Artwork for podcast Translating Proteomics
Are Proteins A Myth? With Special Guest Professor Neil Kelleher
Episode 92nd October 2024 • Translating Proteomics • Nautilus Biotechnology
00:00:00 00:47:34

Share Episode

Shownotes

In our Translating Proteomics episode titled "Harnessing Proteoforms to Understand Life's Complexity", Parag and Andreas discussed why proteoforms are important in a theoretical sense. In this episode, Parag sits down with Northwestern University Professor and proteoform pioneer, Neil Kelleher to dive deep into the biology of proteoforms. They cover:

  • What proteoforms are
  • Examples of the importance of proteoforms
  • The scale of and technological advances needed to meet the challenges of proteoform biology.

Some examples of the power of proteoforms covered in this episode

  • Recent work from Neil's lab showing blood proteoforms can help predict liver transplant success (Melani et al. 2022).
  • Work form Ying Ge's lab showing changes in troponin proteoforms correlate with varying degrees of heart disease (Zhang et al. 2011).
  • The BioTyper - a mass spectrometry-based device that can identify different kinds of microbes.

Additional proteoform resources

Transcripts

Announcer:

On this episode of translating proteomics, host Paraag Malik discusses the importance and challenges of proteoform research with special guest and northwestern professor Neil Kelleher.

Their conversation covers what proteoforms are, examples of the importance of proteoforms, the scale of and technological advances needed to meet the challenges of proteoform biology. To get the conversation started and introduce Neil, here are your hosts, Paraag Malik and Andreas Huma of Nautilus biotechnology.

Parag Mallick:

On this episode of translating proteomics, we revisit a particularly important topic in the world of proteomics, proteiforms. As a reminder, proteiforms are the single molecule variants of gene encoded proteins.

Each proteform is defined by its full set of modifications, regardless of source. And proteiforms are the versions of proteins that we actually find in biological systems.

To get a better understanding of proteiforms and their importance, I had the pleasure of speaking with Professor Neil Callaher of Northwestern University. Neil is the preeminent scholar on proteforms. He literally coined the term.

And I'm very excited to share our conversation, covering concrete examples of the importance of protein forms, the scale of proteiform biology, and also how we can make proteiforms a routine part of biological inquiry.

Andreas Huhmer:

So maybe this is a timely conversation because proteoforms seem to have their moment. There's new technology evolving that specifically allows us to measure proteo forms. So I'm looking very much forward to the conversation with Neil.

Parag Mallick:

It was really fun. Take a listen.

As you've just heard, I am very excited to have the chance to chat with someone who is the world expert on proteiforms. Neil Callaher. Neil is the Walter and Mary E. Glass professor of molecular biosciences, chemistry and medicine at Northwestern University.

He did his graduate work with Tige Begley and Fred McClafferty at Cornell, and his postdoc with Christopher Walsh at Harvard Medical School.

His training in mass spectrometry and entomology really drove his research program of hundreds of publications and a tremendous focus on top down proteomics, creating the consortium for top down proteomics, and initiating the human proteiform project. Neil, on a previous episode, my colleague Andreas and I discussed proteiforms in a. Yeah. Broad, but somewhat theoretical sense.

To start this conversation, I'd like to hear from you directly. What are proteforms?

Neil Kelleher:

Yeah, parag, thanks and great to be with you on your podcast.

And, yeah, so, proteiforms are just basic biochemistry of protein molecules and it is the mature molecular product from a gene that gets made in our bodies into proteins all the time. So it just means to indicate the complete composition of that molecule.

So it would involve decorations to proteins and some other technical words that we'll get into, but it's just the complete compositional molecule of what we call colloquially a protein.

Parag Mallick:

And so let's go back to:

Neil Kelleher:

Yeah, so the word is also a movement. There's a lot behind it, and a lot of people feel strongly about words and semantics.

d came out of a conference in:

There were isoforms and protein variants and, oh, yeah, there's that one with that decoration and that phosphorylation, methylation, acetylation, all these different forms. Oh, and mutations, genetic changes to protein coding regions. Oh, those are also a thing.

And so that word proteiform evolved, and once it was, it was like a moment of magic in the room when that sort of hit, there were different forms of proteins, and you can imagine the moment.

