Structural bioinformatics practical

TP AspRS
In silico directed mutagenesis of the aspartyl-tRNA synthetase

The objective of the TP is to study by molecular modeling the specific recognition between the aspartyl-tRNA synthetase and its substrate Asp. We will try to evaluate the specificity by comparing the binding of the ligands Asp and Asn. We will then seek to identify and model mutations in the active site that could favor the binding of Asn instead of Asp. This is a first step towards an engineering of the genetic code.

AspRS

Introduction

Aminoacyl-tRNA synthetases (aaRS) are an enzyme family implicated in protein synthesis. They are involved in translation by allowing the binding of an amino acid to its transfer RNA. They are very specific of the amino acid concerned and corresponding transfer RNA. Thus there exists one for each amino acid.

We will be interested particularly in the aspartyl-tRNA synthetase (AspRS), the goal being to perform point mutations of this enzyme in order to reduce its affinity for its natural ligand aspartate and favor its binding to asparagine.

For that, we will consider the problem in terms of protein sequences and structures. The study includes three steps:

This analysis will lead to propose judicious mutations of the active site, allowing to modify the AspRS specificity by favoring Asn binding over Asp.

Protocol

A) Analysis of aaRS sequences

  1. Retrieve the E. coli AspRS sequence in the UniProt (http://www.uniprot.org) bank
  2. Obtain homologous sequences: BLAST search

    3. The E. coli AspRS has three domains: the tRNA anticodon binding domain, the catalytic site domain, and a third one inserted into the catalytic site domain.

    Launch a BLAST search. What types of proteins are found?

  3. Identify important residues: make a multiple sequence alignment

    4. Identify strongly conserved regions that can correspond to the active site. Choose some positions that seem distinctive of the AspRS and of Asp binding.

    Which strategy have you employed? Which mutations do you propose to modify the AspRS affinity for aspartate and asparagine?

B) Structural analysis: inspection of the AspRS structure

With the informations previously obtained, propose judicious mutations to modify the AspRS affinity for aspartate and asparagine. We will try to test several of them in the next step of modeling.

Does the active site examination lead you to modify your mutation propositions made from the sequences?

Can we use the structure to verify the sequence alignment?

C) Molecular modeling study

This is the most ambitious and complex part of the TP. There are two steps:

We will follow this protocol with the XPLOR program:

  1. Inspect the files available:

    2. tp_asprs.tar.gz

    asprs.seq
    sequence of the AspRS protein
    asprs.pdb
    experimental structure of the AspRS protein
    asprs.xplor.pdb
    experimental structure of the AspRS protein formatted for XPLOR
    asp.xplor.pdb
    structure of the Asp ligand formatted for XPLOR
    amber.rtf
    topology file for XPLOR
    isolated_aa.rtf
    additional topology file for isolated aa
    amber.prm
    parameter file for XPLOR
    build.inp
    model building of the protein:ligand complex
    minimize.inp
    energy minimization of the complex
    energy.inp
    energy calculation of the complex
    run.sh
    script to drive the calculations

  2. Compare the files asprs.xplor.pdb and asprs.pdb

    3. The PDB files must comply with a particular format to be readable by XPLOR. The segment name has to be written on 4 characters in columns 73-76. We also note that the 3-letter code of histidines has been changed from HIS to HIE. There indeed exists 3 possible protonation states for histidines and it is necessary to tell XPLOR which state is chosen among HID, HIE, or HIP (see the topology file amber.rtf for the definition of these states).

    Does the HIE protonation state chosen for all histidines seem reasonable to you? If in doubt, try other protonation states and evaluate the impact on the results.

  3. Build a model of the AspRS:Asp complex with XPLOR

    4. xplor < build.inp > build.out

  4. Minimize the energy of the complex to improve its geometry

    5. xplor < minimize.inp > minimize.out

  5. Estimate the energy of the AspRS:Asp complex, and then of each partner alone

    6. xplor < energy.inp > energy.out

    What is the affinity of the AspRS protein for the Asp ligand?

  6. Edit the PDB file of the ligand asp.xplor.pdb to change Asp into Asn. We will simply replace one of the carboxylate oxygens of the Asp sidechain by a nitrogen (corresponding to the Asn NH2 group). It will then be easy with XPLOR to position the two missing hydrogens.

    7. What is the affinity of the AspRS for Asn?

    Experimentally, the wild-type enzyme binds Asp considerably stronger than Asn, with an association free energy difference of more than 7 kcal/mol. Do you find the same tendency?

  7. Mutagenesis of the AspRS: choose a mutation among the candidates previously identified.

    8. A simple mutation (for example Asp→Asn or Gln→Glu) can be realized by editing the PDB file, as explained for the ligand.

