UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional heterogeneity of adult mouse bone marrow hematopoietic stem cells Dykstra, Bradford John 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-267073.pdf [ 14.04MB ]
Metadata
JSON: 831-1.0100570.json
JSON-LD: 831-1.0100570-ld.json
RDF/XML (Pretty): 831-1.0100570-rdf.xml
RDF/JSON: 831-1.0100570-rdf.json
Turtle: 831-1.0100570-turtle.txt
N-Triples: 831-1.0100570-rdf-ntriples.txt
Original Record: 831-1.0100570-source.json
Full Text
831-1.0100570-fulltext.txt
Citation
831-1.0100570.ris

Full Text

FUNCTIONAL HETEROGENEITY OF ADULT MOUSE BONE MARROW HEMATOPOIETIC STEM CELLS by BRADFORD JOHN DYKSTRA B.Sc (Hons), Trinity Western University, 2000 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Genetics) THE UNIVERSITY OF BRITISH COLUMBIA December 2006 © Bradford John Dykstra, 2006 Abstract The mammalian blood-forming system sustains physiologically required levels of mature blood cells by supporting their continuous generation from a rare population of undifferentiated, self-sustaining pluripotent hematopoietic "stem" cells (HSCs). Throughout adult life HSCs are located primarily in the bone marrow. Traditionally, the study of HSCs within larger populations of cells has hampered the direct observation of any unique differentiation or self-renewal properties that might distinguish individual members of the HSC compartment. To circumvent this, I analyzed the number and types of progeny generated from single purified HSCs both in cultures initiated with a single cell and in irradiated mice injected with a single cell. In a first set of experiments of this type, I demonstrated that two growth factor cocktails with the same mitogenic and anti-apoptotic activity on HSCs in vitro could have remarkably disparate effects on their concomitant self-renewal behaviour, even within the span of a single cell cycle. In addition, I used high-resolution video monitoring of single purified HSCs cultured in microwell arrays to identify cellular features that were associated with HSC self-renewal in vitro. These parameters include longer cell-cycle times than those of their differentiating progeny and an absence of uropodia on the majority of cells within the clone during the final 12 hours of culture. When combined, these parameters improved by a factor of 2-3-fold the identification of clones found to contain daughter HSCs with longterm in vivo reconstituting ability. Finally, from longitudinal and serial WBC analyses performed on a large number of recipients of single purified HSCs, I found that the adult HSC compartment could be resolved into 4 HSC subtypes, 2 of which stably ii and autonomously propagate their initial unique patterns of WBC reconstitution through many self-renewal divisions in vivo. I also found that, in vitro, HSCs could rapidly acquire less competitive in vivo reconstitution programs although remarkable symmetry was retained in the reconstitution programs acquired by the daughter HSCs generated in the first 4 days in vitro. These findings provide evidence of intrinsically determined heterogeneity in the differentiation and self-renewal properties of individual HSCs. 111 Table of Contents Abstract ii Table of Contents.... iv List of Tables vi List of Figures vii List of Abbreviations ix Acknowledgements xi Dedication xii Co-Authorship Statement xii Chapter 1 - Introduction 1 1.1 - Hematopoiesis and Hematopoietic Stem Cells 1 1.1.1 - Evolution of HSC concept, detection, and quantification 2 1.1.2 - HSC purification 5 1.1.2.1 - Limitations of phenotypic identifiers of HSCs 6 1.1.3 - Classical view of HSC hierarchy in adult mice 7 1.1.4 - in vivo assays for HSCs 8 1.1.4.1 - Limitations and other considerations of in vivo HSC assays 12 1.2 - Regulation of Hematopoiesis 15 1.2.1 - Survival and Apoptosis 15 1.2.2 - Proliferation and Quiescence 17 1.2.3 - Lineage Restriction 18 1.2.4 - Regulation of HSC self-renewal and differentiation 20 1.2.4.1 - Intrinsic Regulators 21 1.2.4.2 - Extrinsic Regulators - HSC niche 25 1.2.4.3 - Studies of extrinsic control of HSC self-renewal in vivo 28 1.3 - Heterogeneity of Hematopoietic Stem Cells 32 1.4 - Thesis Objectives 37 1.5 - References 43 i v Chapter 2 - Different in vivo Repopulating Activities of Purified Hematopoietic Stem Cells Before and After Being Stimulated to Divide in Vitro with the Same Kinetics 62 2.1 - Introduction 63 2.2-Results 64 2.3 - Discussion 71 2.4 - Materials and Methods 77 2.5 - References 91 Chapter 3 - High Resolution Video Monitoring of Hematopoietic Stem Cells Cultured in Single Cell Arrays Reveals New Features of Self-Renewal 97 3.1 - Introduction 98 3.2 - Results 99 3.3 - Discussion 102 3.4 - Materials and Methods 105 3.5 - References 120 Chapter 4 - Longterm Symmetric Propagation in vivo of Functionally Distinct Subtypes of Hematopoietic Stem Cells 124 4.1 - Introduction 125 4.2-Results 127 4.3 - Discussion 133 4.4 - Materials and Methods 138 4.5 - References 154 Chapter 5 - Discussion and Future Directions 159 5.1 - Major Contributions 159 5.2 - Implications and Future Directions 163 5.2.1 - Basis of Functional Heterogeneity in HSCs 163 5.2.2 - Basis of HSC Self-Renewal 167 5.2.3 - Intrinsic Determination of Lineage Output Patterns by Individual HSCs 171 5.3 - Concluding Comments ..175 5.4 - References 178 v List of Tables Table 1.1 - Definitions of long-term multilineage donor repopulation in various studies 40 Table 2.1 - Frequency and type of repopulation in irradiated recipients of single lin~Rho~SP B M cells or their immediate progeny 83 Table 2.2 - Single lin~T<ho~SP cells can generate sufficient daughter HSCs to repopulate secondary and tertiary recipients 84 Table 2.3 - Frequency of HSCs in the Rho* and Rho+ subsets of lin" SP adult mouse B M cells 85 Table 2.4 - LTC-IC assay of single lin"Rho"SP cells 86 Table 3.1 - Application of selection criteria developed from the "training set" of data to results from two additional experiments I l l Table 3.2 - Candidate biomarkers considered 112 vi List of Figures Figure 1.1 - Hierarchical model of hematopoiesis 41 Figure 1.2 - Functional detection of HSC by transplantation in vivo 42 Figure 2.1 - FACS profiles of the lin" Rho subsets of freshly isolated mouse B M SP cells and their first- and third-generation clonal progeny produced in serially transplanted mice 87 Figure 2.2 - Repopulation levels in mice transplanted with single lin Rho_SP B M cells and doubles derived from them in vitro 88 Figure 2.3 - Long-term repopulation kinetics of single cell-transplanted mice 89 Figure 2.4 - Cell division kinetics of linTlho~SP B M cells subcultured in two different cytokine cocktails 90 Figure 3.1 - In vivo repopulation characteristics of sing le CD45midlin"Rho"SP cells or their clonal progeny 113 Figure 3.2 - Description of high-resolution time-lapse array systems and representative culture results 114 Figure 3.3 - HSC activity is associated with smaller clone sizes and longer cell-cycle times 115 Figure 3.4 - Use of behavioural parameters defined by cell tracking to predict HSC-containing clones 116 Figure 3.5 - Purification of CD45midlin"Rho"SP cells 117 Figure 3.6 - Sample calculation of the donor-derived leukocyte level in the PB of a transplanted mouse 118 Figure 3.7 - Calculation of time to a third division exclusion parameter 119 Figure 4.1 - Schematic representation of the overall experimental design 140 Figure 4.2 - WBC outputs in recipients of single LTRCs or their clonal progeny generated in vitro 141 Figure 4.3 - Identification of LTRC subtypes in ternary plots of their lineage-specific contributions at 16 weeks post-transplant 143 vn Figure 4.4 - Clonal propagation of repopulation patterns in secondary and tertiary recipients 145 Figure 4.5 - Intra-clonal comparisons of LTRC progeny 146 Figure 4.6 - Rapid alteration of LTRC distributions in vitro 147 Figure 4.7 - Schematic diagram of the relationships between LTRC subtypes and the mature WBC types they generate 148 Figure 4.8 - Representative examples of mice repopulated with each LTRC subtype illustrating the gating strategies and relevant calculations 149 Figure 4.9 - Ternary plots of the lineage-specific contributions show the same LTRC subtypes in fresh and cultured cells but in different proportions 151 Figure 4.10 - Differential association of secondary LTRC activity with different measures of primary WBC output 153 Figure 5.1 - Hypothesized relationships between LTRC subtypes 177 viii L i s t of Abbreviations 2 - M E 2-betamercaptoethanol A P C Allophycocyanine A T R A All-trans retinoic acid B6 C57B1/6J B M Bone marrow C D K Cyclin-dependent kinase C F C Colony forming cell C F U - S Colony forming unit-spleen C I Confidence interval C L P Common lymphoid progenitor C M P Common myeloid progenitor C R U Competitive repopulating unit F A C S Fluorescence activated cell sorting F G F Fibroblast growth factor F I T C Fluorescein isothiocyanate F L Fetal liver Flt3L Fms-like tyrosine kinase 3 ligand G - C S F Granulocyte-colony stimulating factor G M Granulocytes/monocytes G M - C S F Granulocyte/macrophage-colony stimulating factor G P I Glucose phosphate isomerase H B S S Hank's balanced salt solution H F HBSS plus 2% fetal bovine serum H o Hoechst 33342 H S C Hematopoietic stem cell I G F Insulin-like growth factor I L Interleukin I M D M Iscove's modified Dulbecco's medium K S L cKit+Scal+Lin" IX L D L Low density lipoproteins L I F Leukemia inhibitory factor L i n Lineage markers L T C - I C Long-term culture-initiating cell L T R C Long-term reconstituting cell M - C S F Macrophage-colony stimulating factor M H C Major histocompatibility complex M P P Multipotent progenitor P c G Polycomb group P E Phycoerythrin P R C 1 Polycomb repression complex 1 R A R Retinoic acid receptor R B C Red blood cell R h o Rhodamine 123 S A Streptavidin S F Steel factor SP Side population S T R C Short-term reconstituting cell T C R T-cell receptor TGF-/3 Transforming growth factor beta T N F Tumour necrosis factor T P O Thrombopoietin W41 C57B1/6J W 4 1 / W 4 1 W B C White blood cell X Acknowledgements First and foremost, to Connie: For your foresight in agreeing to take on an enthusiastic but completely clueless greenhorn from a no-name university with no funding. For your infectious love and appreciation of good science For giving me space to do my own thing For just the right mix of encouragement and ass-kicking For making me a priority while dealing with a to-do list a mile long For placing trust and thereby instilling confidence To Rob, Keith, and Kelly for your constructive ideas and positive energy To those with whom I worked in the trenches - Jody, Nao, Oliver, Silvia, Maura, John R., Dave, and Kai - thanks for your attitudes of cooperation and selflessness. Special thanks to Dave, who happily and regularly went above the call of duty for the sake of science To the talented students that worked alongside me over the years - Kristin, Melisa, Merete, Lindsay, and Shannon - thanks for your hard work and positive attitudes and I wish you every success in your futures. To the Timmy's crowd and the Martini's crowd, and all the other good friends I've made over the years - thanks for helping make this a fun and memorable experience. Thanks to all members of the TFL, past and present. You have made it a special place. To my mentors in earlier stages of life - Wendall Ewald, John Krisinger, and Eve Stringham - your love of science impacted me much more than what you taught in the classroom. To my parents, who believed in me and encouraged me at every step along the way.. And to Mollie, for doing everything in her power to get me through this, including working to pay the bills, assisting with late-night experiments, helping out with the thesis preparation, and putting up with many a late and stressful night. XI To my wife Mollie, whose unwavering encouragement and support made this all possible xu Co-Authorship Statement Chapter 2: This article was co-first-authored with Dr. Naoyuki Uchida, a post-doctoral fellow who I worked with in Dr. Connie Eaves' lab during the years 2000 to 2003. Dr. Uchida identified the original purification strategy and designed the limiting dilution, single cell, and serial transplantation repopulation experiments and the single cell LTC-IC assays. I developed the single cell culture and cell cycle kinetic measurement strategies, discovered the reduced frequency of sex-mismatched sublethally irradiated recipients and designed the doublet injection experiments. The LTC-IC experiments were performed exclusively by Dr. Uchida. The single cell cultures, cell cycle kinetic measurements, and doublet transplantations were performed exclusively by me. All other aspects of the research were performed jointly. The LTC-IC analysis, along with the majority of the peripheral blood chimerism analysis was performed by Dr. Uchida, with technical assistance from K. Lyons and F. Leung. I analyzed all of the single cell culture data and all of the doublet transplant data. Dr. Uchida wrote an initial rough draft of the manuscript and Connie Eaves and I were responsible for the remainder of the manuscript preparation. Chapter 3: This article was co-first-authored with John Ramunas, a senior research technician from the lab of Dr. Eric Jervis, Department of Chemical Engineering, University of Waterloo, Waterloo, ON, Canada. We worked together on this project between 2004 and 2006. Mr. Ramunas, with assistance from his colleague April Blaycock, designed and constructed the time-lapse camera system, the silicon microwell array culture system, the single cell manipulation apparatus, and programmed the algorithms for tracking individual cells within the cultures. Together with David Kent, I improved the efficiency of the Lin-Rho-SP sort with the addition of CD45 and designed the in vivo assays. Collectively we developed a live-cell shipping system. xiii David Kent and I performed the initial cell harvest and purification in Vancouver, plus injected single cell transplants as controls. Purified cells were then sent to M r . Ramunas in Waterloo and imaged for 4 days, after which individual clones were harvested and sent back to Vancouver. Together with David Kent, I then transplanted the clones into individual recipients and with the assistance of Lindsay McCaffrey, subsequently collected and analyzed their peripheral blood at regular intervals for donor reconstitution levels. The analyses of the time-lapse images were performed by M r . Ramunas, Er in Szumsky, L iam Kel ly , and Kristen Farn. Lindsay McCaffrey and I performed the peripheral blood analysis of all recipients. The remainder of the analysis was performed jointly. I generated most of the figures and tables, and M r . Ramunas, David Kent, Connie Eaves, Eric Jervis, and I were involved in all other aspects of manuscript preparation. Chapter 4: The work presented in this chapter was the culmination of many experiments and much data collection, distillation, and analysis over a period of several years, from 2002 