Episode#17 Isothermal Titration Calorimetry – Dr. Shashank Chavali

Barkha Yadav-Samudrala: 0:07

Hello, everyone to another episode of nerdRX podcast. And I'm your host Barkha. Today, we have Dr. Chavali, who is going to talk about isothermal titration calorimetry, which I have absolutely no idea about. So I'm going to learn with all of you. Welcome Dr. Sol to the show.

Dr. Shashank Chavali: 0:29

Thank you so much. Vodka. Thanks for having me. It's, it's, this is the first time I'm doing and I'm really looking forward to this. Thank you so much for inviting me.

Barkha Yadav-Samudrala: 0:40

You are most welcome. I'm so glad you could you took some time out of your schedule to be here with all of us. So before, before we jump into the topic, we would love to know about your education background, and how did you get to this point in your career?

Dr. Shashank Chavali: 1:03

Okay, sure. So, I started off my scientific career as an undergrad researcher. I've actually during my 11th and 12th grade, that's when I decided to pursue science. And I really somehow felt that you know, discovering something that no one has discovered ever, or like seeing something that no one has seen before, for the first time, or something that really clicked well with me. And I really wanted I started became becoming interested in research for that reason. So I I wanted to I didn't know what I should research. So the most the most cool thing at that time was research in life science because, like, I used to hear terms like DNA and cloning and like all that, it sounded very cool to me. But I had no clue because like I had a mathematics and physics chemistry background up until 12 standard and Biological Sciences required some background and by the way, all right, right. So that is something I thought maybe I would develop I mean, I had like high school biology. It's not it wouldn't be that bad. But so I was sort of an outlier when I decided to study biology or like, biological sciences, in Specially in India, where I grew up. I'm from Hyderabad in India. Engineering is like, is the thing like everyone else suppose, like, that's, that's, that's the thing.

Barkha Yadav-Samudrala: 2:59

Parent thing is engineering, art medicine.

