"
Team:Brown/Project Histamine Sensor
From 2009.igem.org
| Line 1: | Line 1: | ||
{{Brown}} | {{Brown}} | ||
| + | |||
| + | |||
Revision as of 03:34, 21 October 2009
Histamine Sensor
Alteration of Fusion Protein Tar-EnvZ to Sense Histamine
To synchronize production of rEV131 with fluctuations in histamine concentration, a histamine receptor is necessary. Natural histamine receptors exist only in eukaryotes as G-coupled protein receptors, unusable for our prokaryotic cells. Therefore we have set out to engineer our own receptor. This novel receptor will sense extracellular concentrations of histamine and initiate an intracellular signal cascade that leads to the production of rEV131. rEV131 would in turn sequester histamine and lower the extracellular concentration, thus diminishing transcription of rEV131. By design, this system uses negative feedback and is self-regulating.
We approached the development of a histamine receptor with two different strategies. First, we set out to mutate the Tar periplasmic receptor domain of the Taz chimera protein. This receptor, normally sensitive is normally sensitive to Aspartate. We performed a site-directed mutagenesis of the Aspartate binding pocket. The amino acids that take part in ligand binding have been specified in the paper (Yeh, et al. 1996).
Loren Looger at the Howard Hughes Medical Institute's Janelia Farm campus used his protein design software Chameleon" to calculate mutations that would transform Tar’s aspartate binding pocket to a histidine binding pocket, as the first step on the way to making a histamine binding pocket. His algorithm gave us the top 16 receptor designs and we are currently in the process of creating this library of mutants. We have designed primers for each of these designs and are introducing these mutations by both the “Round the Horn Site-Directed Mutagenesis” protocol on OpenWetWare and the Strategene Mutagenesis II Kit.
Our assay to test these receptors’ affinities for histamine is based on fluorescence. We have constructed a cassette from the registry that places the OmpC promoter gene over the gene for RFP. The Taz receptor, upon binding of its ligand, has an EnvZ intracellular kinase domain that phosphorylates the transcription factor OmpR, which subsequently activates transcription of the OmpC gene. With this OmpC-RFP reporter cassette, we can test both quantitatively and qualitatively the receptor’s affinity for the ligand. We have tested this signaling cascade by transforming the normal Taz receptor with the reporter cassette and introducing the ligand Aspartate to show that it works. The E. coli strain RU1012 was used, as it is an EnvZ knockout strain.
RU1012 with OmpC-RFP + Taz1 receptor
RU1012 with OmpC-RFP
RU1012 with Taz1 receptor
RU1012 with no plasmid
<PHOTO HERE>
Alteration of Endogenous Ribose Binding Protein to Sense Histamine
Secondly, we set out to mutagenize the endogenous ribose binding protein, using our own receptor design computer program.
To design a histamine receptor, we developed a computational approach modeled after that taken in Looger et al., "Computational design of receptor and sensor proteins with novel functions" (2003). The software we used was the Rosetta macromolecular modeling software. We took the existing Rosetta enzyme design functionality (such as used in Röthlisberger et al "Kemp elimination catalysts by computational enzyme design" (2008), and modified it to design an enzyme that would bind to a ligand without actually catalyzing any reaction. The conformational change caused by this binding triggers an intercellular signaling cascade affecting gene expression.
How we designed the protein:
1.) Took the PDB file for the crystal structure of the ribose-binding protein (RBP) cocrystallized with ribose (2DRI) and cleaned it up (removed waters, added missing hydrogens).
2.) Used UCSF Chimera to geometrically search for all the van der Waals interaction contacts between ribose and RBP in the crystal structure. Identified the amino acids in the protein that made those contacts as those most likely involved in the ligand binding pocket of RBP.
3.) Used UCSF Chimera to mutate all the identified residues in RBP to alanine (which has neutral chemical properties, and almost no side chain). This effectively created a "blank" version of RBP that had no specific binding pocket for any ligand (removing its binding affinity for ribose).
4.) Replaced the ribose in the PDB file of the alanine-mutated RBP (polyala RBP) with a 3D structure for histamine in a low energy conformation.
5.) Used Rosetta's Ligand Docking mode on a cluster of 100 servers to dock histamine into the polyala RBP ligand binding pocket. This used Monte Carlo minimization to find the relative orientation of each that minimized steric contacts between the two and keep the histamine within roughly the same ligand binding pocket as ribose was in the original structure. This will output 10000 PDB files of the histamine docked to the polyala RBP.
6.) Sort the 10000 docked PDB files by their interface energy between the ligand and the protein. Select the top 2500.
7.) Take those top 2500 and use them as input into the Rosetta Enzyme Design mode. Run the mode without specifying any catalytic activity that needs to be designed. Specify the residues that were mutated to alanine as those that you want to have the program to design. This will use the RosettaDesign functionality to search through residue mutations in the ligand binding pocket that minimize the total energy between the protein and the ligand. A lower energy between the ligand and the protein means that the combination of the two is more stable, and the protein is more likely to bind to the ligand. It finds these mutations through a probabilistic simulated annealing algorithm, so the final designs are not guaranteed to have the lowest total energy that is possible for the protein. However, by doing thousands of designs in parallel, we can make it more likely that the algorithm will hit upon the mutations that do result in good histamine binding.
8.) Sort through the output designs based on their predicted interface energy between the ligand binding pocket and histamine, how well the protein is predicted to fold, and how many hydrogen bonds the protein makes with the ligand (H-bonds are very good for ligand binding).
The receptor design software output a list of predicted receptor protein designs, which were ranked based on their predicted folding and ligand-binding abilities. The DNA for the top design was synthesized by GeneArt AG. We are still in the process of testing the designed receptor protein.
The receptor itself consists of a modified version of the E. coli ribose-binding protein, which is free-floating and binds to ligands in the periplasmic space of E. coli. When the ribose-binding protein is in its ligand-bound conformation, it interacts with the periplasmic Trg domain of the transmembrane E. coli chemotaxis receptor.
To have this ribose-binding protein/Trg interaction induce gene expression, we created a fusion between Trg and the EnvZ histidine kinase domain. The EnvZ histidine kinase autophosphorylates the OmpR protein, which causes transcription of DNA regulated by OmpC promoter. We tested this fusion protein with wild-type ribose-binding protein as a receptor, and found that it quite effectively transmitted a signal.
When a prokaryotic histamine receptor is obtained, we will place the OmpC promoter over rEV131, thus creating a self-regulating drug factory in the nose.
<PHOTO HERE>
