Helen Piwnica-Worms (U. Texas) 2: Translating Cell Cycle Principles to Targeted Cancer Therapies

Hello I’m Helen Piwnica-Worms, Professor and Vice Provost of Science at the University of Texas MD Anderson Cancer Center And in lecture 2, we’re going to be expanding on what we talked about in lecture 1 So, in lecture 1, we learned about the cell cycle And now we’re going to be talking about how we can translate this fundamental knowledge of cell cycle control to targeted cancer therapies So, in the first lecture, we talked about how the cell cycle is regulated And, in this lecture, we’re going to be talking about how checkpoints interface with the cell cycle machinery to bring about cell cycle delays, and I’ll define checkpoints Then, we’re going to move on to how cancer cells derail these regulatory pathways, and, finally, how we can take advantage of this basic discovery knowledge to selectively kill cancer cells So, let’s think about the problem Our human genome contains three billion base pairs of DNA Only two percent of our genome encodes for the… encodes for proteins, and that’s about 20,000 proteins that make up the 200 different types of cells in our body These cells, then, are organized into organs and tissues, and ultimately form the complete human body And our adult human body is estimated to contain approximately 50 to 75 trillion cells So, the magnitude of the problem… three billion base pairs of DNA in each of these trillions of cells must be accurately replicated and equally segregated to two daughter cells during each cell division cycle over our lifetime And we’re living to a hundred years now So, the question is, how do cells accomplish this amazing task? And- we know that failure to accurately replicate our genomes leads to mutations, genome instability, and one consequence of this is cancer So, let’s take an example We’re all used to living in a world with sunshine So, what happens when you’re out in the sun? Well, the sun emits UV radiation and this UV radiation can penetrate our skin cells What happens then is mutations So, UV radiation can interact with our DNA and create DNA mutations And cells can respond in three ways: they can die through a process known as apoptosis; they can senesce; or they can induce cell cycle delays and allow time for the cells to repair those mutations So, think about… many of you have been out in the sun and gotten a sunburn So, what is a sunburn? A sunburn is basically when your cells have so much DNA damage they basically say, it’s better for me to die than to compromise the entire organism, which is us And so, when you peel after a sunburn, that’s your cells basically undergoing a process known as programmed cell death The second way cells can respond is to undergo a process known as cellular senescence In this case, cells exit the cell division cycle, but they’re still alive, and they’re still active, and they still secrete things The final way cells can respond is to induce checkpoints, and what checkpoints are… are signal transduction mechanisms that basically signal to the cell cycle machinery to bring about cell cycle delays, or makes the cell cycle stop at different stages of the cycle Now, what happens during this time? The cells were equipped with an amazing repertoire of DNA repair pathways that will try and repair this DNA Now, if not all of those mutations are repaired, then we can pass those mutations on to our daughter cells upon division And, again, one consequence of this is cancer And, as one example, melanoma is a type of skin cancer that can happen when our cells do not appropriately repair the DNA damage after this type of exposure So, when we think about cancer cancer arises from the accumulation of several alterations in the cells of our body And this includes alterations in the tumor cell itself, as well as the cells that make up the tumor microenvironment And so, what do all these mutations do? What these mutations do is to change critical cellular pathways in our body Because a cancer cell wants to continue to proliferate and so it is going to mutate genes

that will sustain its proliferative capacity In addition, a cancer cell wants to get rid of all those pathways that are going to shunt it towards cell death, and so you’ll find mutations in any type of gene that will commit a cell to cell death You find mutations in genes that enhance mobility and invasiveness Most patients die from metastasis and we’ll come back to this concept at the end of this lecture And so, in order for a cell to leave its primary site in the body and take up residence in another tissue, it must gain the capacity to move and invade In addition, the cell must sustain its food supply, and it does this by activating processes that are important for angiogenesis Because the cell is continuing to proliferate, it needs to modify or program… reprogram its cell metabolism It needs to ignore the stop signs — anything that’s