How Chemotherapy Drives Metastasis By Causing Tumor Cells to Become Significantly More Mobile - Sarah Amend

December 6, 2023

Sarah Amend delves into her research on Polyaneuploid Cancer Cells (PACCs) and their increased metastatic potential, a study published in Clinical and Experimental Metastasis. Dr. Amend, a recipient of the 2023 Challenge Award, led by graduate student Mikaela Mallin, explores the response of prostate cancer cells to cisplatin, revealing significant morphological changes and increased cell size, indicative of cancer recurrence. This phenomenon, observed across various cell lines and tumor types, is attributed to an alternative cell cycle known as the endocycle. The research highlights the role of Vimentin in cell motility and invasion, and demonstrates that endocycling cells are more deformable, aiding their movement through physiological spaces. This study, corroborated by in vivo findings, suggests that understanding the endocycling state is crucial for developing effective cancer treatments, as it contributes to metastasis, dormancy, and therapy resistance.


Sarah Amend, PhD, Johns Hopkins Medicine, Baltimore, MD

Andrea K. Miyahira, PhD, Director of Global Research & Scientific Communications, The Prostate Cancer Foundation

Read the Full Video Transcript

Andrea Miyahira: Hi, everyone. I'm Andrea Miyahira here at the Prostate Cancer Foundation. Today I'm talking with Dr. Sarah Amend, an Assistant Professor at Johns Hopkins University, about her group's recent paper, "Cells in the Polyaneuploid Cancer Cell, or PACC, State Have Increased Metastatic Potential." This was recently published in Clinical and Experimental Metastasis. Dr. Amend has also recently received a 2023 Challenge Award on this topic. Dr. Amend, thank you for joining us today.

Sarah Amend:
Thanks so much for having me. I'm excited to share our data. Thank you so much for the invitation to come and talk with a broader audience about this exciting work. I am co-director of the Cancer Ecology Center here at the Brady Urological Institute. We'll be speaking today, as you mentioned, Andrea, about some of our exciting work led by a really talented graduate student, Mikaela Mallin.

Our work really started with what I think of as a pretty straightforward question: trying to understand the nature of lethal cancer, specifically what happens when you stress prostate cancer cells, in this case, in vitro. Here we're using PC3 cells, which is a common prostate cancer cell line, and we're treating them with about a lethal dose 50 of cisplatin. You can see right away that there's a morphological change at 72 hours, so those are the cells that survive.

This is where many people who study lethal cancer stop their studies. What's different about our work is that we then permit these cultures to recover in complete media. In this case, I'm showing you 5, 10 days. We let them go even longer in some cases. What I hope you can appreciate is that cell size is continuing to increase. So certainly, we see the cell die-off that you would expect from chemotherapy, but there's this clear shift in morphology. After a period of time, those cells will undergo what we call ploidy reduction, so they will repopulate a typical proliferative population. So really, what we're seeing here is recurrence. What we would see as clinical recurrence is recurrence in a dish.

The key takeaways from these observations are: Number one, these surviving cells increase in cell size, and we now have published data that this is actually upwards of 70-fold the size they were at the beginning of the experiment. They also undergo whole-genome doubling at least out to 16N and likely higher than that. We've recently published that in Neoplasia. Then, again, as I alluded to before, we do see recurrence, at least in vitro.

Importantly, this seems to be a universal phenomenon. So it's not specific to PC3 cells, it's not specific to cisplatin, but it's observed in all cell lines we've looked at and all tumor types we've tested and with multiple classes of drugs. I showed you here cisplatin. We also have data around docetaxel, etoposide, even radiation, and really exciting for the metastasis story, we also see this happen in response to tumor microenvironment stressors such as shifts in pH and oxygen or overcrowding of cells.

One big question that is a larger conversation for another time is how are cells becoming so large, or hypertrophic, and how are they undergoing whole-genome doubling? To answer this question, we used a next-generation FUCCI Reporter, which is just a cell cycle reporter. The important thing for you in this slide is that cells in G1 will fluoresce red, they're briefly double positive in early S, and then through S, G2, and early mitosis, the cells are green.

