SYNOPSIS: The PI notes that while Oct4 and Nanog are two necessary transcriptional regulators of self-renewal programs, their expression per se is not sufficient for self-renewal. It is hypothesized that the mechanism by which OCT4 and NANOG contribute to the two diverse cell programs of self-renewal and differentiation is through alternative splicing, which functions to remove or alter protein domains. This in turn alters the activation or repression of downstream target genes, thereby altering cell fate. The PI predicts that by characterizing the function of each of the protein domains in OCT4 and NANOG, it will be possible to produce recombinant OCT4 and NANOG proteins that will maintain self-renewal and others that will induce differentiation. Thus the PI proposes that splice variants of one or more of these two proteins may be sufficient for self-renewal, or at least separating the activity of specific isoforms may clarify how they work. The PI therefore proposes a detailed study of two isoforms of Oct4 and two of Nanog, in human embryonic stem cells (hESCs), from mRNA expression analysis, ChIPs on specific gene targets, and cDNA overexpression of specific isoforms in hESC culture to studies of HIV-TAT peptide reagents for each. This is a very ambitious proposal.
IMPACT AND SIGNIFICANCE: The proposal aims to identify molecular characteristics of Oct4 and Nanog that are important for self renewal and use this information to generate a novel protein transduction system to stimulate hESC self renewal. Thus, the overall goal is to generate a technical advance in culturing methods that could provide a significant benefit to researchers by ameliorating problems associated with spontaneous differentiation of hESC. If successful, this portion of the proposal could enable the stem cell community to utilize hESC more effectively; however, it does not directly address either a basic biological question or a potential therapeutic application.
Identifying specific transcription factors which can trigger the expansion of hESCs is extraordinarily important, as it would presumably allow direct expansion of these cells without unregulated differentiation and without complex culture environments (all of which make it far beyond GLP). In addition, if self-renewal versus differentiation is regulated at the level of transcription factor isoform expression (or activity), this would be an very important biological insight, in its own right.
The examination of function of different regions of Oct4 and Nanog proteins in hESC could help elucidate basic molecular mechanisms by which these proteins promote self renewal. Although the downstream target genes and protein interaction binding partners have been characterized for Oct4 and Nanog, a structure-function analysis for these proteins is missing from the literature. Similarly, elucidating the activity of alternative Oct4 and Nanog isoforms is another area where this proposal could contribute to our understanding of the functions of these proteins in stem cells and during differentiation. While the proposal touches on aspects of these interesting biological questions, the scope of examination is limited, which significantly limits the potential impact.
QUALITY OF THE RESEARCH PLAN: The work plan is clear and well laid out, but is extremely ambitious in terms of scope. A large array of experimental approaches are described, each of which would require setting up and validating new assays over months to years. Given the enormity of what is proposed, a better rationale for presuming the hypothesis to have some validity would have been very helpful.
Specific comments on each aim:
(aim 1a) The expression analysis is a straightforward approach that will produce information, and determining whether Oct4b is nuclear in hESC is a valid approach considering cytoplasmic localization in embryos. A rationale for examining organelle-specific localization of proteins is missing. There is no cell-biological reason given to support the possibility that an organelle-specific localization of one of the gene products should be expected, nor is there an interpretation for why it could be an important observation.
(aim 1b) ChIP experiments will be used to test binding of Oct4 and Nanog to target genes. The antibodies used for these experiments were not identified, so it is unclear how isoform-specific chromatin binding will be determined. The approach in this subaim is also limited to examine only chromatin binding without also examining effects on transcription of target genes. As a result, it will not be known if a specific isoform actually affects target gene expression.
(aim 1c) The ability of forced expression of each isoform to promote self renewal in the absence of bFGF and/or MEFs will be tested. The idea behind these experiments is good; however, the level of expression of each gene is a potential concern in terms of interpretation of results. The level of Oct4 protein is important for determining if mouse ESCs undergo self renewal or differentiate toward trophectoderm or endoderm lineages. Without being able to control levels of expression, interpretation of results could be difficult.
(aim 2a) A mutagenesis approach can be very effective; however, providing a rationale for the choice of mutations would generate more enthusiasm. Negative regulatory element would need to be removed by mutation to make a Nanog fragment more transcriptionally active (as desired here); there is no reason to believe that such a mutant will be generated with this approach.
