Funding opportunities

Epigenetic determinants of normal and tumor stem cell biology

Funding Type: 
New Faculty I
Grant Number: 
Funds requested: 
$2 187 000
Funding Recommendations: 
Not recommended
Grant approved: 
Public Abstract: 
Stem cells have great promise to treat a variety of human diseases. This type of therapy, called regenerative medicine, helps repair diseased organs through the stem cells’ ability to grow new, healthy tissue. However, administering living cells to a patient is not without risks, the greatest of which is that the stem cells could produce a devastating side effect of cancer instead of or in addition to the healthy tissue growth. At this point, we can neither estimate nor reduce these risks because we simply do not understand why some stem cells are “good,” helping to grow new healthy tissue, while others are “bad,” driving the formation of tumors. The only way to eliminate the risk of cancer is to understand the difference between “good” and “bad” stem cells, which will pave the way to developing methods to remove the cancer causing ability of stem cells while leaving their regenerative function intact, the main goal of this proposal. The difference between a normal and a tumor stem cell must reside in the way they are programmed. Tumor stem cells have many of the same attributes as normal stem cells, but they must also possess unique properties that lead them to cause cancer. What are these properties? Recent studies suggest a critical regulator of stem cells is a protein called Myc, which has dual and opposite roles in stem cell biology: 1) on its “bad” side, Myc is one of the most prevalent and devastating oncogenes, or cancer-causing genes, and is responsible for a staggering number of human cancer deaths; 2) on its “good” side, Myc is essential for normal organ growth that is regulated by controlling stem cell behavior. We hypothesize that a key attribute of tumor stem cells is that they have too much Myc. Studies in mice indicate that eliminating Myc from stem cells disrupts their function and impairs cellular proliferation, which slows organ growth. For example, mice with Myc-less neural stem cells have a failed brain growth called microcephaly. Because microcephaly is also observed in people with mutated Myc, our studies in mice are highly relevant to humans and regenerative medicine. One particularly notable aspect about both Myc-less neural stem cells and those with excess Myc is that their DNA structure, also called chromatin, is profoundly altered, which suggests that Myc may control stem cell behavior by regulating their chromatin. In our proposed research, we will study the difference between the chromatin of normal and Myc-less neural stem cells as well as tumor stem cells with excess Myc. Unlike DNA mutations, changes in chromatin are inherently reversible, making them more attractive targets for drug therapy. Our long-term goal is to develop methods for removing the cancer-causing function of stem cells through altering their chromatin while leaving their normal tissue-growing ability intact, facilitating the production of safe and effective regenerative medicine therapies.
Statement of Benefit to California: 
How do we treat patients with regenerative medicine without at the same time giving them cancer as a side effect? Our overall goal is address this question to improve the safety of regenerative medicine therapies by defining what drives stem cells to cause cancer. Our studies will provide the foundation for developing methods to remove the cancer-causing ability of stem cells while retaining their ability to mediate healthy tissue growth or repair. Enhancing the safety of regenerative medicine therapies would be of great benefit to the State of California both in terms of improving the lives of patients as well as enhancing the knowledge of the stem cell field. It will also further the development of regenerative medicine leading to a new, valuable biotechnology. California should be a leader in developing safe, effective regenerative medicine.