until 2006.1 designed the experiments, with intellectual input from David Kent. Transplants of single cells and in vitro clones were performed by David Kent and me. Assistance with peripheral blood sampling, staining, and F A C S analysis, was provided by Melisa Hamilton, Lindsay McCaffrey, and Krist in Lyons. Michelle Bowie performed the single cell fetal liver H S C transplants, and Lindsay McCaffrey performed the subsequent P B chimerism analysis. I performed the in-depth data analysis and generated the figures. The manuscript was primarily prepared by me, with significant input from Connie Eaves and David Kent. xiv Chapter 1 Introduction 1.1. Hematopoiesis and hematopoietic stem cells (HSCs) . The bloodstream of an average adult mouse contains more than 10 billion blood cells. The majority of these are enucleated erythrocytes. The other specialized "end" cells of the blood system include T and B lymphocytes, natural killer cells, dendritic cells, neutrophilic, basophilic, and eosinophilic granulocytes, monocytes (which mature into macrophages), mast cells and megakaryocytes which fragment into platelets. Many of these are critical for survival of the organism. The lifespan of these various types of mature blood cells ranges from a few hours or days (neutrophils) to a few months (erythrocytes) to many years (memory T-cells). As a result, -100 million mature blood cells need to be produced each day throughout the entire lifetime of the mouse. The process by which these cells are produced is called hematopoiesis. Hematopoiesis is sustained by a rare population of cells called hematopoietic stem cells (HSCs). HSCs have the ability to produce all types of mature blood cells (i.e., they are pluripotent) but also generate daughter cells with the same pluripotentiality through many cell divisions (i.e. they can self-renew). However, it now appears that the irreversible loss of pluripotency is a highly complex process that may span many cell divisions and be accompanied by reproducible phenotypic changes. This has led to the concept of HSCs with indefinite and intrinsically limited self-renewal activity and a forced reliance on a combination of phenotypic markers and functional (progeny output) 1 endpoints to distinguish pluripotent cells with lifelong blood cell producing ability. Operationally, HSCs are assessed by transplanting them into irradiated recipients and revealing their ability to produce detectable numbers of lymphoid and myeloid progeny for at least 4 months in the recipients. Because this is an experimental description of a conceptually defined cell, it is important to note that the cells identified as HSCs in such transplantation assays may vary depending on the exact endpoint chosen to infer their initial presence. For example, these may involve differences in when and how donor-derived blood cell production is measured, the level and/or types of differentiated cells required to infer origin from HSCs, and whether or not limiting dilution strategies are employed. Although there is clearly irrefutable evidence from retroviral marking studies1'2 and single cell transplants3'4 that pluripotent hematopoietic cells with extensive self-renewal activity exist, the widespread use of different criteria for HSC identification in functional assays has clouded their detailed characterization (see section 1.1.4 and Table 1.1 below for further details). 1.1.1 Evolut ion of H S C concept, detection, and quantification In the early 1950's, it was realized that extracts from the bone marrow or spleen could protect mice from radiation-induced death5. However, in the following years, it was hotly debated whether this protective effect was humoral or cellular in origin. The use of transplants of cytologically distinct donor cells allowed this issue to be resolved in the mid 1950's and established that the bone marrow contained cells able to repopulate the 2 hematopoietic system of irradiated mice6. 30 years later, it was finally proven through clonal marking experiments that the lifelong reconstitution of the hematopoietic system in transplanted irradiated mice was attributable to the activity of a small number of long term repopulating cells (LTRCs) capable of self-renewal and multi-lineage differentiation1'7'8. Not surprisingly, functional assays for quantifying primitive hematopoietic cells and methods for their differential purification were developed and modified in parallel with conceptual advances in understanding how their properties changed during their early differentiation. Functional readouts of the various primitive cell types were originally necessary because all shared an identical, "blast" or undifferentiated morphology. Even as strategies were developed to prospectively identify the various primitive hematopoietic cell types, the functional aspect of these assays remained particularly important since the phenotypic identifiers were fortuitously discovered rather than representing known elements of the stem cell status of HSCs and subsequent studies showed that most of those used are also unstably tied to HSC status (discussed in more detail in section 1.1.2.1). The spleen colony assay, originally described by Till and McCulloch 9, was the first quantitative functional assay described for primitive hematopoietic cells and provided a unique tool for testing and refining many of the basic concepts of stem cell biology, including stem cell self-renewal, multipotentiality, commitment and differentiation. Till and McCulloch found that when irradiated mice were transplanted with small numbers of bone marrow cells, discrete macroscopic nodules formed in the spleens of the recipients and the number of these detected was linearly related to the 3 number of bone marrow cells injected. Histological examination of these nodules revealed that they were composed of proliferating undifferentiated cells as well as differentiated cells of the granulocytes/macrophages and erythroid lineages. Because of the linear relationship between cell dose and nodule count, it was hypothesized that the nodules represented the clonal progeny of rare cells present in the original inoculums. The clonal nature of these nodules was later established using cytogenetic markers10. Subsequent experiments demonstrated that the cell of origin, which they termed a colony-forming unit-spleen (CFU-S) could generate all myeloid lineages11 and shared a common origin with a cell that also had lymphoid potential12. Initial evidence of heterogeneity amongst cells identified as CFU-S came from experiments showing that CFU-S with differing self-renewing capacities could be physically separated13. Later, it was discovered that CFU-S could be subdivided into different classes according to the speed with which they produced detectable spleen colonies and their durability once formed. Colonies that appeared between 7 and 9 days post-transplant were transient14, predominantly composed of one lineage15, and had limited self-renewal activity1 4 ; 1 6. Conversely, colonies that were present on day 14 were not visible within 7-9 days, became larger by day 14, contained cells of multiple lineages15, as well as significant numbers of secondary CFU-S 1 6 . Although CFU-S were equated with HSCs for almost twenty years, the concept of a "pre-CFU-S" cell was hypothesized17'18 many years before convincing experimental evidence of a distinct population of this type was obtained19;20. This came from the use of Rhodamine 123 staining and counterflow centrifugual elutriation, respectively, to separate CFU-S from long-term repopulating cells that were able to subsequently 4 generate daughter CFU-S. Assays for the qualitative and quantitative assessment of LTRCs are described in detail in section 1.1.4. 1.1.2 H S C purification Around the same time, a bone marrow subpopulation including primitive hematopoietic cells was prospectively isolated using flow cytometric techniques by selecting cells that lacked any of a panel of cell surface markers expressed on differentiated hematopoeietic cells (so-called lineage or lin markers) but at the same time 21 expressed the newly christened "stem cell antigen - 1", or Sca-1 . This strategy was subsequently refined by the additional selection of cells expressing c-kit22. Nevertheless, both the Lin-Sca-1+ and Lin-Sca-c-kit+ (KSL) populations were shown to contain both LTRCs and CFU-S, suggesting that the populations were heterogeneous. Later experiments showed that the KSL population could be further fractionated into 3 23 functionally distinct subpopulations, using cell surface markers such as CD34 , CD27 , flk2/flt324;25, endoglin/CD105 2 6 , S L A M family receptors CD150 and CD244 2 7, EPCR/CD201 2 8 , and a-2 integrin/CD49b29, or by differences in staining with the vital dyes, Rhodamine-123 3 0 or Hoechst 3334231;32. Cells with the ability to efflux Hoechst 33342 are often visualized using two emission wavelengths, giving rise to the characteristic side population or SP phenotype33, which we have combined with Rhodamine efflux activity19'34 in a sorting strategy not reliant on either Sca-1 or c-kit35. 5 Using these latter approaches, cells with long-term, multilineage reconstituting ability can now be routinely isolated at purities of >20% by a variety of strategies. 1.1.2.1 Limita t ions of phenotypic identifiers of H S C s : In spite of the identification of phenotypes of adult mouse bone marrow cells that are almost exclusive for HSCs, this relationship is known to break down under a variety of circumstances. Thus the use of these phenotypes as surrogate indicators of HSCs can lead to false conclusions because non-HSCs expressing the same marker profile are mistaken for HSCs and HSCs with a different marker profile are missed. For example, murine mast cells and their precursors lack classical hematopoietic lineage markers and co-express c-kit and Sca-1, making them phenotypically indistinguishable from KSL cells, and hence are commonly assumed to be HSCs when HSCs are not, in fact, present36. Many stem cell markers are expressed differently in quiescent (steady-state) HSCs versus activated HSCs, including those present in developing hematopoietic tissues (fetal liver), or mobilized in adults by cytokine administration or after myelosuppressive treatments with drugs or radiation, or when exposed to growth factors in culture. 37 Examples of markers that show such lability in their expression on HSCs include CD34 , CD38 3 8 , Hoechst and Rhodamine efflux activity39, endoglin/CD10540, and Mac-1 4 1. However, recent reports suggest the existence of some markers that may be stable in both steady-state and activated HSCs, including S L A M family markers42, endomucin43, and EPCR (David Kent, personal communication). Recently, two FACS-based strategies have 6 been described that allow the isolation of fetal liver HSCs at >10% purity ' . Perhaps by using these strategies as a starting point, further experiments will result in the development of a purification strategy that robustly identifies activated HSCs. 1.1.3 Classical view of the hematopoietic hierarchy in adult mice Based on the collective efforts of the past 40+ years, a hierarchical model of hematopoiesis has been proposed (Figure 1.1). HSCs are placed at the top of this hierarchy and besides their extensive self-renewal ability, are thought to give rise to all mature blood cell types. These are continually produced through a series of successive differentiation and amplification steps. The immediate progeny of HSCs are multipotent short-term reconstituting cells (STRCs, also called multipotent progenitor cells, MPPs) that retain full lineage potential yet have a relatively limited capacity for sustaining self-renewal divisions. Since these primitive cells are incapable of prolonged self-renewal, they actively produce cells for only a few weeks or months. These STRCs, in turn, give rise to oligopotent progenitors, which are more restricted in their developmental potential. This is viewed as a major branching point in the hematopoietic hierarchy with the common lymphoid progenitor (CLP) 4 6 giving rise only to mature lymphoid cells and a common myeloid progenitor (CMP) 4 7 capable of giving rise only to mature myeloid cells (see figure 1.1). The oligopotent progenitors in turn give rise to more lineage-restricted progenitors from which all of the mature blood cells eventually arise. 7 Although some unanswered questions remain and exceptions to this described hierarchy exist, most agree that the sequential differentiation of HSCs through progenitors to fully differentiated blood cells generally occurs in this fashion and is generally an irreversible process. Because each differentiation choice is regulated, extensive proliferation (or dampening) can occur at every step, with rates of proliferation increasing in more differentiated progenitors48. This permits a huge potential amplification in the numbers of differentiated progeny that can ultimately arise from a single HSC, but also allows the fine-tuning of the system to respond to ever-changing demands for particular cell types. The hierarchical strategy also demands little proliferation from HSCs themselves, which may explain why they cycle so infrequently and are primarily quiescent49. Avoiding the hazards of constant activation, including the risk of mutations during DNA replication, as well as the mutagenic by-products of active metabolism, allows these cells to contribute to hematopoiesis throughout life, without exhausting or becoming neoplastic. 