Parag Mallick:

And did you feel that those other terms to describe proteins and their combination of genetic variants and splice variants and post translational modifications that you really needed to capture that incredible diversity, and it was just not being sufficiently well described in the community at the time.

Neil Kelleher:

, that's what came out in the:

That single term is in the title, and it just. It just encapsulated all the sources of compositional variation that could occur in a protein.

And since then, there's been edge cases and things that we've worked together as a consortium to. To parse. But, yeah, that's. That word now is a thing.

And if we're ever going to get to regularizing the proteome, that is, to more regularly handle, we're going to have to get proteforms under our belt. And first things first, what do we call it?

And now we have five levels that describe different quality levels of proteiforms in the underlying analytical techniques to discover and measure proteforms.

Parag Mallick:

So tell me a little bit more about that. So the five different layers, I'm not familiar with that. What does that. Walk me through that.

Neil Kelleher:

Yeah, just a few years ago, like, a level one proteiform is perfect, you know, all of the atoms that comprise that molecule, and you know that the phosphorylation is exactly at serine 720. And, you know, it starts at this methionine at position one.

And, you know the amino acid at the very end of the c terminus, it's called, of the protein and everything in between. And, you know, oh, it's got that mutation is coating, and it changed a threonine to a tyrosine, for example, and all of the splicing that occurs.

So the underlying base sequence is basically all the atoms are. That's a perfect, perfect level one proteiform.

Level two could be, well, I don't actually know that the phosphorylation is that that serine, say, at position 100, or it could be at position 108. That ambiguity creates a type of level two proteiform, and then it goes on down in terms of quality from there.

Parag Mallick:

I see. And so tell me a little bit, maybe coming up to a slightly higher scale, why are proteoforms important? Why should we care about them?

Neil Kelleher:

It's just basic biochemistry that we just want to understand exactly the molecules that are in our bodies.

We've done it for DNA for the most part, and now we need to do it for our proteins, because we're built of proteins, which is a fundamental concept that is surprisingly difficult to communicate outward, but it's just so foundational that we're going to have to confront this at some point.

Parag Mallick:

Yeah, it's so interesting that you say that, because I've seen in the field people confusing post translational modifications with proteiforms. And so I'd love to get your thoughts on, first of all, why does that happen?

Why are people using this term proteiform to talk about post translational modifications? Simply, how are they different from each other?

Neil Kelleher:

I totally get your vibe right there. It's like, why do I have to go this two step dance to get to ptms that matter? Why do I have to measure protiforms to get to ptms?

Can I just go after the PTms? Yes. And the world has done that for the last decades, trying to find the ptms that matter, but yet we don't have the proteiform perspective.

It's like a bird's eye perspective of a particular protein. And that term protein, it's just too simple to capture our biology.

That's why proteforms matter, and that's why you have to measure them to find the ptms and protein isoforms that matter. And the argument that is a long logic chain to get through that, but it's just true.

Parag Mallick:

I'm going to come back to that.

In a moment, just because one of the things that I always think about is that the PTM is this amino acid modified or not, and the protein form is really the combinatorial explosion on top of that that says, I have three different post translational modifications present on the same molecule, or a combination of a splice variant and multiple post translational modifications. So it's really that pattern beyond any single site that defines what makes a proteiform.

Neil Kelleher:

Yeah, and if you think about it, yes. And let me build on that.

If you had three phosphorylations on a specific isoform and that was the particular drug target to really shut down a signaling cascade that's gone crazy in my cancer. Like, okay, I have three phosphos in combination.

And if you imagine that, that basically elevates you above the noise band, if that is actually the molecule that exists, you'll see it at the proteform level.

But if you try and piece together those four things, you threw out an isoform specific with one, two, three phosphos, all in, which is actually what we just measured in a protein called Mec one in a model of melanoma. Sorry, geeked out there. Couldn't help. Perfect.

Parag Mallick:

That's like I said, and we're going to dive into that example and more in just a second.

Neil Kelleher:

Cool. So that is what I've experienced in my career is like, oh, my gosh, like, this is really 20 years ago. This is hard to measure.

But though the biology is there, the function assignment, the biomarker, relevance. Because biological noise frustrates measurement science all the time.

So this is why prote forms matter and why year by year, this will just continue to grow.

