Read the Full Video Transcript

Andrea Miyahira: Hi everyone. I'm Andrea Miyahira here at the Prostate Cancer Foundation. Today I'm joined by Dr. Jelani Zarif, an assistant professor at Johns Hopkins University. He will discuss his group's recent paper, "Metabolic Reprogramming of Tumor-Associated Macrophages Using Glutamine Antagonist JHU083 Drives Tumor Immunity in Myeloid-Rich Prostate and Bladder Cancer Tumors," published in Cancer Immunology Research.

Dr. Zarif, thank you for taking the time to share this study with us.

Jelani Zarif: Thank you all so much for having me. I'm excited to share our work, so thank you again for having me. I'm very excited to talk about our work on JHU083 and how it drives antitumor immune responses in prostate and bladder cancer, which are tumors that are infiltrated with many myeloid cell types.

Glutamine metabolism fuels a lot of cell functions. As you can see, this is a cartoon showing glutamine, and when it enters the cell, it can have many different roles such as nucleotide biosynthesis, including purine and pyrimidine synthesis through FGAR and GMPS. It can also play a role in hexosamine synthesis for GFAT and glycosylation. Additionally, glutamine can enter the mitochondria for glutaminolysis so that it can be converted into glutamate and impact the TCA cycle to produce lipids, therefore, it has many functions.

These are normal cells as well as tumor cells. The cell type that we study in my group is macrophages, which highly depend on glutamine for their cell fates. There was a manuscript published in Immunity a few years ago where scientists took mouse-derived monocytes and plated them using media that didn't have glutamine and looked at their effects on differentiation in vitro as well as characterizing them thereafter. What they found was that there was a change in the differentiation states of these monocytes if they were plated in the absence of glutamine or in the presence of low-glutamine. As a fellow with Ken Pienta, I repeated some of these experiments using human monocytes, and what I found was an increase in these M1 markers such as CD86, which is a B7 family member, a decrease in the mannose receptor CD206. Shown here in orange is the decrease in the absence of glutamine, a decrease in the scavenger receptor as well as the IL-4 receptor alpha.

These are all human monocytes from healthy donors so we were really interested in the effects of glutamine on macrophages. That interest was shared with other researchers here at Johns Hopkins. One of those researchers was Jonathan Powell, and they worked on a mechanism in which T cells' metabolism was altered by glutamine sulfates. They've since published that glutamine antagonists that I'm going to talk about today increased T cell proliferation through clonal expansion to lead to tumor cell killing as well as increased MDSC, MHC class 1, class 2 presentation to those T cells.

We delved deeper into looking at macrophages. Macrophages are very abundant in metastatic castrate-resistant prostate cancer tumors. As prostate cancer worsens, we found that immune-suppressive macrophages are more abundant, and this was across several TMAs that we worked on with Angelo DeMarzo. Then when we looked at rapid autopsy tissues, we found that there was an abundance of CD206 mannose receptor-positive immune-suppressive macrophages in men who died from prostate cancer.

We wanted to look at targeting glutamine. Again, groups here were working on this, so those groups were the labs of Barb Slusher and Jonathan Powell. This is the amino acid structure of glutamine. There was an older glutamine antagonist called DON. DON was created in the '60s and has gastrointestinal issues, specifically gastrointestinal toxicity. What Jonathan Powell and Barb Slusher's group did is they created JHU083. You can see here that this looks like DON. In fact, it is DON, but it's a prodrug version of DON. When JHU083 is administered, the prodrug moieties are cleaved off and it becomes DON, and this happens in the tumor microenvironment.

Barb Slusher's group previously looked at toxicities as well as the effect on tumor cells, and they found that the prodrug in the tumor cell becomes very active in the plasma, and then the prodrug does not release DON in gastrointestinal tissues as DON does. They've also tested this in animals that have similar glutamine synthesis pathways as humans, notably pigs, so they found no gastrointestinal issues with this agent in pigs. They worked on T cells with this drug, Jonathan Powell did, and they published that in 2019 in Science, I believe.

We're interested in innate immune responses, particularly in genitourinary cancers such as prostate cancer and bladder cancer, and we wanted to test these agents in prostate cancer. We wanted to first make sure that this was a clinically relevant target, macrophages. We mined data from the Kharchenko group where they looked at nine bone metastatic samples as well as used some non-malignant samples from benign marrow and performed single-cell characterization of these samples and published it. We mined that data to look for tumor-associated macrophages in bone metastatic disease of prostate cancer.

Well, the two genes that we were most interested in were glutamine synthetase, which is GLUL, and glutaminase, which is GLS. You can see we looked at the tumor-associated macrophage proportion of these bone metastatic lesions that in the tumor there was the highest amount of glutamine synthetase present compared to benign bone marrow. We also looked for glutaminase, which is a part of the glutamine synthesis pathway. You can see that it is higher in tumor-associated macrophages found in bone metastatic lesions of metastatic castrate-resistant prostate cancer.

We have a lot of data, but what we did was look at syngeneic prostate cancer tumor models. One of those models is B6CaP, which was developed here at Johns Hopkins. It's androgen sensitive, but becomes androgen insensitive upon castration. Essentially, this is the mid-CaP model that's been backcrossed onto C57BL/6 mice, and it expresses AR, it metastasizes to the bone forming osteoblastic lesions. Pretty pathognomonic of what we see in human disease.