1.1.4 in vivo A s says for H S C s Investigations of the hierarchical model just described have been greatly facilitated by the introduction of in vivo repopulation assays that use prolonged outputs of multiple mature blood cell types as endpoints to allow input HSCs with putative lifelong self-sustaining ability to be specifically discriminated (Figure 1.2). The functional aspect of these assays remains particularly important since most phenotypic identifiers can be 8 altered under certain conditions (see section 1.1.2.1). Primitive cells are tested via injection into genetically distinct myeloablated (usually irradiated) recipients, along with some source of radioprotective cells. This is essential to ensure short-term survival, since purified test cells may require several weeks to produce functional progeny20. Several months later, the differentiated hematopoietic progeny are analyzed to determine the presence or absence of myeloid and lymphoid cells generated from the test cells originally injected. Over the years, different strategies have been used to distinguish donor chimerism in the recipient mice including detection of sex mismatches, and differences in isoenzyme and cell surface alloantigen expression. When co-injected with measured numbers of cells with predefined hematopoietic activity, the relative competitive ability of the test cells can be measured based on the proportion of donor cells in their hematopoietic tissues, usually the blood. This allows a semi-quantitative measurement of HSC activity, and mathematical descriptions relating competitive repopulation ability with input HSC have been proposed50"52. However, measurements of competitive repopulation ability of populations of cells cannot distinguish between variations in LTRC number and variations in competitive ability per individual LTRC. This important information can only be obtained by injecting single cells, or by utilizing the statistical power of limiting dilution, using the competitive repopulating unit (CRU) assay. This latter assay involves transplantation of varying donor test cell doses into groups of myeloablated recipient mice, along with radioprotective competitor cells. If test cell-derived lymphoid and myeloid progeny are detected, it can be concluded retrospectively that the test cells contained at least one pluripotent long-term repopulating cell. The proportion of negative 9 animals at each test cell dose is then used to calculate the frequency of CRU in the test cell sample using Poisson statistics and the method of maximum likelihood. The C R U assay as originally described53 utilized female competitor cells whose HSC content had been reduced by two serial transplantation cycles, test cells from male donors, and lethally-irradiated female recipients. Donor-derived male myeloid and lymphoid reconstitution was approximated by evaluating B M and spleen cells, respectively, using Southern blot analysis. Refinements of the C R U assay have since been developed. For example, more sensitive PCR-based techniques to detect smaller numbers of donor male cells in a female recipient have been described54. Alternatively, erythroid repopulation can be monitored using congenic mouse strains bearing allelic differences in hemoglobin or glucose phosphate isomerase (GPI) genes30'55. Most popular is the use of C57BL/6J (B6) congenic strains that express the CD45.1 or CD45.2 allotypes of CD45 (also known as leukocyte common antigen or Ly5), which is found on the cell surface of all hematopoietic cells except terminally differentiating erythrocytes56" . Use of such mouse strains allows detection by flow cytometry of donor-derived nucleated blood cells of various types using combinations of fluorochrome-conjugated lineage-specific and anti-CD45.1 or anti-CD45.2 monoclonal antibodies. Preparation of serially-transplanted cells is costly and labour-intensive and has largely been replaced by the use of 1 to 2xl0 5 normal B M cells or lx lO 6 Seal-depleted cells to function as helper cells to ensure short-term hematopoietic recovery of lethally-irradiated recipients and long-term hematopoietic reconstitution even when the grafts do not contain detectable CRU. C57BL/6J-Kiti¥~41J (W41) mice, a B6 histocompatable strain with a partial loss of c-kit function through the acquisition of a point mutation59, provides a practical 10 alternative to the use of lethal-irradiation and helper cells. The C R U content of these mice is reduced by about 17-fold60 and sub-lethal conditioning of 400 cGy y-irradiation is sufficient to further compromise endogenous hematopoiesis52'61'62. W41 mice can be partially myeloablated and used as recipients without the need for helper cells, and provide an equivalent competitive environment for individual HSC detection as conventional assays in lethally irradiated mice. The measurement of long-term multilineage donor reconstitution ability is key to identifying HSC activity in the in vivo transplant setting. However, exactly how "long-term multilineage reconstitution" is defined has, over the years, been interpreted in many different ways by different groups. As mentioned, the C R U assay as originally described utilized Southern blots to assess the contribution of male donor cells to the hematopoietic tissues of female recipients. In these experiments, a minimum of 5% male cells in both the B M and spleen at 5 or 10 weeks post-transplant indicated that one or more CRU was present in the test cells. In subsequent years, flow cytometric based techniques have become the norm. In particular, the advent of the benchtop flow cytometer combined with the utilization of antibodies generated against the alloantigens CD45.1 (Ly5.1) and CD45.2 (Ly5.2) has permitted convenient and effective distinction of donor-derived and host-derived WBCs with the ability to co-stain with antibodies against lineage-specific markers to demonstrate pluripotency of the test cells. Table 1.1 summarizes the criteria used to define long-term HSC activity in various studies and demonstrates the evolution of this definition over time. 11 1.1.4.1 Limitat ions and other considerations of in vivo H S C assays: H S C s have two def in ing characteristics; mult ipotent ia l i ty and self-renewal. H o w e v e r , there are addit ional requirements that must be met before cel ls c a n read out i n the in vivo assays just described. F o r example, the abi l i ty to h o m e and engraft is an integral part o f the def in i t ion , since without these abil i t ies they w i l l not read out. A n interesting recent ly descr ibed example is the engraftment defect o f H S C s i n the S / G 2 / M phases o f c e l l cyc le . Because o f this engraftment defect, the cel ls w i l l not be detected i n transplantation assays. H o w e v e r , i f these same cel ls are cultured for a short t ime such that 63 they re-enter a next G l phase, they regain an abi l i ty to engraft and be detected as H S C s . S i m i l a r l y , i f recipients are treated w i t h S D F - 1 antagonists, S / G 2 / M cel ls read out as H S C s . T h i s begs the question: d u r i n g the t ime that these cel ls are unable to engraft, does this m e a n they are t e m p o r a r i l y no longer H S C s ? Str ic t ly d e f i n i n g H S C act iv i ty us ing the in vivo assay suggests that these cel ls temporar i ly cease to be so. A n interesting var iat ion o n this theme has been proposed b y Quesenberry, suggesting the existence o f a reversible c o n t i n u u m o f hematopoiet ic states that v a r y w i t h c e l l cyc le p r o g r e s s i o n 6 4 . H i s t o r y shows that care must be taken w h e n interpreting results f r o m in vivo assays. These assays are, b y def in i t ion, s i m p l y surrogate methods to a l l o w the funct ional ident i f icat ion o f a conceptual ly determined ce l l . It should be kept i n m i n d that the assay as def ined might not overlap perfect ly w i t h the c e l l ' s actual properties, or that the concepts used to define that c e l l might be f lawed. F o r example, the spleen c o l o n y assay was thought for m a n y years to be the def init ive assay for H S C s , since C F U - S satisfied the then-current conceptual requirements o f H S C s . Furthermore, the funct ional difference 1 2 between the primitive cells that gave rise to spleen colonies 7-9 days after transplantation and those that did so after 12-14 days was not initially appreciated, and so no distinction between these two cell types was made. As a result, experiments were designed and results were interpreted on the basis of flawed conclusions. Similar issues exist for other in vivo assays. One of the original endpoints used in transplantation assays was survival. While it is true that HSCs are ultimately responsible for long-term survival in lethally irradiated recipients, less primitive progenitor types are responsible for radioprotection at early stages. Therefore, for example, if a small number of highly purified HSCs were injected into lethally irradiated mice, they would likely not survive. Therefore, while survival assays may have some relevance for experiments related to clinical engraftment studies, in many cases they have limited usefulness as a stem cell assay. Another issue is that measurements of overall competitive ability of populations of cells cannot distinguish between variations in LTRC number and variations in competitive ability per individual HSCs. Similarly, limiting dilution assays measure only quantity and not "quality" of HSCs. A possible solution proposed by Nakauchi is to combine the two strategies and determine the activity per HSC 4 ' 6 5 . In limiting dilution or purified single cell transplantation experiments, the precise definitions used to identify HSCs are of partiular importance for the interpretation of results. In general, if long-term and multilineage donor-derived repopulation is seen, it is determined retrospectively that an HSC was present in the test cells. As technology has developed, this definition and the way it is measured have changed over the years (see table 1.1). Nevertheless, certain issues still remain unresolved. For example, with respect 13 to the measurement of HSC self-renewal, it is usually assumed that "long-term" repopulation equates with self-renewal. However, the extent of self-renewal that must be observed before the HSC definition is met is still debated. In other words, how "long-term" is long enough? It could be argued that the only true test for self-renewal in vivo is to test the regenerated cells using the same assay (i.e. secondary transplantation), but this is a time- and labour-intensive procedure. It is therefore of utmost importance to interpret each conclusion in light of the HSC definition used. Another set of issues concerns multipotentiality, the second hallmark feature of HSC activity. Since differentiated cells are identified primarily via FACS, false positives are a possibility. A recent example involves the use of the Gr-1 antibody, which has been found to co-stain a subpopulation of T-cells, resulting in the possibility of classifying recipients as having all their lineages repopulated when in fact no myeloid cells were present66. This is of particular importance when comparing HSCs to multipotent progenitors with limited self-renewal, since some lymphoid cells produced early post-transplant have a relatively long half-life and might give a false positive. In addition, it is commonly assumed that if a few cell types representative of myeloid and lymphoid lineages are produced, that the test cells are fully pluripotent. However, the recent identification of a primitive lymphomyeloid repopulating cell without erythroid or megakaryocytic activity67 raises interesting questions regarding pluripotency and reminds us that in theory, a cell cannot be classified as pluripotent unless all differentiated lineages are identified. 14 1.2 Regulation of hematopoiesis The bone marrow is a complex environment including many hematopoietic and non-hematopoieitic cell types and a host of extrinsic molecules produced locally and systemically. Collectively, these maintain a tight regulation of blood cell production and allow rapid response to altered requirements to maintain homeostasis. Most often, external regulation of hematopoiesis involves intercellular communication via growth factors. Growth factor activated signalling events are interpreted by other intracellular signalling molecules and ultimately with the transcription factor repertoire present and the chromatin status in each target cell the gene expression status of the cell is determined. Thus, hematopoiesis is regulated by a combination of intrinsic and extrinsic factors such that intrinsically unique cells may respond differently to stimulation by the same growth factor(s). Over the years, many cytokines have been identified and characterized, including positive or negative regulators of survival, proliferation, self-renewal, and (lineage-specific) differentiation. While many aspects of these processes are intertwined, available evidence suggests that they are often regulated by distinct mechanisms. 1.2.1 Surv iva l and Apoptosis A well characterized system in which the regulation of survival is paramount is that of the final stages of T-cell development in the thymus. Of the 20-40 million new T-cells produced per day in a young adult mouse, only 2-3% eventually exit the thymus into 15 the blood. This is due to the double selection process that ensures that the newly produced T-cells have rearranged their TCR genes in a productive manner yet are not self-reactive. In the first instance (positive selection), T-cells are programmed to die unless their T-cell receptors can be stimulated by self-MHC-peptide complexes expressed by thymic cortical epithelial cells. In the second (negative selection), T-cells whose receptors bind strongly to self antigens expressed by antigen presenting dendritic cells receive death signals and undergo apoptosis (reviewed in 6 8). Myeloid and erythroid progenitors are constantly being produced in the bone marrow, but are programmed to die prior to producing their differentiated progeny. However, in the presence of sufficient levels of the positively regulating growth factors erythropoietin, GM-CSF, and/or G-CSF, and thrombopoietin, apoptosis is suppressed and differentiated cells of appropriate types are finally produced and released into the bloodstream69'70. In this way, the body is able to maintain appropriate levels of blood cells in a timely fashion. HSCs, on the other hand, appear to be primarily regulated through control of proliferation, mainly through interactions with the HSC niche (see section 1.2.4.2). However, the control of apoptosis has been shown to play a role as well. That HSCs are susceptible to apoptosis has been demonstrated by studies in which the overexpresssion of bcl-2 was shown to decrease the death of HSCs in response to a variety of apoptosis inducing stimuli, including irradiation71, exposure to chemotherapeutic agents72, and growth factor deprivation73. Conversely, the inducible deletion of a related family member, mcl-1, resulted in the loss of primitive hematopoietic cells, including HSCs 7 4 . It was also shown that the overexpression of bcl-2 in combination with Steel factor 16 signalling was sufficient to prevent apoptosis and allow proliferation in serum-free cultures of HSCs 7 5 . Notably, this did not alter the likelihood of self-renewal, indicating that the control of survival, proliferation, self-renewal, and differentiation are distinct processes. However, some growth factors can affect multiple processes. Thrombopoietin, for example, can promote survival76, but also the proliferation of HSCs and 77 differentiation of multipotent progenitors . 1 . 2 . 