Dr. Shashank Chavali: 3:04

Exactly, exactly. So I was also like, one of them. I was, thankfully, like, I wasn't like, pressurized by my parents, like anyone around me to do engineering, but I was also like, I was confused. Like, I wanted research. I didn't know if I should I was supposed to do engineering in like biotechnology or like biomedical engineering or, or whether I'm supposed to, you know, do a BS Bachelor of Science in chemistry or biology or stuff like that. But I ended up doing my engineering. I studied biotechnology in Calle University in India, it's in it's in a place called VZ. Avada in Andhra Pradesh. So, so, I mean, I've, from the time I joined, I was like, super interested in, in jumping into research projects. And I've heard at least like when I was applying to different colleges, that this particular university had a very good biotechnology program and the faculty there and pretty good doctors and postdoctoral experiences. So which is true, so I so when I joined there, I started I didn't know what project to do, but like I slowly know once I got into my sophomore year, I started like, appreciating the science behind drug discovery, the science behind cancer biology, the science and genetic engineering. And so that's when I slowly like started into like mini projects and I collaborated with one of the faculty members there I'm with a structure biology project. And that's where my first, like legitimate research experience started. And for some reason, I really felt like looking at things at an atomic level, like looking at structures of proteins looking at atoms, like it was, it felt really cool. You know, because if, if we are able to connect a function of a phenotype or some some sort of function back to an atomic level, I think that is, in my opinion, a very comprehensive understanding about a particular biological function. So, so yeah, so I became very interested in like, knowing the chemistry of interactions, and then the physics of interactions. And so in my first project, that's how I started into structure biology. And so, after I finished my bachelor's, I was able to be I was I was a co author on a paper that came out like, at the time, I was graduating, and I was like, really happy with with that, because that is something. So, I did not know study like at a top university like an 80 or something in India and to be able to get out, get some something out that, that that yeah, that anyone in the world can read. It can be cited by people I, that that felt like a pretty good achievement for me. Yeah. So so yeah. So from there, then I went on to to my Master's at Rutgers University in New Jersey, where I slowly got into computational biology at that point, because so my, my research work as an undergrad, dealt with computational biology, I did not do any experiments there. Like, I used to do some molecular docking studies and stuff like that. And that really, like made me focus on these molecular interactions, which, because like, they can be seen on a computer, and they can be simulated on a computer, I really found it very cool. So, so I did my Master's in computational biology in practice. And there, I learned something called as molecular dynamics simulations. And this basically, is to simulate proteins or like nucleic acids or lipids, under physical forces, like, they're like, basically, you're just taking apart a tiny little part of a cell and putting it on the computer and simulating it and see what it does. And that, that, that that idea itself was like, so fascinating to me, because you're not just understanding the biology, but you're also understanding the physics there, you have to understand the physics because you're, if you're simulating something, then there has to be all sorts of physical forces that are applied, or that are applicable in a cell on the computer. So, so that that really improved my understanding about biophysics, because that's, that's the first time I appreciated the input. I knew the I knew the term biophysics before that, but I actually appreciated how important biophysics is in order to understand biological functions is when I started learning more about molecular dynamics simulations and yeah, so there I used to, so I worked on membrane proteins and simulation of membrane proteins and understanding how how metabolites pass through membrane channels. And that all of that made me so up until then, I was interested in a particular technique or I was interested in a particular field, like, you know, structure biology, computational biology and biophysics. But after that, I became interested in in like, finding a question to answer. Like, I didn't, I didn't care if it was a if it was, if it was a question that only computational biology can answer it for. There's a question of molecular biology, or if it is question of biochemistry or like structural biology, I didn't, I didn't care about that. I, I wanted to come up with a question. Think about how, how I would approach it, and go from there. So, so then I joined University of Rochester, in upstate New York into the biophysics program. And obviously, during the first year, I rotated in several labs. And I started appreciating the research on RNA. Up until then, like, again, like just how my transition from bachelors to Masters was my transition to masters from masters to my PhD program was also similar, where I didn't know much about RNA research. And so when I started learning about how RNA also like physically behaves like proteins, it also falls into structures, very complex complementary structures. And I was like, completely blown away by how I didn't know about this, like we all study about proteins. We all study about secondary, tertiary quaternary structures of proteins and how enzymes, enzyme inhibitors need to be designed and like or how GPCRs are to be targeted. But targeting RNA targeting RNA structures really is is an under explored field, and Rochester University of Rochester and if any listeners are interested in r&d research, the University of Rochester is one of the best places in the US that does RNA research from any perspective like B, computational cell biology, biochemistry, pharmacology, there's like every sort of approach into studying RNA. So, so, so yeah, so there I started learning more about RNA biology, RNA structure, biology and biophysics. And I joined a lab which develop therapeutics that target RNA. So it's a structure biology lab. My basic interest has always been, like I said, studying things at an atomic level. So I am studying RNA at an atomic level there. Like I did a lot of X ray crystallography, and, like isothermal titration calorimetry that I'm going to talk about other binding experiments, such as surface plasmon resonance, and I also did like a structural bioinformatics project during COVID. Because like, we were not allowed to go into the labs. And so, somehow, this project came up. And I started working on that. And that led to a very, very interesting finding, which was published in Jackson, Journal of American Chemical Society. So, so, so yeah, so I basically try to understand how to target RNA structures. My advisor, Dr. Joe radicand, is an expert in RNA, structural biology and RNA biophysics. And he like, taught me about how to approach designing drugs against RNA and what what sort of things to keep in mind while optimizing things and so I learned a lot in those five years where I started off as a person who knew nothing about RNA biology, but at the end, I, I came out with a lot of knowledge in RNA structures and biophysics and the specific RNA that I was working on was a non coding RNA, which is part of the HIV genome. So yeah, so I I'll talk more about more about that when I'm explaining. But, yeah, so so. So from then as I tend to, the most favorable transition from X ray crystallography for any structural biologist is cryo electron microscopy. So everyone so in 2007, cryo electron microscopy won the Nobel Prize and So that's when, I mean, it was very, very well known in the structure biology field, but that's when people outside of structural biology field started appreciating that. So, so I started, I was also one of them I, I really thought cryo electron microscopy is something that it's a new thing, relatively new thing. But it's still there are so many things that can be done with it. And that's still under explored that and cryo electron tomography, these are two things that are immensely contributing in the scientific community to come up with drugs like even in the case of COVID, there's so many structures that came out, and cryo em pipeline is so well optimized at this point that, like people with good resources and money can crank up structures, like every month. So so that it's really impacting the scientific community, like across different disciplines. So I felt so when I was transitioning into a postdoc, so now I'm currently postdoctoral associate in Yale University, where I'm working on understanding the molecular mechanisms of cytoskeleton and motor proteins. So there is so I found a gap in the field. In terms of muscle related diseases and muscular problems, there is the biophysical and structural aspects of understanding these diseases was like very, it was under explored so so that's when I thought maybe I should learn more of cryo em and, like, learn more about how things move inside the cell, like as an how, how motility cellular motility happens, and what happens when, like molecular motors bind to excite a skeletal elements, how are they regulating a diverse set of functions and how these can be applied in understanding muscular problems and any muscle related diseases, because that is, like specially there are so many genetic diseases in the better, neuromuscular, musculoskeletal, genetic disorders, and the focus has always been to come up with gene therapies and come up with like other other therapies that are not really you know, structure biology based, but there are great, and there's like, a lot of important research going on there. But I think structural biology can also aid this and like, help, or, you know, bolster this whole effort of treating muscular problems. So I think I, I felt that was the gap that I think structural biology can fill. So I wanted to learn, because I had like a very strong background up until now, crystallography, molecular dynamics simulations and other computational techniques, which always all of them dealt with structures. So this is the only thing that I was not like, super well exposed to criterion. So that's when I decided to learn more about that. And, and yeah, I joined Yale last year in July, and I'm still going.