I'm going to draw your attention to a cell sort of in the middle of this video that I'm going to allow to play. Right now it's sort of an orangey-red color and now it's yellow and now it's green. So that's a cell now that has progressed from G1 through S into G2. That cell is going to sit in G2 for a little while. It's still moving around. What you saw just there after a brief colorless period, it went through G1, back through S, and now we're sitting in G2 again. In a moment it's going to leave G2 and progress again straight over into G1. Forgive the image quality there, and now we're red again in G1.

Now, the reason why this is so important is that what we're seeing here is that cells are exiting the mitotic cycle and entering a non-proliferative cell cycle where we see these repeated growth G phases and DNA replication S phases in the absence of mitosis. This alternative cell cycle is known as the endocycle.

Now, this isn't just a phenomenon that we see in vitro. We have also recently published that the presence and number of cells with evidence of this type of whole-genome doubling in primary prostate tumors actually will predict metastatic recurrence. On the right here, what you're seeing are radical prostatectomy samples from a TMA series here at Hopkins, and stained in brown is EpCAM so that we can really delineate the cell borders. That arrow is pointing to a cell with whole-genome doubling. We found, you can see here from the hazard ratio of 2.7, that the total number of cells that had undergone this whole-genome doubling in each TMA spot was predictive for metastasis-free survival. So if you had these cells, if you had more of these cells, you were at increased risk of metastasis and progression.

Taking all this together, this led us to hypothesize that endocycling cancer cells have increased metastatic competency. To start to answer this question, we just did some simple 2D tracking of these cells. On the top, we have parental mitotic PC3 cells and on the bottom, we have cells in the endocycling state. You can see that they're quite large. What you can see from the single-cell tracking data on the right is, even just by eye, cells in this endocycling state are moving much more than the mitotic cells. One question we often get is, "Oh, it's just the rigor cells of the step size must be larger." Even after correcting for increased cell size, we still see this increase.

In fact, we see that cells in the endocycling state are moving greater distances, so from point A to point B in a dish. They're also moving more directly, so we see less wandering movement. They're actually sort of moving in more of a straight line, and you can see that quantification here. Now of course in this case, these are cells that are just in static culture, so they aren't moving towards a chemotactic gradient. That's going to be something that's really important in metastasis, as cells, after they are invading from the primary tumor into the circulation or after they reach a secondary site, how are they actually entering that secondary site? In this case, we also tested the capacity for cells to climb up a chemotactic gradient.

Here we have the parental mitotic controls on the top and the endocycling cells on the bottom and positive FBS, climbing up an FBS gradient, on your right. All of those red tracks are cells that are moving toward that chemotactic gradient. Now, interestingly here, and I'm not showing you the data here, but we saw similar amounts of directness and distance moved in the endocycling cells regardless of whether there was a chemotactic gradient or not. What's different here is that now that directionality, all of those cells are really moving in that one direction of the chemotactic gradient.

In fact, in the timeframe of our environment, the Rayleigh test here, as well as other tests that we perform, indicate that only the endocycling cells are actually responding and showing positive chemotaxis in our dataset. This then begs the question of, how does this happen? These large cells, even by looking at the movies, you can see that their membranes seem to be very highly active.

To look at this more specifically, we looked at mean displacement of beads that are linked to the cytoskeleton via an RGD linker, and those are those little dots that you see in this image. We just recorded it for about 300 seconds. You can see right away that endocyclic cancer cells are moving those beads around more, so they are remodeling their cytoskeleton much more. We find that a key intermediate filament that's important in motility and invasion, Vimentin, is likewise increased in these cells. On the top, we have phase, followed by DAPI, which indicates DNA content, and then Vimentin. What I hope you can appreciate is our non-treated, so those are cells that are mitotic versus the cells 10 days post-treatment. Those are endocyclic cells. Those endocyclic cells have a lot more Vimentin in their cell bodies. In fact, if we use acrylamide to depolymerize Vimentin, we can actually reverse this motility phenotype. So it really seems as if Vimentin is critical for that motility phenotype.

Finally, it's great that these cells are moving flat on a dish, but of course, we know that that's not what happens in a patient, of course. These cells really need to be able to move their cell bodies through space. One important measure of this is to understand how deformable, how squishable, really, these cells are. And so in this case, we're using those same beads that we saw before. They're still linked to the cytoplasm, but in this case, we're applying a magnetic field. So we're looking to see whether cells are resisting that magnetic field with their cytoplasm or whether they're sort of being pulled along. What we can see from this data with magnetic twisting cytometry, is that cells in the endocycling state are less stiff or more deformable, so they're actually able to squish and move through space.