Review Summary: 
SYNOPSIS: The "cancer stem cell hypothesis" casts a pall over the potential of stem cell-based regenerative medicine therapies. A general consensus is that the greatest safety risk for stem cell transplantation protocols is the potential for transplanted stem cells to convert to cancer stem cells. The broad goal of this proposal is to define the molecular basis of stem cell tumorigenic potential, and decipher differences between normal and cancer stem cells, by describing the control of gene expression by N-Myc and its effector GCN5 in normal neural stem cells and in neuroblastoma tumor stem cells. The c-Myc proto oncogene is a key component of the "genetic cocktail" that reprograms fibroblasts into pluripotent ES stem cells. N-Myc is important for neural development: N-Myc null mice die during embryogenesis with an abnormal neuroepithelium, and targeted deletion of N-Myc in neural progenitor cells causes generalized defects in neural development including decreased brain size and impaired development of the neural retina. Conversely, targeted overexpression of N-Myc in transgenic mice causes neuroblastoma. In Aim 1 the applicant proposes to use a ChIP-chip approach to map N-Myc binding sites in neurospheres and H3K9-acetylation in control versus N-Myc KO neurospheres. RNA expression arrays will be used in parallel to correlate chromatin changes with gene expression profiles. In Aim 2, similar comparisons will be made between control neurospheres and “neurospheres” prepared from neuroblastomas that overexpress N-Myc, and the same tumor-derived neurospheres following knockdown of N-Myc. In Aim 3 the applicant will characterize a new, conditional GCN5 KO mouse he has constructed. He will study the effects of the KO on neural development in vivo and in vitro using neurospheres. Also, mirroring the approaches in Aim 1, he will determine the effect of GCN5 deletion on the distribution of H3K9-acetylation and on gene expression patterns in neurospheres. Lastly, the applicant will ask whether GCN5 is required for the tumorigenenic effect of N-Myc overexpression, using either a knockdown strategy or by constructing doubly-mutant mice. STRENGTHS AND WEAKNESSES OF THE RESEARCH PLAN: Understanding the basic differences between ES cells and tumor stem cells represents a very important area of basic research in this field. In addition, this understanding is directly relevant to clinical applications, as the tumorigenic properties of hESCs need to be controlled before cell based therapy in clinic can be seriously considered. The strengths of this application are that it is from a highly productive applicant in a very good scientific environment for stem cell research. That said, there are some flaws that keep this application out of the top tier. Many of the experiments proposed have already been conducted in other cell types, and the sole rationale for much of the work proposed here is that the studies have yet to be done in neural stem cells. Further, the potential for positive impact on problems in public health is not immediately apparent. The importance of c-Myc in ES cell self-renewal is now well-established, although its functional role remains to be fully understood. Some evidence suggests that N-Myc may be similarly important in the stem cells of specific lineages, including neural stem cells. However, this remains to be directly tested. Thus, loss of N-Myc has neural phenotypes in vivo, but whether this is attributable to a stem cell defect, per se, or to some other problem (such as a proliferation defect in committed progenitors) has not been determined. In the face of this uncertainty the emphasis placed here on describing N-Myc’s effects on global patterns of gene expression and chromatin structure seems premature without first having greater insight into the biological context into which those observations should be placed. The author does not consider other levels of activity of N-Myc unrelated to chromatin remodeling. Evidence of why this assumption is so strongly applied by the PI is not supported by preliminary results, and as this assumption might be proven to be unsafe, dedicating two out three aims of a grant seems adventurous. Indeed, there is reason to be concerned. In Aim 1 the applicant plans to prepare neurospheres from N-Myc KO mice, and moreover says that such neurospheres are currently growing in his lab. If N-Myc were essential for neural stem cells then how can these neurospheres exist? It is possible that c-Myc has compensated for the absence of N-Myc, but that would severely compromise the goal of characterizing the impact of N-Myc loss on chromatin structure and gene expression. An additional issue is that the studies, as described, are certain to yield an enormous amount of descriptive information about differences in chromatin structure and gene expression in N-Myc wild type versus null neurospheres. The array of c-Myc induced changes in gene expression and chromatin structure is enormous, and N-Myc is likely to be just as promiscuous. Indeed, the applicant points out that thousands of Myc binding sites and potential target genes have already been identified. He argues, however, that nothing is known about target genes in stem cells, and that there is therefore a strong rationale for this work. However, no guidance is provided as to how this information will be analyzed and how biological hypotheses will be extracted and tested. There are other problems as well. Neurospheres are mixed populations of stem, progenitor and even differentiated cells. Its not clear how global patterns of gene expression and chromatin structure can be accurately determined and interpreted in heterogeneous cell populations. It is not evident that stem cells constitute a majority or even significant minority of the cells. In Aim 2 the applicant proposes to define a "gene expression signature" for tumor stem cells by comparing gene expression patterns in control cells, N-Myc knockout neurospheres, and neuroblastoma cells. This aim is confusing because it is focused on understanding