has been expanded and pruned they provide selection and compound service to academics and industry folks alike I have no special interest in them but I do know more than a few people at that facility and can recommend a contact person on request. Screening X-ray screening is doable although painful. If you can imagine solving ~300 crystal structures of *the same protein* and not be sad when 298 structures come out empty try it. Be prepared to write scripts for data processing, solution, and refinement and make sure that your beam line of choice is equipped with a nice & fast robot for sample changing. This method is ideally suitable for proteins that make reproducible, sturdy, well-diffracting and readily soakable crystals. More common screening technologies include: biochemical assays, NMR, SPR, X-ray induced fluorescense, fluorescense-replacement/quenching, and so forth. Every technique has advantages and (usually serious!) drawbacks. For weak binders confirmation is achieved via combination of two or more techniques because each individual one may give noisy results (typical compromise of HTS). You have to target techniques that can detect weak binding if you want to discover fragments or imperfect ligands. If youre interested in finding tight binders and have a rough idea what set of compounds you should be targeting then methods need to be adjusted (for instance commonly used NMR techniques are often not very good for compounds that bind too tightly). Finding tight ligands is technically easier but is less likely to succed for a protein in general unless you happen to have a reasonably characterized protein class and youre looking for a specific substrate or ligand out of a small group. MS-based techniques for compound discovery can be tricky because various compounds may or may not fly depending on their nature and the experimental conditions. I can recommend an expert in this if you contact me directly. Subtractive MS is particularly tricky unless you couple it with some other form of detection. Youre entering a fascinating field where a huge set of options and techniques is available. The trick is to find the right ones for the job. An hour in the library saves a month of lab work (although this should now be adjusted to the 21st century version: fifteen minutes online saves $5000 spent on a CRO). Artem _____ From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Jürgen Bosch Sent: Tuesday, October 27, 2009 10:53 PM To: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] small molecule library Hi Christopher, have a look at this paper and the references. Using fragment cocktail crystallography to assist inhibitor design of Trypanosoma brucei nucleoside 2-deoxyribosyltransferase. J Med Chem (2006) vol. 49 (20) pp. 5939-46 The SEC after soaking is a bad idea, you will loose your fragment as it is most likely a low micromolar binder. A better way to approach this is to use a filtration device say with 3 kDa cutoff. Concentrate your protein + ligands and in a separate tube just the ligands without protein. Then run both samples over a mass spec and see if you can detect a specific decrease in one or more of your ligands. Jürgen P.S. wasn't sure what you mean by biological relevant DeCode has a library called Fragments of Life, otherwise Zenobia is not bad. On Oct 27, 2009, at 10:58 AM, c.weinert wrote: Dear all. I am looking for a library of small (biologically relevant) compounds to test binding to a protein. Does somebody know, if there is a company that sells such a library for fragment based library screening? (to a reasonable price). And does anybody has experience with screening such libraries? I assume that one approach would be soaking the protein with the compounds, followed by SEC and MS analysis. Anybody tried directly soaking crystals followed by X-ray analysis? Thanks alot already in advance for your help. Sincerely, Christopher - Jürgen Bosch Johns Hopkins Bloomberg School of Public Health Department of Biochemistry & Molecular Biology Johns Hopkins Malaria Research Institute 615 North Wolfe Street, W8708 Baltimore, MD 21205 Phone: +1-410-614-4742 Lab: +1-410-614-4894 Fax: +1-410-955-3655 http://web.mac.com/bosch_lab/ <http://web.me.com/bosch_lab/>
