Alzheimer’s Disease (AD) is a progressive incurable disease that robs people of their memory and ability to think and reason. It is emotionally, and sometimes financially devastating to families that must cope when a parent or spouse develops AD. Unfortunately, however, we currently lack an understanding of Alzheimer’s Disease (AD) that is sufficient to drive the development of a broad range of therapeutic strategies. Compared to diseases such as cancer or heart disease, which are treated with a variety of therapies, AD lacks even one major effective therapeutic approach. A key problem is that there is a paucity of predictive therapeutic hypotheses driving the development of new therapies. Thus, there is tremendous need to better understand the cellular basis of AD so that effective drug and other therapies can be developed. Several key clues come from rare familial forms of AD (FAD), which identify genes that can cause disease when mutant and which have led to the leading hypotheses for AD development. Recent work on Drosophila and mouse models of Alzheimer’s Disease (AD) has led to a new suggestion that early defects in the physical transport system that is responsible for long-distance movements of vital supplies and information in neurons causes neuronal dysfunction. The type of neuronal failure caused by failures of the transport systems is predicted to initiate an autocatalytic spiral of biochemical events terminating in the classic pathologies, i.e., plaques and tangles, and the cognitive losses characteristic of AD. The problem, however, is how to test this new model and the prevailing “amyloid cascade” model, or indeed any model of human disease developed from studies in animal models, in humans. It is well known that mouse models of AD do not fully recapitulate the human disease, perhaps in part because of human-specific differences that alter the details of the biochemistry and cell biology of human neurons. One powerful approach to this problem is to use human embryonic stem cells to generate human neuronal models of hereditary AD to test rigorously the various hypotheses. These cellular models will also become crucial reagents for finding and testing new drugs for the treatment of AD.
Alzheimer’s Disease (AD) is emotionally devastating to the families it afflicts as well as causing substantial financial burdens to individuals, to families, and to society as a whole. In California, the burden of Alzheimer’s Disease is substantial, so that progress in the development of therapeutics would make a significant financial impact in the state. Although there are not a great deal of data about the burden of AD in California specifically, the population of California is 12% of that of the United States and most information suggests that California has a “typical” American burden of this disease. For example, information from the Alzheimer’s Association (http://www.alz.org/alzheimers_disease_alzheimer_statistics.asp) reveals: 1) An estimated 4.5 million Americans have Alzheimer’s disease, which has more than doubled since 1980. This creates an estimated nationwide financial burden of direct and indirect annual costs of caring for individuals with AD of at least $100 billion. Thus, a reasonable estimate is that California has more than half a million AD patients with an estimated cost to California of $12 billion per year! 2) One in 10 individuals over 65 and nearly half of those over 85 are affected, which means that as our population ages, we will be facing a tidal wave of AD. Current estimates are that with current rates of growth that the AD patient population will double or triple in the next 4 decades. 3) The potential benefit of research such as that proposed in this grant application is that finding a treatment that could delay onset by five years could reduce the number of individuals with Alzheimer’s disease by nearly 50 percent after 50 years. This would be significant since a person with Alzheimer’s disease will live an average of eight years and as many as 20 years or more from the onset of symptoms. Finding better treatments will thus have significant financial benefits to California. 4) After diagnosis, people with Alzheimer’s disease survive about half as long as those of similar age without AD or other dementia. 5) In terms of financial impact on California families, the statistics (http://www.alz.org/alzheimers_disease_alzheimer_statistics.asp) are that more than 7 out of 10 people with Alzheimer’s disease live at home. Almost 75 percent of their care is provided by family and friends. The remainder is “paid’ care costing an average of $19,000 per year. Families pay almost all of that out of pocket. The average cost for nursing home care is $42,000 per year but can exceed $70,000 per year in some areas of the country. The average lifetime cost of care for an individual with Alzheimer’s is $174,000. Thus, any progress in developing better therapy for AD will have a substantial positive impact to California.
SYNOPSIS OF PROPOSAL:
This proposal targets Alzheimer's disease (AD), which remains an incurable, progressive and poorly understood disease. One of the reasons for this limited progress is that there are no good animal models of AD. The prevailing current hypothesis, although a number of competing hypotheses exist, is that AD is due to amyloid buildup.
The PI proposes to develop human neuronal models of AD derived from hESCs, and to use these cells to test the prevailing hypothesis of AD cause and progression. It is crucial to do these experiments in human ESCs rather than mouse ESCs, because mouse and human neurons have important biochemical and physiological differences.
IMPACT AND SIGNIFICANCE:
This is an interesting proposal - if the PI is successful it will have a high impact on both the field of stem cell research as well as on the fields that encompass Alzheimer’s disease research as a whole. It is clear that any gain of knowledge of the molecular basis of pathogenesis of AD represents a major endeavor, representing a top priority for the use of hESCs. The proposed experiments and the results anticipated from them provide a foundation for the rational design of cell-based therapies in the clinic.
This proposal will develop human neuronal models of AD derived from human ESCs and use such models to test prevailing hypotheses of AD cause and progression. Thus the proposal carries high clinical significance and may provide insights into new therapeutic treatments for AD and related diseases. The proposal is innovative, indeed it could serve as a paradigm for the use of hESCs in studies on the etiology, pathophysiology and treatment of human disease. At this point in time, animal models for AD do not replicate the disease as seen in humans. The work will extend the PI's extensive experience in the mouse model to the human and should provide crucial reagents required for finding and testing drugs and other therapies. The ability to observe human neurons in a mouse model will be an important addition to the field. Although this work could fall under the purview of Federal funding agencies, the PI has to date obtained unique preliminary results using the hES line, HUES-9 which is not approved.
