Science Signaling Podcast: 15 December 2009

Technical Details

Length: 17 min

File size: 8.2 MB

File Format: mp3

RSS Feed: http://stke.sciencemag.org/rss/podcast.xml

Download Podcast: http://podcasts.aaas.org/science_signaling/ScienceSignaling_091215.mp3

Educational Details

Learning Resource Type: Audio

Context: High school upper division 11-12, undergraduate lower division 13-14, undergraduate upper division 15-16, graduate, professional, general public and informal education

Intended Users: Teacher, learner

Intended Educational Use: Learn, teach

Discipline: Cell Biology, Metabolism, Molecular Biology, Neurobiology, Pharmacology

Keywords: Science Signaling, epidermal growth factor receptor (EGFR), glioblastoma, lapatinib, lipid biosynthesis, sterol regulatory element-binding protein 1 (SREBP1)

Transcript

Host – Annalisa VanHookWelcome to the Science Signaling Podcast for December 15th, 2009. I’m Annalisa VanHook. In this episode, I'll be speaking to Paul Mischel about a paper from his group published in the current issue of Science Signaling about fatty acid biosynthesis and glioblastoma (1). Glioblastoma is an aggressive form of brain tumor that results from the unregulated and invasive growth of glia, which are cells that support and protect neurons. Glioblastomas are the most common type of brain tumor and are often associated with an increase in signaling through the epidermal growth factor receptor, or EGFR. Dr. Mischel spoke to me on the phone from the David Geffen UCLA School of Medicine in Los Angeles.

Interviewer – Annalisa VanHookWelcome, Dr. Mischel.

Interviewee – Paul MischelThank you, I’m delighted.

Interviewer – Annalisa VanHookThe paper that your group has just published in Science Signaling this week is about glioblastomas. Many glioblastomas are associated with mutations that cause increased EGFR signaling, or epidermal growth factor receptor signaling. Is EGFR signaling what causes the glial cells to become cancer cells, or is this a consequence of their becoming cancerous?

Interviewee – Paul MischelThere’s compelling evidence that increased EGFR signaling contributes to glioma formation. There’s also considerable evidence to suggest that increased EGFR signaling needs to collaborate with other genetic lesions in order to cause tumors. So, in human glioblastoma, EGFR amplification and mutations occur in nearly half of patients. Importantly, though, these EGFR genetic lesions occur in tumors that also contain a spectrum of other genetic abnormalities, including losses of critical tumor suppressor proteins, such as PTEN, CDKN2A, and p53. Two recent papers—one by the TCGA, the Cancer Genome Atlas Group, and one by the Johns Hopkins Group beautifully illustrate this point (2, 3). I think information about this also comes from mouse genetic models, where they’ve tried to understand the role of EGFR signaling in glioma formation. And in a variety of models, the introduction of constitutively activated EGF receptor mutants did not lead to tumor formation by itself. However, if the EGFR mutations are introduced into cells lacking these same tumor suppressor proteins, then the increased EGFR signaling collaborates with these losses to form gliomas—including glioblastomas, the high-grade form. So, this suggests that, in fact, it’s a collaboration between EGF receptor signaling and other key pathways that plays a critical role in developing tumors. So, for this reason, I think, and because of the high frequency of EGFR amplification and mutation in glioblastoma patients, EGF receptors emerged as a very compelling molecular target.

Interviewer – Annalisa VanHookThe combinatorial nature of genetic lesions contributing to cancer, is a very common theme for different kinds of cancers, right? I mean, there are very few examples of cancers that result from a single mutation.

Interviewee – Paul MischelYeah, I think that’s absolutely right. And it’s very gratifying, I think, that that kind of data is derived both from looking at human samples and from doing mouse genetic models. It looks to be a pretty strong principle that there needs to be collaboration of pathways for the development of cancer.

Interviewer – Annalisa VanHookIn the paper, you mention that EGFR inhibitors aren’t used to treat glioblastomas. So, I know that EGFR inhibitors are effective in treating some types of cancer. Why aren’t they effective in treating glioblastoma?

