Team:Lethbridge/Modeling

From 2010.igem.org

Revision as of 15:33, 25 October 2010 by David.franz (Talk | contribs)




Contents

Metabolic Modeling

Why Metabolic Modeling

The majority of degredation pathways for oilsands contaminants convert various organic compounds into catechol. Our project this year in the incorporation of the gene xylE into E. coli. The xylE gene encodes catechol-2,3,-dioxygenase, which converts catechol into 2-hydroxymuconate semialdehyde. This semialdehyde can be metabilized by E. coli by means of the glycolysis / Krebbs cycle.

However, little is known about how the effect of increased flux of these metabolites into cell (such as intermediate concentrations and metabolite fluxes). Thus to asses the viability of specialized bioremediation and metabolic engineering it is very important to understand what is happening within the cell. By using computer modeling of metabolism pathways and using kintetic data, we can asses the robustness of the system and what we can do to improve the efficiency of the system.

The Model

The proposed model will include the entire glycolysis / Krebbs cycle and the entry of catecol and 2-hydroxymuconate semialdehyde into it alongside normal metabolites. Databases (such as BRENDA) have detailed all the pathways and should have most if not all the kinetic parameters required.

The Questions/Answers

1) What is the maximal rate of catechol degradation by the cell by this pathway?

2) What are the rate limiting steps and what can be done to increase throughput (i.e. the effect of increasing the expression of one or two proteins)?

3) At excessively high levels of catechol, what is the effect on the system (robustness of the system)?

These a very important questions as many iGEM projects use catechol as their entry point into bacterial metabolism. To enable metabolic engineering an understanding of what is occurring within the cell is crucial.


Homology Modeling

Another aspect of our project is working on the localization of catechol-2,3,-dioxygenase (and other proteins) into the interior of microcompartments. To target the protein into the microcompartment (Lumazine modified to have an even more charged interior) requires the fusion of a polyarginine tail to either the C or N terminus of the protein.

As with any fusion protien, the addition of this polyarginine tail to the protein has the very real potential of blocking (at least partially) the active site of a protein. Thus if one can predict the structure of the protein with the addition of the tail, it is possible to avoid creating inactive/ low efficiency fusion proteins.

Figure 1. Homology model of xylE from Pseudomonas putida. The original structure (1MPY) is shown in teal; the generated homology model is shown in green. An Fe2+ ion near the proposed active site is shown is brown.


The Method

To model the xylE structure, the sequence for xylE from Pseudomonas putida (NCBI accession number NP_542866) was aligned with the primary sequence from the crystal structure of xylE from the same organism (pdb: 1MPY; several differences in amino acid sequence were observed) using CLUSTALW (Higgins et al., 1996). Based on this sequence alignment, a homology model was generated using the alignment mode in SWISSMODEL (Guex et al., 1997; Kiefer et al., 2009; Arnold et al., 2006). To model the placement of an N-terminal arginine tag, the tag was manually added to the N-terminus of the model. Energy minimization was carried out in SWISS-PDB viewer in vacuo utilizing a GROMOS96 energy minimization (Guex et al., 1997). The resulting pdb file was visualized and manipulated using PYMOL, images were taken using the same software (DeLano, 2006).

Figure 2. Energy minimized position of the N-terminal 10X Arg tag (shown in blue).


Results and Limitations

Homology modeling of xylE shows that while the C arginine tag cannot interfere with the active site, the N terminal tag is actually a concern. As shown in figure 2, the arginine tag is located perilously close to one side of the active site and has a real potential for at least partially blocking substrate access.

However, these results are from only using one force field and by holding the structure of the rest of the protein static. The use of simulated annealing methods or even molecular dynamics would greatly increase our ability to predict the effect of these tags.

References

Arnold, K., Bordoli, L., Kopp, J., and Schwede, T. (2006) The SWISS-MODEL Workspace: A web-based environment for protein structure homology modelling., Bioinformatics 22, 195-201.

DeLano, W. L. (2006) PyMOL, DeLano Scientific.

Guex, N., and Peitsch, M. C. (1997) SWISS-MODEL and the Swiss-PdbViewer: An environment for comparative protein modelling, Electrophoresis 18, 2714-2723.

Higgins, D. G., Thompson, J. D., and Gibson, T. J. (1996) Using CLUSTAL for multiple sequence alignments, Methods Enzymol. 266, 383-402.

Kiefer, F., Arnold, K., Künzli, M., Bordoli, L., and Schwede, T. (2009) The SWISS-MODEL Repository and associated resources, Nucleic Acids Res 37, D387-D392.