From nglover-!at!-sfu.ca Thu Feb 12 15:11:34 1998 Received: from gremlin.chem.sfu.ca for nglover {*at*} sfu.ca by www.ccl.net (8.8.3/950822.1) id OAA28360; Thu, 12 Feb 1998 14:39:02 -0500 (EST) Received: from gremlin (localhost [127.0.0.1]) by gremlin.chem.sfu.ca (950413.SGI.8.6.12/950213.SGI.AUTOCF) via SMTP id LAA20000; Thu, 12 Feb 1998 11:33:41 -0800 Sender: nick -8 at 8- sfu.ca Message-ID: <34E34E94.167E&$at$&sfu.ca> Date: Thu, 12 Feb 1998 11:33:40 -0800 From: Nick Glover X-Mailer: Mozilla 3.01SGoldC-SGI (X11; I; IRIX 6.3 IP32) MIME-Version: 1.0 To: chemistry {*at*} www.ccl.net CC: nglover-!at!-sfu.ca Subject: CCL: Salt bridges, H-bonds, aromatic stacking in MD? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit CCLers: I am looking at the interaction of a bound peptide and a target enzyme with EM and MD simulations. The peptide is docked into the active site, and lies, extended, across the protein, plugging a 'hole' on the surface of the protein. I have three queries: --------------- Q1: Salt bridge... My major concern is that I have noticed that a salt-bridge between a surface Arg on the protein and a bound peptide Asp residue lengthens from ~4.0 ang. when docked to > 6.0 ang. during the MD portion of an EM/MD run. It seems to fluctuate by about 1 ang., but over 100 ps doesn't shorten to ~4.0 ang. again. The system is in a large solvated sphere, and calculations are run on Discover 95 with CVFF using the default water model. As per advice from MSI, the cutoffs are set thus (i.e. double cutoffs, no switching function): cutoff=13.00 cutdis=12.00 swtdis=0 cutof2=15.00 cutds2=14.00 swtds2=0 Is this observation considered 'normal' given the conditions of the calculations, or should I be concerned? ------------- Q2: Intermolecular hydrogen bonds... Similar to Q1, I also see certain intermolecular H-bonds lengthen to > 4.0 ang. (acceptor - donor) during MD simulations, and fluctuate about this longer distance (i.e. 3.5 - 4.5 ang). Again, is this considered 'normal' given the conditions of the calculation, or should I be concerned? ------------- Q3: Aromatic stacking... Can anybody indicate how successful the CVFF forcefield is at simulating aromatic-aromatic stacking interactions? I notice that phenyl rings in the active site pocket of the enzyme re-orient with respect to ligand, and appear to adopt 'favourable' edge-on, tilted-T, etc. orientations, depending on the ligand. How 'realistic' (within context) are these observations likely to be? ------------- Can anybody offer any insights into these phenomena? Thanks in advance, Nick ____ Nick Glover Department of Chemistry Simon Fraser University 8888 University Drive Burnaby, BC V5A 1S6 Canada Tel: (604) 291-4531 Fax: (604) 291-3765 email: nglover ^%at%^ sfu.ca