Identification of microenvironment factors that regulate hematopoietic stem and progenitor cells.

University of Minnesota Ph.D. dissertation. January 2009. Major: Molecular, Cellular, Developmental Biology and Genetics. Advisor:Catherine M. Verfaillie. 1 computer file (PDF); xiii, 153 pages. Ill. (some col.)The hematopoietic stem cell (HSC) maintains hematopoiesis throughout life. HSC are used to treat a number of hematopoietic disorders including leukemia and other hematopoietic malignancies, as well as immunodeficiencies. The first successful bone marrow (BM) transplantation was performed over 30 years ago. Even so, we still have an incomplete understanding of what regulates different aspects of HSC behavior. To repopulate the hematopoietic system, physicians initially used bone marrow (BM) as a cell source, and more recently granulocyte colony stimulating factor (G-CSF) mobilized peripheral blood (PB) cells, and umbilical cord blood (UCB). Although HSC transplantation can cure a significant proportion of patients, limitations still remain. Histocompatibility complex leukocyte antigen (HLA) matching is imperative to reduce the risk of graft versus host disease, such that a suitable donor cannot be found for many patients. Near complete matching is required when adult sources of cells are used, whereas, some degree of HLA-mismatch is allowed when UCB grafts are used, even though a low incidence of severe graft versus host disease (GVHD) persists following UCB transplantation. One of the problems related to the use of UCB as cell source is the limited number of HSC per graft, such that one UCB unit is insufficient to treat adult patients. Therefore, many investigators have examined whether HSC can be expanded ex vivo, as this would circumvent the problem associated with the limited number of HSC present in UCB. However, robust methods for ex vivo maintenance, let alone HSC expansion, have not been developed. Understanding the mechanisms that regulate proliferation, self-renewal, and differentiation of HSC may yield means of HSC expansion in vitro, as well as, allow for a single cord blood graft to be used for an adult recipient, and aid in attempts for genetic modification of autologous HSC for therapy of genetic hematopoietic disorders. Theoretically, the potential to expand HSC in vitro could provide a limitless supply of HSC for therapy. In vivo, HSC receive signals from the local microenvironment in which they reside that regulate self-renewal vs. differentiation. In adults, HSC reside in the BM, in putative stem cell niches. The hypothesis of the HSC niche was first proposed in 1978 by Schofield, who proposed that cell contact is required between HSC and cells in the niche for HSC maintenance. Over the last 5 years, two different locations have been identified within the adult BM that may serve as the BM niche: the endosteal niche and the vascular niche. Even less is known regarding putative stem cell niches during development. During development hematopoiesis is initiated in the yolk sac and the aorta-gonad-mesenephros (AGM) region. Around embryonic day (E) 12 in mouse, HSC migrate to the fetal liver (FL) where extensive expansion of the HSC pool occurs and HSC differentiate to the lineage committed progenitors. HSC migrate from the liver to the BM just before birth. As each of these microenvironments plays a unique role in development of hematopoiesis, it is hypothesized that they may possess unique signals that govern different aspects of HSC development, self-renewal and differentiation. However, the nature of these signals is still poorly understood. The aim of this thesis was to identify candidate factors produced by the HSC niche that regulate HSC cell maintenance (and expansion). In Chapter 4 of this thesis I hypothesized that at least some factors produced in HSC niches responsible for HSC self-renewal would be soluble. A number of stromal cell lines have been generated from the AGM region, fetal liver and BM that support HSC in vitro. I used three stromal cell lines that preserve human and / or mouse HSC cultured in direct contact with the feeders, namely UG26-1B6, derived from the AGM region of E10.5 embryos; EL08-1D2, derived from fetal liver of E10.5 embryos; and AFT024, derived from fetal liver of E14.5 embryos. To determine if soluble factors secreted by these feeders could support HSC, I cultured HSC-enriched murine BM cells either in direct contact with or separated by a transwell from the feeder for 3 weeks, and tested the ability of the ex vivo maintained cells to competitively repopulate the hematopoietic system of lethally irradiated mice. I demonstrated that only UG26-1B6 cells support competitive repopulating HSC cultured separated from the feeder, and that UG26-1B6 cells must secrete one or more HSC-supportive factors that are not secreted by the EL08-1D2 feeder. In Chapter 5, I describe studies in which I compared the transcriptome of UG26-1B6 and EL08-1D2 cells, to identify candidate factors secreted by UG26-1B6 but not EL08-1D2 that support HSC maintenance/expansion in vitro. Eighteen candidate-secreted factors were identified. I describe initial studies to test if 3 of these 18 factors, Galectin-3, Tfpi, and SerpinE2, can support HSC ex vivo. In Chapter 6, I evaluated the effect of a fourth factor expressed significantly higher in UG26-1B6 than EL08-1D2 cells, Wnt5a, on HSC support in ex vivo cultures. I describe studies to determine if (1) addition of Wnt5a to the EL08-1D2-based cultures allows maintenance of HSC placed in transwells above the feeders and if addition of anti-Wnt5a antibodies to the UG26-1B6-based cultures prevented HSC persistence; (2) if Wnt5a can also be used in the absence of any stromal support to maintain HSC; (3) and to evaluate the mechanisms via which Wnt5a affects HSC. In Chapter 7 of the thesis, I evaluated the effect of bone morphogenic protein (BMP) antagonists on HSC maintenance / expansion. These studies were not strictly derived from the transcriptome studies described in Chapter 5, but followed from other studies in the lab demonstrating that BMPs play a key role in mesoderm specification and initiation of hematopoiesis during zebrafish development. I describe studies (1) wherein the effect of loss of two BMP antagonists, Twisted gastrulation (Twsg1) and Chordin (Chrd), on adult hematopoiesis is evaluated, (2) and studies wherein the effect of addition of Twsg1 and Chrd to ex vivo feeder-free cultures on the repopulation ability of HSC is being tested