Barkha Yadav-Samudrala: 19:05

Well, in the beginning, you mentioned you did your bachelor's from a very small college, but look it you're now you're one of the best.

Dr. Shashank Chavali: 19:16

That is splendid. When I do look back, I do feel very happy about that. And like, I think like, if I can come across from that point to this point, then I think, like anyone can do it. So I would, I would like to suggest to all the listeners that if you're passionate about something, don't give up.

Barkha Yadav-Samudrala: 19:45

Absolutely. Well, thank you so much for taking over this journey of how you got here, and it's honestly quite inspiring. Like having a paper out of your Bachelor's is a big deal. So, congratulations on all your achievements

Dr. Shashank Chavali: 20:04

Thank you so much. Yeah.

Barkha Yadav-Samudrala: 20:07

So, let's get into the topic now and talk about ITC. So, Shang what is ITC and what about it got you interested in it.

Dr. Shashank Chavali: 20:23

So, I do see like you explain the starting stands for isothermal, titration calorimetry. And so, so, in any sort of interaction to happen between any biomolecule that could be between a protein and a protein, a protein or RNA or protein and a metabolite or RNA or metabolite anything it has to that interaction is has to obey certain laws of physics. So, thermodynamics is one thing that is, is extremely important that to govern certain interactions, the favourability of a particular interaction, and topic or an entropic contribution towards an interaction is extremely important to study because if we understand how to do biomolecules or interacting, when that thing is messed up, or when there is a mutation or when there is something that's not working in the, in the cellular environment, we know how to target them, we know how to repair them. So, so, so when you look, when you zoom into a cell, when you zoom into the cytoplasm, when you look at two particular interactions between two molecules, then they should obey the laws of thermodynamics. So, in order to come up with drugs, that can mimic those interactions in order to come up with drugs that can stop those interactions, there's so many possibilities of targeting these interactions. And so, so, when I worked as a as a doctoral student, when I was trying to design therapeutics that bind to RNA, my goal was to come up with cyclic peptides that target RNA, and because, because these peptides are a small chain of amino acids, right, so the, so when I wanted to design these molecules I want, I had to understand how these two molecules interact. And in order to optimize these interactions, in order to improve their affinity, in order to know increase the specificity towards a particular target, we need to understand how they both interact and understanding that thermodynamics is one of the best ways to, to, to, to basically, like figure out how to approach how to target this particular RNA or protein or enzyme or whatever. So that's the reason I became interested in learning. ITC, I wanted to, I wanted to see how to buy molecules. In my case, its RNA and peptide, I wanted to see how they interacted with what affinity they interact, what is the entropic and the enthalpy contributions towards these interactions? And in order to improve this binding in order to improve the specificity, how should I engineer it? How should I optimize it? So these are the questions I wanted to answer and an ITC is the best way to do that.

Barkha Yadav-Samudrala: 24:07

Okay, so when you are designing an experiment for ITC, what are the steps involved and how long it typically takes from start to finish? Could you go over the steps involved quickly?