2 Proliferation and Quiescence Cell cycle progression of hematopoietic cells is directed by the external binding of multiple types of growth factor molecules to specific receptors on the cell surface, which then activate the appropriate intracellular machinery. Well-defined examples include GM-CSF, G-CSF, M-CSF, and IL-3, responsible for the proliferation of myeloid progenitors (reviewed in 7 8). Other growth factors, including thrombopoieiten, Steel factor, IL-l 1, IL-6, and Flt-3 ligand have been shown to be important for the stimulation 79 of proliferation in more primitive hematopoietic cells (reviewed in ). In certain combinations, these growth factors have been shown to have synergistic (or sometimes antagonistic) effects80"83. Cell cycle progression of the most primitive hematopoietic cells 8 4 can also be negatively regulated by cytokines that inhibit proliferation, such as TGF-(31 8 6 . Quiescence of adult HSCs is thought to be controlled primarily through interactions with the HSC niche (see section 1.2.4.2). The intracellular machinery involved in cell cycle control includes the highly conserved cyclin-dependent kinases (CDKs), cyclins, and C D K inhibitors (reviewed in 17 ). Interestingly, certain components of this machinery are variously expressed in different stages of hematopoiesis (reviewed in 8 8). Examples of proteins involved in the regulation of proliferation in primitive hematopoietic cells include the the transcription factors M E F / E L F 4 8 9 and G f i l 9 0 ; 9 1 , as well as the C D K inhibitors p 21 c i e 1 / w a f l 9 2 , P 16 I N K 4 a 9 3 , and p27K , p l 9 4 , shown to work in cooperation with the M Y C antagonist M A D 1 9 5 . A recent study by Yamazaki et al 9 6 reported that stimulation by thrombopoieitin or steel factor induced lipid raft clustering in primitive hematopoietic cells and that this clustering was essential for proliferation to occur. Quiescent cells whose lipid raft clustering was inhibited stayed in GO and did not divide, regardless of stimulation with mitogenic growth factors. Of note, some of these cells remained viable and functional, and could respond to the growth factor stimulation upon removal of the inhibitor. This suggests that one of the ways that HSC proliferation could be negatively controlled might be through intrinsic or extrinsic modulation of lipid raft reorganization. 1.2.3. Lineage restriction Interestingly, it has been found that primitive hematopoietic cells express a host of genes normally associated with differentiating or differentiated cell types, albeit at low levels (reviewed in 9 7 ) . This has led to the hypothesis that chromatin is generally maintained in an open state in primitive cells, and is thus accessible to transcriptional complexes98. Progressive alterations in chromatin structure then occur during differentiation, resulting in altered transcription factor expression and the subsequent up-18 regulation and down-regulation of appropriate lineage-specific genes in a step-wise fashion". Such alterations in chromatin structure are examples of epigenetic changes, which refer to gene expression alterations that are heritable but not caused by changes in the DNA sequence itself. Epigenetic changes can be initiated and maintained by alterating the methylation status of DNA sequences, which recruit repressive chromatin modification complexes100, or via the methylation, phosphorylation, or acetylation of specific amino acids of histone tails, which can induce or prevent chromatin condensation101"103. However, it has been shown that these changes during hematopoietic differentiation are not absolutely irreversible, since cells of one lineage can switch to another lineage in response to specific extrinsic or intrinsic manipulations that result in changes in the transcription factors active in commitment and maintenance of lineage-specific gene expression programs104. Some lineage-specific transcription factors have a dual functional role, in that they not only promote expression of genes of one lineage, they repress expression of other lineage-specific genes. Recently characterized examples include the master transcriptional regulators GATA-1, which is essential for erythroid development105, and PU-1, which is essential for the development of B- and T- cells, granulocytes, and monocytes106. Interestingly, a functional antagonism exists between GATA-1 and PU.l that is important for determining erythroid versus myeloid lineage commitment. For example, GATA-1 overexpression causes the reprogramming of myeloid precursors into erythroid cells1 0 7, while forced PU.l expression blocks erythroid differentiation108. This block can be overcome by co-expression of GATA-1109. Direct evidence of functional antagonism was shown in 19 experiments in which GATA-1 and PU.l binding sites were placed in front of a reporter gene, revealing that each of the two proteins repressed transcriptional activation by the other1 1 0'1 1 1. Furthermore, it was found that PU.l inhibits GATA-1 target genes expression by recruiting chromatin-modifying proteins to genes to which GATA-1 is bound 1 1 2. The repressed chromatin structure thus created has been shown to be released during erythroid development via the down-regulation of PU. ln}. More extensive models of how lineage restriction occurs and how cell fate decisions are initially made in HSCs are still debated. A number of studies measuring the lineage outputs of clonally derived colony-forming cells suggest that this can be best described using stochastic mechanisms (see section 3, diversity of HSCs, for more details). In this model, extrinsic factors serve only to regulate proliferation and survival, not lineage choice. This concept is reinforced by observations that the overexpression of various cytokine receptors does not alter lineage fate decisions114'115. However, strong evidence of cytokine-directed lineage choice has been observed in some systems116"118, demonstrating that lineage restriction may be a multifaceted process. 1.2.4 Regulation of HSC self-renewal and differentiation In order to self-renew, HSCs require functional proliferation pathways along with repression of apoptosis and differentiation pathways. Therefore, many of the regulators discussed in sections 1.2.1 and 1.2.2 could also impact the likelihood of HSC self-renewal. The decision of HSCs to self-renew or to differentiate is governed by a complex 20 interplay between autonomous signals (intrinsic regulation) and stimuli from the surrounding microenvironment (extrinsic regulation). 1.2.4.1 Intrinsic regulators The molecular pathways involved in H S C self-renewal are intricate networks of receptors, signalling transducers, transcription factors, and other intracellular proteins. These networks have begun to be unravelled using knockout mice, conditional inactivation strategies, constitutively active or dominant negative proteins, and other techniques. Thus far, relatively few have been identified and characterized, but it is clear that self-renewal is a complex process. The following are a few examples. The transcription factor Tel/Etv6 is required for the maintenance of H S C s , but not less primitive cells. Conditional inactivation of the Tel/Etv6 gene in B-cell , T-cell, or erythroid progenitors did not affect their differentiated cell output. However, when inactivation was induced in H S C s , a complete depletion of the entire bone marrow was seen within several weeks 1 1 9 . A s was expected, Tel/Etv6 inactivated bone marrow could not compete effectively against wild-type bone marrow cells when co-transplanted. G f i l is a transcriptional repressor that controls H S C self-renewal by limiting their proliferation. Upon deletion, H S C s proliferate much more rapidly and lose radioprotective ability and competitive reconstitution ability when transplanted 9 1. Its mechanism of action is not completely clear but may work via regulation of the C D K i P21CIPI/WAFI^ w h k h i s d o w n r e g u l a t e d m Gfil-deficient H S C s 9 0 . 21 Pten is a tumour suppressor that is a negative regulator of the A K T pathway, which is involved in many cellular processes, including proliferation, survival, and lineage choice. Besides increasing the likelihood of leukemic transformation, the conditional inactivation of Pten in bone marrow HSCs negatively affected HSC self-renewal. Upon transplantation, Pten null HSCs initially engrafted normally but were unable to sustain long-term multilineage WBC production. A transient expansion of Pten HSCs was seen at early times post-transplant, but over time the HSC pool was exhausted 1 2 0 . Cell cycle analysis revealed that Pten' HSCs were rarely in G o , suggesting a role of Pten in blocking cell cycle progression. Interestingly, Pten deficiency resulted in an alteration of WBC production, with severely reduced B-cell output and a 121 corresponding increase in T- and myeloid cells . The JAK-STAT pathway is a common downstream pathway of cytokine-induced signalling. Constitutively activated Stat5 in HSCs promoted in vitro self-renewal of HSCs and a dramatic amplification of downstream progenitors122. Conversely, its deletion resulted in decreased bone marrow and blood cellularity and a loss of competitive ability when transplanted123. Similar results were seen with Stat3. When its activity was decreased in HSCs through expression of a dominant negative STAT3, the competitive reconstituting ability of the transduced HSCs was reduced. Conversely, up-regulation by the expression of a constitutively activated form of Stat3 enhanced the self-renewal and regeneration activity of transplanted HSCs during the initial period of hematologic , recovery, and this could be recapitulated upon tertiary transplantation124. Collectively, these studies suggest that the JAK-STAT pathway plays an important role in regulating HSC self-renewal. 22 The homeobox gene HOXB4 encodes a transcription factor that is expressed in primitive hematopoietic cells and is rapidly downregulated in more differentiated populations12*126. When HOXB4 was retrovirally overexpressed in HSCs, a remarkable expansion of HSCs was observed in v/vo127"129and in vitrouo, suggesting that it could act as a positive regulator of HSC self-renewal. Interestingly, Hoxb4-deficient mice had no major hematopietic abnormalities and only a mild proliferation defect131, suggesting that its loss of function could be compensated by a different mechanism, possibly Hoxa4 and/or Hoxc4. In addition, the effects of HOXB4 overexpression could be enhanced by downregulation of p21, suggesting the simultaneous modulation of two independent self-renewal pathways132. The Polycomb group (PcG) of developmental regulatory genes were originally described in Drosophila. The PcG gene Bmil is a transcriptional regulator of various homeobox family genes in the mouse133 and is a member of PRC 1 (polycomb repression complex 1). The PcG genes encoding members of this complex, including Bmil, Mell8, and Mphl/Rae28, are expressed in the CD34-KSL B M population (which is highly purified for HSCs) and are down-regulated in more differentiated B M populations134. All three PRC1 members have also been shown to play a role in HSC self-renewal. Mphl/Rae28 deficient HSCs were able to fully reconstitute transplanted recipients with all lineages for several months and Mphl/Rae2'8 overexpression did not result in expansion of primitive hematopoietic cells in vitrom. However, Mphl/Rae28 deficient HSCs were reduced in their ability to expand stem cell numbers, as quantified by limit dilution transplants into secondary recipients135. In addition to a quantitiative decrease, 23 the regenerated Mphl/Rae28 deficient HSCs were qualitatively deficient, as measured by a lower activity per stem cell . Mel 18 null HSCs were shown to express elevated levels of Hoxb4 and the frequency of HSCs, measured as CRU, was elevated. At the same time, however, there 137 • was also a qualitative decrease in repopulating activity per CRU . However, it should be noted that both of these measures may be skewed due to the previously described defects in T- and B- lymphocyte development in Mel18'1' mice 1 3 8, which results in a lower B M cellularity (potentially affecting HSC frequency) and presumably a reduction in lymphoid cell production in the PB (potentially affecting the WBC output of Mell8~'~ HSCs). When Mell8 was overexpressed, the opposite was found, including lower levels of HoxB4 expression, a lower HSC frequency, and a higher repopulation activity per H S C 1 3 7 . Bmil null HSCs were able to generate a normal spectrum of differentiated blood cells, but only temporarily, due to a marked reduction in self-renewal ability 1 3 9' 1 4 0. The lack of self-renewal ability of Bmil'1'HSCs was confirmed by their inability to repopulate secondary recipients just six weeks after the primary transplant139'140 and explains why Bmil null mice die of bone marrow exhaustion within two months of birth 1 4 1. Conversely, when Bmil was overexpressed in HSCs, a marked increase in self-renewal was seen. HSCs overexpressing Bmil were dramatically expanded in vitro, and following 10 days of culture, contained 35-fold greater repopulating ability than mock transduced cells1 3 4. Recently, E4F1 was identified to physically and functionally interact with BMI1 to mediate HSC self-renewal, and its deletion could partially overcome the negative effects of Bmil knockout142. Similarly, the deletion of pl61NK4A and p l ^ restored 24 considerable self-renewal ability in Bmil null HSCs . Interestingly, overexpression of HOXB4 failed to rescue the defective self-renewal of Bmil'1' HSCs, suggesting that Hoxb4 acts upstream of Bmil134. The retinoic acid receptor (RAR) family are nuclear receptors that are members of the steroid/thyroid hormone superfamily of transcription factors. Two members of this family, RARa and RARy, are expressed in HSCs, and RARy, but not RARa, has been shown to be a key regulator of HSC self-renewal144. The loss of RARy results in reduced numbers of HSCs and increased numbers of downstream progenitors, while the loss of RARa did not result in any hematopoietic defects. Activation of RARy by stimulation with its natural ligand, all-trans retinoic acid (ATRA), resulted in increased Notch 1 and Hoxb4 expression, accompanied by enhanced self-renewal. Therefore, loss of RARy results in reduced numbers of HSCs due to increased HSC differentiation, while RARy activation results in increased HSC self-renewal144. 