    A more complex mutation can be performed with the SCWRL program. The mutation choice (for example R10K) is done by replacing in the asprs.seq file the one-letter code of the native amino acid in lowercase by the code of the amino acid chosen for the mutation in uppercase (for example replacing the lowercase ``r'' in position 10 by an uppercase ``K'').

    We then launch the SCWRL program as follows:

    scwrl -s asprs.seq -i asprs.wt.pdb -o asprs.pdb > scwrl.out

    Compare the mutated structure obtained asprs.pdb with the native structure asprs.wt.pdb.

    It is recommended to work in a separate folder for each mutant.

  8. Perform the affinity calculations for the mutated enzyme.

    9. In order to use the structure mutated by SCWRL with XPLOR, it must be ensured that it is correctly formatted. For that, we will use the pdb2xplor program as follows:

    pdb2xplor asprs.pdb A PROT > asprs.xplor.pdb

    What are the affinities for Asp and Asn obtained with the mutated protein?

    Have you succeeded in inverting the specificity?

    Interpret structurally the effect of the mutations.

  9. Which improvements could we bring to the model or the protocol?
  10. Was the AspRS the most judicious target for this engineering?

TP Trp-cage
Structure and stability of Trp-cage

The objective of the TP is to study the structure and stability of a small protein, the Trp-cage.

Trp-cage folded ↔ Trp-cage unfolded

Introduction

Trp-cage is a small artificial protein of 20 amino acids, which has been designed to fold easily. Its amino acid sequence is NLYIQWLKDGGPSSGRPPPS. The protein folding problem is one of the most important challenges of structural bioinformatics. It consists in predicting the three-dimensional structure of a protein from the sequence information alone.

We will employ the methods of molecular mechanics to model the Trp-cage.

Protocol

A) Dynamics at equilibrium of the folded Trp-cage

  1. Inspect the files available:

    2. tp_trp-cage.tar.gz

    folded.pdb
    experimental (NMR) structure of the folded Trp-cage
    unfolded.pdb
    linear unfolded structure of the Trp-cage
    amber.rtf
    topology file for XPLOR
    amber.prm
    parameter file for XPLOR
    build.inp
    model building and energy minimization
    md.inp
    molecular dynamics at 300K
    traj2mpdb.inp
    conversion of the trajectory to the multiple PDB format
    analyze.inp
    analysis of the trajectory
    run.sh
    script to drive the calculations

  2. Model building

    3. xplor < build.inp > build.out

    This script builds a model of the Trp-cage with XPLOR and performs an energy minimization to improve the geometry.

    Inspect the output file and visualize the structures produced.

  3. Molecular dynamics

    4. xplor < md.inp > md.out

    This script performs a molecular dynamics of the Trp-cage during 20ps, assigning random initial velocities and then maintaining the temperature at 300K.

    Inspect the output file and track the energy and temperature as a function of time.

  4. Visualization of the trajectory

    5. xplor < traj2mpdb.inp > traj2mpdb.out

    This script converts the format of the trajectory produced from DCD to multiple PDB.

    We can then visualize the trajectory with PyMOL by loading it as follows:

    load md.multi.pdb, multiplex=0

  5. Analysis of the trajectory

    6. xplor < analyze.inp > analyze.out

    This script reads the produced trajectory (md.dcd) and performs structural or energetic calculations at each step. The results are written in a text file (md.dat). Represent them graphically.

    The analyses included in the script are only examples, it is your task to add more relevant ones with the help of the XPLOR documentation.

B) Unfolding the Trp-cage

C) Folding the Trp-cage

TD/TP Modeller
Homology modeling of mimivirus tyrosyl-tRNA synthetase

Mimivirus is a giant DNA virus. It is larger than many bacteria, and can itself be infected by other viruses. It was discovered that mimivirus possesses certain genes for proteins involved in translation, absent in other viruses that use the host cell's machinery to multiply. These discoveries have fuelled debate about the boundary between living and inert matter.

Mimivirus

Homology modeling aims at building a model of the unknown structure of a target protein, knowing its sequence and the structure of another template protein of homologous sequence. The method can be decomposed into four steps:

  1. template selection
  2. target-template alignment
  3. model building
  4. model evaluation

The aim of this work is to propose the best possible structural model (criteria to be defined) of mimivirus tyrosyl-tRNA synthetase (its structure is assumed to be unknown) by homology modelling with the Modeller program.

  1. Retrieving the sequence

    2. Retrieve the sequence of mimivirus tyrosyl-tRNA synthetase in FASTA format in UniProt (http://www.uniprot.org) database.

