Biophysics: A Role for Non-equilibrium Physics in Determining Protein Structure and Function
An emerging paradigm, which the O’Brien Lab is helping to forge, is that translation kinetics can influence nascent protein behavior. Introducing synonymous codon mutations into an mRNA molecule, which changes the rate at which codon positions are translated by the ribosome but not the amino acid encoded, has been shown to influence whether a nascent protein will fold and function, misfold and malfunction, dimerize, aggregate, or efficiently translocate to a different cellular compartment. The genomes of different organisms use synonymous codons with different frequencies, indicating that mRNA molecules encode an additional layer of information to guide the variation in translation speed across a coding sequence and thereby influence the fate of proteins in cells. Indeed, synonymous mutations that can change translation rates have now been linked to a variety of diseases, including subtypes of hemophilia and cancer. These findings are a shift away from the predominant “thermodynamic control” paradigm, which states that a protein’s amino acid sequence is the primary determinant of its structure and function, to one in which kinetics can have a much greater influence that previously appreciated.
The O’Brien Lab’s efforts in this area include developing chemical kinetic models that accurately predict the influence of translation kinetics on co-translational protein folding and misfolding; applying coarse-grained simulation techniques we have developed to describe in molecular detail how nascent protein self-assembly is altered during synthesis; and understanding the molecular origins of codon translation rates. Our ultimate goal is to understand and quantitatively predict how synonymous codon mutations in cells affect translation kinetics, protein behavior and down-stream cellular functions.
We have made significant progress towards these goals. For example, we have demonstrated that kinetic models can accurately predict co-translational folding curves measured in vivo (Read More); identified the physical origins of “critical” codon positions – that is, positions along an mRNA sequence where a synonymous mutation will dramatically alter co-translational protein folding (Read More); and created a general model that can predict translation-rate effects on protein folding mechanisms of arbitrary complexity (Read More). This research program is advancing the nascent proteome field by examining details of translation that are difficult to measure experimentally, by providing molecular explanations for experimental observations, and by challenging the field’s current paradigms to motivate entirely new research directions (Read More).
Physical Bioinformatics: Combining ‘Big Data’ with Biophysical Modeling
A wealth of data concerning translation is now regularly being generated by Next-Generation Sequencing Methods such as Ribosome Profiling. Next-Generation Sequencing involves sequencing small fragments of nucleic acids a large number of times, and Ribosome Profiling allows the location and number of actively translating ribosome to be measured. Such data offers a wealth of opportunities to estimate codon translation rates, identify the molecular origins of these rates, and develop models that predict codon translation rates from the mRNA sequence alone.
The O’Brien Lab is using its expertise in chemistry and molecular physics to develop new methods to analyze such data in order to accurately identify the location of the A- and P-sites on ribosome-protected mRNA fragments, extract codon translation rates from Ribosome Profiling data, and characterize the relative contribution of different molecular factors on codon translation rates. Ultimately, our goal is to be able to extract accurate measurements of translation initiation and elongation rates from Ribosome Profiling data simply and cheaply.
To achieve these goals the O’Brien Lab has paired up with the Bioinformatics and Genomics Ph.D. program at Penn State. Nabeel Ahmed, a graduate student from this program, is leading the efforts in this area. We believe that combining our molecular/quantitative reasoning with such Big Data approaches will facilitate the development of biophysical analysis tools that will find wide spread use in the field and also provide novel insights.
Theoretical Chemistry: Improving the Accuracy of Calculated Reaction Rates by Accounting for the Ensemble of Reaction Pathways
To predict the rate at which a biochemical reaction will occur a common technique is to use what is known as QM/MM calculations – that is, simulations that treat the non-catalytic portion of an enzyme classically and the catalytic portion quantum mechanically. In this class of models a common sampling technique is the “string-of-states” method to determine the lowest potential energy reaction pathway connecting a reactant and product structure. The transition state barrier height is then determined and an approximate reaction rate estimated. This method can be extended to account for multiple pathways and combined with kinetic Monte Carlo simulations to estimate an overall reaction rate. However, both of these methods have drawbacks. The string-of-states approach is often applied at absolute zero, meaning that thermal effects, important for enzyme motion and catalysis, are neglected in part. The kinetic Monte Carlo method involves numerical methods, which have sampling errors associated with them and take longer to calculate than solutions of analytical equations.
The O’Brien Lab is carrying out research to overcome these limitations. Specifically, our goal is to come up with a QM/MM procedure that maintains the appropriate thermodynamic ensemble (i.e., avoids the absolute-zero problem) and solves for the overall reaction rates using an analytic solution to an underlying chemical reaction scheme that accounts for all reaction pathways. We have derived an analytic solution for single-step reaction schemes involving an infinite number of pathways, and our next step is to test this approach on enzymes for which accurate rates have measured. Achieving these goals will lead to more precise and accurate models for the prediction of biochemical reaction rates.