MONDAY, 5 OCTOBER 2020
Below is a transcript of the latest episode of the BlueSci Podcast, which we have provided to help our podcast reach a wider audience. The script has been transcribed using an audio-to-text converter (otter.ai) and edited by our hosts. If you spot any errors please do let us know! Please do subscribe, follow, like, and review our podcast to show us your support and give us feedback, whether you listen or read the episode.
You can find the podcast on:Understanding protein behaviour at the nanoscale, featuring Dr Jerelle Joseph
AnchorFM : anchor.fm/bluesci-podcast
Apple Podcasts: podcasts.apple.com/podcast/id1358079746
Google Podcasts: podcasts.google.com?feed=aHR0cHM6Ly9hbmNob3IuZm0vcy8zZWRmMTdjL3BvZGNhc3QvcnNz
AnchorFM : anchor.fm/bluesci-podcast
Apple Podcasts: podcasts.apple.com/podcast/id1358079746
Google Podcasts: podcasts.google.com?feed=aHR0cHM6Ly9hbmNob3IuZm0vcy8zZWRmMTdjL3BvZGNhc3QvcnNz
Released 5th October, 2020
BlueSci Podcast 0:05
Welcome to the Bluesci podcast, brought to you by Cambridge University Science Magazine. I'm Ruby and I'm Simone. Every two weeks we speak to local researchers, university staff and students and anyone who works in science to learn about their research and activities, hear about the work that they do, and uncover what goes on behind the scenes. If you want to get in touch with a question, suggestion or just want to be featured on the podcast, just drop us a tweet. Our handle is @bluescipod and you can also email us at firstname.lastname@example.org.
Ruby Coates 0:43
Welcome back to the BlueSci podcast. We hope you enjoyed your little summer break. We definitely enjoyed ours. We're now refreshed and recharged and we have some really cool content coming up for you over the next couple of weeks. So today's episode is a interview with Dr. Jerelle Joseph and she is a postdoc working on cellular liquid liquid phase separation at the Cavendish Laboratory at the University of Cambridge. And in our discussion with Jerelle she tells us all about her PhD, postdoc and the organization Carischolar, which aims to match Caribbean students with mentors all over the world in order to provide a source of guidance and advice for students from the Caribbean wanting to go into academia, which is really interesting to learn about. We hope you enjoy this episode.
The BlueSci podcast is sponsored by Greiner Bio-One who supply laboratory diagnostic and medical products to research institutions, higher education, the NHS and others across the UK. For details of the full product range, visit www.gbo.com.
Ruby Coates 2:04
Thank you so much, again for joining us today. Dr. Jerelle Joseph. So you're a postdoc at the Cavendish lab laboratory, to Cambridge and your PhD was in protein modeling with the Department of Chemistry. And so you're onto your postdoc now. So could you tell us a little bit about your, your current research as well as all of your past research and where it fits into everything?
Okay, so for my PhD, I worked in the Department of Chemistry at Cambridge under the leadership of Professor David Wales. And I was looking at developing computational models to study protein folding. And if you are aware with sort of like the history of proteins and so on, you know, that there is this, this structure-function relationship as it relates to proteins, whereby the function that the proteins perform, is very intricately linked to its structure. And so the protein folding problem essentially emerged, where researchers were trying to reconcile, how is it that you go from the primary amino acid sequence of a protein to a well folded three dimensional structure structure? And so, a lot of the work that has been done over maybe about 70 years or so, has been to predict protein structure from the amino acid sequence. Now, the problem was further complicated somewhere in the mid 1990s, where it was discovered that some proteins actually do not have a very well defined structure. And these proteins have seen have since been characterized as being intrinsically disordered. So they can, some of them can adopt multiple different stable folds, whereas others, they may adopt a fold upon binding to a molecular partner. And then we have these further discoveries that you have a protein that has a particular amino acid sequence that can include two entirely, completely different forms. So we're talking about having a sort of beta, what we described in the literature, as having this sort of like beta barrel sheet sort of structure and alpha helical structure, and anyone who's familiar with protein literature knows that these are very different types of folds and so on. The challenge now for us was, well, when I started my PhD was the how do we modeled and studied this sort of folds. And so I was developing approaches that is based on energy landscape framework to look at how these folds are encoded and how we can predict and look at structural transitions that proteins undergo from one fold to the next. And so this work is important in the context of, if we're thinking about designing proteins that have specific folds, we need to be able to study this different folds, right? We need to be able to have approved approaches that can capture that sort of behavior. And so a lot of my PhD work centered around that.
