On this episode of Translating Proteomics, Parag, Andreas, and special guest Don Kirkpatrick answer questions submitted by the Translating Proteomics community. They cover:

• Needs in plasma proteomics
• How proteomics impacts drug development – with special guest Don Kirkpatrick, Ph.D.!
• How lifestyle impacts the proteome
• How the Nautilus Proteome Analysis Platform is impacting tau and Alzheimer’s disease research

References

Shome et al., 2022 - Serum autoantibodyome reveals that healthy individuals share common autoantibodies

https://www.sciencedirect.com/science/article/pii/S2211124722006489

LaBaer Lab paper investigating autoantibody levels in plasma and their relationship to health.

Sylman et al., 2018 - A Temporal Examination of Platelet Counts as a Predictor of Prognosis in Lung, Prostate, and Colon Cancer Patients

https://www.nature.com/articles/s41598-018-25019-1

Mallick lab paper investigating temporal changes in platelets and their associations with cancer biology.

Krönke et al., 2014 - Lenalidomide causes selective degradation of IKZF1 and IKZF3 in multiple myeloma cells

https://www.science.org/doi/10.1126/science.1244851

Seminal paper describing selective protein degradation caused by lenalidomide.

Fink and Ebert 2015 - The novel mechanism of lenalidomide activity

https://ashpublications.org/blood/article/126/21/2366/34644/The-novel-mechanism-of-lenalidomide-activity

Review of research elucidating the mechanisms of lenalidomide activity

Ndoja et al., 2025 - COP1 Deficiency in BRAFV600E Melanomas Confers Resistance to Inhibitors of the MAPK Pathway

https://www.mdpi.com/2073-4409/14/13/975

Describe links between kinase inhibitor vemurafenib and changes in ETV transcription factor degradation

Song et al., 2022 - RTK-Dependent Inducible Degradation of Mutant PI3Kα Drives GDC-0077 (Inavolisib) Efficacy

https://aacrjournals.org/cancerdiscovery/article/12/1/204/675622/RTK-Dependent-Inducible-Degradation-of-Mutant-PI3K

Use proteomics to discover that inavolisib acts through selective degradation of mutant PI3Kα

Canon et al., 2019 - The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity

https://www.nature.com/articles/s41586-019-1694-1

Covers the development of an inhibitor of KRAS mutant KRAS (G12C).

Schneider et al., 2024 - Feeding gut microbes to nourish the brain: unravelling the diet-microbiota-gut-brain axis

https://www.nature.com/articles/s42255-024-01108-6

Review on the gut-brain axis

Webpage for Johanna Lampe’s Lab at Fred Hutch Cancer Center

https://www.fredhutch.org/en/faculty-lab-directory/lampe-johanna.html

Guseh et al., 2020 - An expanded repertoire of intensity-dependent exercise-responsive plasma proteins tied to loci of human disease risk

https://www.nature.com/articles/s41598-020-67669-0

Work investigating the relationship between exercise, the plasma proteome, and health

Human Proteome Organization (HUPO) webpage with resources focused on proteomics in space

https://hupo.org/Human-Proteomics-at-Extreme-Conditions

Joly et al., 2025 - Large-scale single-molecule analysis of tau proteoforms

https://doi.org/10.1101/2025.06.26.660445

Nautilus preprint using Iterative Mapping of proteoforms to measure tau proteoforms in Alzheimer’s disease and related dementias samples.

Foreign.

Welcome back to Translating Proteomics. On this episode, Andreas and I are excited to answer questions submitted by members of the Translating Proteomics community.

Before we begin, I'd really like to extend a heartfelt thanks to everyone who submitted questions over email on YouTube, on Reddit and elsewhere. It is just fantastic to see so many people engaged with and excited about the future of proteomics.

We'd also like to acknowledge that we got a bunch of really insightful questions about Nautilus technology.

We only have a short amount of time today, so we're going to answer one Nautilus focused question at the end of the episode and three broader, more general proteomics questions before that. If you're hungry for more information about our technology at Nautilus, we no worries.