going to make the cell cycle stop it wants to, again, mutate those genes And it wants to evade immune destruction Think about a tumor growing in your body That tumor expresses… has mutations, those mutations then make mutant proteins, and those mutant proteins, if they’re expressed on the surface of the cell, can be recognized as a foreign body So, in the same way your immune system will fight a virus or a bacteria, it should be able to recognize and then kill that tumor But what tumor cells do… they’re very tricky They figure out ways to not be recognized by our immune system They also want to maintain their telomere length — this gives them the ability to divide ad infinitum And they want to sustain their cell cycle progression So, we’re going to be talking about introducing the concept of ignoring stop signals, and this is the concept of checkpoints So, what are checkpoints? So, checkpoints, again, are basically signal transduction pathways They activate pathways in the cell which eventually interact with the cell cycle machinery to bring about delays So, for example, if a cell is in the G1 phase of the cell cycle and it senses that its DNA has mutations or that there’s replication stress, it will stop in the G1 phase of the cell cycle and not move into S phase If a cell is already in the S phase of the cell cycle, it will no longer fire new replications of origin If a cell is in the G2 phase of the cell cycle, it will delay there, because it doesn’t want to segregate that mutant DNA to the two daughter cells And so those are checkpoints that respond to DNA damage There’s also a checkpoint known as the spindle assembly checkpoint and this monitors… recall, when we talked about the various stages of mitosis in our first lecture, the sister chromatids align on the metaphase plate, and if the microtubules are not attached appropriately to each of those daughter sister chromatids, then the cell will sit there and it won’t move into anaphase until that happens So, that’s called the spindle assembly checkpoint, but I’m going to be focusing on the DNA damage checkpoint in this lecture So, again, the DNA damage checkpoint Its goal is to stop the cell cycle at these various cell cycle phases So, recall, from our first lecture, we talked about how the cell division cycle is regulated by a family of cyclin-dependent protein kinases We spent a long time on the Cdc2 protein kinase — remember, this is the master regulator of mitosis It must be activated to drive cells from G2 into mitosis Another key kinase is the Cdc2 Cdk2 protein kinase This is important for early cell cycle transitions Okay so, how, now, does replication stress or DNA breaks signal to the cell cycle machinery to bring about cell cycle delays? So, what needs to happen is the cell needs to turn off the cell cycle accelerators So, who are these cell cycle accelerators? Well, these are some of the same proteins we talked about in our first lecture So, we have the cyclin-dependent protein kinases that drive the cell cycle forward, and we have a key activator which drives the cell cycle forward So, a normal cell wants to turn the activity of these enzymes off In addition, a cell wants to turn on the brakes And so, what’s an example of a cell cycle brake?

So, one of the most key brakes of the cell division cycle is the p53 tumor suppressor protein This protein is a transcription factor It activates several genes and these genes then encode proteins which are effectors of p53 And they then interact with the Cdks to inhibit their activity In addition, there are two additional brakes, Chk1 and ATR These are both protein kinases So, Chk1 phosphorylates the Cdc25 protein phosphatase, and what it does, then, it licenses it for ubiquitin-mediated prote… proteolysis So, Cdc25A becomes degraded Now, when it degrades, it can’t activate the Cdks, and this is a key way that cells stop their progression, turn on their brakes The ATR is also a protein kinase, and it directly phosphorylates and activates Chk1 So, let’s think about what we’ve learned about the cell cycle and some of these key brakes and accelerators, and how we might begin to target these for cancer treatment So, let’s think about adding a Chk1 inhibitor to a normal cell So, a normal cell has an intact p53 pathway p53 is absolutely essential for cells to stop in the G1 phase of the cell cycle It also has an intact ATR/Chk1 pathway, so cells are able to stop in the S and G2 phases of the cell cycle But if you add a Chk1 inhibitor, now you basically… still have p53, and so it can hold cells in G1, and there is a p53 pathway that helps to reinforce the S and G2 checkpoints And so normal cells, under these conditions, can respond to this inhibition But what about a cancer cell that doesn’t have p53? In this case, if a cell does not have p53, it cannot stop in the G1 phase of the cell cycle However, the ATR/Chk1 pathway is intact, and so that cell can stop in the S and G2 