  2. Template selection

    3. Carefully select a structure to be used as a guide for homology modeling (of course, we won't use the structure of mimivirus tyrosyl-tRNA synthetase, which we assume to be unknown). Retrieve this structure in PDB format.

  3. Conversion of the sequence format

    4. Convert query sequence from FASTA to PIR format (http://salilab.org/modeller/manual, File formats, Alignment file (PIR)) with which Modeller works. Example of a sequence in PIR format:

    >P1;TvLDH
sequence:TvLDH::::::::
MSEAAHVLITGAAGQIGYILSHWIASGELYGDRQVYLHLLDIPPAMNRLTALTMELEDCAFPHLAGFVATTDPKA
AFKDIDCAFLVASMPLKPGQVRADLISSNSVIFKNTGEYLSKWAKPSVKVLVIGNPDNTNCEIAMLHAKNLKPEN
FSSLSMLDQNRAYYEVASKLGVDVKDVHDIIVWGNHGESMVADLTQATFTKEGKTQKVVDVLDHDYVFDTFFKKI
GHRAWDILEHRGFTSAASPTKAAIQHMKAWLFGTAPGEVLSMGIPVPEGNPYGIKPGVVFSFPCNVDKEGKIHVV
EGFKVNDWLREKLDFTEKDLFHEKEIALNHLAQGG*
  4. Inspect the files available

    5. td-tp_modeller.tar.gz

    The Modeller program is launched as follows:

    modeller file.py

  5. Target-template alignment

    6. Align the query sequence with the sequence of the selected guide structure by adapting the align2d.py script.

    modeller align2d.py

    The alignment produced is written in PIR, PAP, and FASTA formats. Examine these files.

  6. Model building

    7. Model by homology the target structure by adapting the model-single.py script.

    modeller model-single.py

    A summary including the PDB file names of the models produced as well as the value of the Modeller energy function and the DOPE score for each model can be found at the end of the output file (model-single.log). Examine the models produced with PyMOL.

  7. Model evaluation

    8. Evaluate the models generated in the previous step by adapting the evaluate_model.py script.

    modeller evaluate_model.py

    This script allows a more detailed evaluation of the models produced by calculating the DOPE score for each position of the alignment. Plot the DOPE score as a function of position (columns 1 and 42) for the different models.

  8. How to define the best model produced?

For further information:

Modeller tutorial: http://salilab.org/modeller/tutorial/basic.html

Modeller manual: http://salilab.org/modeller/manual

TD/TP AutoDock
Amarrage moléculaire de l'Indinavir à la protéase du VIH-1

td-tp_autodock.tar.gz

lock and key

AutoDock est une suite d'outils destinés à l'amarrage moléculaire (« molecular docking »). Le docking consiste à prédire comment des ligands, comme des substrats ou des médicaments potentiels, se fixent sur un récepteur de structure tridimensionnelle connue.

La procédure de docking avec AutoDock se décompose en deux étapes principales :

  1. génération de cartes d'interactions du récepteur avec le programme AutoGrid
  2. amarrage proprement dit du ligand au récepteur avec le programme AutoDock

Nous utiliserons également une interface graphique appelée AutoDockTools (ADT) pour faciliter la préparation du docking et la visualisation des résultats.

La méthode sera illustrée avec le docking de l'Indinavir, un inhibiteur de la protéase du VIH-1, utilisé comme antirétroviral dans le traitement du SIDA.

  1. Préparation du ligand

    2. AutoDock a besoin de connaître les charges et types atomiques de chaque atome, ainsi qu'une liste des liaisons avec libre rotation présentes dans le ligand.

  2. Préparation du récepteur

    3. Grid → Macromolecule → Open... → PDB files → "hsg1.pdb"

    AutoDockTools lit le récepteur et comme pour le ligand effectue les étapes de calcul des charges atomiques de type Gasteiger, fusion des hydrogènes non-polaires et attribution des types atomiques.

    Sauvegarder le récepteur sous le nom "hsg1.pdbqt".

  3. Préparation des cartes quadrillées (« grid maps »)

    4. Il est nécessaire de générer une carte d'interaction pour chaque type atomique du ligand plus une carte pour l'électrostatique et une carte pour la désolvatation.

  4. Docking

  5. Analyse des résultats

Bioinformatics servers

UniProt

NCBI

EBI

ExPASy

PDB

XPLOR

Documentation en ligne de XPLOR 3.1

PyMOL

PyMOL site

PyMOL wiki

PyMOL CheatSheet

Linux

Linux installation in a virtual machine

Linux command line formation

Formation Debian GNU/Linux

Linux Command Line Cheat Sheet

Unix/Linux Command Cheat Sheet

UNIX Tutorial for Beginners