We'll post a video on our YouTube channel shortly where I'll answer many of your Nautilus technology focused questions.

In the meantime, I recommend checking out our recently posted Bioarchive Preprint, our blog and the Resources page on our website where you can learn a ton about the Nautilus proteome analysis platform and its technical capabilities. You can find these links to all these resources in the description of this episode. With that, Andreas, what's our first question?

Well, let's get things started with a question from Bharat Kumar 91.

He submitted the following questions as a comment on our YouTube video channel and he is asking are we missing things in plasma proteomics because we keep searching for needles in the haystack and ignoring the haystack? For instance, are there important insights in the higher bound proteins like albumin?

Do we establish absolute plasma protein quantities with reference intervals across dynamic range of the plasma proteome to help us move from relative trends to absolute numbers? Now, I believe you have a lot of comments and insights in this field, so take it away.

Absolutely.

I think certainly my academic lab has studied biomarkers and the plasma proteome for a long time and so I do have a of lot lot of thoughts just to try and organize them. I'll first start with just high abundance proteins.

It's certainly true that a large number of proteomics workflows, the first thing that's done is to deplete those high abundance proteins. We know for example, that things like albumin are actually known biomarkers. They are used in multiple myeloma, for instance.

They are generally descriptive of general health and so they are valuable. Some of the questions though that we have are really about some of the specificities, even proteins like PSA which are really widely studied.

We don't know terribly much about their post translational modifications, their proteoforms, albumin, same thing, its extent of modification, how it may be truncated and if that's informative or useful.

So I do think that there is, there is a, there's a lot to learn even from variations on high abundance proteins that we, we just, we haven't captured at all yet. And I think the other, other question there is how do high abundance proteins potentially interact with lower abundance proteins?

Albumin is a tremendous carrier. It is great at solubiliz.

There may be signals in proteins that are carried along with albumin and maybe they are particularly carried along in some contexts or under some disease conditions and not in others. So there, you're absolutely right. There's potentially a lot of information in those top 20.

I'll also call out that among the most abundant proteins in the blood are immunoglobulins. And so there is a tremendous amount of research.

Josh Lebear's group has done a lot looking at autoantibodies and their variation and so those signatures are often discarded in common proteomics workflows. So I think there's definitely richness in both the top 20 and the next 10,000.

So maybe a follow up question from my side as you talk through the importance of all the proteins.

Do you think these next generation technologies which allow us now to do high throughput analysis of mini of these tissues of interest, that they will ultimately change a perspective that we only have to study a certain portion of it like the low abundance or because now we have access to all of them.

So I, I think what we will, so I'm just, I'm going to wave my hands around wildly at this point.

I, I, I think when, when we are given the option to look at everything, we know that there are complex interactions, quaternary interactions that occur.

We know that there's information that occurs from not just protein A being up, but protein A cycling over time or the combination of protein A cycling over time alongside protein B rising or falling over time.

And so those temporal trends we may want to look at larger numbers of proteins rather than focusing on just a handful so that we can really elaborate the full complexity of co correlations, co discordances and the temporal variation therein.

Yeah. So hopefully the next time we do our annual physical checkup and the doctor says oh, do we want to do a full blood panel?

Then they mean a full proteome blood panel in a addition to many other other panels.

Yeah, Absolutely. You know, my lab had a paper several years ago where we were looking just at platelet counts, which are considered an incredibly vanilla measurement.

It's done as part of every CBC test. But what we found is that there's associated with cancer. There's an increase in the spiking of platelet counts. There's not a set change.

So you don't see the total number of platelet counts increasing, but you see an increase in the frequency of spikes in platelet counts. And this was data collected from the VA and over millions of patients.

And so I think as we draw on richer data sets, we'll be able to figure out both where is the fullness of data? But we'll also be able to figure out for the clinical test where you actually do want the smallest, cheapest, most compact thing.