Dr. Shashank Chavali: 24:23

Sure, sure. So, one thing that is required in ITC is you basically need purified samples, because, like I said, you're looking at interactions between two distinct biomolecules. We need them to be super pure. So if if I'm working with RNA, I need to purify a particular sequence of RNA. And if I'm working with protein, I have to purify a certain protein that is that that could that could but it has to be extremely pure, if there are any contaminations in that, then we are measuring interactions with other things. So, we to make sure we are basically seeing the interaction between two things, we need to make sure, those are the only two things we are putting into that instrument. So, so that is the first thing, which so if the experiment itself doesn't take too long, but coming to that point takes too long. Like, I'm sure you are like the listeners would appreciate how difficult protein purification is. And RNA purification is another difficult step. Because one thing about RNA purification is, there are like, just like proteases, there are RNAs is everywhere. Like, even as we speak, we're expecting out RNA. And so we have to be extremely careful, while purifying RNA samples. And that takes a while. But after we come up with those, I also worked on purifying proteins I worked on looking at the interaction between protein and RNA. But basically, coming up to that point is, it's the most time taking step but after getting to there, so as an example, like I worked on RNA and protein interactions, so So from there, the the using the instrument is actually not super difficult. It's it's pretty simple, given the understand the basics of how the instrument functions, which we'll go through in a minute, but so, the, so, yeah, so, the time that usually is required to complete this experiment is probably I would say, three to four hours, and now the instruments have gotten smarter. So like some, some experiments, take like a, like an hour, or, yeah, I would say an hour with like with the most state of the art instruments, I'd say it would take about an hour and so, once I have the purified protein and whatever biomolecule then So, the the instrument has two cells cells as in like two chambers, basically, one one chamber is filled with water, just just water, distilled water and the other chamber is where we have our sample. So, one of the samples whether it can be a protein or RNA or any biomolecule usually it is the if you're looking at the interaction with between a protein and a small molecule, then the small molecule the protein will be in the chamber, and there's a syringe in which you load your small molecule. So, in my case, the RNA is in the chamber and the peptide or protein, the one that should go and interact the target usually is in the chamber and the one that is that needs to go in is usually in a syringe and that syringe. So, once we load the RNA in the chamber and protein in the syringe, we basically start the experiment. So, what the instrument does is it releases like extremely small amounts like nano liters of the the sample from the syringe into this chamber and it mixes the syringe mixes the liquid or the solution in the chamber. So, if an interaction happens if if a binds to be and if that is an exothermic reaction, which means if it's releasing heat, then the instrument senses that heat and it utilizes power to increase or decrease the temperature of the reference cell which is which has water so, like I said that two chambers one chamber has water in it, the other chamber has our sample in it. So if there is an interaction that happens inside our sample chamber, then it will either release and heat or it will absorb heat. If it's an endothermic it absorbs heat. So once it absorbs heat or once it releases heat the temperature changes in the chamber. So once that temperature has changed, the sensors recognize that and there is a feed back into the machine that increases the power to do Heat or like cold reference chamber. And once that is done this power is basically integrated over time to generate a heat. So, we basically get from the experiment from the first injection, we get like a heat value. And now, now this process goes on and in an iteration, so, so basically, once a small nano liter amount is injected into the chamber, you're going to inject the same amount again. Now, if there are like 100 copies of your protein or RNA there, and if in your first injection, like 10 copies of them have been occupied by the, by the small molecule or your drug or whatever, then that like few more left, right, so, so we So, in this iterative process, you keep injecting sample, and you keep measuring heats, you keep increasing the, the machine does this, but the machine increases the power to balance the temperatures in the reference chamber and the sample chamber. And slowly the ones once like all of the protein molecules or all of the RNA or whatever biomolecules are saturated with the ligand that you're injecting, then there is a plateau that that is reached. So, basically, you're going to get a curve, that that is called an isotherm. And the heats keep reducing as we go towards saturation, because like there are fewer copies that need to be occupied after each injection. So fewer copies of biomolecules so, so, so, this, basically, the machine basically measures the power required to balance both the temperatures, but that power over, if it's integrated over time gives you the heat, the energy that is calculated, and the slope of this curve gives the binding affinity of, of that ligand to the molecule to the macro molecule. So, this basically gives us two things, one is the binding affinity, and one is the heats of binding. So, once we know more bees to things, we can use laws of thermodynamics to calculate the whole the free energy of binding, which is delta G. And we will know about we already know delta H, and the only thing we need to know is T delta S. So, I just want to remind the formula that we all usually use for the Gibbs free formula, which is the delta G equals delta H minus d delta s. And so, so T delta S stands for entropy, delta H stands for enthalpy and delta G is the Gibbs free energy. And the, the the so that, once we calculate these, the lower the delta G is, the more favorable the interaction is. So, so, so if if, if we have two interactions with one delta G is, let's say, it's like minus 10 kilocalories per mile, and then there is another interaction that is minus five kilocalories per mole. So, the first interaction is more favorable than the second one and that so, so, that is like the thermodynamic understanding about how, so, any anything that is more favorable has less energy, like lower energy. So, yeah, so, are think about it in a more intuitive way, like, if something is favorable, we don't have to apply a lot of energy, it happens effortlessly. Right? So, so, so, it is it is a crude example, it's a crude analogy, but that's, that's, that's a basic thermodynamic understanding of how low energies are, contribute to favorable interactions. So, I can obviously, like, like, we don't have to get into the details of how we don't have to get into thermodynamics here. Thermodynamics, but but sub delta G. Actually, maybe I forgot to mention this, but delta G, we know delta is from the experiment, but delta G is calculated using a formula delta G. So equals negative RT log KD or negative log k and k here is the binding affinity. So, once we get the binding affinity from the slope, we calculate the delta G we calculate delta H. Now, from that we calculate T delta S. And now, magically, we have the enthalpy contributions, which is delta H, the entropy contributions the T delta S, and the overall favourability of an interaction it is delta G. So, so, we don't have to do any of this the software does all of this we just need to inject the samples into the syringe and into the sample chamber and all the magic is done by the by the instrument.