It is an innovative approach and if successful will be useful for gaining a more basic understanding of what is happening in the brains of AD patients as well as a model for potential drug discovery/therapeutic approaches.
QUALITY OF THE RESEARCH PLAN:
This is a very well written grant proposal. The problem and the experiments have been nicely defined. The scientific and technical approaches reflect the talent of the investigator. For each individual experiment the outcome analysis is done to satisfaction. While it can be argued that the PI enjoys Howard Hughes and private funds, and thus did not need to wait for CIRM to come into existence to perform these experiments, his current entrance to the field to hESCs is of great value.
Human ESC lines will be created that carry genetic alterations that are known to cause hereditary AD using homologous recombination methods to introduce point mutations into the genes encoding amyloid precursor protein and presenilin 1. Cells from these lines will subsequently be differentiated to human neurons in vitro allowing evaluation of the differences from or similarities to wild type cells. The molecular approaches involved in previous work in the mouse by the PI will be extended to the human and carry high feasibility, however there is some risk in this aim such that the PI has indicated a 2 year committment to its completion. In other proposed experiments, methods to reduce gene expression will be applied in efforts to enhance the AD phenotype. Additionally, wild-type or knock-out cells will be introduced into the brains of either wild-type or FAD mutant mice. Cellular behavior will subsequently be measured and the effect of an AD environment evaluated. The research plan is logical, adequately described and based on prior experience in animal models or in hESCs and is likely to produce meaningful results. The timetable is appropriate.
It is difficult to predict whether the PI will have success, however given his overall track record and collaborations it appears that he has as good a shot as anyone to accomplish this. The grant is well written and interesting as it challenges some of the long-term basic hypotheses of AD research. There are some caveats - the levels of Aß that the PI is measuring are at or below current levels of detection, but given his experience and collaborations it is clear that he will keep up with advances in Aß detection methods and use them as they become available.
This work is going to be performed by a PI and group with the best track record in the world for APP studies and the molecular dissection of AD. Most of the data generated by this group has been in the mouse, acknowledged by the PI as not a very good model system. So it is more than urgent to bring these studies directly to their relevance in humans.
The PI and the team are highly qualified and experienced, and the clinical relevance of the project is a strength. The appropriate extension of the PI's experience with mouse ESCs to human ESCs is also a strength. These experiments have a high likelihood of producing significant results, and the experimental plan provides a paradigm for acquiring molecular level insights into the etiology, pathophysiology and potential treatment of human disease.
Major strengths include the novelty of the approach and of some of the hypotheses, and the likelihood that this will have significant clinical and basic science impact.
There are two main weaknesses of this proposal that have dampened the enthusiasm to rate this proposal at the highest level. First, targeted homologous recombination in hESCs has been and remains one of the most challenging technical hurdles to overcome. In fact, in the literature there is only a single example of such a targeted mutagenesis. For reasons that are not well understood, this has been the single most important barrier to overcome in the hESC field. The first part of the reseach plan proposes to introduce point mutations into endogenous APP and PS1 genes. The remainder of the research plan rests on the success of creating these mutations, so if this fails the whole grant collapses. The investigator himself points out is likely to be a difficult undertaking. However, as mentioned above, the PI has preliminary data and collaborative efforts that suggest that he well be the right person to undertake the challenge.
Second, those mutations are proposed to be done in hESC prior to their differentiation into neurons. If these two genes are involved in other functions (which they will certainly be) it is not clear that interference with these genes will not affect the differentiation process itself. The presenilins are known to be involved in the cleavage of at least two dozen substrates other than APP - it is not clear, nor is it discussed - whether changes in the cleavage of any of these might effect the differentiation of the hES. Moreover, highly efficient methods for generating neurons from hESCs are not yet available or are variable to the extent that heterogenous populations are the outcome. This may limit the usefulness of genetically modified human ESC lines.
Another concern is that in vivo experiments in the mouse may not provide useful or interpretable results. Mice carrying these mutations generally have not had strong developmental phenotypes, however it would be nice to know that the PI was aware of the possibility.
This is a well-researched, clear proposal where each aim is well-defined. It is important to do this work since there is no real valuable Alzheimer’s Disease (AD) model in mouse. There is truly a problem in mice generating a good AD model because human and mouse neurons have very basic differences. Two important hypotheses will be tested in this proposal – first, that amyloid-beta buildup is the causative agent in AD, and second, the novel hypothesis that axonal transport is involved in AD severity. Studying APP and Presend 1 is important since these are two of the genes implicated in amyloid build up. The effect of kinesin-1 needs to be tested directly in humans because animal models may not be relevant. The applicant proposes to test the contribution of these neurons on brain function in-vivo. There are considerable difficulties in performing homologous recombination (HR) in hESC ("an act of heroism" if it works), and the PI shouldn’t be penalized for taking the risk.
Another potential difficulty is that point mutations in genes involved in AD don’t exist yet, and these will be tough to generate.
The funding of the applicant was a point of discussion. One reviewer felt that the applicant could otherwise get federal funding for this work but chose to do preliminary experiments with non-approved hESC lines. Another reviewer added that HHMI funding might make this application less urgent than others.
There was a brief discussion about the significant risk of collapse of the research plan if the proposed homologous recombination were not successful.
A discussant asked about the prospects for acquisition of relevant samples from patients. While it may be possible to get samples from AD families to support the research, the sporadic incidence of disease would make it difficult to get samples from IVF families.