1.2.4.2 Extrinsic regulators - HSC niche In recent years, much has been learned about the specific environments within the bone marrow that are of particular importance for the survival and self-renewal of HSCs. This so-called stem cell "niche" contains a unique combination of stromal supporting cells, osteoblasts, and extracellular matrix, and is thought to provide a unique environment that protects HSCs from stimuli, including those that would induce apoptotis, differentiation, or excessive mitogenesis145. It is thought that HSCs in this 25 environment are usually quiescent with periodic activation in order to produce daughter HSCs (thus increasing HSC numbers) or multipotent progenitors capable of transient amplification and differentiation in order to replenish functional blood cells as needed (Figure 1.1). It has been hypothesized for decades that the bone medulla surface lining, or endosteum, was important for the maintenance and self-renewal of primitive hematopoietic cells. For example, experiments studying the spatial distributions of primitive hematopoietic cells within the bone marrow showed that the density of granulocytic progenitors increased with distance from the surface of the mouse femur. Conversely, the density of CFU-S decreased towards the bone marrow axis1 4 6. Interestingly, the peak density of CFU-S was approximately 100 microns from the bone surface, which might be expected since CFU-S are likely produced by the HSCs more closely associated with the osteoblasts at the endosteal surface. Indeed, the relationship between HSCs and bone may go even "deeper", as many HSCs can be found within dissociated skeletal bone (Brenton Short, personal communication). Mouse bone-forming cells, called osteoblasts, are located at the endosteum and are an essential component of the HSC niche. When genetic strategies were employed to increase osteoblast number, increases of HSC numbers were also seen. This increase did not extend to downstream progenitors, suggesting that the supportive role of the osteoblast is specific to HSCs 1 4 7 ' 1 4 8 . Conversely, it has also been shown that in a mouse model with decreased osteoblast numbers, hematopoiesis is severely altered 1 4 9 . Combining recently developed HSC markers with immunohistocytochemical localization techniques have confirmed that HSCs are often localized at the endosteal 26 surface, associated with osteoblasts ' . Suzuki et al also monitored this interaction in vivo using Gata2-directed GFP fluorescence, and found that individual quiescent GFP-positive cells were in intimate contact with osteoblasts at the endosteum. Following 5-FU treatment, time-lapse fluoresecent imaging revealed that the few remaining GFP + cells residing at the bone marrow edge did not move, whereas the surrounding G F P - cells were very active151. Osteoblasts have also been demonstrated to secrete factors known to modulate stem cell function, including G-CSF, M-CSF, GM-CSF, IL- l , IL-6, as well as cell cycle inhibitory factors such as TGF-p\ LIF, TNF-a, and TNF-P (reviewed in 1 5 2). Osteopontin, a glycoprotein secreted by osteoblasts, has been shown to have an important role in the modulation of HSC numbers in the niche 1 5 3 . It has also been suggested that Wnt proteins secreted by niche stromal cells contribute to HSC self-renewal by activating the Wnt signalling pathway154. Direct cell-cell interactions between HSCs and osteoblasts are also important. N-cadherin and beta-catenin, which play roles in adherens junctions, were asymmetrically localized between HSCs and osteoblasts148. It has also been shown that the interaction of Tie2, expressed by HSCs, and Ang-1, expressed on the osteoblasts, promotes HSC quiescence150. In addition, interaction between the Jagged ligand, expressed on osteoblasts, and Notch family receptors on HSCs, has been shown to promote self-renewal of H S C s 1 4 7 ; 1 5 5 . In studies of hematopoietic regeneration following the ablation of non-quiescent hematopoietic progenitors using 5-fluorouracil treatment, Heissig et al. observed a shift of clusters of proliferating hematopoietic cells from the osteoblastic niche to the "vascular niche", proximal to the sinusoidal blood vessels in the B M 1 5 6 . That putative 27 HSCs are located at both the osteoblastic and vascular niches has been confirmed using * 27 immunohistocytochemical localization . 1.2.4.3 Studies of extrinsic control of HSC self-renewal in vitro Stringently defined assays are of particular importance when interpreting results of ex vivo manipulated hematopoietic cells. While huge expansions of restricted progenitors and considerable expansions of multipotent progenitors have been described, increases in the numbers of these cells do not necessarily correlate with increases in rigorously defined HSCs 1 5 7 ' 1 5 8 . The first studies to demonstrate rigorously defined HSC self-renewal in vitro used retrovirally marked bone marrow cells grown in stromal cultures for 4 weeks and transplanted into multiple recipients. Expansion of a subset of the input HSCs was revealed by clonal integration patterns shared between the WBCs of multiple recipients up to 7 months post-transplant. Despite the demonstration of occasional expansion, there was an overall net decline of HSC numbers in these cul tures , 5 9 ; , 6 ° . Subsequent studies of in vitro self-renewal, primarily utilizing HSC-enriched cell populations cultured in defined serum-free and feeder-free conditions plus recombinant cytokines, have yielded direct evidence of growth factor determined modulation of HSC self-renewal versus differentiation decisions. Using various concentrations of the cytokines Steel factor (SF) and/or Flt3-ligand (Flt3L) plus IL-6 and/or IL-11, maintenance of HSC activity in 7-21 day feeder-free cultures initiated by HSC-enriched 28 populations of mouse B M has been reported in several studies 8 1 , 1 5 7' 1 6 1' 1 6 2. Conversely, if IL- l , IL-3, or TNF-a were added to similar cultures, a decrease in HSC activity was seen 8 l ; 1 6 , ; 1 6 3 ; ' 6 4 . However, another report suggests that IL-3, albeit in combination with slightly different cytokines, did not have a negative effect on HSC activity over 10 days in culture165. Using various combinations of SF, Flt3L, IL-6, and IL-l 1, net increases of HSC activity have been documented with the use of 100 ng/ml IL-l 1 plus 18 ng/ml SF 1 6 6 , 100 ng/ml IL-l.1 plus 50 ng/ml SF plus 100 ng/ml Flt3L 6 2, and 50 ng/ml SF plus 20 ng/ml IL-6 plus 100 ng/ml Flt3L 5 2. A subsequent multifactorial design analysis using 83 varying combinations and concentrations of SF, IL-l 1, and Flt3L was performed . This study revealed that while FU3L or SF alone can stimulate proliferation, maintenance of HSC activity requires activation of the gpl30 pathway via IL-6 or IL-l 1 stimulation. It was also found that the exact concentration of cytokines was very important for HSC amplification. A negative interaction was found between SF and Flt3L when both were used at high concentrations, and stimulation of HSC amplification by IL-l 1 was observed at a narrow concentration range, with high levels having a negative effect. Experiments using thrombopoietin (TPO)-null mice have demonstrated that TPO is an important component of in vivo regeneration of HSCs following transplantation167. A subsequent study168 reported that these effects of TPO may be due to an increase in Hoxb4 mediated by an increase in its upstream regulator USF-1 1 6 9 . As a single agent, TPO has been suggested to maintain HSC activity over 7 days in serum-containing, feeder-free culture without stimulating their proliferation76. This lack of mitogenic activity was confirmed in serum-free, feeder-free cultures in single cell cultures of CD34" K S L cells cultured with TPO alone65. Combining 100 ng/ml TPO with 10 ng/ml of SF 29 was highly mitogenic yet promoted HSC self-renewal (maintenance) over 3 or 6 days of culture65. Interestingly, similar experiments performed using 100 ng/ml each of Flt3L, SF, and IL-11 showed a dramatic decrease in HSC self-renewal compared to 100 ng/ml TPO plus 10 ng/ml SF 8 2 . This data seemed to conflict with the previous findings of Miller et al. 6 2, but this might be explained by the 2-fold difference in SF concentration used. In addition, the Miller et al. experiments used cultures initiated with 15 Lin-Sca+ cells and measured CRU activity after 10 days, while Nakauchi et al. initiated the cultures with single CD34-KSL cells and measured after 3 days. Thus, at the time of transplant, the latter contained only 1 or 2 cells, while the former would have grown exponentially to tens of thousands of cells, raising the possibility of indirect mechanisms being partly responsible for the difference in effect. A dramatic expansion of HSC activity was observed when unfractionated bone marrow cells were cultured for 3-4 weeks in serum-free media and fibroblast growth factor (FGF) 1 7 0 ; 1 7 1 . Although FGF receptors were shown to be expressed on HSCs, the HSC expansion requires the co-culture of non-HSCs for its effect, since the culture of purified K S L cells did not lead to HSC expansion170'172. Yet, the culture of purified K S L cells along with genetically distinct bone marrow cells led to HSC expansion at the same rate 1 7 2 . Interestingly, the addition of SF+IL1 l+Flt3L to the FGF culture system induced a strong proliferative response and all HSC activity was rapidly lost172. This suggests that the mechanisms of self-renewal induced in the FGF culture system are distinct from those induced by SF, IL11 and Flt3L. Purified and lipid-modified Wnt3a, a ligand for the Frizzled family of proteins, appears to induce significant HSC expansion over 6 days in cultures containing serum 30 and SF, although the long-term repopulation ability of the expanded HSCs was not rigorously examined173. Complementary experiments in Bcl-2 transgenic mice where the Writ signaling pathway was activated using the constitutively active form of P-catenin demonstrated a similar effect174, possibly in conjunction with Notch signaling175. However, very recent experiments have cast doubt on the applicability of these findings in a normal genetic background.176. All-trans-retinoic acid is a hydrophobic vitamin A analogue that is an agonist for the nuclear retinoic acid receptors (RARs). In 7-day and 14-day cultures containing fetal bovine serum, SF, Flt3L, IL-6, IL-l 1, and ATRA, maintenance of HSC activity was observed relative to the starting cells. In the same conditions without added ATRA, a slight decrease of HSC activity was seen. Conversely, the addition of a RAR antagonist abrogated all HSC activity within 7 days of culture177. It has since been shown that ATRA's effects are mediated via the nuclear receptor RARy, as RARy"7" HSCs did not repopulate recipients after being cultured in the presence of ATRA, while RARa"7" HSCs responded similarly to WT HSCs 1 4 4 . Zhang and Lodish characterized a minor population of day-15 murine fetal liver, the CD3 + fraction (-2%), that could support expansion of HSCs in culture. Using Affymetrix microarrays, they then compared the transcriptional profile of these cells with day-15 G r l + cells and with adult splenic CD3 + cells, neither of which could support HSCs in culture, and identified candidate secreted and membrane proteins that were relatively abundant in the fetal liver CD3 + cells178. One protein thus identified was insulin-like growth factor 2 (IGF-2). In cultures containing fetal bovine serum, SF, Flt3L, and IL-6, maintenance of HSC numbers was seen over 3 days. However, when a high 31 concentration of insulin-like growth factor 2 (IGF-2) was added to cultures, a 2-fold expansion of HSCs was observed178. Two additional candidates identified in the original screen, angiopoietin-like proteins 2 and 3, were subsequently shown to support a significant (24 and 30-fold) net expansion of HSCs over 10 days in culture with SF, TPO, IGF-2, and FGF-1 1 7 9 . 1.3 Heterogeneity of HSCs Heterogeneity within the HSC compartment has been observed and discussed since the field was first established. In general, any observable diversity within the HSC compartment can be attributed to alternate self-renewal and differentiation decisions. For example, a HSC that undergoes limited self-renewal divisions would likely exhaust faster than one that self-renews indefinitely. Or, a HSC whose progeny tend to differentiate down the lymphoid pathways would likely contribute less to the erythrocyte pool. The mechanisms that generate diversity in the HSC compartment have been and continue to be much debated. Diversity in CFU-S size and considerable size-independent variability in CFU-S regeneration were observed in day 10, 12, and 14 spleen colonies16. Subsequently, lineage output per colony was also found to be heterogeneous11. Because the frequency distributions of secondary colonies formed per primary colony did not fit a Poisson distribution, it was determined that the diversity observed fell far outside what could be expected due to experimental variation. Therefore, it was hypothesized that the variability 32 was due to either heritable biological heterogeneity, or that the heterogeneity developed during the growth of the colony in a random fashion16. This latter hypothesis was 180 described in detail as a stochastic model of stem cell proliferation . This model was subsequently expanded to include the apparently random choice of lineage restriction 181 during differentiation . The heterogeneity of colony size and self-renewal ability observed within the CFU-S compartment was paralleled by similar (clone size-independent) variability observed in the secondary colony-forming ability of primitive in vitro colony forming cells 1 8 2 ; 1 8 3. Again, the frequency distributions of secondary colonies formed per primary colony did not fit a Poisson distribution, and were thus interpreted to support the 180' 181 stochastic model of stem cell self-renewal and commitment proposed earlier ' . Experiments involving the sequential replating of progenitors derived from primitive in vitro colonies demonstrated that these primitive cells could produce diverse combinations of cell lineages184. When these progenitors were allowed to divide once and separated, comparison of the progenitor pairs revealed a progressive restriction in lineage potentials, but the lineage choices were made in an apparently stochastic fashion 1 8 5' 1 8 6. Similar in vitro paired-daughter cell experiments performed more than a decade later using FACS isolated HSCs also showed dissimilar combinations of lineages and sizes and was also • 187 interpreted to support the stochastic model of stem cell commitment . The advent of retroviral marking of transplantable primitive hematopoietic 7 -8 • 1 8 8 • 1 8 Q cells ' ' ' allowed clonal tracking of differentiated cell output over time following transplantation. Heterogeneity and fluctuations in total output, lineage distribution, and longevity were observed within retrovirally marked clones1'2'190"193. Attempts to explain 33 the observed clonal variations included hypotheses of stochastic mechanisms1'2'193, intrinsic differences in the primitive cells themselves2'190'194, or the successive output of variable but relatively short-lived clones1'1 9 1'1 9 5. A distinction between long-term and short-term pluripotent repopulating cells was made clear using limiting dilution transplants of bulk bone marrow196or enriched populations197'198. Further studies showing that these two types could be phenotypically separated1 5 8'1 9 9'2 0 0 provided compelling evidence that intrinsic differences exist between LT-HSCs and ST-HSCs. Subsequently, multiple additional purification strategies have been found able to phenotypically separate the two cell types (see section 1.1.2 for details). More recently, additional evidence of heterogeneity within the HSC compartment has been obtained from analyses of single highly purified cells, including variations in 201 transcriptional profiles , time to first division upon cytokine stimulation in single cell cultures6 5'1 1 7'1 8 7, variation in WBC reconstitution levels after single cell transplants4'31'65'202"204, repopulation kinetics post-transplant3, in vivo daughter HSC regeneration4^5, and lineage distribution of cell output in vitro187 and in vivo4. Some of the most direct evidence supporting the intrinsic nature of heterogeneity within the HSC compartment has been generated by the Muller-Sieburg group. They generated clonally repopulated mice by initiating stromal cell-containing cultures with limiting dilutions of unseparated B M , harvested positive cultures 4 weeks later, and transplanted entire cultures into irradiated recipients205. These clonally repopulated mice were analyzed for donor repopulation levels and lineage distributions over time and found to be heterogeneous in overall repopulation kinetics and the WBC types produced. 