phases of the cell cycle So, if you come in with a Chk1 inhibitor, now you lose all of your brakes, okay? And so what happens is you can commit these tumor cells that lack p53 to cell death with this treatment So, again, let’s come back How can we take advantage of our basic understanding of cell cycle and checkpoint control for targeting cancer cells? So, it ends up that weakened checkpoints are a universal feature of cancer cells, and that’s because the p53 pathway is disrupted Now, cancer cells however, still need brakes And so let’s take an example of a stop sign So, a normal cell, when it approaches a stop sign, it will balance its accelerators and brakes, and it will stop A tumor cell that lacks p53, it’s got its accelerator revved up and it’s got its brakes revved down Now, a tumor cell, under normal circumstances, can get away with running a stop sign However, if a Mac truck is coming in the other direction, that cancer cell needs to stop And so when I talk about another car coming in the opposite direction, one can think about adding a DNA damaging agent, and now that cancer cell needs to stop to try to repair that DNA before moving forward So, using this as a rationale, clinical trials have been started in patients And, again, the rationale for the clinical trials is that a normal cell exposed to DNA damage… and when we think about patients in the clinic, these DNA damage agents could be things like irinotecan, which induces single-strand breaks, etoposide, which induces double-strand breaks; ionizing radiation, which induces double-strand breaks; sometimes patients are given antimetabolites, these induce DNA replication stress So, all of these chemotherapies in the clinic will induce DNA damage And most cells of your body will respond by stopping However, a cancer cell, like we mentioned, doesn’t have p53 And so it’s going to not be able to stop in G1 It will be able to stop in S and G2, because the Chk1 pathway is intact, but if we come in with a Chk1 inhibitor, now cells will blast through those checkpoints, and it’s a way to preferentially kill p53-deficient cancer cells So, we started a clinical trial when I was a faculty member at

Washington University School of Medicine And 25 patients with a variety of metastatic cancers were treated And they were given a DNA damaging agent, in this case, irinotecan, with a Chk1 inhibitor Now, 4 of 4 of these patients had a special form of breast cancer known as triple-negative breast cancer So, when breast cancers are being treated, they’re typically divided into three different types In the first case, the tumor will express the estrogen and progesterone receptors, and we have nice, targeted therapies that rely on estrogen-deprivation pathways for treating those cancers The second major type are those cancers that overproduce the HER2/neu tyrosine kinase and we have targeted therapies for them as well In the case of triple-negative breast cancer, what that means is it’s what it’s not — it doesn’t express estrogen receptor, it doesn’t express progesterone receptor, and it doesn’t over-produce HER2/neu And if those patients fail chemotherapy, we really have no targeted therapies at present to treat them So, what was exciting about this trial is four of the… 4 of our 4 patients responded, 2 patients with the partial response and 2 patients with stabilization of disease We were able to do correlative studies, and in a very small cohort of patients we were able to show that the tumor response correlated with p53 status So, in other words, if the tumors had a mutant p53, they responded to this therapy There was one patient… patient with… that was… her tumor was estrogen receptor-positive, her tumor was positive for p53, wild-type p53, and that tumor did not respond to the therapy Here’s just an example of one of the patients You can see, when she started the treatment, she had metastatic disease on her chest wall, and after three cycles of treatment a dramatic loss of that tumor to her chest wall was seen Now, what’s very exciting for someone like me, who’s been in the cell cycle field for a very long time, and as a basic discovery scientist, you know, I started my career trying to understand how entry into mitosis is regulated And, as we learned in the first lecture, you know, this required us to understand basic fundamental concepts using frogs, and sea urchins, and yeast But now, the field has progressed where we have inhibitors to many of these cell cycle regulators that we’ve identified and characterized in the field These drugs are now in clinical trials in patients And we use many of them in my laboratory in preclinical studies But I want to quote Socrates And what Socrates says is, “The only true wisdom is in knowing you know nothing.” And what I mean by this is, you know, I am a basic discovery scientist I am not a clinician But when my work became app… applicable to applying in the form of clinical trials to