The Bench Brief - Dr. Lucas Ferrari de Andrade

James Taylor-Mosedale (00:07)
Hello and welcome to The Bench Brief presented by ichorbio your rapid insight into cutting-edge life science research. Today we're joined by Dr. Lucas Ferrari de Andrade from Icahn School of Medicine at Mount Sinai. Lucas will be presenting on selective inhibition of Fc receptor shedding to promote antibody-dependent cellular cytotoxicity. Lucas, thank you for joining us. The bench is yours.

Lucas Ferrari De Andrade (00:35)
Thank you, James. All right, so let me quickly introduce an immune concept called antibody-dependent cellular cytotoxicity, or just ADCC. This is a mechanism of immunity whereby natural killer cells can recognize the Fc region of antibodies by using CD16a, which is an Fc gamma activating receptor. And this engagement triggers the cytotoxicity of natural killer cells.

they release perforin granzymes that will kill any target cell that would be opsonized by an antibody, which you can see here in this ⁓ previous study where tumor cell lines were treated with different concentrations of EGFR, which is an antibody. And as when any case cells were incubated together with the tumor cells, you can see in a dose-dependent manner how NK case cells killed these tumor cells.

And this pathway is physiologically relevant, which you can see here in this other study where mice that were treated with TA99, an antibody against a melanoma antigen, this treatment is able to reduce the amount of tumors in these mice. But when the investigators used CD16 deficient mice, which don't have the receptor for ADCC, this effect was partially lost.

Despite that, CD16a is subject to a post-translational modification called cleavage, where proteases like ADAM17 can cleave the stalk of this protein and release the entire extracellular region, so turning this surface protein into a soluble peptide, which you can really appreciate in this study. There were many studies that showed this. This is one of them. And what you can see is that this is the amount of CD16a on the surface of human and case cells.

and how they downregulate after they have been co-cultured with tumor cells in the presence of an antibody. This is mediated by cleavage because Adam-17 knockout NK cells don't downregulate.

And this protein has to go somewhere, right? Once it is shed from the cell surface, it can be detected in the human serum or plasma. And all of us, people have this in our blood circulation, except for these individuals here who

are CD16a deficient, you can see that they don't have, but healthy donors in general do have this protein shed in the circulation. Even mice that were being humanized of Fc receptor, they can have soluble CD16a in their serum.

Something that I find fascinating about this enzyme is that it can cleave a variety of proteins. I think there's near 100 different proteins ⁓ that are targeted by Adam-17. And CD16a is just one of them. If you would use Adam-17 inhibitor to stop the shedding of CD16a, you would affect the cleavage of a lot of different proteins. So we would need a more selective way to inhibit the shedding.

So that's what my lab recently studied. We wanted to develop a pharmacological means to stop the cleavage in a way that spares the protease but targets the substrate. So look at this illustration here. We have this enzyme that could cleave CD16a and also this other protein. But we developed an antibody, monoclonal antibody, binds to the substrate and prevents the cleavage. So this antibody doesn't affect the cleavage of unintended

We call this antibody F9H4, it's a mouse anti-human CD16a. And the way we discovered that it inhibits the shad was by this assay where we biotinylate the surface of the human primary NK cells. And when we treated them with PMA, a protein kinase C agonist, they shed CD16a from their cell surface, which we can detect using a capture antibody in ELISA plates. This antibody is called 3G8.

Since the protein that was shared is biotinylated, we could use just streptavidin with peroxidase to detect or to quantify the amount of soluble CD16a. And you can appreciate here in this figure how NK cells shared about 300 picograms per mL of CD16a after they have been activated by PMA, but our antibody was able to completely prevent that. And the consequence is if NK cells are no longer shared with CD16a when they are treated with F9H4,

they keep the receptor on the cell surface, which is exactly what you see here. You can appreciate that in a PMA dose-dependent manner, NK cells down-regulate CD16a from the cell surface. But when they're treated with our antibody, this antibody retains that. look at the histograms, for example. This is the amount in blue, how much CD16a NK cells have in the beginning of the assay. And as you add the PMA, and wait one, two, or three hours in black, they down regulate. But when we also gave our antibody, this was retained on the cell surface. CD16a is expressed not only by NK cells, it can also be in myeloid cells, including macrophages. And also, neutrophils express CD16B, which is very similar to A, and our antibody inhibits the shedding of B, inhibits the shedding of A by macrophages.