Do you mind explaining why protein folding is such an important, like problem to solve? And, you know, what kind of information do we get, besides the fact of knowing how the protein folds, in terms of we know why that's an interesting thing to look at?
Well, one of the main sort of one of the main drivers in that field is in the in the development of drugs. So as we know, many, many drugs are designed to bind to proteins and change the behavior of proteins in ourselves. And so, therefore, a lot of a lot of these interactions between drugs and ligands. So essentially proteins are very structure-specific, so there's specific epitopes for example, specific sites, on antigens, which are the drug causing protein proteins in our body that bind specifically to antibodies, right, which drug developers developed to combat certain illnesses and as you can see, now, with COVID-19 it is a very important case where where you have this spike protein, that plays a role in in the progression of the disease. And now all researchers are trying to, you know, to model and to find what's the exact structure of the virus, so that they can design specific drugs to target it. And so, therefore, proteins and understanding protein structure structure from a humanitarian perspective is very important, because it helps us to be able to develop drugs to combat viruses, which are, which are associated with proteins as well as RNA. And so, and so, from that perspective, it is very important for us to be able to understand to be able to understand and predict these these folds. Now, as it relates to sort of designing proteins that have new functions, if you are to design a new protein with a specific function, you also want to understand the principles that is going to allow it to fold a particular way, and hence, ultimately function in a particular aspect. So there is the design aspect that we normally stress on as well as the, in the area of health and pharma and development of drugs. And there are several other reasons why for some proteins, some proteins, you don't fully understand their functions. And the hope is that if you can predict the fold that forms, then that can give you some indication as to which sites which may be important for a particular function with which regions of it are very similar to other proteins and might have a shared functionality with other proteins. And so, yes, so for that reason, it's very important for us to really sort of like be able to understand, and the problem there is that there is sort of like a gap between what we can do on the computer and what experimentalists can, can do. Right? So the gap has closed quite a bit over the last over the last few few decades, whereby the skills that experimentalists and computational persons working in the computational space, have sort of has sort of closed, but there's still a lot of work to be to be done right. We still have a lot of challenges as it relates to how big of a protein that we can simulate on a computer and how long we can simulate, simulate it for and the approach that I worked on during my PhD is sort of saying that we have this this space, this high dimensional space thatthe protein folds in. And we want to sort want to sort of focus in on these specific regions in this high dimensional space. And the idea is that if we can, if we can map out regions of the of the high dimensional space that are important, we can get the important information that we need or not just important, but that can capture the overall behavior of the protein, we can get important information that we need.
Ruby Coates 10:32
Great. So obviously now, you're a postdoc at the University of Cambridge in the Cavendish lab. And could you give us a sort of rundown as to what your, what your postdoctoral work entails?
My path really what happened is that my current lab that I'm working, I'm working in my boss, Dr Rosana Collepardo-Guevara, she works in the area of liquid liquid phase separation, which looks at intercellular organization, and how the organization is governed by phase separation behaviors of proteins and other molecules. And so, therefore, I found that this area of research should be very appealing to me, when I finish my PhD, because it was sort of bringing in my love for proteins and looking at different protein systems in a sort of bigger picture as it relates to the organization. So, if we think about the cell and the way the cells organized and compartmentalize, we often think about this membrane bound compartments, right. So, you have the nucleus, which has like a membrane around it, and you have organelles, such as the mitochondria and so on, but they're also micro-environments inside the cell, that are not enclosed by physical membranes. So, for years, scientists have sort of wondered how did these so called non-membrane bound organelles come about, and towards the earlier parts of this century, there have been important experimental works that have suggested that these environments are sustained by the physical chemical process of liquid liquid phase separation. So, an analogy that we like to use which is very sort of like simplified, is thinking about if you have like a glass that has oil and water in it, the way in which the oil and water separates into two very distinct liquid liquid phases. So, the hypothesis there is that under certain conditions inside the cell, bio molecules in particular, intrinsically disordered proteins (which I happened to work on during my PhD), several multi domain proteins and as well as RNA, they can interact with each other and de-segregate from the from the cytosol or or the nucleus, forming this membrane, this compartments and these are essentially sustained by the physical chemical process of liquid liquid phase separation. And studying this area is very important, because, liquid liquid phase separation inside cells or more broadly speaking, the formation of this biomolecular condensates have been linked to the way the genetic material is organized have been linked to neurodegenerative disorders. And there are several works that have suggested that these compartments play a very important function inside of the cell as it relates to RNA metabolism, the this cope coping with stress and so on. And so there are several implications that can lead to the formation of these, what we call biomolecular condensates or membraneless organelles, there's a lot of names as as you know, when something is new, everybody wants to coin like a new name for something.
