Diabetes mellitus is a disease that results from insufficient insulin production and affects approximately two million people in the state of California. Currently, the most widely-used therapy involves direct insulin injection to supplement the insulin that is naturally produced in the pancreas. This therapy diminishes the quality of life for diabetes patients, particularly because lifelong injections are required. Drugs that could either increase production of insulin-producing beta-islet cells or upregulate endogenous insulin would provide an alternative to current insulin injection therapy and dramatically decrease morbidity. In order to more accurately and quickly identify these candidate drugs, it would be very advantageous to have cells that reliably mimic a developing pancreas “in a dish” for drug screening. Human embryonic stem (hES) cells can provide a renewable source of normal cells that can be differentiated into pancreas-like cells, thus providing a good model for studying the effects of various compounds on insulin production.
The applicant institution has significant expertise and intellectual property in engineering mouse embryonic stem cells to differentiate into pancreatic-like cells that strongly resemble cells in the body. We propose engineering a similar design in human ES cells to facilitate their differentiation into pancreas-like cells. This will allow for scaled production of large quantities of pancreatic-like cells, sufficient for screening drug screening. One key advance in the proposal is the ability to detect the effect of a given drug specifically on insulin production in these cells using fluorescence monitoring. Because small changes in fluorescence can be easily detected, drugs can be optimized to increased potency and reduce unwanted side effects. Another key element in this ES cell design is the ability to easily screen for developmental genes and growth factors that can stimulate beta-islet cell growth and insulin expression. The institution maintains a proprietary set of candidate genes that will be evaluated in this fashion to identify novel therapeutic molecules for treating diabetes.
Statement of Benefit to California:
Diabetes and related diseases constitute a significant portion of healthcare expenditures in California. Direct and indirect costs are estimated to be approximately 16 billion USD a year based on population (http://www.cdc.gov/nccdphp/overview.htm). Furthermore, incidence in minority groups in California is disproportionately higher. Diabetes associated diseases such as stroke and cardiovascular disease are complicated and are often underestimated in cross-sectional assessments of State health.
Tragically, the percentage of adults diagnosed with diabetes continues to rise. In 2005 seven percent were diagnosed with diabetes, compared to 6.1% in 2001 (Diamant et al, 2007). To address this alarming health issue, the pharmaceutical industry needs more efficient and clinically-relevant (i.e. human-based) drug discovery and development tools focused on the biology of beta-islet cell growth and response, and insulin regulation.
The present proposal describes an engineered human cell line that will serve as a first-tier commercial drug discovery and screening platform that could yield many diabetes-relevant drug candidates. The proposal is by a California company that has many years of experience in commercializing human and mouse ES cell-based tools for drug discovery and development, and that has previously developed an analogous mouse insulin reporter ES cell line currently being used in commercial cell-based screening assays with a pharmaceutical partner, which is currently bringing revenue to California. Similar clinically relevant screening partnerships will be sought with the proposed hES line systems, bringing revenue and new jobs to California. In addition, commercialization of new diabetes drugs developed with the proposed hES cell line system will help relieve the burden on healthcare. Intellectual property gains for California are also potentially significant.
Publication of the results derived from the described research project will continue to support the advancement of California and human embryonic stem cell technology and stimulate new academic and commercial collaborations for recruiting and job growth.
The goal of this proposal is to characterize and engineer a human embryonic stem (hES) cell line that will efficiently differentiate into insulin-producing beta-islet cells. Such cells will be valuable tools for drug screening and the development of new therapies for diabetes. Three existing cell lines will be validated and characterized. These cells will be progressively engineered to promote more efficient differentiation into beta-islet cells and to monitor the differentiation process. A novel hES cell line appropriate for differentiation into pancreatic-like cells will be created and evaluated.
This proposal has a significant goal: the engineering of cell lines that efficiently differentiate into insulin-producing beta-islet cells. As diabetes mellitus is an important and increasing health concern, methods for production of pancreatic-like cells would be an important contribution to the development of new therapies.
Reviewers had serious reservations about the feasibility of the proposed study. The proposal suffers from a number of inconsistencies and does not present a clear or coherent experimental design for achieving the desired goal. Several aims lacked convincing rationale, and the logic of particular approaches was unclear. For instance, justification for the choice of the starting cell lines was not apparent. Furthermore, it is not clear that preliminary experiments presented for pancreatic differentiation of mouse ES cells are relevant or translatable to work with human ES cells. Many aspects of the proposal left reviewers with the impression that the applicant’s understanding of beta-islet cell biology was insufficient (e.g. reviewers questioned whether the proposed molecular manipulations would direct hES cells to a beta cell identity).
The application was viewed as not responsive to the RFA. The utility of any cell lines produced is unclear. Reviewers were not convinced that any new cell lines, if produced, would be made readily available to other researchers.
