My project focuses on creating proteins with new form and function. Using the eFx Flexizyme, an RNA enzyme, I charge engineered tRNAs with unnatural amino acids (UNAA) and use in vitro translation to incorporate them into specific sites in proteins. This is accomplished by incorporating the TAG amber suppressor sequence in my protein transcript and the CUA amber suppressor anticodon on my tRNA. I am also optimizing in vitro translation to increase the efficiency of UNAA incorporation by adjusting the concentrations of initiation and elongation factors to enhance their interactions with engineered tRNAs.
I work on several different projects for the Ellington laboratory. Though the bulk of my projects involve using repertoire analysis to create and enhance antibodies for a wide array of targets. An example would be me analyzing a set a b cells which encode a wide array of VH and VL genes that are encoded by mRNA transcripts in b cells. The present mRNA subject to overlap extension RT-PCR to create linked amplicons that can later be sequenced and identified. Using this information I can select for the chains that have the highest representation in the targeted archetype (Antibodies that bind to a certain substance). In combination with yeast display this allows me to select for the antibodies with the highest affinity.
The main focus of my work has been on expanding the capabilities of Flexizymes, ribozymes that catalyze the amino acylation of tRNA with an almost infinite library of amino acids and amino acid analogues. Their substrate flexibility stems from the recognition of a chemical leaving group on the carboxyl, rather than the amino acid itself. Using these ribozymes, you can insert unnatural amino acids into peptides and proteins without the need to evolve an amino acyl tRNA synthetase. However, flexizymes function is limited to in vitro protein synthesis. Using directed evolution, we are attempting to identify substrates and flexizymes constructs that will allow for in vivo function, in order to facilitate future genetic code expansion.
We use Saccharomyces cerevisiae as a platform for studying and engineering eukaryotic proteins. In particular, we are exploring how various classes of human receptors functionally signal in yeast and the minimal components sufficient to reconstitute these pathways across kingdoms. By mutagenizing the ligand binding regions of these receptors and subjecting them to rounds of screening or selection we can endow receptors with novel ligand specificities. These mutant receptors can ultimately be used for chemogenetic activation of specific neural subtypes for neuroscience applications. Our microbial receptor strains also act as inexpensive screening platforms for drug discovery applications, and more broadly, as biosensors for analytes of interest.
Dallas Lee - Systems and Synthetic Biology Approaches to Plastic Degradation
Plastic waste mismanagement was estimated at 31.9 million metric tonnes in 2010, more than ten percent of total worldwide production. Estimates of plastic waste entering oceans was projected to be increasing at the rate of an additional 6-17 metric million tonnes per year.1 There exists the necessity to be able to degrade not only mismanaged plastics, but also the ever increasing burden of plastic production. This is accomplished by utilizing and improving bio-prospected enzymes from bacterial origins. 1 (Jambeck, J,R. et.al. Plastic waste inputs from land into the ocean. Science 347,6223, pp. 768-771. DOI: 10.1126/science.1260352)
Cody McLeland - Evolving Adhesion: Creating a Superior Bio-Glue
This project aims to improve upon the previous work done on biomimetic adhesives derived from mussels such as Mytilus Californianus. By utilizing a DOPA-charging synthetase made in house, we can more efficiently create peptides/proteins which will be displayed on the surface of E. coli. By selecting for adhesion, we create a means of evolving peptides/proteins based on adhesive strength or even selective adhesion.