Parag Mallick:

So somebody said there are no proteins, there are only proteiforms, and I'd just like to get your hot take on this. What does that mean to you? Do you believe that?

Neil Kelleher:

Yeah, hot take on. Yeah, that proteins. I heard someone else say proteins are a myth, that they don't actually exist, that they actually have protein.

I would say the more complex organisms that, um, like. So for example, us versus E. Coli, you can get an E.

Coli gene that, um, you know, will trim the n terminus and create one dominant or two dominant proteiforms. And that's just like the central dogma, and it doesn't get altered, uh, very much or at high abundance. Okay.

But when it comes to humans, I I'm then more aligned with your statement.

So my hot take is, in our biology, we have this escape from the central dogma, where throughout evolution, and we have a proteiform program of life, it has evolved to constrict the creation and regulation of proteforms.

And yes, there's a noise band, yes, it goes awry in disease, but there is this dominant canonical proteiform program that we do not have the tools to read out yet. But, like you, I think we are on the precipice.

You know, within less than a decade, I think we can reach this new plateau where we have tools that match the scale of our biology at the protein and proteiform level.

Parag Mallick:

So I think that's a really great segue. I'd like to come back to a little bit of your history.

What was the very first time, what was the very first protein that you saw a proteiform acting differently than a collection of ptms?

Neil Kelleher:

Yeah, that would be on histone h four. Histone h four through the cell cycle.

That really showed me, like, okay, a ptM, like lysine 20 on this protein, histone h four histones pack DNA into the nucleus. And people around that time were all talking about the histone code and that multiple ptms were combining and were actually the algorithm of life.

And. And that lysine 20 on histone h four gets methylated in one methyl, two methyls or three methyls added to this lysine.

But how does that occur in combination with acetylations elsewhere or a phosphorylation elsewhere? This was quite open.

And just from doing this kind of bird's eye perspective and then lining up ten different proteiform snapshots from the cell cycle, you can just see this progressive methylation happen through the cell cycle. And I was like, oh, well, that's the logic right there. And the acetyls don't really have much to say about that progression.

Parag Mallick:

So that there's a distinctive pattern over time where it starts off with one modification of one type at one position and then adds on a second one at a different site and sort of stacks up from one to multiple modifications, ultimately defining the behavior of the histones.

Neil Kelleher:

Yeah. And there is other biological states where the acetylations matter. Like if you're inhibiting with a cancer drug deacetylation.

Here we come with the proteiform dynamics, then reveal how much those acetyls build up. Because you're inhibiting an enzyme that should be taking them off.

Parag Mallick:

That's great. That's great. Tell me a little bit about a few other examples that people may not be as aware of.

You mentioned Mac one, for instance, or troponin or PSA.

What are the most glaring, important examples that we've seen where protein forms either drive biology or are better biomarkers or are better drug targets?

Neil Kelleher:

Wow. Yeah. That's a great wind up pitch. And let me try and hit what I had three fastballs coming at me, but one is PSA.

You mentioned prostate specific antigen PSA. It's generally known. The New York Times wrote about it years ago that it's a really bad biomarker, that, you know, PSA, the protein.

We know there's over 100 proteforms of this sucker. Right? There's glycans that get attached.

There's polymorphisms that change the amino acid sequence person to person, you know, and so no wonder that it's a bad biomarker. You're just looking at the level of 80 to 150 different molecules, and you're looking at the total level.

We need to look at the levels in prostate patients. So line up 100 men in stage one through four prostate cancer. Line them up. Get the technology lined up.

This is actually a three or $4 million study I'd love to do, and it hasn't been done yet, but the technology is there to do it today, and it wasn't five years ago. So that's an example of, like, okay, maybe five of those proteforms of PSA are really specific to stage of cancer.

That's the kind of thing that has to be done in the biomarker diagnostic space.

Parag Mallick:

So that sounds a little bit like a hypothesis, Neil, as opposed to a place where we have evidence that says, hey, psas that have these three modifications are cancer specific, and psas that have those n modifications are just evidence of enlarged prostate or something that is not prostate cancer. Do we have an example that we can point out today to say this proteiform or the set of proteforms is better?

Neil Kelleher:

right as we wake up today in:

It is the biotyper, and this is a proteiform detection engine that's now in thousands of hospitals to tell you very quickly, very cheaply, what species of bacteria is in your infection. That's the case that, and it would not work with small peptide surrogates or these little pieces of proteins.