Barkha Yadav-Samudrala: 35:47

Wow, that was a very detailed and easy to follow explanation. Thank you so much for that, like I could picture the entire thing in my head and I have no idea. So, another thing I wanted to ask is are there any alternative techniques to ITC?

Dr. Shashank Chavali: 36:12

Right. So, there are and there are not as in so, so one of the main reasons to perform IPC is two main reasons one is to know the binding affinity of ligand to molecule macromolecules. And the other is to know the thermodynamics. But usually, knowing the thermodynamics is not always necessary, like it is necessary to optimize drugs to know. Like, I mean, it depends on the question. Basically, it depends on what if you're, if you're developing a drug, and you want to, if you have like tons of drugs that you want to screen, ITC is not a good method at that point, because you need a lot of amounts of your protein and drug and you have to add one after the other. There are more high throughput ways to you know, filter certain compounds out. And yeah, so So at that scenario, ITC is not the best experiment to do. But if you have like a few heads, and you want to optimize those heads, you want to increase the binding affinity, you want to change the functional groups of a drug molecule, you want to change the protein sequence, you want to change the peptide sequence that we know. And the most useful approach of ITC is how we can relate structure to function. So, we are basically if so, if we look at a crystal structure or cryo em structure, and we see two amino acids interacting, and if we mutate that amino acid, and we perform ITC, we will see that interaction lost in terms of how much delta G is lost. So if one hydrogen bond is lost between one amino acid between two interactions, then we would get delta delta G of negative 1.5 kcal per mole, which is the energy required to form one hydrogen bond. So that is the level of detail we can understand. So it's a bit of a digression, but the alternate to do this is we can perform other binding experiments, which is like surface plasmon resonance, you can perform fluorescence assays to get the KT. Once we get the KT, you can also like biophysics, people will, will not like this. But you can in theory, calculate delta G from KD because all you need to do is plug in the KD value into a formula delta G equals negative RT loc KT. But the biggest problem with that is Katie has units right like cake, like binding affinity has micromolar or milli molar or nano molar or whatever units. And you cannot take logarithm have something that has units. Yeah, so the biggest problem so it's not delta G anymore. It's something called as delta G naught, which means delta G at steady state. And steady state assumes one molar of the solution. So you're basically in that logarithm. You're, you're you have a denominator of one molar, which is the concentration so the concentration values units gets cancelled and now you can take a log for them, but you will not get in therapy can entropic contributions from other binding techniques. And sometimes you don't need them. Like I said, if you want to filter out a lot of compounds, if you have 100 compounds, and you want to filter and come to like 10, then ITC is not a good way to do that more, they're more like very effective high throughput, fluorescence assays and lots of other assays that you can do to, like, filter out the stuff that are not binding to your target. And once you have a few, then you can start performing those and step by step, you know, like, it's like a Lego piece, you're constructing something. And if something doesn't fit there, you're taking that out, you're putting another thing, and you're specifically designing something to bind to that pocket because in ITC experiments, now, here, you're only looking at to two things, you're not looking at anything that is complicated. So, that actually increases chances of coming up with something that is very specific. So, so, yeah, so there are alternatives, there are actually very good alternatives to ITC. But those alternatives, don't ask the same question. Sometimes, sometimes those are important and sometimes ITC is important. So,

Barkha Yadav-Samudrala: 41:26

yeah, it depends on the question you are like the answers exactly for like your experiment. Exactly. So, is this technique user friendly or is there like a big learning curve to it?