Ruby Coates 14:26
So it's it's fascinating how you can sort of develop all of all of this understanding of how proteins function and how they how they work, and how that knowing that information gives you insight in sort of more of a pharmaceutical level in terms of drug interactions, but equally understanding molecular biology in the way that proteins function within within cells, either you know, in membrane bound organelles or non membrane bound organelles and it sounds like sort of the main thread of your research throughout your PhD and into your postdoc has been kind of looking at these protein structures. And I was just wondering, you know, there's an awful lot of sort of computational power going into that and, and sort of on a day to day level, is your interest in research, more sort of centered around developing programs, or softwares or scripts to actually analyze the proteins themselves? Or do you also work in the lab? Like, what's the split if if you do have a split?
Okay, so what I do now is I focus more on developing multi-scale approaches. So when we talk about looking at phase behavior inside cells, or looking at phase separation, in general, this is a very collective process. So whereas for my PhD, I would be focusing on one single protein, and studying that protein very closely, when we talk about phase separation, because it's a collective behavior, that is you have hundreds of proteins coming together and forming this sort of condenses this compound compartments, then that makes the challenge for us as computational chemists or biologists, whatever you want to call it, that makes the challenge for us even more difficult. And so therefore, the approach that I take in my work is a multiscale. One, whereby we develop approaches at different resolutions to tackle different problems. So say, for example, one of the one of the key drivers for this sort of phase behavior is metabolism, so that is one protein can bind to several different proteins at the same time. And so in that multivalency, you have different types of interactions, right? So you have, I don't want to go too much can I go further into like the chemistry of it. So you have, you have interactions that are class classified as hydrophobic, you have electro-static interactions, you have pi-pi interactions, and any anybody in the chemistry space would have heard of this sort of like terms. And so if I want to look at how a particular interaction may be important, I might need a high degree of resolution, right. So I might be, I might need to be able to represent all of the atoms that are involved. And so therefore, I can develop a an approach to study all these specific interactions as a very high resolution. When I say high resolution, I mean that I have more details in my simulations, right, as a very high resolution, I can develop these approaches that can allow me to study that particular interaction, but therefore in but when I'm looking at the entire condensates itself, which is a collection of the proteins, then having every single atom explicitly represented is going to be too computationally expensive. And so therefore, I may move to look at every each amino acid, which is the building blocks of proteins, right. So I might represent protein as a collection of its amino acids, right, instead of looking at the, all of the atoms in each amino acid, right. And when I want to look at bigger things, such as what what if I have a multi component system, where I have very different types of proteins all interacting at the same time, then that again, if I have a protein that is really big compared to one that is smaller than that, again, might pose a problem if I'm trying to represent it in terms of each amino acid in the protein. And so therefore, I might need to go further to a lower resolution, where I essentially have a lower resolution model to study these effects. And so that is what I speak about the multiscale approach. And so here in a day to day setting, it would be more so looking at the problem that I am interested in, and which methods that I want to develop. But in our lab, the good thing is that we work very close with experimentalists. And that is the beauty of this field. Because the field is relatively new, every every every sort of like subfield is being developed at the same time. So the experimentalists rely a lot on the computational work for so that we can inform their their work. And we also do rely heavily on the experiment so that we can develop our model. So my day to day is very sort of different depending on what project that I'm working on. But we have a lot of talks now via Zoom, where we speak about our results or work and so on. And we decide what best approaches that we want to develop to tackle to tackle the particular problem at hand.
It sounds like a super interdisciplinary field as well as having social multiscale approach. And from what you were saying earlier as well, it sounds like you have quite an interdisciplinary background, right? Like your PhD was in chemistry, and now you're working in the physics department. So how is it working with on a problem that is, you know, not unique to a field rather, at like, the intersection of all these different fields together?
Yeah, so I find, I find, I find that to be to be the most exciting part of it. And if I said that I actually planned this out, it would, it would not be true, because actually for my master's, I actually like worked on more quantum mechanics space where you have like this high detail. And then I move over to my PhD, where I was working on more sort of atomistic models, along developing costing approaches. And then now I'm doing my, my PhD, my Sorry, my postdoc, where I'm using, like, all of these different skills, so you really don't realize all of the different skills that you sort of like developing as you go along. But now that I'm working in this space, that that that where it is essential that we look at the problem from different time skills and different length scales, I can see the now that now that I can bring in all of the skills that I have, like sort of developed over the, over the years, and work with different people. And also in our lab, there are a lot of people who have different backgrounds. And so learning from them, and from what for drawing from their expertise, and so on, is it's been like a super rewarding process. And it's been like, very good to sort of like see how the sciences and the different disciplines can come together and work towards like a common problem. Yeah, it's really cool.