What is that smallest, cheapest, most compact thing.

So the second part of the question asks about reference intervals and dynamic range. Any thoughts on that?

So I think there are. Before I dive into that, I want to just quickly comment on that.

When we talk about the abundance of a protein and the difference in abundance of a protein, we often are asking a question about the difference between relative quantitation and absolute quantitation. And relative quantitation allows us to say, for protein A in this patient, it has changed by fourfold.

Absolute quantitation allows us to look across proteins and say, all right, well, albumin was present at 4 nanograms per mil, and IGG was present at 3 nanograms per milliliter. And so I can establish relationships between proteins, not just within one particular protein.

And that's a critical piece of information because we might, for instance, want to look at. I'll come back to what we were talking about before in terms of autoantibodies.

If we look at cases, particularly in autoimmunity, things like multiple sclerosis, we may want to be able to say not only, oh, this viral mimic of myelin basic protein is up, we may actually want to look at the relationship between the viral protein going from 3 units to 10 units alongside the auto antibody going from 10 units to 100 units, and it being the ratio of that 100 units of autoantibody to 10 units of viral protein, that is the ultimate predictor. So that ability to look across proteins is really what is enabled by having absolute quantitation.

Also buried in that question is this question is trends and what does normal look like right now? For some proteins, we have very good understanding of how they vary over time and how they vary in different populations and subpopulations.

But that's not true for most proteins and that's particularly not true for the very low abundance proteins where we really just don't know what normal is. And so it's very hard for us to say what disease is.

Excellent insight, an excellent question. So maybe we can go and look at the second question here. Paragraph.

We have a question here from Pratt K. Who asks, can you share examples where pharmaceutical companies have used protein detection platforms to assay samples, particular cases that led to the key scientific or clinical insights or supported the development of a therapeutic or diagnostic product? Wide ranging question. So can you take a stab at that?

Well, that's a deep question you hear all the time about the potential for proteomics to revolutionize drug and diagnostic development. And what are the real world examples?

I think the best way for us to answer this question is really I'd like to phone a friend who is actively using proteomics for therapeutic development. His name's Don Kirkpatrick.

For many years he was a director at Genentech and then most recently a senior vice president at Zyra Therapeutics where they, according to their LinkedIn profile, seek to rethink the drug discovery and development process from end to end. Don is the, is really the perfect person to talk to about this. So if you don't mind, I'd really love to to phone a friend to answer this.

So let's do that then.

So I'm very delighted to welcome to my phone a friend, Don Kirkpatrick, who has a rich history of first being a leader in proteomics methods and then applying those methods first at Genentech and then at Zyra to really diving deep into developing new drug targets, understanding drug mechanisms. And so for this question I thought Don would be the absolute perfect person to talk to. So Don, thank you so much for being here with us today.

Yeah, it's a pleasure to be with you Parag, and fun to continue the conversation that we've been running in the background for a long time.

Absolutely. Well, so I'm going to turn it over to you and we hear about proteomics being important in pharmaceutical companies and in drug development.

Can you give us a couple concrete examples that share with our listeners? Viewers?

Yeah, for sure.

So I think, you know, proteomics has sort of grown over these, these recent years into having a more and more influential role in drug day to day drug discovery across, across the field and that involves large molecule and small molecule therapeutics.

You know, the ones that I think most actively about are in the small molecule domain really because proteomics has the ability to understand targets in a rich and sophisticated way. So I did some, some thinking about ways and places that it's had a big impact.

And there's a couple of them that are, I think are really like front and center and growing in their, in their influence.

One of the ones that really struck me was a paper that came from Steve Carr's group and the group at the Broad many years ago that really was sort of the gateway into, into Cereblon based targeted protein degradation.

And it was the discovery that lenolinamide led to the degradation of the Icarus transcription factor through an E3 ubiquitin ligase in the Cul4 family with a substrate adapter known as Cereblon. And that was a discovery that was made in a couple of different dimensions. But one of those was mass spec proteomics.

