Question

Your project should be on temperature adaptation or thermostability of proteins.

For a temperature adaptation project select an ectothermic organism with sub-species living in different climates. You will then focus on one protein, preferably an enzyme, for which you need to find amino acid sequences (one for each sub-species). You also need at least one crystal structure of this protein with enough sequence similarity to your selected amino acid sequences to make it feasible to create homology models. Finally, you will make comparisons between the structures or homology models while focusing on structural features that promote protein stability.

OR

For a thermostability project select a thermophilic organism (for example a bacterium thriving in a hot spring) and one protein, preferably an enzyme for which a crystal structure is already available. Next, you will need to search for crystal structures and/or amino acid sequences of your selected protein in other non-thermophilic organisms so that you can make comparisons between the crystal structures (or homology models). Aim to identify structural features that promote the thermostability of the protein from the thermophilic organism.

State the parameters of your project:

Species:

Protein(s):

PDB ID:

Confirm that you will have access to protein sequences and at least one protein structure (as a template for homology modeling) so that you can carry out your project. Post the access codes for the sequences and the PDB-ID(s) here:

Make a list of structural features that promote protein stability:

Note: Except the isocitrate dehydrogenase protein.

example like in the below articke

Answer 1

The ultimate goal of protein modeling is to predict a structure from its sequence with an accuracy that is comparable to the best results achieved experimentally. This would allow users to safely use rapidly generated in silico protein models in all the contexts where today only experimental structures provide a solid basis: structure-based drug design, analysis of protein function, interactions, antigenic behavior, and rational design of proteins with increased stability or novel functions. In addition, protein modeling is the only way to obtain structural information if experimental techniques fail. Many proteins are simply too large for NMR analysis and cannot be crystallized for X-ray diffraction.

Among the three major approaches to three-dimensional (3D) structure prediction described in this and the following two chapters, homology modeling is the easiest one. It is based on two major observations:

1. The structure of a protein is uniquely determined by its amino acid sequence (Epstain, Goldberger, and Anfinsen, 1963). Knowing the sequence should, at least in theory, suffice to obtain the structure.

1. During evolution, the structure is more stable and changes much slower than the associated sequence, so that similar sequences adopt practically identical structures, and distantly related sequences still fold into similar structures. This relationship was first identified by Chothia and Lesk (1986) and later quantified by Sander and Schneider (1991). Thanks to the exponential growth of the Protein Data Bank (PDB), Rost (1999) could recently derive a precise limit for this rule, shown in Figure 25.1. As long as the length of two sequences and the percentage of identical residues fall in the region marked as “safe,” the two sequences are practically guaranteed to adopt a similar structure.

1

2                                                                                                            H O M O L O G Y M O D E L I N G

Percentage of identical residues

100
90
80
70			Safe homology
60
			modeling zone
50
40
30
20	Twilight zone
10
0	50	100	150	200	250
0

Number of aligned residues

Figure 25.1. The two zones of sequence alignments. Two sequences are practically guaranteed to fold into the same structure if their length and percentage sequence identity fall into the region marked as ‘‘safe.’’ An example of two sequences with 150 amino acids, 50% of which are identical, is shown (gray cross).

Imagine that we want to know the structure of sequence A (150 amino acids long, Figure 25.2, steps 1 and 2). We compare sequence A to all the sequences of known structures stored in the PDB (using, for example, BLAST), and luckily find a sequence B (300 amino acids long) containing a region of 150 amino acids that match sequence A with 50% identical residues. As this match (alignment) clearly falls in the safe zone (Fig. 25.1), we can simply take the known structure of sequence B (the template), cut out the fragment corresponding to the aligned region, mutate those amino acids that differ between sequences A and B, and finally arrive at our model for structure A. Structure A is called the target and is of course not known at the time of modeling. In practice, homology modeling is a multistep process that can be summarized in seven steps:

1. Template recognition and initial alignment

1. Alignment correction

1. Backbone generation

1. Loop modeling

1. Side-chain modeling

1. Model optimization

1. Model validation

At almost all the steps choices have to be made. The modeler can never be sure to make the best ones, and thus a large part of the modeling process consists of serious thought about how to gamble between multiple seemingly similar choices. A lot of research has been spent on teaching the computer how to make these decisions, so that homology models can be built fully automatically. Currently, this allows mod-elers to construct models for about 25% of the amino acids in a genome, thereby supplementing the efforts of structural genomics projects (Sanchez and Sali, 1999, Peitsch, Schwede, and Guex, 2000). This average value of 25% differs significantly

H O M O L O G Y M O D E L I N G                                                                                                   3

Template sequence B (arabinose-binding protein, 300 residues) Aligned region

Target sequence A (150 residues)

Step 1 and 2: Template identification and alignment

Step 4 and 5 - Loop and side chain modeling Step 3 - Backbone generation

Step 6 - Model optimization

Figure 25.2. The steps to homology modeling. The fragment of the template (arabinose-binding protein) corresponding to the region aligned with the target sequence forms the basis of the model (including conserved side chains). Loops and missing side chains are predicted, then the model is optimized (in this case together with surrounding water molecules). Images created with Yasara (www.yasara.com).

Step 7: Model Validation