Interviewee – Paul MischelSo, EGF receptor inhibitors, including erlotinib and gefitinib, have been used to treat glioblastoma patients in early-phase clinical trials, and unfortunately the results were quite disappointing, with relatively infrequent clinical responses, sort of on the order of 10 to 15%. And the responses themselves were unfortunately very short. So, there was a real need to try to understand why, exactly as you stated, are the EGF receptor inhibitors so ineffective in patients. So, to try to understand the mechanisms underlying this therapeutic resistance, we showed, in a 2005 New England Journal of Medicine paper, that PTEN loss promoted resistance to EGF receptor inhibitors in glioblastoma patients (4). So, PTEN is a tumor suppressor protein that negatively regulates PI3 kinase signaling, and PI3 kinase is a critical effector of the EGF receptor that promotes many of the malignant features of gliomas. PTEN is also commonly lost in glioblastomas that have EGFR amplification and mutations. And what we showed was that P10 loss limits the ability of EGF receptor inhibitors to turn off PI3 kinase signaling, resulting in clinical resistance to erlotinib. Further, in a 2007 Science paper Ron DePinho's group showed that coactivation of other receptor tyrosine kinases, like MET or the PDGF receptor, in glioblastomas treated with EGF receptor inhibitors can actually maintain this same pathway, thus continuing to have high signaling through the PI3 kinase pathway to promote clinical resistance (5). So, I think these studies highlight the importance of the PI3 kinase pathways, a critical effector of the EGF receptor signaling and demonstrate why it’s been so difficult to affect durable clinical responses in patients with EGF-bearing glioblastomas. So, it looks as if a series of genetic and epigenetic events conspire to maintain downstream PI3 kinase signaling, thus promoting clinical resistance to EGF receptor inhibitors. And, really, it was this set of observations that led us to look for alternative ways to target the EGF receptor downstream of PI3 kinase, including by examining its metabolic consequences. So, in particular, my postdoctoral fellow, Deliang Guo, who’s the first author on this paper, developed a strong interest in trying to identify the molecular circuitry that links EGFR PI3 kinase signaling with altered cancer cell metabolism and then testing whether this could be therapeutically exploited.

Interviewer – Annalisa VanHookHow does the increase in EGFR signaling affect the cancer? Does it make it more invasive? Does it make the cells divide more? How does it actually affect the biology of the tumor?

Interviewee – Paul MischelSo, the EGF receptor is actually one of the most studied oncogenes in cancer, and there’s a lot of evidence that it activates a series of canonical signaling pathways—such as the PI3 kinase pathway, the Ras MAP kinase pathway—that promote cancer cell proliferation, invasion, and survival. For example, increased EGF receptor signaling, particularly the most common mutant form, this EGFRvIII mutant, potently promotes resistance to apoptosis, in response to chemotherapy through modulation of antiapoptotic proteins, and this was shown by Web Cavenee's group (6). So, it looks as if the EGF receptor, through multiple signaling pathways, actually does a lot to promote cancer progression in terms of proliferation, in terms of invasion, and in terms of preventing apoptosis. So, you know, I think it’s, it’s becoming increasingly recognized that EGF receptor, and perhaps other oncogenes and tumor suppressor proteins that regulate these same pathways, can also promote tumor progression by orchestrating major shifts in cellular metabolism, and that’s why we began to think about targeting metabolism.

So, you know, this includes a shift towards anabolic processes to meet the demands for increased biomass that you get by having a rapid ly proliferating tumor. So, you know, work from Lew Cantley’s group and Craig Thompson’s groups have begun to elucidate the molecular circuitry that’s underlying this process (7), and there’s strong suspicions that the very same canonical pathways that drive things like proliferation, invasion, survival—that those pathways may also play a critical role in regulating these metabolic shifts. In fact, there was an interesting paper in a Science Signaling last week that showed how these receptor tyrosine kinases could actually regulate some of the very important metabolic switches (8). So, at the moment the understanding of how these oncogenic signals regulate altered tumor cell metabolism is still in its relatively early phase, but this really interested us because the idea is, if you could understand what was going on downstream, particularly at the interface between altered signal transduction and cellular metabolism, there might be something there that was really therapeutically exploitable and not subject to the same kind of rewiring that maintains the signaling pathway that we’ve seen when we’ve tried to target the EGF receptor itself.