And from there then that ubiquitin ligase and a number of others in that family really have grown in their impact and influence.

And you know, the, the different ways that that groups have gone after trying to utilize them to specifically target certain substrates for degradation.

And what was the proteomics component? Was it discovering the particular ubiquitin ligase? Was it seeing that it was upregulated in a particular system?

How did proteomics play a role in that mechanistic discovery?

And of course I will say, you know, in, in papers, as we all know, the story gets told in a way that helps to make sure that the reader can get it.

So if you take it at face value in terms of the order of events, the first event, the first discovery that was described in that paper was what we call digly proteomics or KGG profiling of ubiquitinated substrates.

And they identified specific ubiquitination events marked by this diglycine remnant of ubiquitin on the Ikaros transcription factor that were induced by linolenomide in cells. And from that then came the sort of follow on discoveries around the full mechanism of how that was playing out.

But the seminal discovery in that paper was the actual mark of the ubiquitination and then the field as a whole. And it's hard to tell how much and who's doing what.

But it's clear that across the field of targeted protein degrad that there are a multitude of groups that are using mass spec in a variety of different ways to try to understand the mechanisms, the targets and the enzyme enzymatic machinery that are Involved.

That's interesting because you could imagine that proteomics in that context would dominantly be used to say, is the protein here or not? Just as sort of a fancy Western really. But what you're describing is actually much, much more sophisticated than that.

It's finding what are the targets, what are the off targets and what are the compensating pathways that are impacted as well.

So those two play nicely together.

The idea that a protein goes away completely is often seen, but because it's series of steps downstream, mechanistically you have to then sort of move in proximal in the time dimension to really understand what the drug is doing. And I think that's why this particular example and this particular piece of data that started that was so compelling.

But yes, I think global proteome profiling as a way to understand how a drug remodels the proteome has been effective.

In fact, there's some really new and interesting work that came out of Vishwadiksik's group at Genentech led by Ada Nadoja that showed that an ERK inhibitor specifically triggers a degradation of an ETV family transcription factor as effectively the one and only protein that's degraded in the first 30 minutes after drug treatment in.

And this is not a targeted degrader, this is an ERK targeting kinase inhibitor leading to a protein degradation.

Exactly.

So it is an ERK kinase activity dependent degradation event that when you inhibit the kinase, you trigger the inducible degradation of this particular transcription factor. And that's a paper that was just published even in the last couple of weeks.

So to say there are these types of activities underlying many of the molecules that have been used effectively in the clinic for a long time.

And as proteomics really sort of positions itself alongside other technologies, we are starting to see the full multitude of mechanisms that are in play when drugs come into contact with the target cells, but potentially also with off target cells and you know, manifesting as off target activities in some of those.

So when we, when we hear people chat in this space, and I'd just like to get your perspectives on this, we hear, oh, hey, proteomics is just used at the front end to see what's different between healthy and disease. And so it's your sort of initial target finding.

But what you're describing is not just that you're describing all throughout the drug development journey that proteomics is playing a role. Absolutely.

And in fact it certainly has a strong role.

In the beginning, I think some of the work that we had Done even in the early days of my drug discovery career was in identifying those markers that could then be used to screen for and ultimately identify those early lead molecules for targets such as the LRRK2 kinase in Parkinson's disease.

But as you start thinking about, you know, programs and the multitude of effects that they have in living organisms, understanding the diversity of effects that that small or even large, large molecule therapeutic have on cells and tissues is perfectly positioned for proteomics to have that impact.

And it's done so repeatedly now in places like Genentech and Zahra and a number of other companies that, that are, you know, investing in the technology.

And so what. So you shared one really beautiful example from your own career. Can.

What is your favorite example of where you used proteomics to inform therapeutic development?

That's a fun question because I spent some time thinking about this recently and there's a molecule that recently went through and received approval and it really was based on an understanding that came from mass spec proteomic results. So there was an idea that one could selectively inhibit mutant PI3 kinase over the wild type protein.

