An interview with Jin Yu on ‘Switching promotor recognition of phage RNA polymerase in silico along lab-directed evolution path’

This content is part of the Researcher Highlights Life Sciences & Biology series of interviews with scientists, focusing on how their research work can have impact on medical practice. Click here to follow 'Researcher Highlights Life Sciences & Biology' - for free - on Researcher for more great interviews.

Could you tell us who you are, where you work, and what you're currently working on?

I’m Jin Yu. I work at the University of California, Irvine. I'm a faculty member.  I’m in the Department of physics and astronomy. I also have a joint appointment with the Department of Chemistry, and an affiliation with the NSF-Simons Center of Multiscale Cell Fate Research, at UC Irvine. . We do research in computational biophysics, and we're interested in protein machinery, its structure, its dynamics, and its physical mechanism. Although the focus would be on the fundamental, physical aspect, we hope our research to be useful to medical research and application as well.

Could you tell us more about your specific research paper, and why the topic is so important?

We studied this RNA polymerase from bacteriophageT7. We use it as a model system, bacteriophage is a virus that infects bacteria, it has simple protein machinery for their transcription. This polymerase has been widely used in lab condition; it doesn't need a transcription factor, so it can self-sufficiently run from initiation to elongation to termination. It's an ideal model to study as the minimum model for transcription. We studied it for its elongation, how it translocates or how it controls transcription fidelity. Recently, we started working on initiation, which is the first step in the transcription process. Once the polymerase recognizes a DNA promoter, it can start transcribing the DNA template information to RNA. It is important that if the sequence is correctly recognized, then the polymerase will produce a transcription bubble to start the process. A particular feature of this work resides in its use of information from directed evolution. This lab-directed evolution is an experimental technology, that can cause many mutations of the protein that lead to desired features of those mutants, or say those mutants carry on some specific property as you wish them to be amplified in the lab condition. It is very cool in the sense that in nature, evolution takes so many years to happen, but in the lab condition, it can happen very fast. In this experimental data set, we got from the previous publication result: they switch the promoter recognition of this T7 RNA polymerase from RNAP’s original T7 DNA promoter to T3 DNA promoter from another bacteriophage T3, which is not originally intended for the T7 RNA polymerase. The wildtype T7 RNA polymerase doesn't really recognize this promoter, but via the lab-directed evolution, which amplifies certain mutations, representative mutants are obtained, from which we simulated six mutants. These mutants are polymerase proteins which carry some mutations on some amino acids near the protein-DNA interface. They started with just one amino acid of mutation, at one of the binding sites of the protein with the DNA, and then they started having a second mutation up to the sixth mutation. From experimental data, you can find that with those mutants as one-point mutant, two-point mutants, etc., and gradually the capability of the mutant polymerase to recognize the T7 promoter just weakens. In the middle way, i.e., for the three-point mutant, the mutant polymerase can recognize both T7 and T3 promoter, but not very well.  In the end, the five-point and six-point mutants can recognize the T3 promoter well. Then we simulated those mutants, as the experiments themselves are very cool but the mechanism is not showing, you cannot understand what happens there. This is where a computational tool can come into play. We have all-atom molecular dynamics simulation as the tool or as the computation microscope. You can then zoom into every detail of the system, atom by atom. We can see what has happened in the atomic structural dynamics. To simulate the system, where each atom follows the Newton's equation, we use the supercomputer to deal with the computation. That says, once we have the software and the molecular force field, we can run the simulations on the supercomputer. 

Do you have any dream outcomes or what is the next phase of this research and how do you think it will be applied?

We want to use this research to see if we can use computational power to find out what kind of mechanism this polymerase protein uses to recognize specific DNA sequence information. It is difficult in the beginning if you wonder about such mechanisms, because computational microscopes using atomic simulation have complicated data output and a limited simulation timescale. Looking at the mutation without a biological context is also hard to decipher mechanisms. That motivates us to take directed evolution experimental data, as they already carry an interesting story there, because they generate products under the evolution pressure, even though it is not from nature, but from the lab condition. We have a detoured way here to learn from nature, from evolution pushed toward certain direction. It’s a combination of the detour and the intention to learn from the experimental data that made us to explore the inside mechanisms. It's like data learning, but we are not just learning a large amount of raw data, we have a selection and some story behind that, and then we learn from those mutants produced by the evolution experiments in details and try to decipher the detailed story there. We found that the polymerase protein prefers to recognize the DNA promoter which has strong electrostatic interactions with the protein. People know that protein binding and stability are important, technically it is related to free energy overall. But here simulations show that some component dominates, which is the protein-DNA electrostatic interaction. That says, the polymerase favours or recognizes the promoter it has strong electrostatic interaction with. We can identify important amino acids residues as well as some key protein structure elements, from which we found an auxiliary helix structure at the protein-DNA binding interface, where it contributes a lot to the promoter DNA recognition and then switch of the recognition to the other promoter. So you’ll see that the purpose here is to try to figure out how polymerase proteins play their functions, in particular at the beginning of transcription, at the stage of DNA promoter recognition.

Is there anything else that we should know about this paper or your research in general and what you’re working on?

We are interested in the protein machinery in the central dogma of molecular biology, those machinery work or function to manipulate the genome or epigenetic change. We have studied RNA polymerase and we also studied transcription factors. If we understand transcription as a process from which many protein factors and enzymes coordinate altogether, and further involving DNA supercoiling and topoisomerases as mechanical feedback and regulation, then we can see with an integrated perspective on the protein and DNA, as how they interact with each other, and bring some story about the secret in the central dogma.

You can read and discover Jin Yu’s research here.

Switching promotor recognition of phage RNA polymerase in silico along lab-directed evolution path is published in ScienceDirect

Photo Credits: Research Gate

Disclaimer: This is a transcript of a video conversation. You can listen to the recording on Researcher.