Funding opportunities

Derivation of Inhibitory Nerve Cells from Human Embryonic Stem Cells

Funding Type: 
Comprehensive Grant
Grant Number: 
Principle Investigator: 
Funds requested: 
$2 507 223
Funding Recommendations: 
Grant approved: 
Public Abstract: 
Statement of Benefit to California: 
Review Summary: 
SYNOPSIS: This proposal will look at generating large numbers of GABAergic neuronal progenitor cells from human embryonic stem cells (hESCs) and from adult neurogenic sources. The aims are to generate the cells with a medial ganglionic eminence (MGE) phenotype, using proper culture and growth factor conditions, and to transplant them into specific regions of the brain in neonatal and adult mice, and compare them with those transplanted into rodent models of transient lobe epilepsy (TLE - using kainite lesions) and Parkinson's Disease (PD using 6-OHDA lesions). IMPACT & SIGNIFICANCE: New protocols are needed to generate many populations of neurons and glia needed for replacement and protection protocols for neurological disease. In this proposal, hESCs will be primed to make large numbers of “pure” GABAergic neuronal precursor cells for applications in epilepsy (to increase GABAergic inhibitory tone in a rodent TLE model) and Parkinson’s Disease (to introduce GABAergic inhibitory neurons into striatal targets that have lost dopaminergic-induced GABAergic striatal efferent tone). The clinical significance is clear. This proposal will address a novel cell-based therapeutic approach towards the treatment of PD and epilepsy from the generation of inhibitory interneurons (GABAergic neurons) from hESCs. No such therapy currently exists and there is reasonably good rationale that the generation of inhibitory neurons could have potential clinical utility. Inhibitory neuron grafts are proposed to suppress regions that are abnormally active in PD similar to the approaches currently in clinical use based on STN stimulators. A similar approach has also been pursued in a controversial first clinical gene therapy trial for PD where a GAD expression vector was introduced into the STN area. Preliminary data are based on primary mouse MGE cells and show some improvement of symptoms in a specific PD model. Preliminary data are also provided to demonstrate integration of primary mouse MGE cells in adult cortex and hippocampus causing an overall increase in IPSCs. This study may also provide general insights into the molecular mechanisms guiding the development of GABAergic neurons in general. This would be far different then current therapies of drugs which modules these neurons, or surgical aspiration of brain tissue, an approach used in intractable epilepsy. The additional novelty in this proposal is that the investigators will attempt to develop methodologies for better generation of GABA interneurons since many ESCs in cultures head towards a glial fate. QUALITY OF THE RESEARCH PLAN: This is a very logical, well-designed research plan. Four specific aims are proposed, using different in vitro and in vivo bioassays to develop stable human MGE progenitor cell lines and establish their efficacy in rodent transplant studies. The aims of this proposal are to generate a series of stable human neural stem (NS) cell lines from hESCs that will start in year one and be completed by year two. This is a reasonable timetable for this reagent-generating aim. Specific Aim II will be the production of MGE cells from human NS cells that will begin in year one/two and continue until completion, anticipated to be year two/three. The human NS lines from Professor Smith appear to be in place to help meet the milestones of this aim. Naturally they will be phenotyping these cells with appropriate neuronal and glial markers. They have logically planned out a series of external manipulations with sonic hedgehog and FGF, and other appropriate factors to treat cells attempting to differentiate them to an MGE-like progenitor cell line. Alternative approaches including the effects of WNTs and neuregulins as well as EGF, FGF, and sonic Hedgehog. Of course this aim is the most critical aim in that success of these steps are critical for the generation and ultimate plan of using these inhibitory neuron stem cells in their planned therapeutic endeavors. In the last two aims,they will take these cells to in vivo preparations first to graft them into the brains of rodents, then to follow their differentiation and engrafting characteristics including synapse formation. Specific Aim III will be the transplantation of human MGE progenitor cells to the cortex and striatum. And, lastly they hope to use these successfully generated cells in actual models of human disease including a mouse model of temporal lobe epilepsy as well as a standard rat model of PD using 6-OHDA. Specific Aim IV is the exploration of the therapeutic potential of human MGE progenitor grafts in animal models