It really has to function at the entire proteform level I revert from what is biotypert?

Parag Mallick:

What is it measuring in particular?

Neil Kelleher:

Yeah, it's measuring about 50 ribosomes are what make proteins. It measures about 50 ribosome proteiforms from a bacterial species, from a bacteria that was grown for a few hours in a dish, from an infection.

That's the assay. And it was launched by Brooker Deltonics many years ago. But it kind of proves the bot business and science case for those who want a mature example.

Yes, we have in more complex disease. So detecting a bug that's in a person.

Okay, that maybe is a little easier than, okay, now I have cancer, or I have another example I'll give you, is in transplant. In transplanting organs, we have a panel of 24 proteforms, and so there's a few prote forms of, like, chemokines. This is in Pb.

These are in the white blood cells, the immune cells of your body, which are the first to react and reject a transplanted liver in this case.

Parag Mallick:

So, Neil, these are great examples.

Neil Kelleher:

Of.

Parag Mallick:

A biomarker and in the case of histones, something that is driving biological function. Are there other places where, when you think of proteiforms, that if we don't understand proteiforms, if we can't characterize them, we're limited?

Neil Kelleher:

Yeah.

The troponins are a great example in cardiology, the work of yinga in Wisconsin, she really clearly showed that as your heart, as you get enlargement of the heart in pig models and in humans, that you have this protein troponin, that has just a simple story. Three proteiforms, and there's two serines right next door to each other that go pop, pop. And that's the. That's what happens in healthy bio.

That is the healthy proteform program running in my. In my chest. That's what you want.

And then as you get enlarged in the heart, that phosphorylation, those phosphoproteforms go down and to an alarming degree, in very enlarged hearts. And so that actually tracks progression of disease.

It's quantitative now that you know that you then could go back to other types of assays and build assays that are proteiform informed.

Parag Mallick:

Yes.

Neil Kelleher:

And so there's a distinction between, okay, how do we discover this biology, which in retrospect, is really clear, the signal's really clear at the proteform level versus proteform resequencing or proteoform informed biology? And that distinction, we could get into but I think it's just so it's that bird's eye perspective of just what's going on to your proteins.

Parag Mallick:

So I think this is a great opportunity to talk about the goal of building a proteiform atlas and a human proteform project. So this is something that you've been driving with support throughout the community.

Tell me a little bit about the goals of this project and what are you hoping to achieve with it?

Neil Kelleher:

ust described, something like:

And of course that means different organs like the kidney would have something like 80 cell types, and we would determine it at a given depth, equivalent conceptually to the four or five x coverage. So you'll recall the human genome project had a four x coverage when it was first done.

Now it's 40 or 50 x coverage, because look what happens after you regularize a biomolecular landscape in the human body and create all these economies of scale after it. That's what the project is meant to do.

ut the project itself is say,:

Parag Mallick:

Think I realized there were twelve fluids in the human body.

Neil Kelleher:

So there's a lot of, there's a lot of side fluids and some of them, your viewers, your listeners may not wish to hear all twelve body fluids.

Parag Mallick:

Fair enough. Let's not, let's not go into that.

Neil Kelleher:

But yeah, that's the Atlas or the human proteiform project would create along with the technologies that finally would at least regularize, domesticate, sequence the proteome and to make it not so enigmatic that we as a species, we domesticated plants 10,000 years ago. We can do this as well.

And after all, precision medicine 2.0, it, it really requires us to confront the complexities and harness the proteome, which right now is enigmatic and open ended and just makes drug discovery and diagnostics just really challenging.

Also, for the business sector, it's really difficult to establish drugs and do so with anything more than 5% success rate through all the way through clinical trials. So this is what the project is meant to do, level up our understanding of ourselves at the protein level.

Parag Mallick:

Great. So Neil, I'd like to ask two quick follow ups to that. One is the project that you've just described. Sounds immense. Is this 100 years of work?

Is this, what is the scale that we're talking about to complete the human protein form project?

Neil Kelleher:

o seemed quite challenging in:

How are we going to possibly sequence that within ten years? It was done. And that's what happens when you bring the best minds in the world to the problem and the belief that it can be done.

