IMMUNOLOGY2026™: Day 3

Brandon and I have been having such a great time at the meeting, especially getting to meet so many of you, and the fun's not over yet. If you want to keep up with what we're doing here at the conference, you can follow us on X at Adamino Podcast. Or over on Blue Sky at Immunology Podcast dot com. And if you missed our earlier recaps, you can catch the episodes covering days one and two at ImmunolyPodcast.com or wherever you listen to your podcast.

Today was an extra big day for us. We hosted a live show of the podcast in front of an audience here at the meeting. Doctor Charles Sentman and Julia Escobar joined us to talk about new strategies for tackling one of the biggest challenges in immunotherapy, treating solid tumors.

That episode drops on April twenty first, so make sure you don't miss it. For this episode, we'll be diving into the science that really stood out to us on day three of the meeting. We're gonna be kicking off things in just a minute, but before we get to that.

Stemcell.com slash sample dash setmate dash lympoprep. Terms and conditions apply.

But then try to figure out the type of fat. So they did all these different types of fats. They found animal fats, in particular butter was the worst, and palm oil, a saturated vegetable fat, was the least bad at this effect, which was interesting. So not all not All the mice were equally fat, and they said there were no systemic metabolic changes. They all had insulin resistance. They all looked the same, both fatty and labalized.

But this T cell effect was worse than the butter. So they tried to figure out which metabolites were different. And long chain acyl carnitines, which are animal only proteins, were more in these CDH cells accumulating in the butter, because duh, there's more of them there. So maybe that's it. Um and they found that these cells that have a bunch of this live longer but can't kill at all.

And the longer the chain, the worse. Like, okay, so maybe it's this, but let's really control this. These cocoa butter. And they were really worried cocoa butter was fine. And cocoa butter has a lot of these steric acid types of things. So it's one of the things that has acids like this, like the these fats like the animals do. So don't worry, chocolate's okay.

And they're like, well, okay, so it's not just the fat, but the fat's important. What else is there? And the answer is, well, cholesterol. Which right. Which they're like, we're, you know, clearly everyone knew this, but we didn't remember this because we're not like lipid people.

They go up and there's more IL-18 in the butter diet, you know, more flamboxone activation. So they're saying that these proteins make your T cells suck. But also create inflammation. And sucky T cells plus inflammation is cancer.

If you do an EM on a neutrophil, they have glycogen granules and are often derived not from glucose uptake, but amino acid uptake and metabolism. So when a neutrophil enters a cell, it's just ready to go with all its energy to do its job and die. They found in this study that lipid uptake is at in neutrophils is regulated by TLR signaling and it's lipid class specific.

So certain lipids are up driven by certain signaling of TLRs. So there's this crosstalk between your pathogen sensing for neutrophils and uh lipid uptake. And then they found this was altering actin. Both by proteomics and if they looked at microscopy, the actin on the neutrophils look different. The more fat you fed them microscopy.

And then they showed at the end that if you look at atherosclerotic plaques, which everyone is just laden with macrophages, right? You and I know, hopefully you know the macrophage, athlosclerosis thing. But there's neutrophils too.

Gamma delta T cell mediated pathway that is IL-1 driven. And uh bacterial ceramides are important in driving this process. So bacterial metabolite and ceramides help drive this process. So again, bacteria for the way. Then I went to some sessions on T cell interactions of the microbiota. I'm having a very good conference. There's a lot of microbiome magic in in my life this this week, which is with not a not every immunology conference has that, but this one does.

All right. So uh t this w first one is Maria Cecilia Campos Caneso. Who also acknowledged her email for work is really long.

And um what they showed is they use SFB as a prototypical antigen. But if it's on a bacteria and then you swallow that bacteria, you get T regs. And the antigen presenting cells shift from D uh DC two to DC one. But if it is the normal antigen presentation, like in your blood, you get activated T cells.