And if you could do that, then you could get around some of the side effects that come from broadly inhibiting PI3 kinase in normal cells and tissues.

And an investigator, Lori Friedman at Genentech, came one day and said, hey, I have a molecule and has some interesting effects and it's only working in mutant cells. You know, is there a way to show that in fact it is having an effect on the mutant protein selectively?

And so we did that because several of the different pi3 kinase mutants have proteotypic peptides, a peptide that will tell you that the mutant that will tell the difference between the mutant protein and the wild type protein.

And indeed you could show that in cells that expressed one or another form of mutant pi3 kinase, that only the mutant protein as read out by that one mutant selective peptide was being degraded where the wild type protein was being spared when the cells were treated with this particular molecule that's now known as Involisib.

Wow, that's, that's really a beautiful story of how the specificity is, is critical in understanding and how the proteomics played a key role.

Because if you, if you, again, if you not to pick on westerns, but if you just had a western against the total protein, you wouldn't be able to see that mechanism at all.

Yeah, in that case there's one specific residue that differs between the wild type and the mutant and it confers a multitude of different effects downstream. But as a readout, that's the one and only readout that you would have.

And you'd need to have a detection reagent, an antibody, or some other probe that specifically allowed you to do that. But how does that work?

If you have a mutant, a kinase like pi3kinase, that has different mutations across its spectrum, you'd need a mutant assay for each.

But it's interesting to think a little bit about because, you know, in the context of mutant specific degradation, this same sort of process has played out with KRAS, particularly starting with the KRAS G12C where Amgen has the G12C specific KRAS inhibitor.

And that again started with an assay that allowed them to show that that molecule specifically engaged cysteine 12, which is the mutation in the KRAS molecule, and, you know, led to the cellular effects that they ultimately were able to demonstrate. So that one proteotypic readout became the foundational piece of the entire KRASG12C inhibition story.

Don, we've talked a lot about how proteomics has helped you so far, but there are probably areas where you wish you could further apply proteomics or where you wish technologies were more advanced to be able to ask different questions. Tell me a little bit about how you view the future and the unexplored.

Yeah, there's a few dimensions that I'd love to be able to dive into and that that are starting to emerge from technologies like the ones being developed at Nautilus.

And I think the one that I've thought about a lot, and certainly you and I have talked a lot about, is around proteoforms, because while we think of an MRNA and we think of a protein and maybe a post translational modification or two on a protein, in fact, any given protein molecule can exist in a multitude of different states, what we refer to as proteoforms.

And to date it has been nearly impossible to characterize the proteoform repertoire of any protein, let alone the proteoform repertoire of a given cell or cell type. And I think those are the kinds of approaches that are coming to bear from groups like Nautilus.

And so I'm quite excited to think about how would we, how would we use the information about proteoforms if we had access to it and if we could do those types of analyses systematically, whether it be for the drug targets that we care about specifically, or in a grand sense for the entirety of the proteome of a cell tissue or organism.

Yeah, Mel, that's a, that's a really beautiful future and definitely one that, that we've been working on. And I'm, I'm excited about having folks like you take advantage of that knowledge to push the boundaries on our next generation of therapeutics.

The day is here.

So Don, one last maybe contentious question to just put you on the spot.

We often hear people saying, oh well, I don't really need proteomics, I can just RNA seq everything and that's good enough for my therapeutic development. I'd love to hear your thoughts.

Yeah, no, RNA seq has a clear, a clear place in understanding the cellular phenotypes that we are monitoring in the context of drug discovery. But you can't get away from actually measuring the protein itself.

All kinds of post translational controls that are occurring on cells and on the proteins within cells are those that are directly being impacted by our drugs when they go in and hit their target and ultimately manifest this downstream activities.

And you can see that with both intracellular targets, but also with the targets of, of biotherapeutics where you're engaging a specific protea form, maybe a modified form or a proteoform that's in a specific three dimensional confirmation.