We even did this experiment where we analyzed tumor samples from patients with cancer and we saw that in tumor explants, supernatants, we detect soluble CD16a by ELISA. But when we treat these samples with our antibody, we were able to completely stop that. And as a consequence, there was more CD16a on the surface of these tumor-infiltrated leukocytes. Most of them were myeloid cells. There were few any case cells here.

but this provided the proof of concept that TILS also sheds CD16a. This was done in vitro, we didn't treat patients, we treated their tumors in vitro.

But CD16a is the receptor that triggers ADCC. And since we are increasing the amount of CD16a that is on the surface of NK cells, the logical consequence is that they can kill tumor cells better by ADCC, which is exactly what we see here. We use the cetuximab as the opsonizing antibody. And you can see how well NK cells degranulate against tumor cells that work.

with cetuximab and we're treated with our antibody. We see more degranulation, more interferon gamma production, and also more target cell killing. We also did in vivo experiments where we reconstitute this mouse strain with the human NK cells and also inoculate the human tumor cell line that develops metastases in the lungs. And this model is, ⁓ it's really hard to do. There's a lot of work in order to...

to actually conduct this model. And what we established is that when we inoculate N K cells seven days before we inoculate tumor cells, this is the amount of mass, which is much less than mice without any case cells. And there's this intermediate point here when any case cells are inoculated 10 days after tumor cells. So we did this model using this setting here. And we were able to detect that there were fewer tumors in the lungs when the mice were treated with our antibody plus

cetuximab See how cetuximab by itself was ineffective compared to the isotype control and also how our antibody did not cause a major impact by itself. It's because in this case there was no tumor cell opsonizing antibodies. We need the combination in this model in order to achieve a biological effect. On the other hand, it's a little different when we use Fc gamma receptor humanized mice This is a mouse strain that is immunocompetent.

and we inoculated these mice with a mouse tumor cell line expressing the antigen. We treated them with cetuximab plus or minus our antibody and we were able to detect this significant inhibition of tumor growth when we combine our antibody with cetuximab. Our antibody by itself also had a biological effect in vivo and we believe that this is caused by endogenous antibodies that would be generated by...

the immunity, these are immunocompetent mice. And as a proof of concept, we analyzed blood N K cells and we saw that they have more surface CD16a in the group that was treated with cetuximab and plus our antibody, just suggesting how our antibody inhibits the shedding in vivo.

So the working model for this approach is that, for this project, is that ADAM17 cleaves CD16a from the cell surface. And there are many studies showing that this reduces the ability of NK cells to engage with the Fc of opsonizing antibodies. So this is a mechanism that could prevent the ADCC. But we developed a monoclonal antibody, F9H4, which binds to CD16a.

It doesn't block the receptor. It's not an agonist. Also doesn't cause cell depletion because we engineered the Fc with mutations. And this antibody stops the cleavage by presumably blocking access by the protease, although we haven't formally demonstrated the exact mechanism. And the consequences that antibody retains CD16a on the surface of N K cells that they can better engage with the Fc to kill the target cells.

So we think F9H4 is a key reagent for basic immunobiology that we can use to study CD16a shedding in vivo. This is a few questions we have, one address in follow-up with studies. And also we think F9H4 is a new opportunity of cancer immunotherapy because there are so many antibodies that are used for cancer treatment and many of them are Fc enabled antibodies. F9H4 can increase the efficacy of

tumor cell opsonizing Fc enabled antibodies. So we wanna explore new combinations of F9 with different types of antibodies in follow up research. I must thank Bruna Bertoletti, who was the postdoc who worked in this project. A fantastic postdoc worked really hard to complete this project, as well as all of our members who helped it, and our collaborators who provided the key reagents, as well as the funding sources.

Thank you very much.

James Taylor-Mosedale (10:52)
Super, thank you Lucas for that really interesting presentation. To round things up, what is the best way that the Bench Brief community can support your work following this

Lucas Ferrari De Andrade (11:03)
We really think that this antibody F984 could be used in combination with treatments in order to enhance efficacy. So we would be delighted to work with biotechnology companies or pharmaceutical companies to help do the transition to the clinical trial phase. And we are also very interested in collaborating with academic investigators that could be interested in CD16A shedding

perhaps understand the immunobiology of this pathway.

James Taylor-Mosedale (11:30)
Well I'm sure lots of biotechs will be interested in working with you.

Thanks to everybody for listening to the Bench Brief. You can find all the details and contact information on Lucas's final slide, and we'll make sure to include it in our show notes. You can find all our episodes and transcripts on icor.bio forward slash the Bench Brief, and please subscribe on YouTube to ensure that you never miss an update. Join us again next time for more essential insights by scientists for scientists.