That's not to say that it is not. It's a significant challenge. You're talking about doing a project, but not until.

And this is where the private sector conceptualization gets involved. The genome project didn't go to production scale until it was a dollar per base, and that was 19, 94, 95. Really?

And so a dollar per proteform is the price threshold. And at that threshold for defining the reference set of human proteforms, that's not to say all in the universe.

You're looking at the same arc and scope as the genome project. So you're looking at ten years and something on the order of three to $4 billion.

Parag Mallick:

I mean, that seems very tractable, so I'd love to double click for a moment on the question. How many purdy forms are there? Because one could imagine there's a protein like tau that's 85 phosphorylations. There could be.

Neil Kelleher:

No, it's so complicated.

Parag Mallick:

Two to the 85 possible combinations of just those, and that is a terrifyingly large number.

Neil Kelleher:

Yeah.

You know, back when we were talking about the example that had one isoform due to a splice variation and then three phosphos and all those things occurred in combination, and then I jumped on that and I said, oh, there's a filtration there, that it's not that biology populates all the possible proteforms in the universe. No, no, no.

There's a great constriction of proteiform explosion that could happen, or that can happen in someone's computer on their desktop, but doesn't happen in the body. And your example of Tau with 65 phosphos and six isoforms and all you could mathematically create six times two to the 65 power.

Oh, my God, it's so complicated. That's why I reacted, because it's different than a lot of the ways people react to that complexity.

And I will say that the Tau proteiform landscape, if you react the way a computer would and propose all of them or explode them, you say it is too complicated because two to the 65 is a ginormous number. Yes, but the biology doesn't create all those combinatorial events. Why would it?

If you had that much leakage of ATP and the kinases were dropping phosphos, that's an enormous waste of energy. Therefore we have a proteform program of life that makes a constriction a reality.

Lowering the number of prote forms, allowing us to actually create tools to domesticate, regularize and deep sequence the proteiform landscapes in our body. And Tau is a great first example. But I can expand more on that. Like how many proteins are there?

Parag Mallick:

I would love to hear, because we had a paper that asserted there were about 6 million proteiforms, and we've just established that tau alone could have more than 6 million prote forms. So how do we get from our current state of knowledge to an estimate of 6 million?

Neil Kelleher:

So that would be from an entire cell, like a single cell sitting there and doing its proteform program.

And 6 million is a good estimate of, if you were developing technology to determine the prote forms in a cell or a type of cell, a kind of cell, like maybe you needed 10,000 or 100,000 of those cells of the same type, you could set a threshold of, say 6 million or 1 million proteiforms that you would want to sequence in such a cell type. Now, it just so happens that the world is creating an ontology or a known set of all human cell types.

you know, something of around:

So you would want to know in a deep human proteiform project, the 880, excuse me, 80 cell types, a deep version would be 6 million in each cell type.

Parag Mallick:

Oh, I see. So it's not 6 million in the universe.

So the universe of proteomic forms might be much larger than that, but within a given type of cell, about 6 million.

Neil Kelleher:

And the 6 million is a deep version. I think a million is a good doable number. And let me also make one other pro tip expert moment.

There are redundant proteo forms, like the form of histone h four that I mentioned earlier, that has a dimethylation at lysine 20 and an acetylation at lysine 16 and, oh yeah, an acetylation at the clipped end terminus. Okay, that proteiform, it's going to exist in many different types of cells. So then it causes you to have to say, oh, do you mean unique proteiforms?

Like a new molecule added to the atlas of all proteiforms in the human body that we wish to measure at a defined threshold? Or do you mean non, sorry, redundant or unique prote forms?

Those are two different counts, but I think the number you're after is, is about 100 million. In a deep version of a project, 100 million prote forms, unique proteforms would be a deep scale of.

It'd be the equivalent of the human genome project, but at the proteo form level.

Parag Mallick:

Yeah, maybe just one follow up question there. You've mentioned that one of the things. That'S going to have to happen technologically.

In order to enable a project of this scale is bringing down the cost of discovery. What are the other things that need to change, the other holes in technology that are required to bring protein form analysis into routine use?

Neil Kelleher:

So there you go with the chicken and the egg question. That's perfect, right?

ished at the time points into:

So it's very similar here that protiform discovery itself, which I think your question was about the discovery.

