SYNOPSIS: The applicant proposes to produce and characterize hematopoietic stem cells (HSC) that are derived from human embryonic stem cells (hESC). Their long-term goal is to use these cells instead of bone marrow, peripheral blood, and cord blood-derived stem cells for the treatment of patients with blood disorders including cancer. In this way they hope to get around the problem of limited donor availability, especially for patients from minority populations. Their focus will be on techniques to scale up production of HSC and also to characterize such stem cells, especially with regard to the expression of human leukocyte antigens (HLA) that may influence the engraftment capabilities of these cells. The characteristics of hematopoietic reconstitution will be studied in a mouse xenograft model involving NOD/SCID mice. Finally, they propose to identify new cell lines with hematopoietic progenitor characteristics using a system that isolates cell lines from embryoid bodies.
IMPACT AND SIGNIFICANCE: This grant is aimed at developing methods to derive hemangioblasts (common precursors for blood and endothelial cells) from hESC. These cells could be used as cell replacement vectors for multiple disorders of the blood and vascular systems. The hope is that such cells would be infinitely expandable without differentiating in vitro, and therefore they would provide a reliable and inexhaustible supply of HSC. The proposal is innovative in its use of a novel panel of hES-derived "progenitor" cell lines that will be provided by the applicant's collaborator, Advanced Cell Technologies; however, details about these progenitor cells are rather sparse in the application, so the true benefit to use of these cells is difficult to assess. If successful, the primary impact of this work would be an improvement in HSC reconstitution for hematopoietic-related diseases in humans by production of transplantable HSCs derived from hESCs. It certainly has the potential to ease the usual issues of searching for match donors for bone marrow transplantation. The applicants' ultimate aim is to advance work on hES-derived hematopoiesis to a point where its use in clinical hematopoietic cell transplantation is within reach. Insights into the directed differentiation of hES along the hematopoietic/angiogenic lineage could also provide insights into directing the differentiation of hES into other mesodermal and non-mesodermal lineages.
QUALITY OF THE RESEARCH PLAN: The research plan is logical and relatively straightforward, but overly vague in its description of experimental strategies. The applicant proposes four aims: 1) to define minimal cell culture parameters for derivation of hemangioblasts from hES, 2) to characterize the cell surface antigens expressed by these cells, 3) to test their engraftment in NOD/SCID mice, and 4) to determine ways to maintain them in vitro indefinitely in an undifferentiated state. These are very ambitious goals, and the strategy for achieving them is not clearly described.
Much of the work will be done at the Advanced Cell Technology (ACT) Company, which is located both in Massachusetts and in Alameda. The company is experienced in cell culture and already has a library established of 200 clonal and stable differentiated cell lines from embryoid bodies derived from both NIH-registered and proprietary hES cell lines. They propose to use an iterative matrix-type experimental approach of cell culture with the eventual aim of carrying out bulk processing and reducing the complexity of the cytokine cocktails to the minimum necessary. It seems reasonable to assume that this all will be done at a high level of expertise, but the details of how exactly the culture conditions will be optimized or what the readouts will be are not clearly laid out.
The second specific aim appears straightforward in evaluating expression and function of histocompatibility and other cell surface antigens on hemangioblasts and HSC. Some of this work has already been done. For the histocompatibility antigens, they will focus on HLA antigens and analyze expression as well as functional studies on the cells' allostimulatory capacity in MLC assays. The applicant does not clearly discuss how cell surface molecules will be identified, or how they will be tested for enrichment of hemangioblasts (i.e., is there a functional assay that will be used?). Also missing is an approach to evaluate minor antigens such as the hematopoietic antigens.
The third specific aim will characterize in vivo hematopoietic differentiation potential in a xenograft model involving NOD/SCID immunodeficient mice, which, in addition, will be given sublethal total body irradiation. The TBI doses of 0.2-0.3 Gy seem low. Preliminary data is lacking for these in vivo studies, and the details of how many cells will be injected, what effects exogenous cytokines may have, and what alternative approaches may be pursued are lacking. This is particularly concerning in that the applicant indicates that the hES-derived cells have "yolk-sac" properties, and YS HSC do not engraft adult animals. Also, the applicant seems to assume that establishment of an immortalized, undifferentiated hemangioblast cell line will be straightforward, but details of how this will be accomplished and analyzed are lacking.
The fourth specific aim seeks to produce hemangioblasts or stem cells from clonal and stable cell lineages that have previously been derived from hES cells. These cell lines will be evaluated for their capacity for further induction to definitive hemangioblasts or hematopoietic cell lineages using an iterative process similar to that described in Specific Aim 1. The applicants believe that this approach eventually will result in the production of cells for clinical studies under relatively less complex and costly conditions than cells directly derived from embryoid bodies.
