Difference between revisions of "Part:BBa J58105:Design"

Latest revision as of 04:17, 30 October 2006

Synthetic periplasmic binding protein that docks a vanillin molecule

Assembly Compatibility:

10
COMPATIBLE WITH RFC[10]
12
COMPATIBLE WITH RFC[12]
21
COMPATIBLE WITH RFC[21]
23
COMPATIBLE WITH RFC[23]
25
COMPATIBLE WITH RFC[25]
1000
COMPATIBLE WITH RFC[1000]

Design Notes

We apply an automated protein design protocol to construct a protein from its constituent amino acids. The method uses the high-resolution atomic structure of a protein in order to find all possible sequences stabilising the corresponding fold (this is the so called inverse folding problem). Either the complete set or a subset of sidechains are allowed to vary randomly. The score used to rank the mutant sequences is an approximation to the folding free energy, which is computed by modelling both an unfolded and folded state, and using a molecular mechanics force field. This modelling step can be viewed as a structure optimization step, an NP-complete problem that is the bottleneck of the design procedure. Thus, we explore the sequence/structure space associated with a given active site or with a given protein-ligand interface. We have followed the work of Prof. Hellinga [1] but using the methodology of [2].

Open and closed conformation of synthetic vanilline binding protein designed using an automatic computational design procedure.

We have used a structure for the closed rbsB (D-ribose binding protein, periplasmic) which has an x-ray structure with pdb 2DRI (271 residues, [http://www.rcsb.org/pdb/cgi/explore.cgi?pdbId=2DRI see here the PDB entry]). The vanilline molecule has a small structure and is not so different from the TNT and other ligands used by the designs of Hellinga.

When the vanillin binds the computationally designed protein at the periplasm, a signal transduction pathway formed by a trg-EnvZ fusion phosphorylates the OmpR transcription factor, which can then induce transcription.

We have considered the 2DRI pdb structure and we have removed all side chains corresponding to the aminoacids surrounding the original ribose. We have used a computational protein design approach [2] to search among all possible sequences the ones that stabilise (in the sense of a folding free energy) the given atomic structure. Our combinatorial search couples the side chain placement problem, with combinatorial mutagenesis and with the docking problem. Therefore, it explores the best binding mode. It also tries to find solutions with all possible ligand-protein h-bonds satisfied, which would confer a high specificity for the ligand.

Source

Gene synthesis (Epoch Biolabs).

References

[1] L.L. Looger, M.A. Dwyer, J. Smith, H.W. Hellinga. Computational design of receptor and sensor proteins with novel functions. Nature, 423, 185-190 (2003).

[2] A. Jaramillo, L. Wernisch, S. Hery and S.J. Wodak. Folding free energy function selects native-like protein sequences in the core but not on the surface. Proc. Natl. Acad. Sci. 99 (2002), 13554-13559.

@@ Line 6: / Line 6: @@
 ===Design Notes===
+We apply an automated protein design protocol to construct a protein from its constituent amino acids. The method uses the high-resolution atomic structure of a protein in order to find all possible sequences stabilising the corresponding fold (this is the so called inverse folding problem). Either the complete set or a subset of sidechains are allowed to vary randomly. The score used to rank the mutant sequences is an approximation to the folding free energy, which is computed by modelling both an unfolded and folded state, and using a molecular mechanics force field. This modelling step can be viewed as a structure optimization step, an NP-complete problem that is the bottleneck of the design procedure. Thus, we explore the sequence/structure space associated with a given active site or with a given protein-ligand interface. We have followed the work of Prof. Hellinga [1] but using the methodology of [2].
 [[Image:vanbp2.png|thumb|300px|Open and closed conformation of synthetic vanilline binding protein designed using an automatic computational design procedure.]]
-We have used a structure for the closed rbsB (D-ribose binding protein, periplasmic) which has an x-ray structure with pdb 2DRI (271 residues). The vanilline molecule has a small structure and is not so different from the TNT and other ligands used by the designs of Hellinga.[[Image:vanilline.png|thumb|100px|Vanilline molecule.]]
+We have used a structure for the closed rbsB (D-ribose binding protein, periplasmic) which has an x-ray structure with pdb 2DRI (271 residues, [http://www.rcsb.org/pdb/cgi/explore.cgi?pdbId=2DRI see here the PDB entry]). The vanilline molecule has a small structure and is not so different from the TNT and other ligands used by the designs of Hellinga.[[Image:vanillin-taste.png|thumb|200px|When the vanillin binds the computationally designed protein at the periplasm, a signal transduction pathway formed by a trg-EnvZ fusion phosphorylates the OmpR transcription factor, which can then induce transcription.]]
 We have considered the 2DRI pdb structure and we have removed all side chains corresponding to the aminoacids surrounding the original ribose. We have used a computational protein design approach [2] to search among all possible sequences the ones that stabilise (in the sense of a folding free energy) the given atomic structure. Our combinatorial search couples the side chain placement problem, with combinatorial mutagenesis and with the docking problem. Therefore, it explores the best binding mode. It also tries to find solutions with all possible ligand-protein h-bonds satisfied, which would confer a high specificity for the ligand.
 ===Source===
-We have followed the work of Prof. Hellinga [1] but using the methodology of [2].
+Gene synthesis (Epoch Biolabs).
 ===References===