Jordan Mackay and Dr. Julianne Grose, Department of Microbiology and Molecular Biology
Characterizing the interacting protein partners of Per-Arnt-Sim (PAS) kinase has proven to be a formidable task. Since the beginning of 2012, I have been working hard with graduate student Desi DeMille on confirming the putative interactors uncovered by genetic screening done the previous year. This endeavor has yielded both successes and failures.
Successes in this project includes one instance of successfully confirming an interacting partner with PAS kinase as a true substrate. This was done slightly differently than originally proposed. By collaborating with Dr. John Prince of the BYU Chemistry Department, we performed mass spectrometry in addition to coimmunoprecipitation (co-IP) and kinase assays. The MS data show us that PAS kinase and a substrate identified in our screens appear to interact via kinase-substrate phosphorylation. Another success includes successfully cloning several of our interactors into the appropriate plasmid vectors harboring myc tags. This took many weeks of work, since these genes were difficult to manipulate and clone properly into the vectors, but nonetheless appear to have been successfully completed.
The screening process still has not been completed due to the introduction of new genetic libraries. A large portion of the time spent on this project was used to saturate these libraries in order to have statistically sound results for the screens. This part of the procedure must be satisfactorily complete for a truly comprehensive analysis of the metabolic relevance of PAS kinase. Many potential interacting partners of PAS kinase have been identified, but the statistical significance of the number of hits from these screens remains to be confirmed. In order to confirm these hits, independent tests must be done on these genes to be sure they are not false positives.
The screening process itself did not proceed particularly smoothly. Many obstacles had to be overcome in order to reach this point; the screening process did not turn out to be simple, since the libraries we used were not initially cloned very efficiently. While this may not count as a failure since reliable results were obtained, there were still more stumbling blocks than one would wish in this project.
Among the failures seen in this project, the most glaring is the lack of success in performing the co-IPs. This issue stems from the inadequate amounts of initially purified protein. Overexpressing certain proteins can cause varying amounts of stress to the cell, and it is very dependent on which proteins are being overexpressed. Changing certain variables in the protocol may allow for healthier cells, resulting in a greater amount of purified protein. This would lead to conclusive co-IP results. Providing conclusive results of physical interactions is very important in characterizing protein-protein interactions, even interactions as transient as kinase and substrate.
Other pitfalls in our experiments have largely resulted from issues with contamination and unreliable reagents used in reactions. Due to the carelessness of many undergraduate researchers (and occasionally myself), many of our yeast cultures were contaminated. Many of the confirmatory tests performed after screening turned out to be unreliable due to the fact that incorrect strains were used for the confirmatory testing. Therefore, these tests must be carefully redone and carefully documented in order to complete this project in a timely fashion.
Future research must be done surrounding the newly identified interacting partners. Experiments still need to be done confirming the actual interaction of PAS kinase and the newly discovered interactors. By carefully cloning these interactor genes into appropriate plasmid vectors and purifying tagged and overexpressed protein with an optimized protocol, the co-IPs will be successful. Experimentation is also needed to confirm how the phosphorylation of substrates by PAS kinase affects their activity. This will be done by individualized phenotypic assays in which each substrate is overexpressed with PAS kinase or knocked out. Specific phenotypes normally produced by that gene in vivo will be assessed with respect to PAS kinase activity.
Our results have been shared with the scientific community and the general public through several poster presentations, including one at a meeting of the American Society of Microbiology, which took place in Pocatello, Idaho. A similar presentation was given at the National Conference for Undergraduate Research (NCUR) held at Weber State University in Utah. Our preliminary data was also presented at a Yeast Genetics conference held at Princeton University over the summer by Desi DeMille, the graduate student with whom I have been working. This research was also presented by poster to the President’s Leadership Council (PLC) of Brigham Young University. Presenting this research gave me opportunities to share what I have learned with others in and out of the discipline as well as opportunities to accept suggestions regarding how I could approach some aspects of this project.
Currently, the data that has been gathered is insufficient for publication. However, the data gathered up to this point are very promising, and given more time will inevitably result in publication. We expect our data to be complete in the next few months with results planned to be submitted for publication by April. Following the publication of our screen results, the other data specific for each characterized interactor will be published in separate articles.