It it is the fact that it's presented into the gut on a bacteria because they make these chimeras that seem to retrain it to be, oh, you should be tolerant to this antigen, not kill it. So that was interesting. More more on olert oral tolerance, which is a long done thing. Then we're going to go to my multiple sclerosis with Kate Stack. So the role of microbial tryptophan metabolism in TH17 activation and multiple sclerosis using the EAE model.

I have to give a shout out to her in particular. She's a first-year grad student who presented a bunch of data and then some of her slides had the data missing when she tried to present it. And she talked through the whole thing, didn't drop a beat, didn't like lose it. For a first year grad student.

So this inhibitor of TH17 cells is low. So they did ketogenic diet drives up lactobacillus, and they wonder if you could put mice on ketogenic diet and prevent EAE, which it does. Then ILA is involved in this process. It's one of the other metabolites. And if you administer that endogenously, it also works.

And then there's a supplement that's been like big in the field called 13 B D, which induces ketogenesis in people. Like gets your ketone bodies early because you always have to have a, you know, biohack in the world. And if you use that to induce ketogenesis, It also works. So ketosis, which leads to more lactobacillus and more of this metabolite. Um inhibits TH17 cells and improves at least experimental MS model.

I like this talk because talked about a a topic that I think a lot about and that I'm very exposed to, which is understanding engineering immune receptors and ways of improving the uh characteristic of immune receptors. And a lot of the work that he showed was focusing on

antibodies, uh so B cell receptor, B cell uh immunity, but he also has some work on T cell receptors. And basically a lot of his of the research that he presented and he kind of the work of his lab is uh using mutagenesis to make better binding antibodies, broadly neutralizing antibodies, improving the functionality of these molecules in order to, for example, treat disease.

or to guide vaccination. He also talked about T cell receptors and how really using large data sets to understand the diversity of T cell receptors. But I'm gonna just highlight two little stor kind of two highlights of the stories that he told. And the first one is about thinking of monoclonal antibodies for treatment and uh particularly for treating diseases like chronic diseases. And he was talking about particularly malaria.

And malaria is of course a very difficult disease that has all this the the the parasite has this very complex life cycle. Usually the antibodies that people develop against malaria are in a way directed against kind of the wrong part of the malaria's life cycle because what you really wanna find is an Achilles heel or a uh the part in which is the limiting

step in the infection and he made the point that it this is probably the sporoside stage in which is what comes out of the mosquito when the mosquito bites you. And it's gonna l uh release the sporosides that will then will take a residence in your liver. And then every once in a while the sporosides will release and at the next s uh life cycle stage of the of the of the parasite and that

uh kind of continuous the infection. But if you could really protect people from that sporos initial sporocyte infection, that would get you a really long way and to generate a more protective immunity. So he made the point that there are people do generate antibody response against certain surface proteins in this in this poroside. But the problem is and I think one of the reasons why it's so hard is like malaria can have also some level of mutations or like an ad um antigenic uh changes.

So oftentimes antibodies that are very well neutralizing for kind of this one version of the protein will be uh rendered uh useless when you have slightly changed versions of this protein with more or less repetitions of the repeats that are recognized so then the antibody doesn't bind so well. So what he did in the studies he kinda tested variants, he did a lot of affinity uh changes, uh he mutated the variable domains of the antibody just to change the the what it's how it binds.

And what I thought was very interesting that what he found, he ended up finding a very uh nicely kind of neutralizing antibody that could. you know, very well bind to the original version of these uh sporoside protein, this CSP, but and could also m uh bind to these rare variants that have different amounts of these repeats that are recognized by the antibody.

But the funny thing is that what he shows, and he looks to show some um structural data, is that the way that the antibody does this is by generating this structures this uh what he calls multi axis fab to fab homotypic interaction. And basically you have two antibodies that are interacting with each other through their fab portions or their variable domains. What is very unusual is that they're doing it through t in two different kind of uh planes, in two different parts.