The kinds of readouts that you can get from proteomics that are encoded, but they're encoded not just in the transcript level of the target itself, but really in the full understanding of what's happening functionally within a cell.

Yeah, that makes great sense.

Some of the examples you've talked about about changes in protein degradation rate, those are not going to be well captured by the transcript changes in its extent of phosphorylation or modification with ubiquitin.

Well, so many of these are interplaying with one another that phosphorylation is often controlling the engagement of enzymes like ubiquitin ligases that ultimately control downstream effects such as ubiquitination and the turnover of the protein. None of which would be directly measurable at the level of the mRNA, since all of that is happening downstream at the protein level.

Awesome.

Well, I think we've answered our question, but I think we could probably chat about this topic for hours and probably have four hours and so I reserve the right to invite you back on at another time.

Yeah, it was great to see you today, Barag, and thanks for, thanks for checking in.

Yeah, thank you so much for joining us today, Don. Huge thanks for such an insightful answer to this very important question.

Yeah, this was really great and I can't wait until the Nautilus platform is used in a similar effort.

Yes, definitely. But in the meantime, let's get to our next question.

All right. The next question focuses on lifestyle and how it impacts the proteome.

Go for it. And if I have anything to add, I'll jump in later.

So the proteome was a focus of in food was a focus of the episode 17 where we talked about how proteomics can contribute to food security, security, food safety, but also in authenticity testing. So I encourage you to listen to that particular episode.

What I would also do or like to do is connect that food aspect to the last part of your question, which is knowledge, sorry, feelings. And there is actually a very well established connection between food intake and what's called the gut brain axis.

It turns out the food you intake is tightly linked to emotions because the food is digested in your gut by trillions of bacteria, about 3,000 species, 3,000 different species in your gut and they produce small neurotransmitters that ultimately communicate with the brain and the immune system. So you definitely affecting or the proteome that you take in as food is definitely affecting your feelings and your thoughts.

So as people often say, what you are, what you eat is holding true in this particular case. And then there's the strong connection how the connection between brain and immune system essentially affect the rest of the body.

And so in this case I strongly encourage you to again listen to episode 17 to learn more about how proteomics and food is connected. Any other thoughts on that?

Just that I think our physiology is such a complex adaptive system and so the arpidium, we just talked about this, the serum and circulating proteome and I am 100% certain that my circulating proteome is going to be impacted whether I have pizza versus quinoa and so how it's impacted. There's some really great researchers.

The Lampe group at the Fred Hutchinson Cancer center has done a number of studies putting people on controlled diets and observing how that impacts their, their cancer progression.

As well as we've seen a lot of work looking at anti inflammatory factors and it would be really interesting to see can we observe a clear proteomic indication from the anti inflammatory diet versus not. I think there's a lot of very fun science to be done there.

Yes, I agree. The second part of the question here is around the proteome and physical exercise.

Now this could be its own podcast series essentially, but I think from a very large perspective what you can see is that obviously the proteome is affected through physical exercise.

And there's plenty of research that shows that essentially exercise affects your TCA cycle, the mitral functions, obviously all the muscle related functions.

And people have shown over and over again in many, many different studies that the benefits of physical exercise clearly manifest itself in the changes in the proteome.

In fact, it's now being used as markers in some of the plasma proteomics studies that are merging now, where there's a lot of indication that a plasma proteome from a person that's healthy and exercising looks very different from a person that is not exercising and might actually have the potential for disease. Clearly, physical exercise changes you, changes the proteome and in many cases all the indications are that it actually affects you positively.

Is there a control group? And I think Parag, I think you're familiar with that.

But we have sent people to space where there's no gravity and so it's very difficult to have physical exercise.

And again, research shows that when you look at these astronauts that come back after days, weeks and often months being in space, they show distinct changes to their proteome, particularly the proteome associated with muscle, muscle activity. But then they also show some changes in their defensive functions like in oxidative stress.