tumor stem cells, but the applicant offers no data demonstrating that neuroblastomas isolated from mice are comprised entirely or largely of such tumor stem cells. If they are not, then how are the stem cells to be identified, isolated and studied? In addition, comparison to the neurospheres in Aim 1 will be difficult because neuroblastomas derive from peripheral neurons, whereas the neurospheres in Aim 1 will be derived from the CNS. The gut or sciatic nerve stem cells are probably a closer match to the neuroblastoma cells but even these cells are very specialized and probably much diverged from the neuroblastoma cell-of-origin. Also, the applicant states that by integrating these expression profiles with functional genomics data, he can identify a critical subset of genes wherein expression tracks with N-Myc and the targets are directly bound to N-Myc. This may be optimistic. Functional analysis is the only way to determine if a gene is critical and in all likelihood, there will be too many candidate genes for functional analysis. Aim 3 is based on a conditional GCN5 KO mouse constructed by the applicant. The proposed studies are very descriptive in flavor; however the applicant argues that work is important because GCN5 has not been well studied outside of yeast. The concern here is that GCN5 mediates the effects of many transcriptional regulators, and conversely Myc recruits many interesting chromatin modifiers in addition to GCN5. Thus the effects of GCN5 deletion are going to extend well beyond Myc biology, and on the other hand it is something of a stretch to think that much of Myc biology would be explainable in terms of GCN5 alone. Overall there is some merit in studying N-Myc in the ways proposed here. However, it is rather contrived and/or premature to assert that these studies are relevant to stem cell biology or tumor stem cell biology. This, plus the technical issues discussed above, decreased the reviewers’ enthusiasm for funding this proposal. QUALIFICATIONS AND POTENTIAL OF THE PRINCIPAL INVESTIGATOR: The early stages of this PI’s career bode well for future success. Dr. Knoepfler is a talented investigator with many important contributions to his field as shown by his list of publications and invited presentations at Gordon conferences and at the Cold Spring Harbor laboratory. One of his publications (with postdoctoral mentor Robert Eisenman) was an important contribution to the literature on Myc transcription factors and forms the foundation of the research proposed here; thus, he and his team are perfectly qualified to perform all the specific aims of this grant. The PI received a Bachelor's degree in English literature from Reed College in 1989 and his doctorate in Molecular Pathology from UC San Diego in 1998. He conducted postdoctoral research at the Fred Hutchinson Cancer Research Center from 1998 through 2006. He was very successful as a postdoctoral fellow, publishing 7 first author papers and multiple middle author papers, all in top journals. In 2006 he became an Assistant Professor in the Department. of Cell Biology and Human Anatomy at UC Davis, and an Assistant Investigator in the Institute for Pediatric Regenerative Medicine at the Shriner’s Hospital. He has received a KO1 career development award, an award from the Brain Tumor Society, and a grant from the Shriner’s Hospital. All are focused on the role of Myc function in neural progenitors and nervous system tumors. The applicant has provided a comprehensive plan for his future career development. Multiple aspects of his professional life (lab management, scientific development, funding and teaching) have all been considered and addressed in a satisfactory way. INSTITUTIONAL COMMITMENT TO PRINCIPAL INVESTIGATOR: The Executive Associate Dean for academic affairs at UC Davis has provided a strong letter documenting institutional commitment, and one reviewer feels that UC Davis is the best institution for these studies, encompassing all of the necessary resources and centers to successfully execute the proposed specific aims. The PI has received a “strong” start-up package and occupies 750 square feet of independent lab space. It would have been helpful if additional details about the start-up package had been provided, including descriptions of salary and research support. He has access to excellent shared resources through the UC Davis stem cell program, and a joint appointment in Shriner's Hospital of North California will facilitate the studies proposed. Stem cell research as UC Davis is outstanding and the program is growing such that key faculty members of the stem cell program at UC Davis and other centers will provide mentoring as his career unfolds. DISCUSSION: One reviewer felt that the proposal is narrow in its approach and not particularly innovative. There is not enough biology to contextualize results because Myc is involved in many processes and both Myc and GCN5 recruit many other molecules, so equating Myc with GCN5 as proposed is a mistake. This reviewer had problems with the biological basis of all 3 aims; thus, the application was not as strong as it could have been. Another reviewer pointed out that while the study is exploratory, this is a strong PI with good publications, good training, and strong institutional support for the candidate and stem cell science. PROGRAMMATIC DISCUSSION: A motion was made to move this application from Tier 2 – Recommended for funding if funds are available to Tier 3 - Not recommended for funding. One discussant felt unenthusiastic about the science as the proposal is descriptive in nature. Moreover, this discussant felt that in taking a purely basic science approach, the proposal was not strongly oriented towards the spirit of the RFA in developing future leaders and lacked relevance to any specific stem cell disease target. Another discussant agreed regarding the quality of the science, but did not agree with the other programmatic issues. The motion to move this application from Tier 2 to Tier 3 failed.

© 2013 California Institute for Regenerative Medicine