And this uh complex is what is very good at binding the two variants of the CSP protein. This is something that he has never uh has never been reported on like natural antibodies. This is a consequence of the mutations that he introduced in this antibody to make it better at Neutralizing this protein. And he showed kind of the structure. And I thought that was very cool. A very unexpected way of making better antibodies, is making them good at kind of

assembling with each other and then they can in a way have a better avidity, a combined avidity for the the target protein. So I thought that was really cool. And then he he also talked about how in his lab he was improving and doing more high throughput analyses with better systems and also trying to do these very extensive mutational libraries in which basically he can mutate every amino acid in this fab s sequence.

uh uh and exchange it for any other amino acid and that way he can really try the whole potential landscape of p of of of antibodies of variants. And that really helps find unexpected things. And so he was the example of uh neutralizing antibody for SARS-CoV-2 and how trying to improve the eff effectiveness or the avidity of this antibody for its target.

in that sometimes the mutations that he saw actually really improved these uh functionality these neutralizing capacity were not located in the antibody what you would think as the antibody kind of the as the antigen binding area. Was for example, he found that in some of the in some instances the best mutation was literally one mutation in the first amino acid and the far end of the of the uh antibody.

that that didn't really interact with the target protein at all, but it helped with like the structure of the general structure of the antibody and that's how it helps. So I think it's really nice because it gives you a platform to really m in a more unbiased way analyze all of this the different options I thought it was really cool and it's really important for us. I think particularly antibodies are working on affinity and because they have this therapeutic potential of being a treatment

by themselves. Of course T cell receptors is a lo it's a different uh it's a different kind of uh area. uh affinity maturation in Tol is a bit more more problematic. But for antibodies it seems like, you know, we can do all a lot still in this field. And uh the other talk for us later in the evening, so David Masipus

who has been uh is very well known for his work on uh memory and memory residency and uh he's at the University of Minnesota and Um he kind of talked a bit a bit about the resident memory and how his lab contributed to the understanding of memory and residence. In the tissue. He talked about new work that they're doing, which they're expanding resident memory cells from human explanation.

And this is kind of interesting because of course the cells from the blood and the cells in the tissue are completely different. And it's easy I mean, you can get cells from a mouse from the tissue, you just you know, get the mouse and you just get the tissue, but in humans it's a lot more difficult. But they have this really nice setup in which they get mu they get tissues from transplants and from

um tissue donors and they have actually been able to expand and keep these cells in culture for many months, I think up to a year by now. So this is really interesting because uh resident memory T cells are known for being very finicky. So I'm very impressed that they got around to do this. One of the things they are looking into is the role of rest tissue resident memory cells that are virus specific in, for example, uh tumor immunotherapy.

And how maybe if you target so one of the things they ask is, what if we have a tumor, we know for a fact that inside that tumor there's gonna be T cells? tissue resident memory cells are actually a specific for m a previous viral infection that this person had. What if we activate the cells? What what would happen to the tumor? So they have a mouse model for this. And they show that actually if they stimulate with viral peptides.

these tumors, they actually does help with the immune response and does activate the T cells. But it's not and they were trying to understand the mechanism through this and it's not because they so what f first thing they thought is well maybe if you're putting all this viral peptide it's being loaded on the and on the tumor cells and then the T cells are recognizing that and mistaking them for, you know, infected cells.

But that's not the case because it works even if the tumors don't have MHC. It is not activating indirectly T cells that are tumor specific. but rather kind of just activating the whole environment in a more kind of innate antigen agnostic mm uh way. So I thought that was very interesting and they're actually starting a clinical trial with what they call then peptide alarm therapy in which they use peptides that are from viruses and they are well known immuno immunogenic peptides from viruses.

to see if they can activate tumors uh and make them cold uh hotter, so from cold tumors to hot tumors in human subjects.

And they could still they would pick up the cells from, you know, congenic markers. So it was always the original cell. They could differentiate from the endogenous cells of the mouse. But then the T cells they kept proliferating for years and years. Like Several times the lifespan of the original mouse they came from. So I thought that we discussed that paper a couple of years ago when the result came out. I thought it was.

Kind of funny, kind of kind of like very simple but very profound at the same time and he spoke a bit more about that too.