34 When HSCs regenerated in the primary recipients were transplanted into secondary and tertiary recipients and similarly analyzed, inter-clonal heterogeneity was still seen, but daughter HSCs derived from any given clone behaved similar to each other in lineage distribution and overall repopulation pattern, and tended to exhibit behaviour similar to the HSC(s) originally injected206. These same mice were then analyzed together with a larger group of recipients clonally repopulated with limiting numbers of fresh B M or single purified cells, for the shape of their repopulation kinetics post-transplant. Only a subset of all the possible repopulation curves was observed, suggesting that the hematopoietic stem cell compartment consists of a limited number of distinct HSC subsets207. Collectively, these results suggest that HSC heterogeneity is due to the 208 existence of epigenetically predetermined HSC subtypes . Significant evidence has accumulated to affirm the concept of a HSC-supporting niches within the bone marrow (see section 2.3.2). The existence of multiple niche locations (i.e. osteoblastic and vascular) provides a possible mechanism for alternative extrinsic regulation of HSC behaviour. For example, it has been hypothesized that the osteoblastic niche provides signals to maintain the quiescence of HSCs, while activated HSCs have been associated with the vascular niche2 0 9. This is supported by observations of the migration of actively dividing hematopoietic clusters from the osteoblastic to the vascular niche following the ablation of cycling progenitors with 5-fluorouracil treatment156. It has thus been speculated that there may be multiple distinct niches within the bone marrow, and that the differences in these environments might induce alterations 210 211 in HSC fate decisions , although this is still controversial . 35 Functional differences in the behaviour of HSCs from different stages of development have also been demonstrated. In mid-gestation (-14.5 days post-conception), the primary hematopoietic organ is the fetal liver. Thereafter, HSCs gradually migrate from the liver to the bone marrow until shortly after birth, at which point the role of the liver in hematopoiesis is greatly diminished. When compared with adult B M HSCs, fetal liver (FL) HSCs exhibited faster regeneration rates of daughter H S C s 2 1 2 ; 2 1 3 , spleen colony forming cells2 1 4, and more differentiated cells 2 1 4' 2 1 5 upon transplantion into irradiated recipients. Another distinct functional difference between fetal and adult HSCs is their cycling status. FL HSCs are exclusively cycling, while most steady-state adult HSCs are quiescent63. Parallel alterations in differentiation properties have also been described, in which FL HSCs produce elevated proportions of myeloid45 or lymphoid2 1 6 cells compared with their adult B M counterparts. Interestingly, it has recently been suggested that an intrinsic switch from these fetal-like to adult-like HSC properties occurs rapidly, between 3 and 4 weeks post-birth45'63. Several studies have also implicated differences in bone marrow HSCs from young adult versus older adult mice. In the commonly used C57B1/6J strain, it was observed that absolute numbers of B M HSCs increased with age216"221 yet have poorer homing efficiency216'221. Another well-documented observation is that B M HSCs from older mice tend to produce higher proportions of myeloid cells and lower proportions of 218 222 lymphoid cells (particularly B-cells) than their counterparts from younger mice " . Comparison of HSC numbers and other HSC properties between C57B1/6J and other mouse strains has revealed that these characteristics must be largely genetically determined217. 36 1.4. Thesis Objectives The overall aim of my thesis work was to further characterize the properties of HSCs and the extent to which intrinsic and extrinsic mechanisms may alter their subsequent self-renewal or differentiation behaviour. This goal was based on the hypothesis that the well documented heterogeneity in HSC behaviour might be attributable to a previously unrecognized substructure within the HSC compartment. However, because most of the studies to date have made use either of enriched populations of HSCs, or HSCs present at limiting dilutions in a mixture of other cells, it has not been possible to determine whether these were masking an additional level of HSC heterogeneity. I was therefore interested in developing an experimental approach where I could accurately and repeatedly track the clonal behaviours of HSCs in vitro (by studying cultures initiated with single HSCs) or in vivo (by tracking reconstitution obtained in mice injected with a single HSC). To accomplish this I initially helped to develop and validate a strategy to isolate HSCs at very high purities (>30%) and to use these efficiently to initiate short-term clonal cultures and transplant irradiated mice with single cells or in vitro clones. This is presented in the first part of Chapter 2. I then used these powerful tools to explore the extent to which exogenous growth factors can differentially affect HSC self-renewal decisions in vitro, and whether these decisions were necessarily tied to alterations in cell cycle duration. Extrinsic control of cell differentiation programs is well established, and a link between differentiation and the duration of G l has been described (reviewed in 8 7). In addition, a number of studies of primitive hematopoietic cells have suggested that their proliferative potential may be 37 inversely correlated with their rate of cell cycle entry ' " . However, the link between control of cycling and differentiation in HSCs is complex, since the time taken for highly purified quiescent HSCs to complete a first cell division can be influenced by the combination of growth factors to which they are exposed65'82.1 explored this further by comparing initial cell division kinetics and HSC self-renewal in short-term cultures initiated with single purified cells cultured in two different growth factor combinations. The results, also presented in Chapter 2, indicated that combination growth factor signals can differentially alter HSC self-renewal decisions independently of alterations in cell cycle duration. By analyzing clonal cultures in conditions that support HSC self-renewal, I then ' asked whether I could identify new behavioural traits of cultured HSC with functionally validated long term multi-lineage repopulating activity in vivo. As discussed in section 1.1.2.1, most of the markers used to isolate HSC-enriched populations from steady-state mouse B M are not directly associated with HSC functional potential, since these phenotypes are altered when HSC are activated or stimulated to divide. To search for new identifying properties of HSCs self-renewing in vitro, I used a novel microwell array video imaging system to visualize clones derived from individual HSCs over a 4-day period under the conditions I had just confirmed to support HSC self-renewal divisions. Each clone was then recovered and assayed for the presence of HSCs with long term multi-lineage in vivo repopulating activity. Time-lapse video images of these assayed clones were then used to correlate visible characteristics of the cultured cells with those that had produced functionally defined daughter HSCs. The results of these studies are described in Chapter 3. 38 Traditionally, HSCs with rigorously defined long term repopulating activity have been envisaged to comprise a relatively homogeneous population. However, significant variation in the outputs of individual HSCs transplanted in vivo is a well recognized hallmark of their behaviour. Nevertheless, the extent to which this heterogeneity in HSC behaviour reflects a predetermined intrinsic diversity among these cells2 0 8, or their chance exposure to different environments ' ' , or stochastic events affecting intrinsic 1 fif)• 9 9 f i 9"\0 pathways that regulate their behaviour ' " remains unresolved. Therefore, I set out to characterize the spectrum of in vivo output activity within the HSC compartment, and the extent that this diversity is intrinsically, extrinsically, or randomly regulated. To address this, I performed single cell transplants to examine longitudinally the clonal WBC output of more than a hundred single HSCs (or their in vitro generated HSC progeny), by monitoring the peripheral blood of primary recipients over a 6-month period and, in some cases, also of secondary and tertiary recipients. The results, presented in Chapter 4, provide convincing evidence of intrinsic programming of HSCs and challenge the prevailing linear-branching model of HSC differentiation. 39 Table 1.1 - Definitions of long-term multilineage donor repopulation in various studies 4^ O Publication Host Competitor Cells min % repopulation Lineages Weeks PT Szilvassy/Eaves PNAS 199053 8.5Gy female B6/C3 1-2x10(5) compromised detectable by Southern, -5% male DNA in BM and thymus 5 to 10 Morrison/Weissman Immunity 1994199 split 9-9.2 Gy B6 2x10(5) 0.3% of each lineage B220, Mad, CD3, Gr1 16+ Rebel/Lansdorp Blood 1994157 9-9.5 Gy B6 2x10(5) compromised or 4x10(5) Sca- 20% None - Ly5.1 only 11 to 15 Trevisan/lscove Blood 19966' 3Gy W41 None "detectable", -0.1-1% erythroid (via gpi shift assay) 32 to 52 Osawa/Nakauchi Science 19963 9.5Gy B6 500 CD34 lowKSL "detectable" Mac1/Gr1, Thy1/B220 (not required) 12 +survival Miller/Eaves PNAS 199762 9Gy B6 or4GyW41 10(5) or nothing 1% Gr-1+, SSClow 16+ Sudo/Nakauchi JEM 2000218 9.5Gy B6 2x10(5) 1% Gr1/Mac1, CD4/CD8, B220 12 Ema/Nakauchi JEM 200065 9.5Gy B6 2x10(5) 1% Gr1/Mac1, CD4/CD8, B220 12 Adolfsson/Jacobsen Immunity 200124 9.5Gy B6 1.5-2x10(5) 0.5% of total, 0.1% myeloid Gr1/Mac1 16 Szilvassy/Eaves ExpHem 2003231 split 9Gy B6 2x10(5) 5% Gr1/Mac1,Thy1,B220 5,10,17,26 Uchida/Eaves ExpHem 200335 4Gy W41 None 0.05% of each lineage B220, CD5, Gr1/Mac1 16+ de Haan/Miller Dev Cell 2003170 10Gy B6 2x10(5) 2% of each lineage Gr1/Mac1, Thy1, B220 8 Matsuzaki/Okano Immunity 200431 10.5Gy B6 2x10(5) 1% Gr1/Mac1, B220/CD3 12 Kiel/Morrison Cell 200527 split 11 Gy B6 2x10(5) above background (0.1-0.3%) Gr1/Mac1, B220, CD3 16 Ema/Nakauchi Dev Cell 20054 9.5Gy B6 2x10(5) 1% Gr1/Mac1, CD4/CD8, B220 12 to 16 Zhang/Lodish Blood 200540 split 10Gy B6 1-2x10(5) 1% Thy1, B220, Mad, Gr1, Ter119 16 Rossi/Weissman PNAS 2005220 split 9.2Gy B6 3x10(5) "detectable" Mad, B220, TCR0+ 28 Dykstra/Jervis PNAS 2006232 4GyW41 None 1% of total® 16wk, 1 % of each lineage at any time Gr1/Mac1, B220/CD5 4,8,12,16 c I ^ Q ) ^ Hematopoietic Stem Cell (HSC) S I C CFU-S 6 © > Multi-potent Progenitors 4 CLP Myelopoisis MEP CMP GMP I \ \ \ \ \ Lymphopoisis T / \ \ \ i • I RBC platelets granulocytes monocytes/ macrophages \ \ \ t \ \ Pro-T Pro-NK Pro-B © o © I I I e T-cells NK-cells B-cells Figure 1.1 - Hierarchical model of hematopoiesis (adapted from Bryder et al ). HSCs are placed at the top of this hierarchy and are the only cells with extensive self-renewal ability. Cells of all mature blood cell types are continually produced through a series of successive differentiation and amplification steps. Cells defined LTRC, STRC, and CFU-S (Sections 1.1.1 and 1.1.2) are thought to represent cells in the top levels of the hierarchy, as shown. 41 'test' cell(s) (CD45.1) Test cell-derived myeloid cells CD45.1 o ' ' ' A ; / it inject into irradiated recipient (CD45.2) Col lect b lood sample CD45.1 Determine presence or a b s e n c e of test cel l -der ived (CD45.1+) cel ls of var ious b l o o d cell types (usual ly done by F A C S ) F i g u r e 1.2 - F u n c t i o n a l de tec t ion o f H S C by t r a n s p l a n t a t i o n in vivo. HSCs can be retrospectively identified by their ability to produce multiple WBC types for an extended period of time following transplantation into an irradiated recipient. In this way, the HSC's hallmark properties of self-renewal and multipotentiality can be functionally assessed. Test cell(s) congenic for a single cell surface marker (CD45) are injected into irradiated recipients. After an extended period of time (usually 16 weeks), a blood sample is collected and the tested by FACS for the presence of donor-derived WBC. If WBC of both myeloid and lymphoid lineages are present at this time, it is assumed that the original test cells must have contained a HSC (with self-renewal and multipotent abilities). 42 References 1. Lemischka LR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell. 1986;45:917-927. 2. Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of mice reconstituted with retrovirus-infected stem cells. E M B O J. 1987;6:3955-3960. 3. Osawa M , Hanada K, Hamada H, Nakauchi H. Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science. 1996;273:242-245. 4. Ema H, Sudo K, Seita J et al. Quantification of self-renewal capacity in single hematopoietic stem cells from normal and Lnk-deficient mice. Dev.Cell. 2005;8:907-914. 5. Lorenz E, Uphoff D, Reid TR, Shelton E. Modification of irradiation injury in mice and guinea pigs by bone marrow injections. J.Natl.Cancer Inst. 1951; 12:197-201. 6. Ford CE, Hamerton JL, Barnes DW, Loutit JF. Cytological identification of radiation-chimaeras. Nature. 1956;177:452-454. 7. Dick JE, Magli M C , Huszar D, Phillips RA, Bernstein A. Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/Wv mice. Cell. 1985;42:71-79. 8. Keller G, Paige C, Gilboa E, Wagner EF. Expression of a foreign gene in myeloid and lymphoid cells derived from multipotent haematopoietic precursors. Nature. 1985;318:149-154. 9. Till JE, McCulloch EA. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat.Res. 1961;14:213-22.:213-222. 10. Becker AJ, McCulloch EA, Till JE. Cytological demonstration of the clonal nature of spleen colonies derived from transplanted mouse marrow cells. Nature. 1963;197:452-4.:452-454. 11. Fowler J, Till JE, McCulloch EA, Siminovitch L. The cellular composition of hemopoietic spleen colonies. J.Cell Physiol. 1967;69(l):65-72. 12. Wu A M , Till JE, Siminovitch L, McCulloch EA. Cytological evidence for a relationship between normal hemotopoietic colony-forming cells and cells of the lymphoid system. J.Exp.Med. 1968;127:455-464. 43 13. Worton RG, McCulloch EA, Till JE. Physical separation of hemopoietic stem cells differing in their capacity for self-renewal. J.Exp.Med. 1969;130:91-103. 