Executive Summary The goal of the proposed research is to generate human embryonic stem cell (hESC) lines from multi-pronuclei zygotes, aberrant products of in vitro fertilization that are normally discarded. Following microsurgery to remove the extra pronucleus, cells will be cultured to blastocysts from which hESCs will be derived. Resulting normal, androgenetic, and parthenogenetic hESC lines will be characterized and differentiated into neural lineages. Subsequent analysis will examine the effect of paternal and maternal imprinting on neural cell differentiation. This application proposes an innovative and interesting source of starting cells for the derivation of hESCs. However, reviewers were unenthusiastic about the rationale and justification for the approach or its perceived usefulness for the generation of stem cell lines. In particular, the required microsurgery is technically cumbersome and the use of abnormal zygotes needlessly complicates the derivation process. Although reviewers noted the value of the experiments in dissecting paternal and maternal contributions to neural cell differentiation, the approach was viewed as unlikely to yield embryonic stem cell lines that would be generally acceptable for clinical applications, a stated objective of the proposed work. The proposed research program is entirely feasible. The PI has the appropriate experience with human stem cells and neural differentiation and has had adequate productivity. A named collaborator is a world expert in stem cell biology. Reviewers were concerned that the investigations designed to assess the role of genetic imprinting on neural development were too limited in scope and lacked adequate experimental detail. The proposal was responsive to the RFA, but (as noted above) reviewers were unconvinced that the proposed study and methods developed would add significantly to generation of clinically useful cell lines. It is unclear that the proposed research would meaningfully advance the field toward translational applications. Reviewer Synopsis This proposal aims at the generation of normal and abnormal hESC lines (carrying mutations for genomic imprinting disorders) from multi-pronuclei (3PN) zygotes normally discarded by IVF. The extra pronucleus is proposed to be removed by microsurgery, and subsequently cultured in vitro into blastocyst embryos for hESC derivation. The “corrected” 2PN zygotes provide an opportunity to create hESC lines that are androgenetic or parthenogenetic, and can therefore be used for study of human genetic disorders such as Angelman and Prader-Willi Syndrome. Reviewer One Comments Significance: This proposal has three specific aims. The first is to correct 3PN zygotes into 2PN zygotes by micro-manipulation and culture corrected 2PN zygotes into blastocyst embryos for hESC derivation. The second specific aim is to derive and characterize normal, androgenetic and parthenogenetic hESC lines. Finally, the third aim is to direct in vitro differentiation of these hESCs into neural lineages (neural precursor/stem cells, neurons and glial cells) and determine the effect of paternal and maternal imprinting defects on neural cell differentiation. While innovative, it is hard to see why the investigators plan to make an already daunting task of derivation even more difficult by using abnormal zygotes, purely for nonscientific reasons. While interesting, the justification for the study of paternal and maternal genetic disorders seems largely post facto. Feasibility: The PI, Dr. Guoping Fan, has a good track record of publications. The research plan includes collaboration with Dr. Nissim Benvenisty and Dr. David Hills at Cedars-Sinai Medical Center. The facilities at UCLA are adequate for the goals of this proposal. Responsiveness to RFA: This proposal, if successful will produce hESC lines that can differentiate into all three germ layers. The possibility of translating this to the clinic is questionable, however, given that abnormal zygotes were initially used. This proposal has adequate plans to share any generated hESC lines by making them available to researchers and depositing them in the stem cell bank at UCLA (headed by Dr. Jerry Zack) and the International Stem Cell Institute at Cedars-Sinai. Reviewer Two Comments Significance: This is a nice idea, but it is not clear to me what the new cell lines will add. The derivation of human embryonic stem cells for the great majority of translational and clinical applications will, I think, need to be done from normal zygotes. I don’t think cells from which one pro-nucleus has been removed would be acceptable for clinical applications, even if repeated in vivo tests failed to reveal any abnormalities of proliferation or differentiation. The scientific question behind maternal or paternal genetic contributions is fascinating, and here the cells will be very valuable. However, this is not a novel idea to explore this question, and the need for large numbers of new cell lines for this specific application seems less pressing. My argument here is that the questions surrounding the basic science do not need to consider the issue of natural variation between different human cell lines to the same extent as those where more translational applications are being considered, hence fewer lines are needed. Feasibility: The research program described is entirely feasible. The Principal Investigator has a very strong CV and the track record of the co-investigators is good. All the necessary facilities and collaborations are in place. On the science, the assessment of pluripotency seems adequate. On the neural differentiation, I would have liked to have seen more detail on the methods to be used to promote and direct differentiation and the assays of function. In particular, more sophisticated studies, eg using transplantation, would have been valuable. Consideration of how the cells might be used to study human diseases of imprinting is also required. I wouldn’t disagree with the comment in the last sentence that the knowledge gained in this study will be valuable for addressing the role of genomic imprinting, but some detailed discussion as to exactly how cells in which expression of the entire maternal or paternal genetic contribution might have differences as well as similarities with diseases in which only one or a few genes are affected is needed. Responsiveness to RFA: The project will yield pluripotent human embryonic stem cells as requested, but see my comments above as to their utility. Reviewer Three Comments Significance: The investigators propose using three pronuclei (3PN) embryos, which are typically discarded, as an alternate source of embryos from which to derive hES cell lines. Through a microsurgical procedure to remove one pronucleus, a heteroparental, parthenogenetic, or androgenetic blastocyst can be generated. 3PN embryos represent 5-15% of all IVF embryos, therefore representing a significant source of embryos for deriving new lines of hESCs cells. I don’t agree with the investigators that these embryos represents a significant source of clinical grade hES cell lines. Though this is an innovative method, it is also cumbersome, in that it requires expertise in embryo manipulation. However, both parthenogenetic and androgenetic hES cell lines are of research interest and this is addressed by Aim 3: To direct in vitro differentiation of these hESCs into neural lineages determine the effect of paternal and maternal imprinting on neural cell differentiation. Feasibility: The experience of the investigators and the facilities seem adequate. The methodology for reducing 3PN embryos to 2PN embryos, derivation and characterization of the lines are well described. The impact of maternal vs. paternal imprinting defects on neural (and other types of) development is interesting, but the experiment outlined are too limited in scope to adequately address this question. Responsiveness to RFA: Though pluripotent human stem cell lines will likely be derived by this method, I don’t think this method adds significantly to the goal of deriving lines that will have future clinical use or contribute to the field as is intended under this RFA. Though the imprinting ‘defects’ of the generated lines have utility for research, it is more basic science in nature, than clinical/translational.