Tanner Dean and William McCleary, MMBIO

Introduction

The purpose of this experiment was to identify the physical interaction site between two proteins in Escherichia Coli. These two proteins, PhoU and PhoR, are two proteins involved in Phosphate signal transduction. When these two proteins are mutated, so they no longer interact, the pathway is broken and signaling is inhibited. To break this pathway, we mutated one protein (PhoR) until we developed several mutations that no longer interact with PhoU. This loss of interaction is caused by a physical change to the site of interaction between the two proteins. We hypothesized that we could find a naturally occurring PhoU mutant that would compensate for the known PhoR mutation, thus restoring signaling. Through genetic sequencing of the PhoU mutant, we could then identify potential allele specific interaction sites between the two proteins allowing us to create a genetic map of the physical interaction site.

Methodology

To test interaction we used a bacterial two-­‐hybrid system. This system connects one domain of Adenylate Cyclase to one of our Pho-­‐proteins and the other Adenylate Cyclase domain to the other protein. If there is interaction between the two target proteins, the two domains of Adenylate Cyclase interact allowing for the digestion of maltose. This digest is manifest by turning the colonies a dark red color on MacConkey Maltose plates. Two proteins that don’t interact, or maintain a broken connection, (such as the known PhoR mutant and wild type PhoU) remain a white color on the MacConkey Maltose plates.

To develop natural occurring PhoU mutants, we put the plasmid that had PhoU and the other Adenylate Cyclase domain into a strain of bacteria that has mutations to the DNA proofreading mechanisms. These mutations allow for replication without high fidelity, thus allowing for mutations to arise. We developed a screen for these potential mutations by transforming this plasmid into a strain with our known PhoR mutant. This transformation was accomplished through electroporation of the plasmid into a stain that contained the known PhoR mutant. We then would select those that appeared red on the MacConkey maltose plates indicating a strong interaction between Adenylate Cyclase domains thus restoration to the PhoR-­‐PhoU interactions. We then would isolate this PhoU mutant plasmid and sequence the phoU gene. Sequencing would allow us to compare the mutant to the wild-­‐type gene and see which base pairs changed to restore interaction, giving us a potential interaction site. These compensatory mutations could then be used to map potential interaction sites.

Results

After a few electroporations, we had gathered eight mutant phoU that seemed to repair function according to our bacterial two-­‐hybrid system. We assumed they were compensatory because the mutants maintained a constant dark red color. They maintained a similar phenotype through purification of the plasmid and retransforming into our known PhoR mutant strain. These steps were a check to make sure the mutation was on the plasmid rather than the chromosome of our bacterial vectors. Sequencing of these mutants showed that each mutant had zero mutations in the phoU gene. They were all exactly similar to the wild type.

Discussion

These results indicate a screen that was not capable of selecting only for mutants that repaired the broken phoU-­‐phoR interaction. The likely source of false positives was the mutations obtained in the low fidelity strain were possibly to the plasmid promoter genes. Increase in plasmid promoter sequences would allow of greater expression of the plasmids and cause more maltose to be digested, giving a false positive result in the screen. This is merely an idea and was not substantiated through any more tests, but could be the source of our false positives.

Conclusion

Though it is possible to identify which sites lead to physical interaction between PhoR and PhoU, the process of forming compensatory mutations seems to be a difficult way of reaching any conclusions. This screen may eventually lead to a true compensatory mutation, but the likelihood of just selecting for promoter mutations over and over again seems very likely. A possible way to identify sites is to perform a bioinformatics analysis of these potential interactions sites in a coevolution study. This is currently under way in our lab. This study is important because a positive identification of interaction sites can lead to manipulation of phosphate signal transduction and potential use of that system for water treatment or even manipulation of virulence.