14. Magli M C , Iscove NN, Odartchenko N. Transient nature of early haematopoietic spleen colonies. Nature. 1982;295:527-529. 15. Curry JL, Trentin JJ. Hemopoietic spleen colony studies. I. Growth and differentiation. Dev.Biol. 1967;15:395-413. 16. Siminovitch L, McCulloch EA, Till JE. The distribution of colony-forming cells among spleen colonies. J.Cell Physiol. 1963;62:327-36.:327-336. 17. Schofield R. The relationship between the spleen colony-forming cell and the haemopoietic stem cell. Blood Cells. 1978;4:7-25. 18. Hodgson GS, Bradley TR. Properties of haematopoietic stem cells surviving 5-fluorouracil treatment: evidence for a pre-CFU-S cell? Nature. 1979;281:381-382. 19. Ploemacher RE, Brons RH. Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hemopoietic stem cell compartment following irradiation: evidence for a pre-CFU-S cell. Exp.Hematol. 1989;17:263-266. 20. Jones RJ, Wagner JE, Celano P, Zicha MS, Sharkis SJ. Separation of pluripotent haematopoietic stem cells from spleen colony-forming cells. Nature. 1990;347:188-189. 21. Spangrude GJ, Heimfeld S, Weissman IL. Purification and characterization of mouse hematopoietic stem cells. Science. 1988;241:58-62. 22. Okada S, Nakauchi H, Nagayoshi K et al. In vivo and in vitro stem cell function of c-kit- and Sea-1-positive murine hematopoietic cells. Blood. 1992;80:3044-3050. 23. Wiesmann A, Phillips RL, Mojica M et al. Expression of CD27 on murine hematopoietic stem and progenitor cells. Immunity. 2000;12:193-199. 24. Adolfsson J, Borge OJ, Bryder D et al. Upregulation of Flt3 expression within the bone marrow Lin(-)Scal(+)c-kit(+) stem cell compartment is accompanied by loss of self-renewal capacity. Immunity. 2001;15:659-669. 25. Christensen JL, Weissman IL. Flk-2 is a marker in hematopoietic stem cell differentiation: a simple method to isolate long-term stem cells. Proc.Natl.Acad.Sci.U.S.A. 2001;98:14541-14546. 26. Chen CZ, Li M , de Graaf D et al. Identification of endoglin as a functional marker that defines long-term repopulating hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 2002;99:15468-15473. 44 27. Kiel MJ, Yilmaz OH, Iwashita T et al. S L A M family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell. 2005;121:1109-1121. 28. Balazs AB, Fabian AJ, Esmon CT, Mulligan RC. Endothelial protein C receptor (CD201) explicitly identifies hematopoietic stem cells in murine bone marrow. Blood. 2006;107:2317-2321. 29. Wagers AJ, Weissman IL. Differential expression of alpha2 integrin separates long-term and short-term reconstituting Lin-/loThyl.l(lo)c-kit+ Sca-1+ hematopoietic stem cells. Stem Cells. 2006;24:1087-1094. 30. Benveniste P, Cantin C, Hyam D, Iscove NN. Hematopoietic stem cells engraft in mice with absolute efficiency. Nat.Immunol. 2003;4:708-713. 31. Matsuzaki Y, Kinjo K, Mulligan RC, Okano H. Unexpectedly efficient homing capacity of purified murine hematopoietic stem cells. Immunity. 2004;20:87-93. 32. Pearce DJ, Ridler C M , Simpson C, Bonnet D. Multiparameter analysis of murine bone marrow side population cells. Blood. 2004;103:2541-2546. 33. Goodell M A , Brose K, Paradis G, Conner AS, Mulligan RC. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J.Exp.Med. 1996;183:1797-1806. 34. Bertoncello I, Hodgson GS, Bradley TR. Multiparameter analysis of transplantable hemopoietic stem cells: I. The separation and enrichment of stem cells homing to marrow and spleen on the basis of rhodamine-123 fluorescence. Exp.Hematol. 1985;13:999-1006. 35. Uchida N, Dykstra B, Lyons KJ, Leung FY, Eaves CJ. Different in vivo repopulating activities of purified hematopoietic stem cells before and after being stimulated to divide in vitro with the same kinetics. Exp.Hematol. 2003;31:1338-1347. 36. Drew E, Merkens H, Chelliah S, Doyonnas R, McNagny K M . CD34 is a specific marker of mature murine mast cells. Exp.Hematol. 2002;30:1211. 37. Sato T, Laver JH, Ogawa M . Reversible expression of CD34 by murine hematopoietic stem cells. Blood. 1999;94:2548-2554. 38. Tajima F, Deguchi T, Laver JH, Zeng H, Ogawa M . Reciprocal expression of CD38 and CD34 by adult murine hematopoietic stem cells. Blood. 2001;97:2618-2624. 39. Uchida N, Dykstra B, Lyons K et al. A B C transporter activities of murine hematopoietic stem cells vary according to their developmental and activation status. Blood. 2004;103:4487-4495. 45 40. Zhang CC, Lodish HF. Murine hematopoietic stem cells change their surface phenotype during ex vivo expansion. Blood. 2005; 105:4314-4320. 41. Rebel VI, Miller CL, Thombury GR et al. A comparison of long-term repopulating hematopoietic stem cells in fetal liver and adult bone marrow from the mouse. Exp.Hematol. 1996;24:638-648. 42. Yilmaz OH, Kiel MJ, Morrison SJ. S L A M family markers are conserved among hematopoietic stem cells from old and reconstituted mice and markedly increase their purity. Blood. 2006;107:924-930. 43. Matsubara A, Iwama A, Yamazaki S et al. Endomucin, a CD34-like sialomucin, marks hematopoietic stem cells throughout development. J.Exp.Med. 2005;202:1483-1492. 44. Kim I, He S, Yilmaz OH, Kiel MJ, Morrison SJ. Enhanced purification of fetal liver hematopoietic stem cells using S L A M family receptors. Blood. 2006;108:737-744. 45. Bowie M B , Kent D, Dykstra B et al. Identification of a New Intrinsically Timed Developmental Checkpoint That Reprograms Key Hematopoietic Stem Cell Properties. Proc.Natl.Acad.Sci.U.S.A. 3006 46. Kondo M , Weissman IL, Akashi K. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell. 1997;91:661-672. 47. Akashi K, Traver D, Miyamoto T, Weissman IL. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature. 2000;404:193-197. 48. Hodgson GS, Bradley TR, Radley JM. The organization of hemopoietic tissue as inferred from the effects of 5-fluorouracil. Exp.Hematol. 1982;10:26-35. 49. Cheshier SH, Morrison SJ, Liao X, Weissman IL. In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 1999;96:3120-3125. 50. Harrison DE. Competitive repopulation: a new assay for long-term stem cell functional capacity. Blood. 1980;55:77-81. 51. Harrison DE, Jordan CT, Zhong RK, Astle C M . Primitive hemopoietic stem cells: direct assay of most productive populations by competitive repopulation with simple binomial, correlation and covariance calculations. Exp.Hematol. 1993;21:206-219. 52. Audet J, Miller CL, Rose-John S, Piret JM, Eaves CJ. Distinct role of gpl30 activation in promoting self-renewal divisions by mitogenically stimulated murine hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 2001;98:1757-1762. 46 53. Szilvassy SJ, Humphries RK, Lansdorp PM, Eaves A C , Eaves CJ. Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy. Proc.Natl.Acad.Sci.U.S.A. 1990;87:8736-8740. 54. Ogata H, Bradley WG, Inaba M et al. Long-term repopulation of hematolymphoid cells with only a few hemopoietic stem cells in mice. Proc.Natl.Acad.Sci.U.S.A. 1995;92:9432. 55. Trevisan M , Iscove NN. Phenotypic analysis of murine long-term hemopoietic reconstituting cells quantitated competitively in vivo and comparison with more advanced colony-forming progeny. J.Exp.Med. 1995;181:93-103. 56. Komuro K, Itakura K, Boyse EA, John M . Ly 5: a new T lymphocyte antigen system. Immunogenetics 1975;1:452-456. 57. Scheid MP, Triglia D. Further description of the Ly-5 system. Immunogenetics 1979;9:423-433. 58. Thomas M L . The leukocyte common antigen family. Annu.Rev.Immunol. 1989;7:339-69.:339-369. 59. Nocka K, Tan JC, Chiu E et al. Molecular bases of dominant negative and loss of function mutations at the murine c-kit/white spotting locus: W37, Wv, W41 and W. E M B O J. 1990;9:1805-1813. 60. Miller CL, Rebel VI, Lemieux M E et al. Studies of W mutant mice provide evidence for alternate mechanisms capable of activating hematopoietic stem cells. Exp.Hematol. 1996;24:185-194. 61. Trevisan M , Yan XQ, Iscove NN. Cycle initiation and colony formation in culture by murine marrow cells with long-term reconstituting potential in vivo. Blood. 1996;88:4149-4158. 62. Miller CL, Eaves CJ. Expansion in vitro of adult murine hematopoietic stem cells with transplantable lympho-myeloid reconstituting ability. Proc.Natl.Acad.Sci.U.S.A. 1997;94:13648-13653. 63. Bowie MB, McKnight KD, Kent D G et al. Hematopoietic stem cells proliferate until after birth and show a reversible phase-specific engraftment defect. J.Clin.Invest. 2006;116:2808-2816. 64. Quesenberry PJ, Colvin GA, Lambert JF. The chiaroscuro stem cell: a unified stem cell theory. Blood. 2002;100:4266-4271. 65. Ema H, Takano H, Sudo K, Nakauchi H. In vitro self-renewal division of hematopoietic stem cells. J.Exp.Med. 2000;192:1281-1288. 47 66. Bryder D, Sasaki Y, Borge OJ, Jacobsen SE. Deceptive multilineage reconstitution analysis of mice transplanted with hemopoietic stem cells, and implications for assessment of stem cell numbers and lineage potentials. J.Immunol. 2004;172:1548-1552. 67. Adolfsson J, Mansson R, Buza-Vidas N et al. Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment. Cell. 2005;121:295-306. 68. Starr TK, Jameson SC, Hogquist KA. Positive and negative selection of T cells. Annu.Rev.Immunol. 2003;21:139-76:139-176. 69. Koury MJ, Bondurant M C . Erythropoietin retards DNA breakdown and prevents programmed death in erythroid progenitor cells. Science. 1990;248:378-381. 70. Williams GT, Smith CA, Spooncer E, Dexter T M , Taylor DR. Haemopoietic colony stimulating factors promote cell survival by suppressing apoptosis. Nature. 1990;343:76-79. 71. Domen J, Gandy K L , Weissman IL. Systemic overexpression of BCL-2 in the hematopoietic system protects transgenic mice from the consequences of lethal irradiation. Blood. 1998;91:2272-2282. 72. Domen J, Weissman IL. Hematopoietic stem cells and other hematopoietic cells show broad resistance to chemotherapeutic agents in vivo when overexpressing bcl-2. Exp.Hematol. 2003;31:631-639. 73. Domen J, Cheshier SH, Weissman IL. The role of apoptosis in the regulation of hematopoietic stem cells: Overexpression of Bcl-2 increases both their number and repopulation potential. J.Exp.Med. 2000;191:253-264. 74. Opferman JT, Iwasaki H, Ong CC et al. Obligate role of anti-apoptotic MCL-1 in the survival of hematopoietic stem cells. Science. 2005;307:1101-1104. 75. Domen J, Weissman IL. Hematopoietic stem cells need two signals to prevent apoptosis; BCL-2 can provide one of these, Kitl/c-Kit signaling the other. J.Exp.Med. 2000;192:1707-1718. 76. Matsunaga T, Kato T, Miyazaki H, Ogawa M . Thrombopoietin promotes the survival of murine hematopoietic long-term reconstituting cells: comparison with the effects of FLT3/FLK-2 ligand and interleukin-6. Blood. 1998;92:452-461. 77. Sitnicka E, Lin N, Priestley GV et al. The effect of thrombopoietin on the proliferation and differentiation of murine hematopoietic stem cells. Blood. 1996;87:4998-5005. 78. Metcalf D. The molecular control of cell division, differentiation commitment and maturation in haemopoietic cells. Nature. 1989;339:27-30. 48 79. Ezoe S, Matsumura I, Satoh Y , Tanaka H , Kanakura Y . Cell cycle regulation in hematopoietic stem/progenitor cells. Cell Cycle. 2004;3:314-318. 80. K u H , Yonemura Y , Kaushansky K, Ogawa M . Thrombopoietin, the ligand for the Mpl receptor, synergizes with steel factor and other early acting cytokines in supporting proliferation of primitive hematopoietic progenitors of mice. Blood. 1996;87:4544-4551. 81. Yonemura Y , K u H, Lyman SD, Ogawa M . In vitro expansion of hematopoietic progenitors and maintenance of stem cells: comparison between FLT3/FLK-2 ligand and KIT ligand. Blood. 1997;89:1915-1921. 82. Nakauchi H, Sudo K, Ema H. Quantitative assessment of the stem cell self-renewal capacity. Ann.N.Y.Acad.Sci. 2001;938:18-24; discussion 24-5.:18-24. 83. Audet J, Miller CL, Eaves CJ, Piret JM. Common and distinct features of cytokine effects on hematopoietic stem and progenitor cells revealed by dose-response surface analysis. Biotechnol.Bioeng. 2002;80:393-404. 84. Keller JR, Mcniece IK, Sill KT et al. Transforming growth factor beta directly regulates primitive murine hematopoietic cell proliferation. Blood. 1990;75:596-602. 85. Soma T, Y u JM, Dunbar CE. Maintenance of murine long-term repopulating stem cells in ex vivo culture is affected by modulation of transforming growth factor-beta but not macrophage inflammatory protein-1 alpha activities. Blood. 1996;87:4561-4567. 86. Larsson J, Blank U , Helgadottir H et al. TGF-beta signaling-deficient hematopoietic stem cells have normal self-renewal and regenerative ability in vivo despite increased proliferative capacity in vitro. Blood. 2003;102:3129-3135. 87. Zhu L, Skoultchi AI. Coordinating cell proliferation and differentiation. Curr.Opin.Genet.Dev. 2001;11:91-97. 88. Furukawa Y , Kikuchi J, Nakamura M et al. Lineage-specific regulation of cell cycle control gene expression during haematopoietic cell differentiation. Br.J.Haematol. 2000;110:663-673. 89. Lacorazza HD, Yamada T, Liu Y et al. The transcription factor MEF/ELF4 regulates the quiescence of primitive hematopoietic cells. Cancer Cell. 2006;9:175-187. 90. Hock H , Hamblen MJ , Rooke H M et al. Gfi-1 restricts proliferation and preserves functional integrity of haematopoietic stem cells. Nature. 2004;431:1002-1007. 49 91. Zeng H, Yucel R, Kosan C, Klein-Hitpass L, Moroy T. Transcription factor Gfi 1 regulates self-renewal and engraftment of hematopoietic stem cells. E M B O J. 2004;23:4116-4125. 92. Cheng T, Rodrigues N, Shen H et al. Hematopoietic stem cell quiescence maintained by p21cipl/wafl. Science. 2000;287:1804-1808. 93. Janzen V, Forkert R, Fleming HE et al. Stem-cell ageing modified by the cyclin-dependent kinase inhibitor pl6LNK4a. Nature. 2006;443:421-426. 94. Cheng T, Rodrigues N, Dombkowski D, Stier S, Scadden DT. Stem cell repopulation efficiency but not pool size is governed by p27(kipl). Nat.Med. 2000;6:1235-1240. 95. Walkley CR, Fero M L , Chien W M , Purton L E , McArthur GA. Negative cell-cycle regulators cooperatively control self-renewal and differentiation of haematopoietic stem cells. Nat.Cell Biol. 2005;7:172-178. 96. Yamazaki S, Iwama A, Takayanagi S et al. Cytokine signals modulated via lipid rafts mimic niche signals and induce hibernation in hematopoietic stem cells. E M B O J. 2006;25:3515-3523. 97. Akashi K. Lineage promiscuity and plasticity in hematopoietic development. Ann.N.Y.Acad.Sci. 2005;1044:125-31.:125-131. 98. Cross M A , Enver T. The lineage commitment of haemopoietic progenitor cells. Curr.Opin.Genet.Dev. 1997;7:609-613. 99. Bonifer C. Epigenetic plasticity of hematopoietic cells. Cell Cycle. 2005;4:211-214. 100. Bird A. DNA methylation patterns and epigenetic memory. Genes Dev. 2002;16:6-21. 101. Vermaak D, Ahmad K, Henikoff S. Maintenance of chromatin states: an open-and-shut case. Curr.Opin.Cell Biol. 2003;15:266-274. 102. Peters A H , Schubeler D. Methylation of histones: playing memory with DNA. Curr.Opin.Cell Biol. 2005;17:230-238. 103. Shogren-Knaak M , Ishii H, Sun JM et al. Histone H4-K16 acetylation controls chromatin structure and protein interactions. Science 2006;311:844-847. 