Dr. Shashank Chavali: 41:44

Yeah, so, so, the technique itself is user friendly, but the most challenging part that is kind of underappreciated is how these results can be interpreted. So, So, usually, if you know one compound binds to one protein molecule, the binding stoichiometry is like one is to one that is simple, that that is something ITC has has figured out. So, that is a simple model. So if you get data points, it is easy to you know, come up with a curve that fits all the data points, if the binding is one is to one, but when the binding gets complicated, that is if it's like cooperative binding, or if there is sequential binding, or if there are two ligands binding to one, one protein molecule on two different sites, or so. So that is when the data looks different. It's not that straightforward. And interpreting that data coming up with actually like people develop algorithms like statistical algorithms to fit these complicated data points. And because everything that we understand that we take out from an ITC experiment, the most important thing is binding affinity and thermodynamics. And this is heavily dependent on how good your fit of the data is. And if your fit is like noisy, if it doesn't cover all the points, then your value is not believable. So you're then that is that is the thing that is very challenging for someone to understand. So the technique itself is very easy to master. You can like once you beautify your samples, like injecting sample, you have to be a little careful about using the instrumentation, which is the case with any other like biological experiment. So that that is all you need. But once you like get the results, making sense out of it, and coming up with good facts, understanding, believing those facts, like doing multiple replicates to improve noisy data. All of this is it's time consuming and it's a difficult thing to master but it's it's, it's, it's achievable. But if your system is complicated if your binding is complicated, if something fishy is happening inside of that chamber between two molecules if it's coming on and off. So there are so many things like that. And that is when it becomes a little tricky. But if if if it is simple, if mining is simple, you want to just add you know that that's something binds one is to one you just want to optimize it. Then you do the experiment you get a decent looking curve. The software fits it Are you you will get a KD, you will get a delta G, you will solve everything. That is great. So, but that's not the case easily. Because some things are complicated. But yeah, so yeah, to answer your question, it is user friendly. But it can also be challenging when you're analyzing the data.

Barkha Yadav-Samudrala: 45:20

Yeah, if science was easy, I think everyone would do it. Exactly excellently. I think we spend more time in troubleshooting than doing the actual experiment is than doing the actual action. And that's how we all learn that era. Yeah, you're absolutely right. If everything works, magically, and you have a one is to watch documentary and like, everything is great. There's nothing you have learned other than like loading things onto the machine that which anyone can do once you show them how to do. So. So the, the actual crux of the science or actual crux of the problem comes when you're analyzing that data. And you're inferring the mining thermodynamics and what functional group is doing mining there? What, maybe there are two binding sites, maybe these two binding sites are cooperative. And these are so I mean, I can like point you towards so much literature that does a lot of cool stuff. And in coming up with these data, fitting algorithms, and yeah, so like you said, it's the troubleshooting that teaches us more than like, doing that experiment. Yeah, absolutely. But that is fun to

Dr. Shashank Chavali: 46:48

write. That is that is fun, sometimes frustrating, but it is fun.

Barkha Yadav-Samudrala: 46:52

Yeah, it is very frustrating. But when you are doing troubleshooting, and you work it out. That is I think, a very

Dr. Shashank Chavali: 47:03

dark sense of accomplishment is yes, I Yeah, we have all been there.

Barkha Yadav-Samudrala: 47:10

Yeah, I've I've been working on one of my experiment from I would say last three years and very well. It's called invivo. calcium imaging. It's kind of it's not kind of it is very tricky. Where you image

Dr. Shashank Chavali: 47:28

I have done in vitro calcium imaging,

Barkha Yadav-Samudrala: 47:32

oh, no, this is in vivo. And it's quiet, a unique technique where you inject and it's live. So you see the neurons in a live animal moving around in the cage. That is so cool. And it's so yeah, so that took me like three years and I finally optimized it like last month, and when I got it, I literally screamed oh my god, I think

Dr. Shashank Chavali: 48:09

it's a great achievement. Yeah, I can like I saw I did it in one of the rotations and I thought invitro itself was so difficult. And I can totally imagine how in vivo imaging can like yeah, I Yeah. You should be proud of that.

Barkha Yadav-Samudrala: 48:29

Yeah, it's it's pretty fun. But yeah, I think the more challenges the sweeter the victory feels.

Dr. Shashank Chavali: 48:37

Right. Absolutely.

Barkha Yadav-Samudrala: 48:40

Well, so, getting back to this, what would you say are some of the advantages and disadvantages? I know you covered a few of them. Just to give you an overview.