Again, the mitochondria very much affected how they're being remodeled from a lipid, but also from a proteome perspective. And there in fact is a very, very good resource on the hupo.org G web page that you can go to and read a lot of interesting articles about it.

But I think Bharag, you have done some work in that area. So any thoughts?

Well, so I, first off, we'll definitely link all of these resources that Andreas mentioned down below in the description and the papers of that primary data.

I think the, the research that we've done in our lab has, we've been asking questions initially about diagnostics and how exercise affected the proteome in ways that might confound diagnostics.

And one of the things that motivated this was actually work that was done in, in circulating tumor DNA where they found that there's a tremendous amount of cell death that occurs with, with, you know, aerobic exercise that led to spikes in circulating tumor DNA.

Well, circulating general DNA showing up in the blood at higher levels and that that could then be confounded for circulating tumor DNA when it's really just from. From cell damage.

The what Our early studies seem to suggest that the levels of protein shed by exercise, particularly ones that looked like epithelial cell overall, seemed much more modest relative to what you were seeing from the CTDNA confounder. And so that was already sort of an interesting early finding. We haven't published that data yet, though.

Yeah, I think the other aspect that is often mentioned in the literature is that if you do exercise, it actually activates the mechanisms that actually contribute to the regeneration of the proteome. So damaged proteins, proteins that are no longer needed or required, are then being efficiently degraded. And I think that's really.

Exercise is really an effort to take the trash out of your cells and rejuvenate your proteome. And so there's lots of interesting articles on that. And so I encourage people to read up more about this particular benefit of physical exercise.

All right, our final question for the day comes from George Blumenthal, who I believe both of us have met at various conferences. Hey, George, thanks so much for your question.

I think we can each take different aspects of this question.

Yeah, absolutely.

So maybe I'll just start by talking a little bit about proteoforms in general and why we feel that they are potentially really have an important role in Alzheimer's disease.

First off, as just a reminder for our listeners that proteoforms are those the variants of proteins that arise through a combination of alternative splicing and potentially multiple post translational modifications.

So for instance, in tau, in Alzheimer's disease, the protein tau, which is very influential, has six different alternative splicing forms and has been shown to have potentially dozens of different phosphorylations and methylations, et cetera, and you might see combinations of those present on the same molecule, you know, two, three, four or more different phosphorylations. Now, the challenges that we often face in studying neurodegenerative diseases and even studying tau, we have very clear endpoints.

We can tell when cells have tau tangles, when extracellular regions have a beta plaques, where we have cell death, but those are extremely late measures of the process of Alzheimer's disease. One, there are really two places where I'm very Excited about our platform helping get to a new stage in Alzheimer's disease.

The first is just being able to maybe look at the process as early as possible that led to disease. Where did those early polymers come from before they became tangles? What happened in cellular physiology?

Did we see changes that were associated with oxidative stress or damage control that are associated with particular forms or sets of proteforms? And to date that's been very hard to measure.

It's been hard to say, oh, there's this triple phosphorylation, that's a higher abundance alongside this quadruple phosphorylation, and that is associated with middle stage Alzheimer's disease and later stages, a quadruple or a six tuple phosphorylation.

And so I'm excited about our platform being able to help us untangle, pun intended, all of this complexity of the variation in tau to really understand the progression of the disease and alongside that, understand how different therapeutics might abrogate that progression.

Yeah, let me add to the work that we just have recently published on, on tau proteoform. So we essentially showed that we could develop an assay probing for about seven different phosphorylation sites.

In addition, we directly probe for three different isoforms, obviously concluding from the measurements the different or six different types of isoforms. And then we applied that assay to a variety of different tissues.

Maybe I'll just talk a little bit about one of them, which is the organoids, the cortical organoids. These are little mini brains, call them little brains, that are basically derived from IPSC cells.

Those cells, IPSC cells, come from actual patients and so they contain mutants. In this case, those mutants are known to cause early frontotemporal dementia. And then we also had access to their isogenic controls.