Reviewer One Comments
The goal of this application is to characterize 3 hES cell lines developed by this company but not yet characterized and then to use the best to generate a line that could be used for testing drugs and compounds for stimulating insulin expression. While such a reporter line may be useful for the pharmaceutical industry, the application is rather naive about how to get it and how useful it will be. The reason for using one of their own hES lines, as stated by the applicant, is purely commercial, and does not seem to include consideration of scientific criteria. The use of these lines also is somewhat redundant since they will first test their strategy in a line already partly engineered.
This application from Vistagen is to generate a new reporter hES line with both an insulin reporter and inducible expression of the transcription factors PDX1 and Ngn3 to facilitate “reliable and inducible differentiation of pancreas-like cells”. This line is to be used for screening drugs for boosting insulin expression but there is some confusion over whether this would be used to indicate replication/differentiation of new beta-islet cells or whether it would be stimulating transcription in already formed beta-islet cells.
The first of 4 aims is to characterize 3 hES cell lines generated by Vistagen but not yet characterized due to “lack of NIH funding” (which is a surprising reason for a company that may typically use sources other than NIH for funding). Then, the best of these lines will be used in aim 4 so that the company owns the rights to this platform to enable partnering with pharmaceutical companies, a goal stated by the applicant. Aims 2 and 3 are step-wise engineering of a cell line already partially engineered by Dr Gordon Keller’s group (who is Chair of the Scientific Advisory Board). Initially the floxed reporter gene driven by the ROSA 26 promoter will be replaced by a BAC of brachyury as proof of principle for maintenance of expression through differentiation of this recombineered ROSA 26 locus. The choice of brachyury is due more to Dr Keller’s interest than to the projects goal, and much of this work may be done in NYC in Keller’s lab. In Aim 3, a tetracycline-inducible reporter gene that is flanked by FRT sites will be added to the original cell line for testing of inducibility and ease of recombination through the FRT sites. This approach for making stable engineered hES cell lines seems promising from Dr. Keller’s publications; this is the strength of the project.
Finally in the last aim these techniques will be combined in the selected hES cell line from Aim 1 to produce cells with inducible PDX1-NGN3 and a sensitive insulin reporter. It is difficult to understand why they think the optimal goal is “not to produce a population of pure beta-islet cells; rather it is to closely mimic pancreatic tissue in vitro …in a simplified, practical and reproducible hES differentiation system”. Concerns here are that 1) the protocol they suggest they are using to induce this differentiation was developed in mouse; 2) purified beta-islet cells should be the goal or at least endocrine progenitors rather than pancreas-like cells for either the goal of generating beta cells or for testing drugs on insulin expression; 3) the synchronous expression of pdx1 and ngn3 may be misguided since usually ngn3 is transiently expressed on a background of continuous pdx1 expression; 4) the preliminary data on the expression of “pancreatic lineage genes” after induced PDX1 and ngn3 in mES cells. This last point perhaps illustrates the naiveté of the group on beta-islet cells. A number of the genes induced are not found in beta-islet cells (Nhlh1,2,Ebf2,3,LMO1) whereas they don’t test important beta-islet cell transcription factors such as Nkx6.1, MAfA. Additionally their comparison by microarray to BTC6 cells (with a considerably lower insulin mRNA) instead of islets suggests a greater success than might have been achieved.
In summary, the strategy to make this reporter hES cell line is intriguing and potentially useful, but there are many concerns about the project as presented.
Responsiveness to RFA:
Technically it is responsive to this RFA but seems a bit forced.
Reviewer Two Comments
The goal of this proposal is to characterize and engineer human embryonic stem cell lines that will efficiently differentiate into insulin producing beta-cells. As Diabetes Mellitus is an important and increasing health concern, methods for directing the differentiation of hES cells towards a beta-cell identity would be a significant achievement and likely aid in the development of new therapies.
This proposal suffers from a number of inconsistencies and does not present a clear and coherent experimental design for achieving the desired goal. First, it is entirely unclear why it is necessary to perform the first aim as there is nothing about the proposed evaluation of the three proposed hES cell lines that would make them a more optimal ‘base’ cell lines than any other currently available hES cell line for use in the following aims. The second aim of engineering the human ROSA26 locus to carry a green fluorescent protein reporter of the Brachyury gene is not properly put into the context of how this would help achieve the goal of directing the differentiation of hES cell lines to a beta-cell identity and appears to be a rather misdirected ‘proof of concept’ experiment. Finally, the last aim of creating hES cell lines that inducibly express NGN3 and PDX1, while directly related to the proposed goal, is fraught with problems and pitfalls. For instance, it is not clear that the over-expression of NGN3 and PDX1 would direct hES cells to a beta-cell identity. Thus the feasibility of this proposal is questionable.
Responsiveness to RFA:
This proposal is not responsive to the RFA.