Parag Mallick:

It was about discovery. So just starting with discovery, what are the barriers to discovery that we need to overcome?

Neil Kelleher:

Yeah, it's about 100 fold off in cost and speed of where it needs to be.

roject had, if you go back to:

We don't have that.

So, yeah, that's what's required, is a platform that can emerge to do it at a dollar per protein form for discovery, resequencing of known proteforms that is significantly less expensive even today. And there's new technologies on the horizon that you're well aware of that are well positioned to do that. Resequencing the discovery part of it.

You're right. Yep. It needs to be a hundredfold drop in cost and a hundredfold increase in the speed and throughput.

Parag Mallick:

What is the barrier, either sociologically perceptually, between where we are today, where frankly, most biologists haven't even heard the word proteiform, to that future world where everything is done with proteoforms?

Neil Kelleher:

This is, I think, in decades. You first find a way to move the project forward that has use cases. You remember in the genome project you had basically genetic mapping.

There was this stage of the where they created a genetic map. And so what were they doing? They got families together that were all stricken with the disease and they were sampling those people.

So early on, people were like, oh, this is directly around disease, right? This isn't, that's what the project was. But then it went on to the different kind of mapping.

I think this is where we're at, that I think brain diseases, neurodegeneration, it really is unfortunate that there is technology in the world that could lead us to really show use cases, turning them to success cases in neurodegeneration, and do that with urgency because right now most of the resources are not proteinform informed in neurodegeneration. And I will stake a very large amount of my personal wealth on that, which is not very much on that bet, because proteiforms will shine.

do that, but that itself is a:

Otherwise it's going to be a slow burn until we every year by year we're getting closer, it's making more sense to more people year by year. And that's all I can promise your listeners, is that the more that you drill down or double click into this, the more sense that it makes.

Parag Mallick:

That's great.

I think from my perspective, it really is about getting a few very clear examples where the protein level alone, the PTM alone was insufficient and getting, I mean, if we look at, for instance, immunotherapy, there was, it was for decades being worked on. You hear Carl Jun talk about his journey where people thought it was crazy talk and he had trouble getting funding.

But then there was a moment where a series of independent studies came together and said, wow, this is notably different and better than what we saw before. And I think proteiforms just hasn't had that use case that is so clear and so simple and understandable by every biologist.

I also think accessibility is a big part.

So I think if your lab comes up with a use case and you're the only lab in the world that can do it, it's simply not going to penetrate all of biology. It needs to be something that is accessible to everybody alongside compelling.

Neil Kelleher:

Yeah. And what are our options?

Right, where it's the chicken and the egg problem, where every technological plateau that you want to level up to has this chicken and the egg problem. And yeah, it is a challenging measurement throughout. For the last 20 years, it's been challenging to make the measurements.

And so, like our prostate specific antigen, the PSA test for prostate cancer in men, classic example, where I can see exactly what needs to be done, there is technology in the world that can do it, but you're still. You've got to get the patient cohorts together, you've got to get the sampling done, you've got to then do the study and then report it out.

And that's a five year, $5 million, $4 million effort right there. Times all the other use cases you would want to show. And the tech just isn't widespread yet. Very pocketed, very siloed in certain labs.

Parag Mallick:

Well, that's a great opportunity for the world, is to first work on those demonstrations and then in parallel work on the technologies that can democratize access to proteoforms, not just proteins.

Neil Kelleher:

But, yeah, I just. I think just on basic, foundational, fundamental grounds, first principles, it's that simple.

It's just, you want to know the exact molecule that is produced in my body.

I think that argument, if people can slow down enough in a very busy world, in a big marketplace of ideas, to just confront that and all of the transformational things that would occur were you to able to have the technology that matches the biology in our bodies.

Parag Mallick:

Fantastic. Thank you so much, Neil. This has been incredibly informative. I really appreciate you taking the time to chat today.

Neil Kelleher:

Prague, thanks for the chance. I always want to talk about protein forms, and good luck to us in the future.

Parag Mallick:

So, Andreas, I have my own ideas, but what would you say is your key takeaway from the conversation?

Andreas Huhmer:

So I stuck my head out and publicly went hobo on the podium and said, there's no proteins, really, it's only proteiforms.