In this case, the CRISPR technology was used to remove those mutants and restore the natural sequence of tau. And then we compared or we simply analyzed those cell lines, those organoids, three dimensional cell cultures and compare them.

The first insight that we actually had was that in theory, we could detect over 700 proteoforms with this particular assay. But in fact, what we see is that in mutant organoids, we see about 63 proteforms. In the wild type, we see about 52 proteoforms.

Clearly, not all the proteforms are theoretically possible, are actually found in those organoids.

And so I think that's an important takeaway that when it comes to the combinatorial space that nature keeps ready, we only observe a small portion of it. And we could clearly quantify and compare those across those different organoids.

When we looked at the mutants, what was actually very surprising there is that some of the mutants are known to express the longer form of tau, so the 4R tau at a higher percentage than the 3R tau. So typically, those two forms of tau are balanced in the human brain, but in the case of this mutant, they're actually shifted towards the 4R.

So there's more 4R Tau forms. What we could actually observe as those organoids grew and matured, we could actually see those longer forms of tau evolve and become more abundant.

And that was very important for us to see, because I think that has been demonstrated using traditional Western blotting techniques. And the fact that we can very well recapitulate this particular biology was really very insightful.

I think the other part that was interesting to us is that in those organoids, we see a lot of phosphorylation events that are more associated with the fatal state of cells.

And so apparently what happens in fatal development, tau is being used at a highly phosphorylated state to really drive the growth of neurons within a developing brain. And so These organoids, these 3D organoids, perfectly recapitulate these particular developmental aspects of brain development.

And so it was exciting to see that we actually can measure them and we could actually pinpoint those combinations of phosphorylation events that are critical for these development efforts.

I think that's such a.

Such an important point that you're raising, Andreas, which is that developing model systems to both test and discover the next generation of drugs and model systems that help us understand where the next biomarker might come from is incredibly challenging. There's a statement which is all models are wrong. Some models are useful.

And so one of the areas where I'm really hoping that we can use the tau proteform assay to drive a step change is to understand where different models best describe the progression of the disease, and accordingly, which ones are altered by different therapeutics. I think that's a huge opportunity for the assay to give a little bit more color beyond the gross phenotype of just cell death.

Yeah. And maybe one of the other areas that was surprising for me to see that in the end, the assay sensitivity was not the limiting factor.

So it was really the biology that limited the number of protoforms that were observable. And I think that speaks for the robustness of the assay and the fact that we have the potential to measure many Many more. How proteforms.

As we look into many more tissues in understanding either the developmental aspects of organoids or as you say, how do we utilize these model systems for drug development efforts?

Well, I'm very, very excited about the potential impacts of proteoform focused studies and folks who are interested in learning more about Nautilus's platform and the ability to measure Tau proteforms, I recommend you check out our preprint at the link below and please do let us know what you think.

I want to make sure that we also thank our collaborators at the Neural Stem Cell Institute in Albany.

They put a lot of care and effort in growing those 3D cell cultures and provided us with, you know, organoids from many different patients and different ages. So thank you for their efforts.

Yeah, they really have have built a tremendous model system and done such diligent work in, in both the creation and of that model system and also in ensuring that it's available to the broader community. Well, that's all the time we have to have today. Before we go, thank you all again for submitting your fantastic questions.

I'm hoping we can do this kind of thing relatively regularly. So keep submitting your questions in the comments on social media and by emailing us translating ProteomicsNodilus bio.

And as a reminder, I'll be answering many more questions about the Nautilus proteome analysis platform in a separate video on the Nautilus YouTube channel.

We'll get that video to you as soon as possible and in the meantime, be sure to check out all the Nautilus focused resources on our website at Nautilus Bio. Thanks again for joining us on this episode of Translating Proteomics. I look forward to seeing you again on our next episode.

We hope you enjoyed the Translating Proteomics podcast brought to you by Nautilus Biotechnology. To contact us or for further information, Please email translating ProteomicsNautilus bio.