Interviewer – Annalisa VanHookIn the current study, then, you looked at the link between EGFR signaling and metabolism by looking at how EGFR signaling affects fatty acid biosynthesis. Why did you choose fatty acid biosynthesis?

Interviewee – Paul MischelSo, rapidly dividing cancer cells require a supply of fatty acids for the formation of new cellular membranes, which is obviously something that is very important in cancer. And, in addition, fatty acids appear to play a very important role in regulating signal transduction, and they also provide an alternative energy source for cancer cells. So, it appears that de novo fatty acid synthesis—that is, synthesis of fatty acid from glycolytic precursors—is the preferred route of cancer cells. So, this suggests some kind of specificity that could potentially be therapeutically targeted, and it also suggests an important link between cancer progression and fatty acid synthesis. I think the idea that targeting fatty acid synthesis could be an effective way to block cancer is something that researchers have already begun to think seriously about. However, the molecular circuitry linking oncogenes and tumor suppressor proteins and the pathways that they regulate, with these metabolic changes has yet to be elucidated. And understanding the molecular links between oncogenes, such as the EGF receptor, and fatty acid synthesis could lead to new treatments for cancer. I think, importantly, it could also begin to help us identify subsets of patients that are most likely to benefit from treatments that target fatty acid synthesis.

Interviewer – Annalisa VanHookHow did you look at fatty acid biosynthesis as it relates to glioblastoma?

Interviewee – Paul MischelSo, we examined three aspects linking EGFR signaling with fatty acid biosynthesis in glioblastoma. First, we determined whether EGFR signaling could promote activation of the transcriptional regulator of fatty acid synthesis—sterol regulatory element binding protein 1, or SREBP-1. So, SREBP-1 is a master switch for fatty acid synthesis—it’s activated through cleavage and translocation of its amino terminal to the to the nucleus, where it acts as a transcription factor for a series of fatty acid synthetic genes, including fatty acid synthase itself. So, SREB[P]-1 activation is thought to be regulated primarily in response to low levels of cholesterol in the cell. But, we wanted to see whether EGFR signaling itself could directly regulate SREBP-1 activation, and we did this by studying tumor tissue from glioblastoma patients who were treated in a an EGF receptor inhibitor clinical trial. This was with a new drug called lapatinib and it was a very cleverly designed clinical trial conducted by the North American Brain Tumor Consortium that allowed us to ask what the drug was actually doing to the signaling pathways in the tumor tissue of patients. We also performed extensive studies in glioblastoma cell lines and in a mouse model.

Second, we used genetic and pharmacological approaches to try to identify the signaling pathways that the EGF receptor uses to promote SREBP-mediated activation and fatty acid synthesis. And we looked for therapeutically targetable nodes along that pathway. Third, and this was perhaps the most interesting to us, we asked whether abundant EGF receptor signaling makes glioblastomas more dependent on fatty acid synthesis, and, if so, whether inhibiting fatty acid synthesis—either by blocking activation of SREBP-1 or by blocking fatty acid synthase to preferentially kill EGF bearing tumors, including an in vivo model.

So, we found that EGF receptor signaling promoted cleavage and nuclear translocation of this master transcriptional regulator of fatty acid synthesis, SREBP-1, and that it increased its transcriptional activity on downstream target genes, and that it increased the amount of intracellular fatty acids. We also found that this was mediated through a primary effector of PI3 kinase called Akt, but, rather surprisingly, not through mTOR complex 1. So, this begins to define the molecular circuitry linking EGF receptor with increased fatty acid synthesis. More importantly, we found that abundant EGF receptor signaling makes glioblastoma cells more dependent on fatty acid synthesis. So, as a consequence of that, interrupting fatty acid synthesis—either by blocking SREBP-1 cleavage or by blocking the activity of the fatty acid synthase enzyme—caused massive tumor cell apoptotic cell death in tumors bearing abundant EGFR signaling but not in glioblastomas with little EGF receptor signaling. And this was also seen in an in vivo model. So, it’s exciting to us because it identifies a previously undescribed EGFR-mediated prosurvival metabolic pathway. And I think it suggests a new therapeutic approach to try to treat EGFR-activated glioblastomas and perhaps other cancers bearing abundant EGF receptor activation.