STRENGHTS: The research plan is direct and clear, and focused around a significant health problem (blood disease). It begins with the generation of HSCs from hESCs and finishes with studies of transplantation in an immunodeficient animal model. The strength of the proposal resides largely in the collaborations with Dr. Martin, who has considerable experience in the molecular aspects of hematopoietic cell differentiation, and ACT, who appears to have reduced the complexity of cell line derivation from hES cells by utilizing the unexpected clonogenicity of hES-derived embryonic lineages. More than 200 clones have been generated from colonies with symmetric morphology. Gene expression studies have shown that a subset of these cell lines have significant expression of genes potentially associated with hemangioblasts and early erythroblast development as compared to hES cells and other clonal hES cell lines. ACT has efficiently and reproducibly generated in vitro equivalents of hemangioblasts from hES cells using an in vitro differentiation system. Blast cells from these cultures were shown to clonally differentiate into erythroid, myeloid, and multilineage hematopoietic cell lineages. The use of these novel cell lines can be considered both a strength, and a weakness, of the proposal.
WEAKNESSES: The research plan is exclusively pragmatic with a vague experimental design and little emphasis on hypothesis-driven research of a basic nature. In general, the applicants underestimate the difficulties that will arise when hES cell derived hemangioblasts are being used for transplantation into unrelated individuals. The NOD/SCID mouse model has shown very poor engraftment of hES-cell derived hematopoietic stem cells. It is not clear whether this is due to problems inherent in the stem cells or minimal residual NK-cell activity in the murine hosts. Even if the applicants could overcome this basic problem, the mice will not provide the answer as to the durability of grafts accomplished with hES cell derived hemangioblasts. First, the observation period will be short, and second (and most importantly), mice are small compared to humans, and the demands on hematopoietic stem cells are more than three orders of magnitude less than in a human being. For example, a mouse makes as many red cells in a lifetime as a human in one day, a dog in 2.5 days, and a cat in 8 days. Thus, observation periods beyond the life of a mouse would be required to make reasonable predictions for the fate of such cells in humans. Moreover, the explicit assumption that hESC-derived HSC will be immunologically naive and therefore will not be rejected by human recipients is not supported by existing data. It is safe to assume that multipotent hematopoietic stem cells express both HLA and non-HLA minor antigens (the latter are also called hematopoietic antigens), and therewith elicit immune reactions on the part of the transplant recipients, which will lead to rejections. The current research plan, while attempting to assess for the presence of HLA antigens on these early cells, will neither address the issue of hematopoietic antigens nor will it or can it investigate the issue of cellular immune responses to such antigens, which are likely to involve both host NK cells and T cells. In fact, no studies are planned to use immunologically intact animals to test these assertions, and the selection of hESCs with defined HLA antigens to finesse the issue of allorejection and GvHD is not supported by experimental data.
Another major weakness is the Principal Investigator whose CV indicates no laboratory experience. The PI has not published on hES, and his last publication was in 2003, where he was co-author on a multicenter, predominantly European study on related umbilical cord blood transplantation in patients with thalassemia and sickle cell disease. The second to last publication was in 2001 in the journal Pediatric Pathology and Molecular Medicine on sibling donor cord blood banking.
Despite the accomplishments of ACT, an overarching concern in this proposal is the use of the ACT progenitor cells, which appear to have been characterized thus far predominantly by virtue of their global gene expression profiles. The proposal lacks clear preliminary data to indicate the functional capacity of the ACT-generated hES-derived progenitor lines; thus, there is no evidence that these cells provide a significantly better option than HSC or hES-derived HSC. How homogeneous or heterogeneous are these cells in terms of their phenotype/function? Preliminary data provided in this regard was too small to comprehend. If heterogeneity is a potential problem, then the use of conditioned supernatants from these cells is concerning in that they are very poorly characterized and effects will likely reflect combinatorial action of multiple secreted factors.
DISCUSSION: The aim of this work is to differentiate hESC into HSC, with the long-term goal of making HIV-resistant cells. The application has more weaknesses than strengths. The PI is mainly a clinician who has no real track record in the field (the last publication was in 2003, and in collaboration), and much of this work will be done in collaboration with ACT. Two main areas of concerns are 1) the NOD/SCID mouse experiments, which many not work as proposed because the issue of allogenicity is ignored, and 2) the poor characterization of the ACT lines, where the general consensus is that the researchers don't know what they are working with.
Should future funding opportunities arise, reviewers suggest that more information about the cells to be tested would be very useful in evaluating the potential impact of these studies. Preliminary data reflective of the feasibility of in vivo models also would be helpful. In order to answer the question of whether hemangioblasts generated by the work in this proposal can be used clinically, the applicants would need to develop, in parallel, a large animal model with a known track record of translation into the clinic.