(aim 2b) With the exception of share concerns with aim 2c (below), this aim is straightforward and has potential for success only if the transducible proteins function as desired.
(aim 2c) If a rec-Nanog protein functions as desired in subaims a&b, and the TAT-fusion protein can be transduced, and it regains activity, a valuable reagent for stem cell researchers could be developed. It should be pointed out that it is dangerous to propose experiments for which reagents are far from being developed and rely on novel reagents.
Since Aim 3 is essentially a reiteration of Aim 2 with a different protein, the same overall critiques from Aim 2 apply to Aim 3.
STRENGTHS: The project begins to address a very interesting and important question concerning which regions of the Nanog protein are important for its ability to activate transcription and promote self renewal. Many of the approaches include the expertise of experienced collaborators, which will help the PI through several aspects of the proposed research. Nice preliminary data and the experience of this group demonstrate excellence in the culture and lentiviral transduction of hESC. The proposal identifies appropriate molecular methods to answer specific questions.
WEAKNESSES: The overall goal of the project is relatively risky in that it relies on the creation of a biologically active compound without significant preliminary data supporting a probability of ultimate success. The depth of examination of the structure-function relationship of Nanog and Oct4 is limited. Several flaws in the experimental plan also decrease overall enthusiasm.
The rationale for thinking that the two isoforms of each transcription factor differentially regulate hESC self-renewal is not provided. In the same vein, the motivation for undertaking deletion mutation analysis of the Oct4 protein to identify regions critical for its transactivating activity are not clear. There are likely multiple such regions, but how will this information tell us how Oct4 is necessary but not sufficient for hESC self-renewal?
The molecular assays of isoform activity--transactivation of reporters in heterologous cells--will likely not predict self-renewal per se. We already know this. Antibodies to the different isoforms may be in hand, but they are not described--so specific Western's, IPs, ChIPs, etc. may not be able to be performed. Additionally, siRNAs specific for each isoform have not been produced and validated.
HIV-TAT peptide approaches have been described, and claimed to be straightforward. They are not, and will take a tremendous amount of work, and good fortune, to work in any reasonable time frame, once a clear candidate is chosen.
The fact that there are many Nanog genes and pseudogenes was not discussed. What is the implication of these for studying the regulation of this transcript?
Much is known from the modulation of Oct4 expression in mouse ESCs. The likelihood, or not, of this being recapitulated in human was not discussed. For instance, TAT-Nanog makes sense based on mouse, but not sure that TAT-Oct4 does. Overexpressing Oct4 in mouse cells DOES NOT lead to self renewal but differentiation into primitive endoderm.
DISCUSSION: The proposal aims to study the role of Oct4 and Nanog in self-renewal of hESCs. Since some cells express both, but do not self-renew, the PI suggests that splicing variants might be responsible. This is an important problem not only for hESCs, but also applicable to adult stem cell self-renewal. The main issue with the proposal, according to one reviewer, is that this is a weak hypothesis and the work plan to address it is very ambitious. The PI does not have the reagents in hand and may not understand what they are or how to make them. Each specific aim proposed requires a huge amount of work. Another reviewer felt that there were two main problems. First, the goal of producing tagged Nanog would be beneficial, but there are many steps where the project might fail. The tagging is glossed over and there are not enough alternative methods proposed (e.g., what to do if the protein misfolds). Second, the structure-function studies are important and interesting, but here they are not being done logically because the isoform expression data doesn’t predict a difference in function. Also, there isn't any real rationale for the production of mutants for the structure-function studies and no new informative data would emerge. A third reviewer felt that the Nanog fusion protein would be a good resource since Nanog overexpression in mouse cells is sufficient to maintain pluripotency, and unpublished data also suggests that mouse Tat-Nanog gives similar results as overexpressing the Nanog cDNA. The structure-function approach is standard fare. Even though many of the reagents appear to not be in hand, the preliminary data argues that they have made them in the past. The reviewer felt that while the Nanog experiments are appropriate, the Oct4 experiments may not be because modifying Oct4 in the mouse, even subtley, doesn’t have the same effect as Nanog; thus this proposal is a little naïve. There was a question about whether the differing functions for different isoforms has any support, and it was pointed out that it hasn’t been looked into very well. Nanog in the mouse is more difficult because there are many pseudogenes.