Read the Full Video Transcript

Andrea Miyahira: Hi, everyone. Thank you for joining us today. I'm Andrea Miyahira here at the Prostate Cancer Foundation. With me today is Dr. Ping Mu, an assistant professor at UT Southwestern Medical Center. He'll be discussing his group's latest paper, "UBE2J1 Is the E2 Ubiquitin-Conjugating Enzyme Regulating Androgen Receptor Degradation and Anti-Androgen Resistance,” which was published in Oncogene. Thank you so much for joining me today, Ping.

Ping Mu: Thank you so much for your invitation. Hello, everyone. Thank you very much for taking the time to listen to my talk about this very brief introduction of our recent work. As Angela has mentioned previously, my lab is focused on one very central thing, which is to understand why prostate cancer is gaining resistance to anti-androgens. As we all know, we can actually manage primary prostate cancer pretty well with surgical, radiation, and castration. However, many patients will develop resistance to castration and become mCRPC.

Although we have, like many people in this community, developed second-generation anti-androgens, including enzalutamide, and it's a huge clinical success, but a lot of patients will still develop resistance to anti-androgens, which becomes one of the biggest challenges to manage this disease. We are trying to understand why patients still get resistance. As we know, there are several mechanisms already that have been shown to cause resistance to anti-androgen, including AR mutations, splice variants, amplification, or AR bypass. And a common result of all these mechanisms is that cancer cells, prostate cancer cells, somehow either bring back AR signal or make AR signal be amplified, so they no longer are sensitive to anything we are targeting to AR signals. However, the amount of patients who carry all those known alterations is actually pretty limited. For example, AR mutations are in a very small percent of patients.

So, there is a large amount of patients for whom, when they've been treated with anti-androgens, they have a restored or amplified AR signal, but we don't know how that happens. To find out all the molecular mechanisms happening in those resistant patients, we conducted a very comprehensive in vivo library screening a couple of years ago and we found several of the targets, and we have reported previously. One of the genes which came up as a topic from this library is this gene called UBE2J1, which is a story I want to share with you today.

What exactly is UBE2J1? From the name, you can easily make a guess. It is actually an E2 ligase. As we know, our cell or our body has this machinery, which uses ubiquitination to get rid of all the proteins we don't need, by a cascade of enzymatic reactions to let E1, E2, and E3 enzymes to put ubiquitin onto the protein needing to be degraded. And then this protein will be degraded. As you can see, if anything happens or any of this dysregulation happens on any of those enzymes, including E1, E2, or E3, it's going to lead to the accumulation of a protein which we don't need. Most likely, this protein could be oncogenic, and this happens in many, many cancers, including prostate cancer.

However, there's a big knowledge gap here. Although several E3 enzymes have been reported in prostate cancer, including SPOP, MDM2, or CHIP, as they regulate this process. But what is the E2 ligase actually, which is responsible for AR degradation? We still don't know. This is a very big knowledge gap because we know the E2-conjugating enzyme is the one to attach this ubiquitin onto the side of a protein, which is the specificity of this degradation, and we don't know which protein is the E2 enzyme for AR degradation.

When we got this UBE2J1 protein out as a topic of our library, we first checked the percentage of genomic alteration of the gene in prostate cancer, and we saw that around 5-15% of prostate cancer patients actually carry a homozygous deletion of this protein. That's actually a pretty high number if you consider PTEN is only like 40%. And then we confirmed the role of this protein by knocking it out in two different cell lines, LNCaP/ar and MDA-PCa-2b. We saw the same thing, if we knock out UBE2J1. You get a very significant resistance to antiandrogens, and this is also confirmed in vivo by xenografting this tumor into mice. You can see that UBE2J1 leads to very strong resistant tumor growth.

But most importantly, we noticed that this resistant tumor has a very strong induction of AR protein itself, as well as AR downstream genes. This also happens at the RNA level. You can observe it with those tumors. You see a significant induction of androgen response signals, which is very surprising because we're treating those tumors with anti-androgens. On the other hand, we also confirmed the results by qPCR, showing a robust induction of androgen signals. And why is that? Because UBE2J1 is not an E2 enzyme. So, our initial hypothesis was that this protein is the long-sought-after E2 ligase for AR degradation. How can we confirm that?

