From owner-chemistry@ccl.net Thu Apr 9 09:48:00 2009 From: "Moreau Yohann yohann.moreau.^^^.lcp.u-psud.fr" To: CCL Subject: CCL: transition states of enzymes Message-Id: <-39048-090409060053-8171-LDunhSfTTEZ+CigbSnRioQ^^^server.ccl.net> X-Original-From: Moreau Yohann Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="iso-8859-1" Date: Thu, 9 Apr 2009 11:15:10 +0200 MIME-Version: 1.0 Sent to CCL by: Moreau Yohann [yohann.moreau..lcp.u-psud.fr] Hi, Alexandra,=20 Here's my 2cents : QM/MM methods are widely used so as to take into account all the environme= nt=20 effects on the active site and the reaction. If you wish to study a particular reaction, you have to (once you have=20 obtained your structure) define an active part. This part includes all the= =20 atoms described by QM (but not only) and is of course located at the active= =20 site of your enzyme. The remaining atoms (i.e. the "non-active" part) are=20 frozen during all the following calculations (this is environment). Performing a PES scan is then possible by letting the active part to relax= =20 "around" the coordinate you explore. The active part is supposed not to cha= nge=20 drastically of shape during the scan, provided you already optimised it bef= ore=20 (which ensures you start from a -at least local- minimum of the PES). If you perform a scan with sufficiently small steps, the highest step in=20 energy will give you a quite good idea about the barrier height you wish to= =20 determine as it can give you a good idea of the transition structure. Starting from this point, you can look for the TS. If your active part is=20 small enough, you can compute the hessian of the active part in the field o= f=20 environment directly and then go on with usual TS search algorithms.=20 If your active part is large (this is generally the case) you should perfor= m=20 calculations on a reduced set of atoms, including the very few ones directl= y=20 involved in the reaction. This can work very well if you starts from a=20 structure you know to be very similar to the TS. You have to choose the goo= d=20 atoms. To make a clear answer to your question : characterising a TS in an enzymat= ic=20 reaction is feasible and of course important. You will be interested by onl= y=20 few frequencies if not only (the imaginary) one, the remaining ones should = not=20 be relevant. I would add, that, using the static approach used for studying the reactivi= ty=20 of enzymes by the mean of QM/MM, computing frequencies for the whole system= is=20 totally meaningless, of course. My answer is quite long, however, I hope this will help you. Best regards,=20 Yohann : > Sent to CCL by: alexandra.marques=3D-=3Dfc.up.pt > Hi, > > I have been read some articles about the search for transitions state > structures in enzymatic reactions but I still have some doubts: > 1) It seems that when the full enzyme is intended to be used the most > widely used approach is to use QM/MM to model the enzyme and perform a > relaxed PES scan along some reaction coordinate. Is this correct or > there are better methods? > 2) Imagining that an approximate transition state can be identified in > a PES scan with the full enzyme, then, which method shall be used to > prove that this is the correct transition state? I am asking that > because I think a frequency calculation or even the IRC method cannot > be applied for a large system? > Can anyone help me please. > Thanks a lot > Alexandra > > > ------------------------------------------------------------------------- > A FCUP utiliza o sistema open source de webmail Horde/IMP (www.horde.org) > Visite: http://www.fc.up.pt/ http://info.fc.up.pt/ > > > =2D-=20 Yohann Moreau Laboratoire de Chimie Physique Universit=E9 de Paris-Sud B=E2timent 349 91405 Orsay cedex =46rance Tel : +33 (0) 1 69 15 73 98 =46ax :+33 (0) 1 69 15 61 88