Parag Mallick:

Right. And I think you might have stolen that from somebody.

Andreas Huhmer:

Possibly, but I think it's worthwhile, a serious conversation.

Parag Mallick:

Yeah. So I think proteiforms are clearly important, but I also think that we really just don't have enough information yet.

We have a couple really clear examples of where proteiforms matter, but on the other hand, we have tens of thousands of proteins.

We believe there are proteiforms four, but what they actually do, how they function, how they drive biology, to me, it's just completely unknown so far.

Andreas Huhmer:

And I think portion of what fascinates me in the conversation with Neil is that he termed technologies. We have today as proteome unaware.

And I think what we need to focus on moving forward is we need to create more technologies and use technologies that are protein form aware. And I think that's a critical next step for the research community.

Parag Mallick:

I think that makes sense because, for instance, if I look at a band on a gelatinous, it just tells me the mass of a protein. It hides a lot of detail that I might see in a western, where I see phosphorylation, for instance, the band on gel is still useful.

But if I know that there are six different phosphorylations, that gives me a little bit more information and allows me to interpret and understand what's going on. I sense that's what you're saying with proteoforms as well.

Andreas Huhmer:

Yes, and it's not the, you know, it's a fact that we have been able to observe proteoforms. 2d gels are a good example.

It was just very difficult to actually understand the individual protea form that actually is being separated on these 2d gel spots.

Parag Mallick:

I think what you're saying is that 2d gels are terrible.

Andreas Huhmer:

2D gels are terrible in that sense, that they give us a lot of information that we can't interpret because we really don't understand what is in the spot itself. So I think it goes back to, we need technologies that really are aware of protein forms and able to resolve these complex structures.

Parag Mallick:

So how do we think about the data that's been collected, the decades of research that's been done with 2d gels and westerns and other, where we didn't have a knowledge of the prote forms.

Andreas Huhmer:

Some of them results will still be valid. And I think in many cases, the results will be updated to include the perspective of proteoms.

I mean, part of the conversation with Neil is how big is that space? And if you just theoretically speculate about it, is astronomically big.

I view this from a perspective of there's potential that every cell has a different proteo form, proteome.

What's more important, though, with respect to the number, is can our organism particular humans, create sufficient protein forms to actually adapt to the environment? So it's not the total number that's important, it's the total possible combinations that can actually help us adapt to the environment.

Parag Mallick:

So it's really about flexibility. It's about to recognize, to deal with the unknown and the cell having a lot of levers available to do that.

Andreas Huhmer:

Correct. And I think the challenge is going to be because there's so many proteoforms where do you stop?

And do you have the technology to find the ones that are relevant?

Parag Mallick:

Yeah, I guess my perspective is a little more nuanced than that. I think that there are sets of modifications that are really just there for mechanical purposes.

So if I have a giant glycosylation on something another protein maybe can't approach because there's a giant blockade on that protein. And whether that glycosylation is on site one or site two or site three, maybe it doesn't matter as much.

On the other hand, there may be other proteiforms which are defined by very narrow changes in the charge at a site. And that can, whether it's at that site or a site that's further away, may be incredibly important.

And that's our challenge, is how can we take this landscape that is so infinitely large and differentiate the things that are functionally important from the ones that. Are just kind of there?

Andreas Huhmer:

So do you think we're going to be able to have proteo form aware technologies in the next five to ten years?

Parag Mallick:

Oh, absolutely. In fact, I think much sooner than that. I think technologies are emerging today that are so exciting and are going to allow us to.

To unpack all of this complexity and really understand how proteiforms drive biology.

Andreas Huhmer:

I'm looking forward to that.

Parag Mallick:

Thanks so much for joining us on this episode of translating proteomics.

Today we had a vibrant conversation with Neil Kelleher, learning about proteoms, their history, their impact, their importance, and I'd love to hear from you. Have you ever heard of proteiforms before this episode? What do you think? How important are they? Where have you seen them in your life?

That the best explanation for the biology you saw came from a proteform? Chat with us in the comments or send us an email. We'd love to hear from you.

Announcer:

We hope you enjoyed the translating proteomics podcast brought to you by Nautilus biotechnology. To contact us or for further information, please email translatingproteomicsautilus bio or scan the QR code on your screen.

Chapters

Video

More from YouTube