Dr. Shashank Chavali: 48:52

Sure. Sure. Yeah. Yeah. So, the advantages are, you will know basically, the thermodynamics of binding between two biological macromolecules or even like small molecules. The like I said, the biggest advantage here is how we can relate these results into structures. Now, we have crystal structures, we have criterium structures, the protein databank which is like a database of all structures that have been solved in date, like you can like randomly take one structure out, you can like Express purify that protein, you can mutate that protein. And if that to eliminate a particular interaction, and you can see that interaction gone in an ITC experiment. And I mean, see as a not visually but mathematically, you can see that thing is gone. So that I think is the most powerful application of IDC because because if you, if you, if you understand that, then and if you have a complementary structure, then you're like I said, you're like you're constructing a drug, you're like engineering a drug, you're engineering something that is so unique to that particular target that, like it, it has to work. And so so that I think is the best use of ITC is how we can relate structure and thermodynamics at the same time with one experiment. So that, and once I sometimes we don't even need the structure, if we have like a structure of a starting structure of something. And, okay, we have a starting structure of a protein, but we don't have a drug. So we, if we have narrowed down to like a list of 10 drugs, now we are like, basically docking experimentally, one drug after the other in my case, actually, it helped me so much. So I have the RNA, and I'm targeting it with a peptide, right. So, so when when I target. So when I use the first peptide, I got like a binding affinity, I got like, some enthalpy contribution, and we sat down and we thought, okay, we can still increase the enthalpy, we can still increase the favourability of binding. So we we first engineered the way the peptide is linked. So it's a cyclic peptide, it's not a linear peptide, it has to be linked. So it was linked through a disulfide bond first. So that two ends of the peptide have sustains cysteine residues and these two cyclized. So so that produced some results, but we slowly optimized that cycling strategy. And so we collaborated with a cyclic peptides expert. His name is Dr. Judy Farsan. He is in the chemistry department at the University of Rochester who has this extremely great sense of chemical intuition. So, so they basically engineered these linkers, where we took out the disulfide, we added a methyl group, then we added an ethyl group, then a butyl group, then like Xiling group, like we basically increase the length of the carbons that link these peptides. And we see improvement in binding, we see improvement in the enthalpy contributions. And then we have Okay, now we have the peptide with a good linker that connects that cyclize Is it now we changed the amino acids slightly. So because RNA is like negatively charged, we have like cyclic peptides that are full of positively charged residues so that they can bind to each other electrostatically. But like you don't want it to bind to every RNA in the body, right? Like if you if you do that, like you're basically it's a it's a canister, so So, so we engineered it in such a way that we added one arginine, or one a positively charged residue at each place. And we saw how it affects the binding we did some other renal controls so that we had some commonly seen RNA inside the body that we tried to bind our drug with. And that didn't bind to that. So, so we tested the specificity of this is in theory called structure activity relationship. So basically you have a structure, you have the activity of your compound, and you're optimizing it with the help of ITC and like the structure, so, SO SO, SO Sunday, sometimes you don't even need the structure like you you can have like a protein or an RNA and you can have your drug and you can slowly optimize it and once like you have one drug that is binding while and it has great anthropic entropic contributions and everything. Now, you can try to solve the structure of the drug along with your RNA or protein and you will see those interactions there. So, so that that is an extremely complimentary way and like it is, to me at least it's like fascinating how you can optimize things that is like so translatable into coming up with drugs. Yeah, so this is one of the biggest advantage and the disadvantage is like as So the, the binding becomes complicated. The fitting is the analysis of data and fitting is a very big problem. And sometimes you have to write your own, like fitting code like statistical, you need to come up with your own algorithm to fit that data. And I've seen that so one of my colleagues when I was a doctoral student had this very cool project where he had like an RNA molecule that has two ligands that are stacked over each other that bind in one pocket. And the so there were like two dips in the curve that designed the ITC experiments. And it's difficult to fit with all the statistical algorithms we have, it's difficult to fit some random data. And even if we fit, we don't fit, fit can be believed. So we have to collaborate with computational group to come up with a particular algorithm to fit that data that is believable. So that is one big disadvantage. If the if the mining becomes complicated, it's difficult to analyze. And the other disadvantage is, you need like a lot of amounts of your biomolecules like you need to purify a lot of protein, or you need to purify a lot of RNA, you need a lot of compound, small molecules or peptides or whatever you need, like large amounts of it. So that that sometimes is a rate limiting step, because you spent like weeks in purifying protein. And you may like, especially for someone if they're starting out in this and they may make some mistake and you use up like 50% of your protein in one experiment, and you have to do at least three experiments, right to make sure like to replicate so that like your answer is more believable. So So yeah, so then it becomes a little difficult, but these are the two main disadvantages.

Barkha Yadav-Samudrala: 57:02

Okay, so let's talk about the cost involved here is this instrument and even the consumables like expensive or any regular lab can set this up?