Phi29 DNA Polymerase is a highly processive strand-displacing polymerase that has become paramount as a tool for molecular biology in performing whole genome amplification (WGA). Phi29 is extremely processive in that it can polymerize up to 70kb fragments in a single synthesis event.1 Secondly, Phi29 retains high fidelity as it has 3’-5’ exonuclease proofreading activity.2 These properties make Phi29 the preferred tool of choice when performing WGA via multiple displacement amplification (MDA) as there is reduced amplification bias and allelic dropout compared to other WGA methods.3 However, Phi29 has low thermostability as MDA reactions are performed at 30°C. This leads to amplification bias when dealing with regions of high G/C content. If one were to engineer a thermostable Phi29, the MDA reaction could be performed at an elevated temperature which would presumably lower the amplification bias due to G/C percentage. Secondly, a thermostable Phi29 would benefit from faster amplification reaction kinetics due to increased reaction temperature. Due to reasons cited above, it is highly desirable to engineer a thermostable Phi29 DNA polymerase. We are however exploring alternative routes, such as engineering KOD-Phi29 chimeras by fusing the TPR2 loop from Phi29 onto KOD DNA polymerase. This TPR2 loop has been shown to confer both processivity and strand-displacement activity.4 Therefore, we would be engineering a thermostable strand-displacing polymerase, as KOD is a thermostable itself.
Jonny Riggs - DNA Synthesis Techniques
Under the guidance of Dr. Randy Hughes, my current focus is to learn de novo DNA synthesis techniques. DNA, like any element of nature, can be broken down and analyzed in terms of its chemical constituents. However, building and manipulating DNA can be extremely challenging for molecular biologists. Using modern techniques such as column-based and microchip-based oligonucleotide synthesis, we are able to synthesize single-strand DNA oligonucleotides that overlap to form larger DNA synthons. These synthons can be further assembled to allow us to synthesize them into longer stretches of DNA that may be inserted into expression vectors. The resulting constructs are then sequence-verified and transformed into compatible host cells to assay for downstream applications. The timeline from design to testing of these constructs represents the essential foundation of synthetic biology. As we adapt new methods and technologies, we aim to truncate the timeline of novel DNA synthesis while accelerating and maximizing construct throughput.
I am deeply interested in exploiting big data techniques to better learn and design biology. One avenue currently being explored is the effectiveness of artificial intelligence algorithms in learning and predicting important protein features. We take advantage of growing biological databases and apply maturing machine learning methods. In doing so, we can design more effective proteins while reducing the time and search space where eventual selections are performed.
Changes in protein function are driven by the interplay between their sequence and structure over evolutionary timescales. For protein engineers, one option for designing new protein functionality is to expand the set of monomers used for protein synthesis, perhaps with reactive or exotic new chemistries. This strategy is also used by nature in the form of non-canonical amino acids (selenocysteine and pyrrolysine) which enable novel catalytic activity, and a myriad of post-translational modifications which are used to fine-tune protein properties. I work on several projects with a common theme of developing the infrastructure to incorporate reactive non-canonical amino acids into proteins and designing quantitative reporters for measuring incorporation in vivo. In a recent example, we used a custom pathway for selenocysteine incorporation to synthesize proteins containing unnatural diselenide bonds which conferred significant resistance to reducing conditions compared to native labile disulfide bonds.
The organic synthesis of L-DOPA, a valuable pharmaceutical precursor, currently suffers from poor yield and low enantioselectivity. I am adapting molecular evolution techniques to create a biosynthesis module and dedicated microbial strain for the fermentation of L-DOPA. We aim to use compartmentalized partnered replication (CPR), an in vivo selection technique developed by the Ellington lab, to evolve a 4-hydroxyphenylacetate 3-monooxygenase for greater specificity and activity towards L-tyrosine. Additionally, I am involved in developing orthogonal ligand-gated ion channels. Neuroscientists are faced with the tradeoffs between using optogenetics versus chemogenetics. Creating orthogonal ligand-receptor pairs would expand the neuroscientist toolkit by enabling modular agonism. Finally, I am engineering DNA polymerases to incorporate nucleotides with unnatural nucleobases as part of an artificial genetic information system.
Phosphorothioate (PS)-modified oligonucleotides form less stable duplexes than their unmodified phosphodiester (PO) counterparts. Combined with the possibility of PS-modified oligo synthesis by chemical and enzymatic means, this property has led to the use of PS modifications in nucleic acid diagnostics circuits to improve signal amplification. We are working on a nearest-neighbor model of hybridization for PS-PS and PS-PO DNA, which we envision could enable precise tuning of components used in DNA nanotechnology via PS modifications.