We're able to test this functionally by using a microfluidics device. In this case, we have FBS-high on the top, and these are physiologically meaningful channels. You can see that endocycling cells that were seeded on the bottom are actually able to deform and move through those spaces at two different widths, we have both 10 microns and 20 microns.

Importantly, this isn't just in vitro. We've also seen this in vivo. In this case, these are circulating tumor cells taken from the circulation of a prostate cancer patient with metastatic castration-resistant prostate cancer. And really drawing your attention here to the DAPI channel, where you can see within a single pan-cytokeratin-positive CTC, that we have high genomic content, indicating a cell that had undergone whole-genome doubling. And while it's not quite ready for primetime yet, these data are consistent with our mouse in vivo studies.

Just to wrap up and think about where we're going next, really, we're interested in understanding lethal cancer and cancer metastasis is lethal and incurable. This idea of metastasis resistance from an ecological perspective is that these two responses have a shared adaptive strategy. It's a stress response that would be true whether that stress was experienced in the microenvironment or from therapy. And so, really, what we're focused on now is understanding the shared adaptive strategy of the endocycling state. It was a bit of a whirlwind tour of what we've been working on and thinking about. This is the great team that makes it actually happen with a special acknowledgement to Mikaela Mallin who led this metastasis work as well as our funding sources, most especially the Prostate Cancer Foundation. Thanks.

Andrea Miyahira:
Thanks, Sarah. That was pretty cool. I loved all the videos. Do the larger sizes of PACs impact their movement through vessels and tissues, and have you looked at CTCs to see if you can find them in there?

Sarah Amend:
Yeah, I think this is a really important question, especially when we're considering the incredible cell size of these endocyclic cancer cells. They're 70-fold larger, and one thing that we found in data that I wasn't able to show you is that they're actually more stiff around their nuclear area, the perinuclear space. The reason why that's important is that that means that this represents what's called a hyperelastic material. And so that changes how cells are able to respond under stress, so if something's squishing them, or under strain, if something's pulling them. And so these cells are actually able to both protect their nuclear integrity, which would be really important for cells moving through the vasculature and intravasating and extravasating, but they're also able to manage this huge cytoplasm and navigate those borders. In in vivo work in mouse studies, we have found these cells both in circulation as well as disseminated tumor cells, so we actually do find them at secondary sites. So they get there.

Andrea Miyahira:
Okay, interesting. One interesting question would be if you've started to do any single-cell genomics or omics to look at mechanisms of how they form and how they could be targeted.

Sarah Amend:
We have just begun those single-cell studies in these cells, in part because we sort of needed to optimize how we can even capture these cells because they are so large. In some of our very early, very pilot single-cell omics studies looking at copy number variations, so in this case, we'd be looking at more structural variants rather than In-Dels or SNPs. What was really surprising to us is, actually, compared to the mitotic population they come from, these endocyclic cancer cells are nearly identical. So it really does seem to have high fidelity of that DNA replication, there's just a lot more DNA. What this means for its resistant cell state, because we do know that these cells are also highly resistant, they're the ones that survive therapy, they're also not susceptible to, if we hit them with an additional class of therapy, we're not quite sure yet. But work is underway to figure all of that out.

Andrea Miyahira:
Okay, really interesting. What are the next key steps in your studies?

Sarah Amend:
As I alluded to a little bit before, we really think that this endocyclic cancer cell state is the keystone that is the linchpin of lethal cancer. That it's important for metastasis, it's important for dormancy, it's important for therapy resistance. And so a lot of our work now is understanding both that convergent phenotype, in particular related to that endocycle.

If you can imagine for a moment, if we use chemotherapy as we did in these studies and we kill the mitotic cells, chemotherapy works. It's really important. That's why it's in our toolkit for cancer patients. But those cells that remain, if they're adopting this endocyclic state, can we identify those therapeutic vulnerabilities that are unique to these cells and then identify targets, either by repurposing drugs or identifying new drugs, where then we can specifically target those cells? Then we can actually start to think about a durable cancer response.

Andrea Miyahira:
Well, thank you and good luck with your next studies. This is so interesting and I'm very interested to see what you guys will find next, so thank you.

Sarah Amend:
Thanks so much.