Dr. Shashank Chavali: 57:17

Yeah, so it is slightly on the expensive side. It's not as expensive as an electron microscope, which is like a million dollars or something. But it is it is slightly on the expensive side. And usually, like I've seen that labs, like, like a department has one ITC so that like people can use it, not everybody wants to use it every day. So like, it can be a departmental thing where certain labs can buy, buy together, sometimes like so in our lab, we used it extensively. So in Rochester when RPI got a grant, and he so we had our own ITC machine, I would say it is on the expensive side. But if it is something that someone wants to use, and if they see a lot of application for it, then it's a great investment.

Barkha Yadav-Samudrala: 58:16

Okay. So let's talk about fight facts. Does ITC have any fun facts?

Dr. Shashank Chavali: 58:26

Right, right. So it does. And I think it's pretty cool. So like, AIDS was such a big, I mean, it is still a big problem. There is no functional cure for AIDS, but like the it was, it was much worse back in like the 1990s or early 2000s, where there was no antiretroviral therapy. So the antiretroviral therapy is like a combination of drugs that basically target different points in the HIV lifecycle. So all of the drugs that were developed to target these different points in the HIV lifecycle were like targeting the HIV protease, for example. The drugs that target the HIV protease were actually developed with the help of ITC. Like they were optimized, yeah. So they had the they narrowed down from like a bunch of drugs to a few. And once they had a good head, they like functionally improved the binding of the drug binding thermodynamics, they improved the enthalpy and entropy contributions. And so that actually is helping a lot of people right now. So we don't hear a lot of HIV AIDS related deaths and now die like how we used to hear probably two decades ago. And antiretroviral therapy, even though obviously, like it, it does induce side effects for sure. But it still is like, improving the quality of life for a lot of people. And, and this this, this therapy wouldn't have been as good without the efforts of ITC. And like without the contributions of this, this methodology, so that, according to me is like super fun.

Barkha Yadav-Samudrala: 1:00:34

Yeah. Like I work in, in in an HIV lab, and I had no idea for this. So thank you so much for sharing.

Dr. Shashank Chavali: 1:00:43

Yeah, yeah. My pleasure. Yeah, it's really, yeah. Thanks for doing this. So that like, many people would now appreciate how, how ITC is involved in drug discovery. And maybe they also want to use it, they have some molecules and they want to optimize. They want Yeah, yeah, it's, you can you can collaborate with people if you feel if you're not an expert in that. That's the purpose of science.

Barkha Yadav-Samudrala: 1:01:12

Absolutely. And I think that is the purpose of this podcast, connecting people and getting to know more things. So my last question for you would be, could you suggest any interesting articles or protocols regarding ITC, so that I can link it down in the description for our listeners to look up?

Dr. Shashank Chavali: 1:01:39

Sure, so there are two or two papers, I want to suggest, both from our lab, which I think covers like a good set of applications of ITC. One is my own paper, it's called cyclic peptides with a distinct arginine for motive recognize the HIV trans activation response RNA, in vitro and in cells. So, this is basically part of this published in Journal of Biological Chemistry JBC. And so, this has the story of how we optimized peptides to mine RNA with the help of ITC results. So, you will see like a ton of ITC thermo grams in that paper, that is one paper and the other paper is a more challenging paper from from from the same lab, which, which, which I talked about briefly. It's called a small RNA that cooperatively census to stand metabolites in one pocket for Gene control. This was published in Nature Communications, and this is the one that has a complicated mode of binding. And in this paper, like you will see how the author's came up with a way to fit these complicated data. So I think these these two papers, I'm being selfish and promoting. But I think I think it captures the essence of ITC pretty well.

Barkha Yadav-Samudrala: 1:03:26

Yeah, yeah. Okay. Well, with this, I am going to end today's episode. And thank you so much Assange for being here and sharing all everything about ITC.

Dr. Shashank Chavali: 1:03:41

Absolutely, thank you so much for doing this. And I wish you all the very best and I hope this continues for very long. And thank you for like integrating and uniting like scientists across across the world.

Barkha Yadav-Samudrala: 1:03:56

Well, thank you so much. And it has only been possible because of scientists like you who are willing to participate in it. So thank you so much. Of course, yeah. Thanks so much. Yeah. And listeners. I will catch you next week on another episode of Nerdrx podcast and in meanwhile, if you have any suggestions for the topics that I should cover, or if you would like to join me on the podcast, and share something interesting, please email me at Barkha@Nerdrxpodcast.com. And remember, it's good to be a nerd, bye.