May 2002 Research Report

Conference:

I presented a Poster Presentation at 16^th Darmstaedter Molecular Modelling Meeting on May 7-8, 2002 in Darmstaedt, Germany.

Methods I have tried:

Brownian Dynamics

Should I compute BD simulations for the SRP and receptor? We have decided no since the reaction rates are very slow in the absence of the RNA. Also, usually these calulations are done when the bound and unbound conformations are both known. If we were to calculate, it would be interesting to examine the ionic strength dependence of the protein-protein association in BD.

Quanta

I tried to use Quanta to evaluate hydrophobic patches of the two proteins but it is difficult to visualize. It appears only polarity and amino acid type can be colored.

Grid20

I am using Grid to dock the Mg²⁺ and GTP to evaluate whether the original superposition placement is okay or if there could be a better fitting conformation of the metal-cofactor.

I have set labels for each of the GTP atoms so Grid will parameterize these from grub.dat. I have probed the proteins with the corresponding probes to fit the GTP. I will try the Autogroup automated docking script.

Grasp

I tried to use Grasp at EMBL to evaluate hydrophobic patches of the two proteins but, one I read the manual and figured out how to implement r=hyd to visualize this, the program was crashing (10 minutes open each prtotein, 5 mins to move, and then it crashed... amybe something with Flicker).

Amber7

I minimized the complex (apo-form)using Amber. I set the restraint to 32.0 (using Ting's input files) and just minimized the Hydrogen placements for 4000 cycles. There were still clashes when viewed with Molsurfer. I will minimize again retraining only the Calpha at 5.0 for 200 steps and the H at 10.0 for 50 steps both the apo-forms and with Mg²⁺ and Mg²⁺-GTP. We don't want to compensate for the tight restraint with movement of the side-chain and backbone to accommodate the H's.

SDA

I ran SDA and got many possible interaction surfaces for Ffh and FtsY, most probably not reasonable because the proteins must act as GAPs for each other and so the GTPase regions must be paired. To examine only this possiblity, I set restraints to 10 Angstoms within the GTP to look for a possible surface between Ffh and FtsY in the nitrogenase iron protein positions. The surface looks possible.

Molsurfer

I have looked at the Ffh and FtsY superimposed on the nitrogenase structure with Molsurfer. There are clashes at the interface, so I minimized the complex (apo-form)using Amber. I set the restraint to 32.0 (using Ting's input files) and just minimized the Hydrogen placements for 4000 cycles. There were still clashes when viewed with Molsurfer. I will minimize again retraining only the Calpha at 5.0 for 200 steps and the H at 10.0 for 50 steps both the apo-forms and with Mg²⁺ and Mg²⁺-GTP. In Molsurfer, the red values represent hydrophilic patches and the blue hydrophobic patches. It looks like the hydrophilic match is strong on the residue level. On the atom level, the hydrophobic complementarity dominates.

NACCESS

Computed hydrophobic surface area for E. coli Ffh model and E. coli FtsY with GTP-Mg²⁺ bound. Results show that Ffh is more non-polar than FtsY. Is this meaningful?

.rsa file- relative accessibilities, 1.40 Angstrom probe

E. coli Ffh model

Absolute sums over all chains

All-atoms Total-Side Main-Chain Non-polar All polar

TOTAL 15100.5 13389.1 1711.4 9059.8 6040.7

E. coli FtsY model

Absolute sums over all chains

All-atoms Total-Side Main-Chain Non-polar All polar

TOTAL 13128.9 11554.9 1574.1 7218.7 5910.2

New Paper by D.M. Freymann (I.V. Sheptotinovskaya and D.M. Freymann, Conformational change of the N-domain on formation of the complex between the GTPase domains of Thermus aquaticus Ffh and FtsY, Biochimica et Biophysica Acta, 1597 (2002) 107-114.)

Indicates that Ffh and FtsY form a complex, albeit very slowly, in the absence of the 4.5S RNA. Their data shows that the N domain of FtsY rearranges when a complex with Ffh is formed. Complex formation is Mg²⁺ dependent. To prohibit homodimer formation, Ffh and FtsY were first incubated in the presence of GMPPCP and then were added together in the presence of Mg²⁺ and the complex was formed. Complex formation was inhibited when GDP was bound. Thus GDP makes the dissociation of the complex irreversible. The GTP dependent interaction between the SRP and receptor is mediated primarily through the NG domains.