of disease and should be completed by the end of year four. The overall plan is reasonable though the proposal suffers from a significant lack of preliminary data on hESC differentiation. No evidence is provided on the generation of hESC-derived GABAergic neurons, or the induction of foxg1b or Nkx2.1 expressing cells. The proposal is based on the hypothesis that stable hNS cell lines derived from hESC cells will retain the potential to differentiate into all the various GABAergic interneuron subtypes without providing preliminary evidence. MGE interneurons fates are thought to be determined early in development within a foxg1b+ domain. There was no clear plan to test for foxg1b expression and maintenance of this key marker over time. However, it is assumed that the outstanding team of investigators, including Dr. Rubenstein an expert in MGE & interneuron development, should assure that such studies are ultimately included. Strategies for the genetic modification of interneuron differentiation are interesting and well designed. STRENGTHS: Strengths are many including 1) logical flow-plan, 2) necessary proof of principle that transplantation of MGE cells is possible in adult mice and rats with the effective phenotypic differentiation markers 3) the ability to derive human and neuro stem cells from hESCs. The key strength of the proposal is the outstanding expertise of investigators. Although there seems some less experience regarding hESC studies, the participation of Dr. Smith mitigates this concern though his role in the studies is not very well defined. This is a great team of some of the best stem cell researchers in the world coming together bringing complementary skills and insights; they should definitely be able to achieve the goals of this proposal. UCSF hESCs will be used to generate MGE cells, in addition to a human NS lines that has already been generated from hESCs (provided by Prof. Austin Smith in Cambridge). This is a strength of this proposal because it meets the criteria of the RFA but will also afford comparison of new hESC-derived MGE-like GABA cells with an already proven adult NS line. An interesting set of approaches has been proposed for increasing interneuron yield via genetic means (overexpression of key transcription factors). The project is very interesting – there is a clear need for a better understanding of interneuron development. There is also a need for the development of interneuron-based cell transplantation approaches particularly in epilepsy. WEAKNESSES: Some of the proposed follow-up experiments are a bit open-ended. For example, the PI states that hESC-derived MGE cells may not survive in the host environment, and if immunorejection is detected that they will "empirically test neurotrophic factors (e.g. BDNF, NT3, NGF, or GDNF) that could support survival." This may prove to be more problematic than anticipated/discussed since cells may not integrate due to other non-biotrophic-related issues. Also ,when it is mentioned that if the mouse host environment proves restrictive for survival or integration of human cells, they will substitute macaque monkeys for the transplant experiments; there is no real reason to assume that monkeys would be better than SCID-rodents. The PI also states, “…Grafted cells may develop into teratomas, if they are not sufficiently differentiated or if they are contaminated with undifferentiated hESCs. If so, in vitro differentiation of MGE cells will be extended and modified to obtain a more homogeneous population. Alternatively, FACS will be used to further enrich the population of cells expressing MGE markers. It is also possible that the timing of differentiation is longer for human MGE cells compared to rodent cells. Longer survival periods may be required to reach full differentiation of hESC-MGE cells. …” It should be assumed that any one undifferentiated hESC that is grafted has the potential to form a teratoma; better isolation or enrichment or differentiation in vitro can never assure a “homogeneous population”. Assuming the investigators will see teratomas, then what? Finally, recent studies by others have shown that you can facilitate production of MGE-like GABAergic neuron precursors by exposing ESCs to particular matrix molecule substrates at particular times in vitro. This might be worth thinking about. The NKX2.1/Isl-1 population is certainly amenable to expansion as this other study has shown, and thinking about such growth conditions, it might be possible to avoid the generation of non-GABAergic phenotypes. It’s not at all clear that the proposed series of transcriptional factors and growth factors will reliably convert large populations of the human NS cells into these inhibitory interneuronal phenotypes. Other labs, such as the Jessell lab have spent many, many years understanding the requisite transcription factors and employing them in a series of necessary steps to create specific neuro populations such as motor neurons (and the Macklis lab – in a similar way for upper motor neurons). Therefore, it is not at all clear that over a four-year period the laboratory will be able to effectively develop the critical transcription factors and growth factor steps necessary to direct differentiation of inhibitory interneurons, and this is the most critical step in this proposal. Other specific concerns raised include: 1. Significant lack of preliminary data for hESC work. 2. No discussion of critical patterning windows for MGE interneuron development, no discussion even for the need to optimize forebrain yield as determined by foxg1b or Six3 expression. 