Monali Praharaj, a graduate student in my group, led these studies. We first tested JHU083 in B6CaP models, and we've repeated this several times. What we found was that after tumors are palpable around 32 days, JHU083 administration leads to tumor growth inhibition. We also used another syngeneic prostate cancer model called RM-I, which is very aggressive. When tumors become palpable, you can see that there is a tumor growth inhibition phenotype relative to control after JHU083 treatment in vivo. MB49 is a syngeneic bladder cancer model that has significant macrophage infiltration, and JHU083 also leads to tumor growth inhibition.

We also conducted some adaptive transfer experiments. We took monocytes or macrophages by flow sorting from tumors in C57BL/6 mice, whether they were treated with JHU083 or untreated, then flow sorted those macrophages out, and mixed them 1:1 with B6CaP or MB49 cells and put them into a mouse that did not have any tumor. This is an adaptive transfer experiment to see if we observe sustained immunity after JHU083 treatment. What you can see here in the MB49 model is that these tumors grow fine.

If we take macrophages from mice that did not receive JHU083 and mix them 1:1 with MB49 cells and put them into a new mouse, the tumors grow similarly to the MB49 cells alone. However, when we take macrophages treated with JHU083 that have been mixed 1:1 with MB49 cells and put them subcutaneously into a new mouse, you can see that there is a tumor growth inhibition phenotype, and we see the same with infiltrating monocytes. Monocytes traffic to tumors, differentiate into macrophages, become polarized, and promote immune suppression, which prevents T cell trafficking and T cell proliferation in tumors, which is a big problem in many different tumor types, especially those that don't respond to checkpoint blockade.

We see the same phenotype when we flow sort out tumor monocytes from mice that have been treated with JHU083, mix them 1:1 with MB49, put them into a new mouse, and we observe tumor growth inhibition.

Andrea Miyahira: Well, thank you so much, Dr. Zarif, for sharing this really interesting study with us today. Does glutamine blockade with JHU083 and TAMs resemble any normally occurring phenotypes? For instance, what we would see during certain immune responses? Are these changes permanent, or are they reversible?

Jelani Zarif: Yeah, great question. The treatment of macrophages with JHU083 actually resembles inflammatory macrophages. I didn't get a chance to get to that. But what we found was when we performed heavy glucose tracing, we discovered that JHU083 led to a broken TCA cycle, which resembles the TCA cycle in inflammatory macrophages, known as M1, for in vitro purposes where they produce IL-1 beta. The immune-suppressive macrophages depend on oxidative phosphorylation and glutaminolysis, so they have a full TCA cycle, not a broken one. We found that JHU083 led to a broken TCA cycle, which changed the phenotype of the cells and their function.

Andrea Miyahira: That's really interesting. By what mechanisms do you think that JHU083-treated TAMs are inhibiting tumor growth? Could this be through the reduction of tumor growth factor production, altering tumor cell metabolism, enhancing phagocytosis, or improving antitumor T cell activity or tumor infiltration?

Your text is coherent but needs minor adjustments for grammar, clarity, and consistency in tense:

Jelani Zarif: Yes, excellent question. We did address those. We found that there was an increase in phagocytosis of tumor cells. Once these macrophages are reprogrammed, they phagocytose the tumors using a MYOE1 mechanism for forming a phagophore. We did some really cool IF images and flow cytometry. We RFP-labeled PC3 cells, and then we labeled a macrophage with a different color, and you can see the overlap of phagocytosis.

There was another mechanism that was altered, which was the increase in TNF as well as their mitotic spindles. We looked at differential gene expression in tumors that were treated at an early time point and a late time point, and we saw that there was an increase in TNF signaling, mitotic spindles, inflammatory response, G2-M checkpoint, and a host of other differentially expressed genes in JHU083-treated macrophages versus the control.

We did see an increase in natural killer cells and long-lived T-cell markers, which would indicate that these were CD62L, CD44 positive T cells, which have effective memory phenotypes relative to the control, so we did see an influx of that. That was through single-cell RNA sequencing as well as stem-like CD8s that we observed. We think that this is a sustained immune response, especially given the adaptive transfer experiments that I showed. It's several mechanisms: phagocytosis, increased adaptive immune responses through many mechanisms, and an altered TCA cycle, which alters the overall function of the macrophages.

Andrea Miyahira: Okay, very intriguing. I think you mentioned that JHU083 was developed at Hopkins. What is the clinical potential for this drug, and are there anticipated or known toxicities for it, especially compared to the other one that failed?

Jelani Zarif: Yes. DON failed before my lifetime. It had systemic gastrointestinal issues because it blocks all glutamine substrates, whereas JHU083 is a prodrug and it gets cleaved in the tumor microenvironment, so the gastrointestinal issue is dramatically reduced. This is not my data, but it is data from others who've worked on it. This compound is in clinical development by a company called Dracen, and they found in Phase I, at the time of this recording, no toxicities. It appears to be safe, and we're excited about where it can potentially go next.

Andrea Miyahira: Okay, thank you. What are your next steps?

Jelani Zarif: Great question. We hope to elevate this to clinical trials. I'm working with Dr. Cathy Marshall here at Johns Hopkins to see if we can work this into a Phase I or Phase II clinical trial for men with castrate-resistant prostate cancer, potentially combining it with checkpoint therapy given that JHU083 enhances antitumor immune responses and, through that, attracts the adaptive immune response.

Andrea Miyahira: Thank you again for sharing this really interesting study with us, and I encourage anybody who's interested to read the paper because there's much more data in there.

Jelani Zarif: Thank you again for having me, UroToday. Thank you so much, Dr. Miyahira, for reaching out and allowing us to talk about our work.