Ruby Coates 20:34
Yeah, it's really exciting. And, and considering that you've kind of moved quite dynamically between departments, and now you're working on something that is so interdisciplinary, regarding that, what are your sort of hopes for your future? Would you like to stay in academia? You know, with the field itself, do you think it's a field that is gonna get bigger and bigger and that you can continue to work in?
Yes, it's gonna get bigger, it's definitely going to get bigger. And there's a lot of research going on in that area now. And yes, I hope to stay in academia. That is that is, that is my dream, I can't tell you that I have like this grand 10 year plan, because I don't. But I want to say like in academia, and yes, I definitely see the field continuing to grow more and more persons were working in different sub sub disciplines, and now like tuning in, and now paying in interest to the idea of this space separation process inside of the cell. And yes, it's super exciting. And I can only I'm very excited for where it's going to go. And the fact that there are a lot of like implications of this sort of research research, as it relates to how our genetic material is expressed how or genetic materials access, how it relates to the progression of several neurodegenerative diseases, I think that this field would only grow from from from here.
And are there specifically like questions that you'd be interested in looking at, or I guess, like, is that kind of where you want your research to go to look at these links between the protein folding and our, you know, gene expression and things like that?
Yes. So I, I'm looking at several questions now. So I'm tackling several different different topics, and working with persons who work on human proteins as well as proteins that are in other species. And what has been happening is what's been happening is that we have experimental collaborators who are working in other fields and other systems. And they're beginning to see this sort of phase separation behavior in their systems. And so therefore, there's what I really want my work to, to do is to develop a set of general rules, or to identify patterns that are shared among different species and in different systems that can create this sort of generalize frame framework where whereby we can predict how these behaviors can be tuned inside cells. So I'm not so specific. I'm not not so interested on one particular problem as it relates to phase separation and condensates more. So looking at a wide variety of problems, and to see commonalities between them and trends and patterns, and that that gets and help us to have these predictive rules.
Ruby Coates 25:15
Yeah. And, and I guess, like you said, by working sort of, from it one step back, you can identify sort of common behaviors of these types of proteins, which sounds like, it's, it's a really cool way of looking at it, because I saw the universe tries to show us these things on a scale, isn't it? And then, in terms of how, work closely with the sort of more experimental side of the side of all of this and how does how does the process work? If you want to go into a cell and find out how how this phase separation is happening, what sort of data comes out of the experiments that you then put into your model and how does that sort of flow of information work?
okay. So, so, because this process is driven by changes that occur inside inside the cell, an example that I could give you is say for example, the concentration of salt ions inside of the cell changes due to some chemical reaction that that is taking place. And so, therefore, what the experimentalist would do is that they will modulate the salt concentration in their experiments, and they will look at the formation of these condensates, and how that records in the images is that you see this sort of spherical droplets that form right. And so, therefore, they can, they can continue to change the conditions, and they can see under which conditions, both in terms of the concentration of the proteins that they're using, and the concentration of salt, they can see under which conditions do we have this spherical droplets that are formed, not just spherical drop droplets, but they still maintain the liquid like behavior. So, they're very dynamic, so, if you use sort of like a laser and you you zap this, these this spherical droplets you can use what is called fluorescent tags. So, the drop droplets can be tagged with fluorescent probes for for fluorescent, and if you zap zap them, you can see over time that the fluorescence is recovered right, which is a common experiments called FRAPP, which is often using experimentalists to look at whether the entity entity remains dynamic. And so, therefore, they can get all of this data, where they have the different conditions and the images that show these droplets form and qualification for for this for the systems and then you know, as simulations now, we can try to emulate the salt effects and see whether we get for the same proteins on the different salt conditions, do we get this condensates for me and, and to how the salt effects the formation of this con condensates? And then we can go a step further because we're in the computer and say, What are the specific interactions that the salt modulates that causes the condensates to form or dissolve, right? And so we we can do this work on the computer. And furthermore, it is, it's much cheaper and much easier for us to do things like mutations, right? So we could switch one amino acid for another and see how a mutation say, for example, affects the formation of these con condensates, and then come up with specific sequences or specific mutants that may alter phase separation behavior, and then the experimentalists can go back and test it and they can say, oh, when we tested that, it turned out that yes, it did, in fact, promote phase separation, or it didn't hinder face to face separation. So it's sort of like a feedback loop where they give us information on what they see. And we develop models that that can recap recapture this behavior, and also go a step further and look at mutations, we can also look at how the condensates are organized, which is very hard to sort of like in in experiments, you can see a droplet, but it's very hard to see the orientation of the proteins inside of that drug droplet right whereas in the computer, we can see that and we can we can see how they're arranged and if all models are accurate, then that can give a good picture as to why- why- the why that particular con- condensates forms and what properties that component set may have. So say for example, if it has like a protein that stays in the quantum state the entire time, you know, as to, you know, in solution, then that can give us some indication as to the disparities in dynamics between different different proteins in this call condensates and how that may change over time. And so, that kind of information that we can give back and so on.