We did this very classic experiment to measure the half-life of a protein, which is AR in this case. So you can see that in normal cells, in wild-type cells, AR degradation happens in around like 12 hours. AR protein would go to be degraded internal. But when UBE2J1 was gone, you don't see AR go degraded at all. So AR degradation was specifically stopped. And then we did another very classical experiment using a drug. This drug basically specifically stops every ubiquitination and degradation, and what happens is when we put on this drug this difference in this amplified AR protein disappeared, which means if you don't have AR degradation, then there's no induction of AR protein in the cell. As many follow-ups, and I'll just show you one due to the time limitation, we actually find K48 is a very wild observed ubiquitin site on protein on AR. And we can see that there is actually a strong dip here of K48 UBE2J1 is lost, which means, really, the loss of UBE2J1 caused inefficient degradation or ubiquitination of AR and the loss caused the accumulation of AR protein.

But more importantly, we want to see if we can reverse this resistance, as first we actually checked some of the patient cohorts, including standard Stand Up to Cancer. You can see that when the patient doesn't have UBE2J1, they develop resistance to antiandrogens much quicker than the wild type. On the other hand, if we use a protein degrader, basically as a proteasome, which is called AC176, we can boost AR degradation, because when you've lost the UBE2J1, you have a very low efficiency in degrading AR. But now, we can, on the other side, boost up degradation by an AR degrader, and we can actually reverse the resistance and make the cell be sensitive again. We also confirmed this efficacy of this AR degrader in a patient-derived organoid. You can see the downstream gene of the AR-targeted gene is also re-suppressed in this case, which confirmed this on-target effect.

This is just a brief summary of our study. We think we finally found one of the real E2 ubiquitin-conjugating enzymes for AR degradation, which is this protein called UBE2J1. While it was lost in 5-15% of patients, it leads to inefficient AR degradation and accumulation of AR protein, which, in turn, leads to resistance to antiandrogens, and by using an AR degrader, we can actually boost AR degradation and reverse this process. I think one additional take-home message is, perhaps we could use UBE2J1 as an early biomarker to predict the response to anti-androgens. Thank you very much.

Andrea Miyahira: Thank you for sharing this with us, Ping. You observed no changes in AR-V7 levels with versus without UBE2J1. Did you confirm if UBE2J1 ubiquitination occurs in the AR ligand-binding domain and whether or not it ubiquitinates AR-V7? And did you look at whether AR-V7 or other LBD variants are differentially expressed, whether there are UBE2J1 wild-type versus mutant cells?

Ping Mu: Wow, Angela, that's a very, very good question, and you're absolutely right. Although we're still working on very detailed research to find out what exactly we could incite on AR, but most likely the site is in the ligand-binding domain because we do not see any difference between V7 or any other AR variant that doesn't have ligand-binding domains. But you know, there's a lot of work that needs to be done to find exactly where the site is. For the second part of your question, we don't see, actually, a huge difference on the protein level about the AR variant between the wild-type or between UBE2J1 knockout cells, which is understandable because it doesn't really degrade those variants.

Andrea Miyahira: Okay, thank you. And so the AR degraders that you used and those that are under clinical development, do they target similar sites on AR for ubiquitination as UBE2J1, or do we know even what E2 ligases are recruited by the PROTACs?

Ping Mu: That's an awesome question as well. First, as I mentioned, we haven't found out exactly which site on AR yet, and we're working on it, so I can't really tell you if it's really on the same side. But on the other hand, PROTACs need E2. So why we say this PROTAC, actually this AR degrader, could actually overcome this is there are two possible explanations. One is UBE2J1 is one of the E2. It doesn't mean it's the only E2. There's a possibility there are other E2s which are not the predominant ones at the beginning, but when UBE2J1 was lost, this guy kicked in and compensated partially for the AR degradation and could be recruited by this PROTAC. On the other hand, many patients do not have a complete depletion of UBE2J1. Although there are 5-15% of homozygous deletions, a much larger population has shadow deletion, which means they have impaired AR degradation but they still have some UBE2J1. So when you use an AR degrader, you basically make this E2 ligase more efficiently compensate for the deletion of that one copy. So those are two possible explanations.

Andrea Miyahira: Thank you. Are there translational next steps for these studies?

Ping Mu: Yes, yes, definitely. One of the things you actually mentioned is that both of these AR degraders are in clinical development right now, they're in clinical trials, and we are closely collaborating with the company developing them. On the other side, we're thinking of using UBE2J1 as an early biomarker because we can identify those patients who might benefit better from AR degraders from an early stage.

Andrea Miyahira: Well, thank you so much again, Dr. Mu, for coming on and sharing this study with us.

Ping Mu: Thank you, Andrea, for the invitation, and thank you very much for coming to listen to my short talk.