3. No discussion of the various GABAergic neuronal subtypes that need to be distinguished and generated selectively. No plan for purification of interneuron progeny in these studies. 4. The applicants suggest that immunosuppression may not be necessary at all in their study. However, there is overwhelming data that grafting into adult hosts (as proposed here for several experiments) will require immunosuppression while grafting up to early postnatal stages may not. 5. One reviewer had concerns about the practical applicability of this approach in the treatment of PD. It will be very difficult to fine tune the system perfectly for clinical use in PD and to achieve results better than with current treatment such as STN stimulator. As proposed, the grafts will be significantly contaminated with non-GABAergic cells and inappropriate GABAergic cells. Furthermore the extensive migration of GABAergic neurons all over the brain may be a disadvantage in a disease such as PD and could cause unexpected side effects in other brain regions. 6. There is no published feasibility of ES-derived MGE-type interneurons (beyond Nks2.1 expression) even from mouse ESCs. Aims III & IV are completely dependent on Aim I and a backup strategy using mouse ESCs may be appropriate. 7. There are some concerns with the PD preliminary data given the very small numbers of apomorphine rotations/min. Typically such scores are observed in the case of incomplete lesioning. Administrative Issues - No justification use for the Lhx6GFP mice to be ordered - Funds for equipment seem to be larger than the maximum amounts listed in the instructions DISCUSSION: The aim of this proposal is to create a GABAergic inhibitory tone by making huge numbers of GABAergic neurons (the default for hESC)and then to use them in disease models for PD and other areas of the basal ganglia. The interneurons will also be used in the hippocampus and cortex for epilepsy. The PI will use Austin Smith's adult neural cells as a basis for comparison, where the default for adult cells is also GABAergic (99% of the time). The PI will use macaques instead of SCID rodents in these studies. The PI is an expert in interneurons. The approach is analogous to Tom Jessell's work with motor neurons, and reviewers thought that this is the right way to carry out this work. The problem is that there are many types of GABAergic neurons, and the PI assumes that any GABAergic neuron will do; here, they will need MGE-specific GABAergic neurons. In the Jessell work, the investigators focused on and found the specific type of motor neuron needed. In this case, however it may be OK since we don't really know if the inappropriate GABAergic neurons wouldn't be functional. There is almost zero preliminary hESC data as the PI only shows primary cell data. The PI makes a strange statement about not needing immunosuppression because stem cells show no rejection. Using GABAergic neurons for PD may not be the best model because they'll just graft differentiated neurons, and there is no reason to think that the grafted cells will go specifically to sites where they can function to treat the symptoms. In fact, in the case of PD, cells may go to cortex and make matters worse. The PI assumes but never shows that forebrain neurons are being generated since no forebrain markers are used in the studies (PF1, SG3?). This is naive from this PI who should know more about interneurons. One reviewer agreed that lineage diversity is an issue, but with the mouse, Nkx2.1 islet cells can be used to generate GABAergic cells, and hopefully these would go to the forebrain. The proposal would be much stronger with relevant preliminary data from hESC-derived hNS cells (GABA, Foxg1b, Nkx2.1, Lhx6…). Plan should take into consideration the possibility that the fate potential of hNS cells will change over time in culture. A better discussion on how to generate various specific GABAergic subtypes and how to isolate / purify the cells would be an improvement. One of the reviewers believes that overall this is a good proposal, but it could have been better. There is quite a bit of money included for equipment

© 2013 California Institute for Regenerative Medicine