Yeah,that sounds super interesting. And again, it's, it's like what you were saying about having all these different approaches in terms of like the interdisciplinary aspect, right, having the computational versus experimental approach, and then using that to kind of build on each other, and have like, the combination be stronger than, you know, each individual part. I think that's, that's such an interesting kind of environment to be in and must be really interesting. I was gonna ask as well, because besides the work that you're doing for your research, you're also the founder of Caricholar, do you want to explain to our listeners, what the program is about? And maybe why you founded it?
Yes. So Carischolar is a mentoring organization. And it was developed to help Caribbean students. And essentially, when I started doing my PhD, or if I go back a little bit, when before I went to undergrad, I was a high school teacher. And then when I started doing my PhD, because of the Gates -Cambridge scholarship, and so and there was like, a lot of publicity around that. And then I, I started getting a lot of requests from students back home, who were asking me questions about, you know, PhD and applying for applying to university, and so on. And I always felt that I could help the students who were in the sciences much better than other students. But I was fortunate in the fact that I went to, I did my undergrad in the Caribbean, at the University of the West Indies. And they I met a lot of students who were studying very different disciplines, and remained friends over the years. So I was like, maybe I could get these friends to help these, this the students. And so that was that was the sort of like more motivation that drove me to create Carischolar. So essentially, essentially, we have a network of Caribbean professionals and academics, who have experienced going to school in the Caribbean, and are not necessarily living in the Caribbean right now. They may be all all over the worlds right? right now. And we have this net, this network of this act, this academics and professionals who are helping Caribbean students, or more specifically mentoring and guiding students in their chosen fields of study. So if a student decides that they want to do international law, and they really want to mentor in international international law, and so they can apply to us, and we will find a mentor, who is from the Caribbean, not necessarily living in the Caribbean, who is an expert in international national law, and we will connect them together so that they can help mentor the student and help guide them through the process. Because it can be like, it can be extremely sort of like daunting to, to map out your professional trajectory. And for me, like I call myself a biophysicist, now a computational chemist, whatever you may now, but growing up, I didn't know anybody who was working in like, who was a biophysicist, I didn't know anybody like that growing up. I didn't even know that that was like a possibility for career, right. And I was fortunate enough to be mentored by lecturers during my undergrad, right? So I would have sit in lecture lectures come up to me and talk to me and asked me what I wanted to do and so on. And if it wasn't for these conversations, I would be completely lost, right? And so I really see the value of mentorship and having people who have who are successful who have walked the path that you are currently on and they're ready know what the challenges are, where the challenges maybe help you and give you advice. I think that is the most valuable thing. And we have this huge like diaspora of Caribbean nationals all over the world, right, who are skilled in many different areas and if they find it super rewarding to be able to give back and paid forward and yes, so that was the whole idea with Carischolar and that is what like, to connect students with Caribbean mentors in their field of study.
Ruby Coates 35:04
Wow, it sounds like a wonderful program and like highly beneficial. And if anyone listening wanted to get involved with that, how would they be able to get involved with Carischolar?
Right. So So if persons wanted to serve as mentors, they can go on carischolar.com. And they can click on become a mentor, and they can fill out the application form there. And equally, if students are searching for mentors in their field, they can also go to carischolar.com. And they can apply to be mentored by or Carischolar mentors. Yeah.
Ruby Coates 35:46
So yeah, it's been really interesting to chat with you, because it seems like you've been massively successful, both in the science world and then also organizing a program that is really empowering people in like, helping people connect and communicate, like, you know, really efficiently and, and just forming all of these really beneficial networks. So it's been a real pleasure talking to you. And yeah, thanks for taking the time. Yeah.
Yeah, thank you for having me.
Ruby Coates 36:21
We hope that you enjoyed today's episode, and we'll look forward to seeing you next time for the next episode of BlueSci. If you want to get in touch please email us or find us on Twitter. Our handle is email@example.com and see you next time.