Synonymous Variation and Fitness in Yeast
Variation in a gene’s synonymous codon usage can result in differences in either protein production or protein folding, which can in turn have phenotypic consequences. However, contexts and mechanisms by which codon usage impacts translation are not well defined. We are pursuing two projects that employ high-throughput sequencing technology to investigate factors important to synonymous codon usage and protein production in eukaryotes.
Project 1 - We are using yeast HIS3, an enzyme required for the synthesis of histidine, as a model gene for examining the impact of codon usage on protein production. Having constructed four plasmid libraries of synonymous HIS3 variants, we will submit yeast cells carrying these synonymous variants to a growth assay in media lacking histidine. In the course of this competitive growth assay, cells in the population with efficiently and accurately translated HIS3 transcripts should increase within the population, whereas cells carrying detrimental synonymous codon variation should proportionally decrease within the population. After recovering plasmids from population samples, we will use high-throughput DNA sequencing to measure the relative abundance of each variant and to calculate variant enrichment scores. With these data, we can begin to explore the relative impact of factors such as adaptation to tRNA pool abundance and mRNA secondary structure.
Project 2 - We are collaborating with the Grayhack laboratory at University of Rochester to identify insertions in a GFP reporter that impair translation efficiency. To this end, the Grayhack lab has carried out a Fluorescent Activated Cell Sorting (FACS) assay on a randomized library of codon insertions into the superfolder GFP variant. We are sequencing library variants from FACS expression bins to identify codon insertions with cell distributions skewed toward lower GFP expression values. These types of insertions may represent specific codon pairs or codon combinations that are translated less efficiently.
Deep Mutational Scanning of a tRNA
tRNAs are of fundamental importance in translating the information contained in our genes into cellular and organismal function. A tRNA must adopt a specific and conserved three-dimensional structure in order to interact with the ribosome, with elongation factors, and with its corresponding amino acid tRNA synthetase (Fig. 1). A good deal of cellular energy is also devoted to extensively modifying the bases of a tRNA during tRNA maturation. Despite these constraints on tRNA shape and sequence, significant sequence diversity exists both within the ~500 human tRNAs and ~275 yeast tRNAs, as well as between species. In order to define the set of all functional tRNAs and to determine the extent to which a particular tRNA can tolerate mutation, in collaboration with Eric Phizicky’s and David Mathews’s labs at the University of Rochester, we adapted deep mutational scanning to the study of tRNA function.
The assay relies on the ability of a suppressor tRNA, which recognizes a stop codon, to allow the ribosome to read through the stop codon on an mRNA instead of stopping translation at the stop codon and releasing the mRNA. We modified yeast by the addition of two genes: a Green Fluorescent Protein (GFP) reporter and a tyrosine tRNA that was modified to recognize the ochre stop codon (UAA). The GFP reporter contains an ochre stop codon near the N-terminus of its amino acid sequence, resulting in its failure to be translated into a functional protein unless a working suppressor tRNA is also present (Fig. 2). In this way, the function of a mutant version of the ochre suppressor tRNA can be assessed by observing the fluorescence of the yeast. The ochre suppressor tRNA was mutated extensively, and each mutant gene was transformed into yeast containing the GFP reporter. The transformed yeast were sorted by a Fluorescence Activated Cell Sorter into four bins based on their fluorescence. We sequenced the population of suppressor tRNA genes present in each bin on an Illumina MiSeq. For a given mutant, the percentage of MiSeq reads in each bin, along with the average fluorescence of the bins, can be used to determine the average fluorescence of a yeast cell containing that mutant tRNA. The weighted average fluorescence was used to stratify the mutants by function.
To date, we have obtained functional data for every possible single mutation, for about 13,000 double mutations, and for about 30,000 more highly mutated variants. Surprisingly, 43% of the single mutants show wild type levels of fluorescence, and the majority of the single mutants retain at least some function. In addition, large numbers of double mutants (21%) show wild type levels of fluorescence, indicating that despite the requirements for tRNA modifications and structural constraints, tRNA function is relatively robust to mutation. We are examining the double mutants in order to gain insight into the relationships between positions within the tRNA. We have seen some expected relationships; for example, deleterious mutations that abolish base pairing in one of the stems are rescued by changes that restore base pairing. Other interactions can occur within or between loops. By examining positional interactions in various backgrounds, we hope to gain a greater understanding of the determinants of tRNA structure and function.
•David Young
Investigating the HIV-1 Tat-TAR interaction
The HIV-1 Tat protein is integral to the viral life-cycle as it can induce efficient transcription of the virus after binding a folded element of the HIV LTR called TAR. Previous studies have elucidated the effects of some mutations of Tat, but the overall depth and density of the studied mutations is low. We are investigating the Tat-TAR interaction using deep mutational scanning, a high-throughput technology recently developed in the lab.
By creating a library of hundreds of thousands of variants of Tat and selecting for binding to TAR using a yeast three-hybrid assay, we are examining the relationship between the sequence of Tat and its TAR-binding function at an unprecedented resolution. The Tat-TAR interaction is thought to be driven by an enrichment of basic residues in the core of the protein rather than a specific amino acid sequence, but it is not known if point mutations outside of this core region can affect the TAR interaction. This study of mutations that affect the affinity of Tat to TAR can contribute to our understanding of both protein-RNA interactions, as well as the mechanism of HIV transcription and activation.
•Daniel Melamed, Matt Rich & Christina Miller
|
       |
|
|