patients, it was my first experience really working with humans as a model organism And what I realized very quickly is that you cannot do the type of controlled experiments with humans as we like to do in the laboratory So, for example, as a scientist, I would have loved to run a clinical trial where I could have treated patients with only the DNA-damaging agent, only the Chk1 inhibitor, the combination, and some perhaps with vehicle But you know we can’t do that with patients And so… and the other thing that was very difficult is you want to be able to collect those tumor samples as the patient is being treated, to go back into the laboratory and say, well, this was my hypothesis, but is that really what’s happening in the tumor cell when we treat it So, we couldn’t do that, so we went back to the laboratory and we said, well, what’s the next best thing What can we do to create model systems, where we can do the types of controlled studies that we like to do as scientists? And so, what… what my laboratory started doing was building certain models, and these models are called patient-derived xenograft models And what we do is we obtain samples directly… tumor samples directly from the patients, and we take those tumor samples and we implant them into the mammary fat pads of special mice, which lack immune systems Now, we need to do this because if we put a human tumor in a mouse, the mouse immune system will attack that as a foreign body So, we need to work with special mice that lack certain components of their immune systems And so this is how we do the experiments What we first do is we humanize the mouse mammary gland So, how do we do that? Well, we obtained cells… and this was from a patient undergoing a reduction mammoplasty, so we take her normal mammary fibroblasts and we immortalize those cells And we then irradiate them to activate them to secrete certain things, and this helps,

then, for those cells… we put those into the mammary gland of the mice and this creates, then, sort of a humanized, welcoming environment, so that when our… our clinician collaborators call us and say, we have a patient with breast cancer, we’re going to take a biopsy, please come and we… we go to the surgical suite, or to the radiology suite, and we collect these samples, and we immediately bring them back to the lab, and we dissociate them into these organoids, and we then put them into these recipient mice And we generate these mice and, once these tumors take in the mice, we can then propagate these tumors ad infinitum for all the types of scientific studies we want to do And here’s an example of what some of these tumors look like So, here, they’ve been implanted into the mammary gland and we’ve pulled the tumors out So, now we have a lot of material to answer the questions that we want to answer Now, we know these are human tumors because we can now take the mammary glands and, in this case, this particular tumor metastasized in the mouse to the lung, and we know they’re human because we can take a human-specific antibody and we can then probe this tissue And so anything in brown is human So, you can see, here, a human tumor in the lung of this mouse All of the blue is the mouse lung tissue So, now, we can set up what we call preclinical trials in mice So, we can put mice on trials that we… that we ultimately would like to move to the clinic, if they work in the mice And so, here, what we did is we took some models, so, these are two different models from two different patients One of the patients had a tumor that was wild-type for p53; the other patient had a tumor that was mutant for p53 We engraft those tumors into the mice and then we set up big cohorts And then we’re able to say, okay, let’s treat one cohort with the vehicle — so, this is what we dissolve our drugs in — let’s treat another cohort of mice with the DNA-damaging agent, another cohort of mice with the Chk1 inhibitor, and then let’s do the combination And let’s actually look to see how these tumors respond So, here’s one example We have tumors in the mice and, if you look over here, you can see that the tumor volume there… it’s growing at the same rate, no matter how we treat the mice So, that was the p53 wild-type tumor In this set, we have the mice engrafted with the p53 mutant tumor So, the red and black lines are superimposable The black line is the tumor… the mice have been treated with vehicle only, and the red with the Chk1 inhibitor So, there’s no difference there But if we treat with the DNA-damaging agent, you can see a slight slowing in tumor growth And if we treat with the combination — this is the purple line — we see an even slower slowing of growth of the tumor And in some cases, where you see this fall off, this tumor actually necrosed and fell off the animal Now, this translates into improved lifespan So, again, the top group is a… the mice engrafted with the p53 wild-type tumor No matter what we treat