Interviewer – Annalisa VanHookIn terms of thinking about being able to use this finding as a potential basis for clinical therapy, it feasible to think about blocking fatty acid synthesis? Would that be possible to do locally or just in the tumor? Because I would imagine that blocking fatty acid synthesis might cause massive problems elsewhere throughout the organism.

Interviewee – Paul MischelYeah. Those are very good questions. I want to answer that actually first by raising a philosophical issue, which is, you know, there’s beenfrustration with the idea about targeting oncogenes in cancer because of the multitude of mechanisms that cancer cells seem to have to rewire their circuitry—you know, to maintain the key effector pathways—and that’s what really got us engaged in this problem. And it's led people to begin to think about looking for therapeutically exploitable synthetic lethal interactions. Let me explain what I mean by that. That is that oncogenes can make cancer cells surprisingly dependent on some downstream biochemical enzymes, which aren’t themselves actually apparent or aren’t themselves essential unless the oncogene is present.

So, this could lead to some really surprising and specifically very effective therapeutic strategies. I think Bill Kaelin summarized this work beautifully in a couple of perspective articles (9, 10). And Bill Hahn’s group, he basically showed using an siRNA screening approach to identify synthetic lethal partners, identify the synthetic lethal partner for the K-Ras oncogene and showed, and I think Tyler Jacks did, as well, that, you know, this may be a very effective way of beginning to find treatments that are much more specific and effective for tumor cells bearing a specific oncogene (11,12).

So, you know, I think what our data showed was that there is this therapeutically exploitable synthetic lethal interaction between the EGFR oncogene and fatty acid synthesis—that is, that this pathway becomes important for survival when EGFR is constitutively activated but seems to be much less important when it’s not present in cells. And I think the reason that that’s important is because it suggests the idea of therapeutic efficacy, and that’s really how it ties into the question that you’re asking—which is, if the normal cells have much lower requirements for de novo fatty acid synthesis, then inhibiting this pathway ought to have a preferential effect on the cancer cells.

Interviewer – Annalisa VanHookPresumably, in normal cells, then, there are other ways to activate fatty acid biosynthesis.

Interviewee – Paul MischelThat’s exactly right. And the demands for it are actually less to begin with. So, we think this is a therapeutically reasonable strategy. But, with regard to the issue that you’re talking about—getting it to the clinic—I think there’s a number of steps that need to be taken to move these compounds into trials for glioblastoma patients. You need to be able to develop compounds that block fatty acid synthesis or block SREBP-1 cleavage, and I believe such compounds are under development and even in some early-stage clinical trials for some cancers. However, the development of highly specific inhibitors of fatty acid synthase or drugs that block SREBP-1 cleavage would really be of great value, and I think it’s actually an area of very strong investigation in the pharmaceutical companies, and we’re hoping that this paper will further stimulate these efforts.

I think another aspect of it is that specificity for these is probably very important because some of the toxicities are due to off-target effects on other enzymes that regulate various aspects of lipogenesis—so weight loss, for example, is a significant problem. The compound that was used in vitro in our experiments led to 20% weight loss. You know, that’s not surprising because some of these drugs were originally looked at as weight loss agents, you know, obesity drugs. So, there’s work to do to develop compounds that are clinical-grade that are specific. And then, in brain tumors there’s always the issue of getting them across the blood-brain barrier, although in glioblastomas the blood-brain barriers is quite leaky, so it needs to be empirically examined how well these compounds actually do get across in patients.

Interviewer – Annalisa VanHookDr. Mischel, thank you for speaking with me.

Interviewee – Paul MischelMy pleasure.

Host – Annalisa VanHookThat was Paul Mischel discussing a paper from his group published in today's issue of Science Signaling. That paper is titled “EGFR Signaling Through an Akt-SREBP-1–Dependent, Rapamycin-Resistant Pathway Sensitizes Glioblastomas to Antilipogenic Therapy.” (1).

music

That wraps up this Science Signaling Podcast. If you have any questions or suggestions, please write to us at sciencesignalingeditors{at}aaas.org. This show is a production of Science Signaling and of AAAS—Advancing Science, Serving Society. I'm Annalisa VanHook, and on behalf of Science Signaling and its publisher, the American Association for the Advancement of Science, thanks for listening.