A cloud of my keywords:

word cloud

powered with wordle

beta-turn prediction Experimental and theoritical characterization of type VIII beta-turn
turnpred typeVIIIeq
Orientation and dynamics of membrane peptides under hydrophobic mismatch
Comparison with solid state NMR data
Modeling of non ionic surfactants
Comparison with SAXS data
walp CiEj_SAXS
MD study of the interfacial peptide ALPS  

beta-turn prediction: COUDES

turnpred beta-turns are small protein motif which allow the protein backbone to turn back on itself. They are very important as they play a key role on the orientation of alpha-helices and beta-strands. Moreover they are frequently involved in linear epitopes, being thus important targets of anti-bodies. Predicting their presence from a protein sequence has been a bioinformatic challenge. In this project, I developed a method called COUDES that predicts the presence and the type of beta-turn from a protein or peptide sequence. The principle is to use beta-turn propensities of the different amino-acids (which is simply the trend of each amino-acid to belong to beta-turns) weighted by evolutionary information generated by the PSI-BLAST program. To enhance the efficiency classical secondary structure (alpha, beta, coil) predicted by the PSIPRED program is used. COUDES reaches a good accuracy compared to methods based on artificial neural networks (ANN) though the concept is considerably simpler. Since then, new methods based on support vector machines (SVM) were published.

Check out my COUDES web server which is hosted on the RPBS bioinformatics platform. There one can make a prediction of beta-turns from a protein sequence.

This work was done in collaboration with Alain J.P. Alix from the University of Reims Champagne-Ardenne.

Ref: Fuchs P.F.J. and Alix A.J.P. (2005)
High accuracy prediction of beta-turns and their types using propensities and multiple alignments.
Proteins: Structure, Function, Bioinformatics, 59, 828-839.(DOI)

Back to top

Experimental and theoritical characterization of type VIII beta-turn

FE landscape GDNP In this work we characterized type VIII beta-turn experimentally and theoretically using a designed peptide of sequence GDNP. CD and NMRstudies reveal that this peptide exists in equilibrium between type VIII beta-turn and extended conformations. Extensive MD simulations give a description of the free energy landscape (see figure) of the peptide in which we retrieve the same two main conformations suggested by the experiments. The free energy difference between the two conformational states is very small and the transition DG between them occurswithin a few kT at 300K on a nanosecond timescale. The equilibrium is mainly driven by entropic contribution, which favors extended conformations over beta-turns. Our results clearly show that the XXXP motif can only fold into a type VIII b-turn, which is consistent with its fairly strong propensity for this type of turn. This important finding may help for peptide design and is in line with recent studies on bioactive elastin peptides.

This work was done in collaboration with:

Ref: Fuchs P.F.J. Bonvin A.M.J.J., Bochicchio B., Pepe A., Alix A.J.P. and Tamburro A.M. (2006)
Kinetics and thermodynamics of type VIII beta-turn formation: a CD, NMR and microsecond explicit molecular dynamics study of the GDNP tetrapeptide.
Biophysical Journal, 90, 2745-2759.(DOI)

Back to top

MD Study of a membrane (WALP) peptide under positive mismatch, comparison with solid-state NMR data

WALP23 tilted This work deals with the orientation of a transmembrane model peptide (WALP23) under positive mismatch, assessed by atomistic molecular dynamics simulations. Emphasis was given to link our results to deuterium solid state NMR data of the same system under the same mismatch conditions. So far, small tilt angles were extracted from the experimental quadrupolar splittings using a geometric analysis, called the GALA method. The back-calculation of these NMR quadrupolar splittings from our simulations showed a good fit with experimental data only if several hundred of nanoseconds trajectories were considered. Some coarse-grained simulations allowed us to reach the NMR time scale (a few microseconds) and led to the same observation. For both types of simulation we found that some averaging effects may affect the interpretation of NMR data, and thus larger tilt angles than previously estimated are likely to occur.

This work is done in collaboration with

Ref: Özdirekcan S., Etchebest C., Killian J.A. and Fuchs P.F.J.. (2007)
On the orientation of a designed transmembrane peptide: towards the right tilt angle?
Journal of the American Chemical Society, 129, 15174 -15181.(DOI)

Back to top

MD Study of CiEj surfactants, comparison with SAXS data

CiEj bilayer In this work I'm studying non ionic surfactants CiEj that are very relevant to membrane proteins purification and crystallization. They are also used as templates for ultra sound agents as they can form a variety of vesicules of sizable radii. So far my goal was to build a set of parameters for simulating those molecules under the GROMOS G53a6 force field, in order to reproduce the experimental area per molecule that has been measured by SAXS experiments (small angle X-ray scattering). The next goal will be to embed membrane peptides and proteins in order to interpret experimental results at the moleculare scale.

This work is done in collaboration with:

Back to top

MD study of the interfacial peptide ALPS

ALPS (ArfGAP1 Lipid packing Sensor) is a motif involved in membrane recognition identified in the golgi-associated protein ArfGAP1. This motif is structured as an amphipathic alpha-helix once it binds to the membrane, and is capable to sense the positive membrane curvature of the outer leaflet induced by vesicle formation. ALPS adopts a parallel orientation with respect to the membrane plane, and seems to recognize defects in lipid packing as a result of the mismatch between the actual curvature of the membrane and the lipid geometry. ALPS helix contrasts from a classical membrane-adsorbing helix in the abundance of Ser and Thr residues and the paucity of charged residues in its polar face. Since this motif was observed in several proteins with different functions, it has been proposed as a general proteic motif to detect membrane curvature.

We performed molecular dynamics simulations of ALPS embedded at the water/lipid interface of explicit homogeneous phospholipid membranes. In order to mimic local packing defects, we also performed simulations within mixtures of phospholipid / diacylglycerol membranes. Our results allowed us to better understand (i) ALPS curvature sensing mechanism, (ii) the influence of the lipid packing defects upon ALPS membrane binding and adsorption, (iii) how the network of peptide-membrane interactions favor the ALPS helical structure stabilization . This work gives useful insights in the lipid packing sensor mechanism by amphiphatic helix and contributes to a better comprehension of protein-membrane interactions.

This work is done with Paula Gonzalez-Rubio (Ph.D. student) in collaboration with Romain Gautier, Guillaume Drin and Bruno Antonny (Institut de Pharmacologie Moléculaire et Cellulaire, UMR CNRS / Nice University).

Back to top

Past projects

Here are some of my past projects:

  • Folding and dynamics of peptides: Structure / function relationship of some elastin derived peptides
  • Folding and dynamics of peptides: Preleminary studies of abductin peptides
  • Study of Acyl-Phosphatase unfolding by molecular dynamics (MD)

Back to top