those mice with, it doesn’t extend the lifespan of these mice But look at down here: this is a tumor that was mutant for p53 And you can see that the DNA-damaging agent alone, this teal color, extended the lifespan somewhat, and the combination therapy even more So, this, from a scientist’s point of view, we can make the conclusion, then, that p53 status does help predict how these tumors are gonna respond to this combination therapy So, now let’s move to metastasis So, metastatic breast cancer is cancer that spreads from the breast to other parts of the body And, in the case of breast cancer, they will go to the lymph node and then four major parts of the human body: the brain, the lung, the liver, and the bone Now, this is what ends up killing patients If the tumor remains in the breast of the patient, surgery… surgeons can remove that tumor and women will be fine But it’s when the tumor moves, this is ultimately what kills our patients And so we need to understand the metastatic process and we need to learn how to target that metastatic process So, what is the metastatic cascade? Well, a tumor originating in the breast has to change and get certain properties Remember, earlier in the lecture, we talked about all the properties that tumor cells will gain So, they need to be able to invade and move throughout the body And so what they do is they acquire properties which help them leave the tumor, move through

the basement membrane, they can then spread to the lymph node, they get into the circulatory system… once into the circulatory system, they need to then extrav… extravasate and get out of that circulatory system, and then take up residence in a second organ, colonize that organ, and be able to grow So, it’s a pretty complicated system and only the most fit cells will ultimately be able to do this So, we, again, can use these mouse models to study this process And the way we do this, again, is to take our human… the human tumors directly from the patient, and we’re able to express a gene encoding firefly luciferase So, many of you as children, when you were growing up, might have seen fireflies and tried to capture them Fireflies have an enzyme that basically can emit light And we can capitalize on this to create tumors that, then, will emit light that can be visualized through bioluminescence And so we take these cells that are now expressing the firefly luciferase, we put them into mice, and then we can monitor metastases So, on the left, you can see mice where we’ve implanted into the mammary fat pads, and then you can see the tumor has moved to the lung and up to the brain So, if we now take these tumors, we can open up the mice and you can see… we can see bioluminescence in the lung, in the liver, in the bone, and the brain So, these tumors, human breast tumors, leave the mammary gland of these mice and they go to the same places that a human tumor would metastasize in a patient with breast cancer So, we’re… have a great model, now, to try to dissect that So, what do we do? Well, we collect the tumors, we use bioluminescence imaging, and we open up the mouse, and we take these tumors out of the brain, out of the lung, out of the mammary gland And now, using several different techniques, we can ask, what mutations become enriched in these metastatic lesions, relative to the mammary gland from which they came? We can ask, what genes are expressed at higher or lower levels? We can ask, what proteins are more or less abundant? And we can ask, what proteins have been turned on or turned off? And then we can ask, are these differences targetable? Do we have drugs that we can now use in these mice to try to treat these metastatic lesions? And if drugs are not already available, we can now set up conditions to try to find inhibitors of these pathways, to try to target these metastatic lesions So, in summary, then, I would like to end by saying that, you know, I am a discovery scientist And, recall I mean, in my early career, we started out trying to just understand how the cell division cycle is regulated This led us to do biochemical experiments, to ask how the Cdc2 protein kinase, the master regulator of mitosis, is regulated Once we had the basic cell cycle machinery, we moved up to this checkpoint control, to ask how things like DNA damage or replication stress signal to the cell cycle machinery to bring about delays We could then… once we had fundamental knowledge of how normal cells regulate cell cycle progression, we could then begin to understand how cancer cells derail these pathways And, with that knowledge, then, we could think about clinical trials, coming back, running preclinical trials, and then ultimately we come back into the lab from what we learned And so it’s bench to bedside and bedside to bench And it’s really been a very wonderful opportunity to be able to see this discovery science now impacting patients with the cancer problem So, thank you