Charlie Murtaugh, Ph.D. published a study in the journal eLife showing that sometimes an accumulation of mutations is not enough. He and first author Nathan Krah talk about their results suggesting that in the context of pancreatic cancer, what pushes cells over the edge is a loss of cellular differentiation: mature cells lose the instructions that tell them how to do their designated job in the body. The authors describe their findings, how they might apply to other cancers, and how their work could potentially lead to a new approach toward cancer therapy.">

Jul 13, 2015 — Cancer is thought to typically start from an accumulation of harmful genetic mutations that cause cells to break from their routine and divide uncontrollably. Charlie Murtaugh, Ph.D. published a study in the journal eLife showing that sometimes an accumulation of mutations is not enough. He and first author Nathan Krah talk about their results suggesting that in the context of pancreatic cancer, what pushes cells over the edge is a loss of cellular differentiation: mature cells lose the instructions that tell them how to do their designated job in the body. The authors describe their findings, how they might apply to other cancers, and how their work could potentially lead to a new approach toward cancer therapy.

Interview

Interviewer: A different way of thinking about how cancer starts. Up next on The Scope.

Announcer: Examining the latest research and telling you about the latest breakthroughs. The Science and Research Show is on the scope.

Interviewer: I'm talking with Dr. Charlie Murtaugh. Professor of Human Genetics at the University of Utah and with Nathan Krah. They've just published a paper in the journal "E-Life."

Dr. Murtaugh, as far as I understand it, your results really kind of put the initiation of cancer in a different light.

Dr. Murtaugh: The standard model for how cancer starts is that you have a cell that accumulates a bunch of different mutations until that cell becomes sufficiently abnormal that it turns into a cancer cell and stops behaving like a normal cell. And we were interested in how that applied in pancreatic cancer. And we and others have found that the genetic mutations that occur in pancreatic cancer in the model that we use are not enough to actually trigger the beginning of the disease. So we knew there's something else had to be going on. And what we think is going on now is that the differentiation state of the cell.

So how the cell knows what it normally is supposed to do. In this, case the cells are cells that make digestive enzymes to break down food. And they are normally focused 100% on that task. They're not really dividing. But then, a change happens and those cells forget what they're supposed to do and then in that context then those mutations that can cause cancer can become hyperactive and the cells start to become transformed and move toward cancer.

Interviewer: Nathan, maybe you can tell me why you're focusing on pancreatic cancer in the first place.

Nathan: Yeah definitely. So pancreatic cancer is one of the deadliest solid tumors that patients can have. It has one of the poor survival rates of just about any cancer with a five-year survival of only about 5%. So really, any clues that we can get that push toward earlier detection or better treatment options would be good.

Interviewer: You're implying principles of development to understand cancer. What does that mean exactly?

Dr. Murtaugh: Right. I have a long-standing interest in the process of differentiation. This maybe does go back 20 years. So it just how cells learn to adopt the mature fates that they're supposed to achieve, starting in the embryo and the differentiation continues in the adults. So, of course, we are continually shedding skin so we have to differentiate new skin all the time, new hair, etcetera. And it was through studies of differentiation in the pancreas and trying to understand how the different cell types, the endocrine and exocrine cells, how they get to differentiate that we began to be interested in how that might apply in cancer.

Because it seemed, based on classical work, that it was possible. That in fact cancer might start from not only genetic mutations that we know are important, but changes that are non-genetic like loss of differentiation.

Interviewer: So kind of the ideas that when a cell is differentiated, it's mature and is thought to be just sort of fixed that way, kind of stable that way. And so if you disrupt that, things go wrong?

Dr. Murtaugh: Yeah, exactly. Exactly.

Interviewer: And how did you move that into the cancer paradigm? I think you've been researching a gene, PTF1, for a while?

Dr. Murtaugh: Right. So PTF1 is a gene that's called a transcription factor. And transcription factors are proteins that regulate other genes. So a lot of what differentiation involves is turning on expression of genes that are important for the carrying out the function of the cells. So, for example, the exocrine cells that we study, they make hundreds of digestive enzymes. Like huge concentrations.

And so what PTF1A does normally, in a mature cell, is it activates those genes. So without PTF1A during development, you can never make the cells that normally digest food. And what we found is that once you are a mature cell you still require PTF1A in order to continue making those enzymes and in order to repress alternative choices.

So cells tend to not like to do nothing. If you take away a factor like PTF1A that forces them into a specific identity, if you take that away they will kind of cast around and look for an alternative identity. And in a healthy cell, that might not be a big deal. And in fact, we find that just taking PTF1A away is not enough to cause cancer.

But in a cell that has an underlying cancer-causing mutation, when the cell starts trying to change its mind about what it's supposed to do, those mutations can kind of express themselves and then the cell very, very quickly becomes abnormal.

Interviewer: You've done these experiments in an environment where there's already something wrong? And so this is like one more step that breaks the camel's back?

Dr. Murtaugh: Yeah, exactly.

Interviewer: Nathan, what do you see when you do that?

Nathan: When we take away PTF1A in the context of cancer-causing genes or oncogenes, we see that these cells are rapidly transformed into pre-cancerous lesions. And what's really interesting is that if we just express these oncogenes or cancer-causing genes alone, that's really not sufficient to do too much. They occasionally form these pre-cancerous lesions, but in the context that PTF1A loss, almost every cell that loses PTF1A will give rise to cancer.

Interviewer: And is there any indication that this happens in human pancreatic cancer?

Dr. Murtaugh: Right. So that's a really good question. So one of the things that we were able to show is that in fact in the lesions of humans, these early stage precancerous lesions, the PTF1A is also being shut down. And in fact, there are some other studies out there that didn't focus on PTF1A, but have all sort of implied the same thing that this must be happening at a very early stage in cancer. And so it does look like it probably does happen to humans and one of the things we're really interested in going forward is sort of doing the reverse, which is if we could turn it back on in human cells, would we be able to stop the cancerous cells from growing.

Interviewer: Right. I mean, will that point to a different kind of therapy?

Dr. Murtaugh: I think so. I mean, most cancer therapies in pancreatic and other cancers are targeted to cell division. And, of course, that's what conventional chemotherapy does and it has limitations because it can kill the normal dividing cells. And there are other therapies that are targeted at the mutated signals that occur in the cancer and in some cases like lung cancer, those have been really effective therapies. But in pancreatic cancer, it's been very challenging to do that. So the oncogene, the cancer-causing mutation that is so central to pancreatic cancer is one that is in a protein that is very hard to make drugs against.

And so this has been a long-standing problem in the field. But we do feel that differentiation is something that there's almost no disadvantage. In theory, at least, there are no side effects from having too much differentiation in the pancreas because it's an organ that is normally very differentiated. So if there were a way to reactivate the normal differentiation of these cells, that by itself shouldn't have any major side effects. And the question is whether it would be effective as a therapy.

Interviewer: Yeah, fascinating. I mean, is there any indication that this happens in other cancers as well?

Nathan: Yeah, that's a really good question. So one thing I think that's exciting to us is if we really look throughout the GI tract, there are a lot of cancers that undergo this process called metaplasia or a change in cell fate, which almost always precedes cancer and is usually associated with chronic inflammation.

And so one good example is in the esophagus, a lot of people have the acid reflex and that can lead to changes in the cell populations in the lower esophagus. And it's thought that those changes are actually necessary for cancer formation. And so we show a very similar story in the pancreas. We think that this actually a broadly applicable principle where you need a loss of differentiation in order to actually initiate cancer.

Announcer: Interesting, informative, and all in the name of better health. This is the Scope Health Sciences Radio.