104. Graf T. Differentiation plasticity of hematopoietic cells. Blood. 2002;99:3089-3101. 50 105. Fujiwara Y, Browne CP, Cunniff K, Goff SC, Orkin SH. Arrested development of embryonic red cell precursors in mouse embryos lacking transcription factor GATA-1. Proc.Natl.Acad.Sci.U.S.A. 1996;93:12355-12358. 106. Scott EW, Simon M C , Anastasi J, Singh H. Requirement of transcription factor PU.l in the development of multiple hematopoietic lineages. Science. 1994;265:1573-1577. 107. Kulessa H, Frampton J, Graf T. GATA-1 reprograms avian myelomonocytic cell lines into eosinophils, thromboblasts, and erythroblasts. Genes Dev. 1995;9:1250-1262. 108. Rao G, Rekhtman N, Cheng G, Krasikov T, Skoultchi AL Deregulated expression of the PU.l transcription factor blocks murine erythroleukemia cell terminal differentiation. Oncogene. 1997;14:123-131. 109. Rekhtman N, Radparvar F, Evans T, Skoultchi AL Direct interaction of hematopoietic transcription factors PU.l and GATA-1: functional antagonism in erythroid cells. Genes Dev. 1999;13:1398-1411. 110. Zhang P, Behre G, Pan J et al. Negative cross-talk between hematopoietic regulators: G A T A proteins repress PU. l . Proc.Natl.Acad.Sci.U.S.A. 1999;96:8705-8710. 111. Nerlov C, Querfurth E, Kulessa H, Graf T. GATA-1 interacts with the myeloid PU.l transcription factor and represses PU.l-dependent transcription. Blood. 2000;95:2543-2551. 112. Stopka T, Amanatullah DF, Papetti M , Skoultchi AL PU. 1 inhibits the erythroid program by binding to GATA-1 on DNA and creating a repressive chromatin structure. E M B O J. 2005;24:3712-3723. 113. Rekhtman N, Choe KS, Matushansky I et al. PU. 1 and pRB interact and cooperate to repress GATA-1 and block erythroid differentiation. Mol.Cell Biol. 2003;23:7460-7474. 114. Goldsmith M A , Mikami A, You Y et al. Absence of cytokine receptor-dependent specificity in red blood cell differentiation in vivo. Proc.Natl.Acad.Sci.U.S.A. 1998;95:7006-7011. 115. Stoffel R, Ziegler S, Ghilardi N et al. Permissive role of thrombopoietin and granulocyte colony-stimulating factor receptors in hematopoietic cell fate decisions in vivo. Proc.Natl.Acad.Sci.U.S.A. 1999;96:698-702. 116. Kondo M , Scherer DC, Miyamoto T et al. Cell-fate conversion of lymphoid-committed progenitors by instructive actions of cytokines. Nature. 2000;407:383-386. 51 117. Takano H, Ema H, Sudo K, Nakauchi H. Asymmetric division and lineage commitment at the level of hematopoietic stem cells: inference from differentiation in daughter cell and granddaughter cell pairs. J.Exp.Med. 2004;199:295-302. 118. Metcalf D. Lineage commitment in the progeny of murine hematopoietic preprogenitor cells: influence of thrombopoietin and interleukin 5. Proc.Natl.Acad.Sci.U.S.A. 1998;95:6408-6412. 119. Hock H, Meade E, Medeiros S et al. Tel/Etv6 is an essential and selective regulator of adult hematopoietic stem cell survival. Genes Dev. 2004; 18:2336-2341. 120. Yilmaz OH, Valdez R, Theisen B K et al. Pten dependence distinguishes haematopoietic stem cells from leukaemia-initiating cells. Nature. 2006;441:475-482. 121. Zhang J, Grindley JC, Yin T et al. PTEN maintains haematopoietic stem cells and acts in lineage choice and leukaemia prevention. Nature. 2006;441:518-522. 122. Kato Y, Iwama A, Tadokoro Y et al. Selective activation of STAT5 unveils its role in stem cell self-renewal in normal and leukemic hematopoiesis. J.Exp.Med. 2005;202:169-179. 123. Snow JW, Abraham N, Ma M C et al. STAT5 promotes multilineage hematolymphoid development in vivo through effects on early hematopoietic progenitor cells. Blood. 2002;99:95-101. 124. Chung YJ, Park BB, Kang YJ et al. Unique effects of Stat3 on the early phase of hematopoietic stem cell regeneration. Blood. 2006;108:1208-1215. 125. Sauvageau G, Lansdorp PM, Eaves CJ et al. Differential expression of homeobox genes in functionally distinct CD34+ subpopulations of human bone marrow cells. Proc.Natl.Acad.Sci.U.S.A 1994;91:12223-12227. 126. Pineault N, Helgason CD, Lawrence HJ, Humphries RK. Differential expression of Hox, Meisl, and Pbxl genes in primitive cells throughout murine hematopoietic ontogeny. Exp.Hematol. 2002;30:49-57. 127. Sauvageau G, Thorsteinsdottir U, Eaves CJ et al. Overexpression of HOXB4 in hematopoietic cells causes the selective expansion of more primitive populations in vitro and in vivo. Genes Dev. 1995;9:1753-1765. 128. Thorsteinsdottir U, Sauvageau G, Humphries RK. Enhanced in vivo regenerative potential of HOXB4-transduced hematopoietic stem cells with regulation of their pool size. Blood. 1999;94:2605-2612. 52 129. Antonchuk J, Sauvageau G, Humphries RX. HOXB4 overexpression mediates very rapid stem cell regeneration and competitive hematopoietic repopulation. Exp.Hematol. 2001;29:1125-1134. 130. Antonchuk J, Sauvageau G, Humphries RK. HOXB4-induced expansion of adult hematopoietic stem cells ex vivo. Cell. 2002;109:39-45. 131. Brun A C , Bjornsson JM, Magnusson M et al. Hoxb4-deficient mice undergo normal hematopoietic development but exhibit a mild proliferation defect in hematopoietic stem cells. Blood. 2004;103:4126-4133. 132. Miyake N, Brun A C , Magnusson M et al. HOXB4-induced self-renewal of hematopoietic stem cells is significantly enhanced by p21 deficiency. Stem Cells. 2006;24:653-661. 133. van der Lugt N M , Alkema M , Berns A, Deschamps J. The Polycomb-group homolog Bmi-1 is a regulator of murine Hox gene expression. Mech.Dev. 1996;58:153-164. 134. Iwama A, Oguro H, Negishi M et al. Enhanced self-renewal of hematopoietic stem cells mediated by the polycomb gene product Bmi-1. Immunity. 2004;21:843-851. 135. Ohta H, Sawada A, Kim JY et al. Polycomb group gene rae28 is required for sustaining activity of hematopoietic stem cells. J.Exp.Med. 2002;195:759-770. 136. Kim JY, Sawada A, Tokimasa S et al. Defective long-term repopulating ability in hematopoietic stem cells lacking the Polycomb-group gene rae28. Eur.J.Haematol. 2004;73:75-84. 137. Kajiume T, Ninomiya Y, Ishihara H, Kanno R, Kanno M . Polycomb group gene mel-18 modulates the self-renewal activity and cell cycle status of hematopoietic stem cells. Exp.Hematol. 2004;32:571-578. 138. Akasaka T, Tsuji K, Kawahira H et al. The role of mel-18, a mammalian Polycomb group gene, during IL-7-dependent proliferation of lymphocyte precursors. Immunity. 1997;7:135-146. 139. Lessard J, Sauvageau G. Bmi-1 determines the proliferative capacity of normal and leukaemic stem cells. Nature. 2003;423:255-260. 140. Park IK, Qian D, Kiel M et al. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature. 2003;423:302-305. 141. van der Lugt N M , Domen J, Linders K et al. Posterior transformation, neurological abnormalities, and severe hematopoietic defects in mice with a targeted deletion of the bmi-1 proto-oncogene. Genes Dev. 1994;8:757-769. 53 142. Chagraoui J, Niessen SL, Lessard J et al. E4F1: a novel candidate factor for mediating BMI1 function in primitive hematopoietic cells. Genes Dev. 2006;20:2110-2120. 143. Oguro H, Iwama A, Morita Y et al. Differential impact of Ink4a and Arf on hematopoietic stem cells and their bone marrow microenvironment in Bmil-deficient mice. J.Exp.Med. 2006;203:2247-2253. 144. Purton L E , Dworkin S, Olsen GH et al. RARgamma is critical for maintaining a balance between hematopoietic stem cell self-renewal and differentiation. J.Exp.Med. 2006;203:1283-1293. 145. Moore K A , LemischkalR. Stem cells and their niches. Science. 2006;311:1880-1885. 146. Lord Bl , Testa NG, Hendry JH. The relative spatial distributions of CFUs and CFUc in the normal mouse femur. Blood. 1975;46:65-72. 147. Calvi L M , Adams GB, Weibrecht K W et al. Osteoblastic cells regulate the haematopoietic stem cell niche. Nature. 2003;425:841-846. 148. Zhang J, Niu C, Ye L et al. Identification of the haematopoietic stem cell niche and control of the niche size. Nature. 2003;425:836-841. 149. Visnjic D, Kalajzic Z, Rowe DW et al. Hematopoiesis is severely altered in mice with an induced osteoblast deficiency. Blood. 2004;103:3258-3264. 150. Arai F, Hirao A, Ohmura M et al. Tie2/angiopoietin-l signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell. 2004;118:149-161. 151. Suzuki N, Ohneda O, Minegishi N et al. Combinatorial Gata2 and Seal expression defines hematopoietic stem cells in the bone marrow niche. Proc.Natl.Acad.Sci.U.S.A. 2006;103:2202-2207. 152. Taichman RS, Emerson SG. The role of osteoblasts in the hematopoietic microenvironment. Stem Cells. 1998;16:7-15. 153. Stier S, Ko Y, Forkert R et al. Osteopontin is a hematopoietic stem cell niche component that negatively regulates stem cell pool size. J.Exp.Med. 2005;201:1781-1791. 154. Rattis F M , Voermans C, Reya T. Wnt signaling in the stem cell niche. Curr.Opin.Hematol. 2004; 11:88-94. 155. Stier S, Cheng T, Dombkowski D, Carlesso N, Scadden DT. Notchl activation increases hematopoietic stem cell self-renewal in vivo and favors lymphoid over myeloid lineage outcome. Blood. 2002;99:2369-2378. 54 156. Heissig B, Hattori K, Dias S et al. Recruitment of stem and progenitor cells from the bone marrow niche requires MMP-9 mediated release of kit-ligand. Cell. 2002;109:625-637. 157. Rebel VI, Dragowska W, Eaves CJ, Humphries RK, Lansdorp PM. Amplification of Sea-1+ Lin- WGA+ cells in serum-free cultures containing steel factor, interleukin-6, and erythropoietin with maintenance of cells with long-term in vivo reconstituting potential. Blood. 1994;83:128-136. 158. Spangrude GJ, Brooks D M , Tumas DB. Long-term repopulation of irradiated mice with limiting numbers of purified hematopoietic stem cells: in vivo expansion of stem cell phenotype but not function. Blood. 1995;85:1006-1016. 159. Fraser CC, Eaves CJ, Szilvassy SJ, Humphries RK. Expansion in vitro of retrovirally marked totipotent hematopoietic stem cells. Blood. 1990;76:1071-1076. 160. Fraser CC, Szilvassy SJ, Eaves CJ, Humphries RK. Proliferation of totipotent hematopoietic stem cells in vitro with retention of long-term competitive in vivo reconstituting ability. Proc.Natl.Acad.Sci.U.S.A. 1992;89:1968-1972. 161. Yonemura Y, Ku H, Hirayama F, Souza L M , Ogawa M . Interleukin 3 or interleukin 1 abrogates the reconstituting ability of hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 1996;93:4040-4044. 162. Matsunaga T, Hirayama F, Yonemura Y, Murray R, Ogawa M . Negative regulation by interleukin-3 (IL-3) of mouse early B-cell progenitors and stem cells in culture: transduction of the negative signals by betac and betaIL-3 proteins of IL-3 receptor and absence of negative regulation by granulocyte-macrophage colony-stimulating factor. Blood. 1998;92:901-907. 163. Traycoff C M , Cornetta K, Yoder M C , Davidson A, Srour EF. Ex vivo expansion of murine hematopoietic progenitor cells generates classes of expanded cells possessing different levels of bone marrow repopulating potential. Exp.Hematol. 1996;24:299-306. 164. Bryder D, Ramsfjell V, Dybedal I et al. Self-renewal of multipotent long-term repopulating hematopoietic stem cells is negatively regulated by Fas and tumor necrosis factor receptor activation. J.Exp.Med. 2001;194:941-952. 165. Bryder D, Jacobsen SE. Interleukin-3 supports expansion of long-term multilineage repopulating activity after multiple stem cell divisions in vitro. Blood. 2000;96:1748-1755. 166. Holyoake TL, Freshney M G , McNair L et al. Ex vivo expansion with stem cell factor and interleukin-11 augments both short-term recovery posttransplant and the ability to serially transplant marrow. Blood. 1996;87:4589-4595. 55 167. Fox N, Priestley G, Papayannopoulou T, Kaushansky K. Thrombopoietin expands hematopoietic stem cells after transplantation. J.Clin.Invest. 2002;110:389-394. 168. Kirito K, Fox N, Kaushansky K. Thrombopoietin stimulates Hoxb4 expression: an explanation for the favorable effects of TPO on hematopoietic stem cells. Blood. 2003;102:3172-3178. 169. Giannola D M , Shlomchik WD, Jegathesan M et al. Hematopoietic expression of HOXB4 is regulated in normal and leukemic stem cells through transcriptional activation of the HOXB4 promoter by upstream stimulating factor (USF)-l and USF-2. J.Exp.Med. 2000;192:1479-1490. 170. de Haan G, Weersing E, Dontje B et al. In vitro generation of long-term repopulating hematopoietic stem cells by fibroblast growth factor-1. Dev.Cell. 2003;4:241-251. 171. Crcareva A, Saito T, Kunisato A et al. Hematopoietic stem cells expanded by fibroblast growth factor-1 are excellent targets for retrovirus-mediated gene delivery. Exp.Hematol. 2005;33:1459-1469. 172. Yeoh JS, van Os R, Weersing E et al. Fibroblast growth factor-1 and -2 preserve long-term repopulating ability of hematopoietic stem cells in serum-free cultures. Stem Cells. 2006;24:1564-1572. 173. Willert K, Brown JD, Danenberg E et al. Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature. 2003;423:448-452. 174. Reya T, Duncan AW, Ailles L et al. A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature. 2003;423:409-414. 175. Duncan AW, Rattis F M , DiMascio L N et al. Integration of Notch and Wnt signaling in hematopoietic stem cell maintenance. Nat.Immunol. 2005;6:314-322. 176. Kirstetter P, Anderson K, Porse BT, Jacobsen SE, Nerlov C. Activation of the canonical Wnt pathway leads to loss of hematopoietic stem cell repopulation and multilineage differentiation block. Nat.Immunol. 2006;7:1048-1056. 177. Purton L E , Bernstein ID, Collins SJ. All-trans retinoic acid enhances the long-term repopulating activity of cultured hematopoietic stem cells. Blood. 2000;95:470-477. 178. Zhang CC, Lodish HF. Insulin-like growth factor 2 expressed in a novel fetal liver cell population is a growth factor for hematopoietic stem cells. Blood. 2004;103:2513-2521. 179. Zhang CC, Kaba M , Ge G et al. Angiopoietin-like proteins stimulate ex vivo expansion of hematopoietic stem cells. Nat.Med. 2006;12:240-245. 56 180. Till JE, McCulloch EA, Siminovitch L. A stochastic model of stem cell proliferation, based on the growth of spleen colony-forming cells. Proc.Natl.Acad.Sci.U.S.A. 1964;51:29-36.:29-36. 181. Korn AP, Henkelman R M , Ottensmeyer FP, Till JE. Investigations of a stochastic model of haemopoiesis. Exp.Hematol. 1973;1:362-375. 182. Humphries RK, Eaves A C , Eaves CJ. Self-renewal of hemopoietic stem cells during mixed colony formation in vitro. Proc.Natl.Acad.Sci.U.S.A. 1981;78:3629-3633. 183. Nakahata T, Ogawa M . Clonal origin of murine hemopoietic colonies with apparent restriction to granuclocyte-macrophage-megakaryocyte (GMM) differentiation. J.Cell Physiol. 1982;111:239-246. 184. Suda T, Suda J, Ogawa M . Single-cell origin of mouse hemopoietic colonies expressing multiple lineages in variable combinations. Proc.Natl.Acad.Sci.U.S.A. 1983;80:6689-6693. 185. Suda T, Suda J, Ogawa M . Disparate differentiation in mouse hemopoietic colonies derived from paired progenitors. Proc.Natl.Acad.Sci.U.S.A. 1984;81:2520-2524. 186. Suda J, Suda T, Ogawa M . Analysis of differentiation of mouse hemopoietic stem cells in culture by sequential replating of paired progenitors. Blood. 1984;64:393-399. 187. Nakauchi H, Takano H, Ema H, Osawa M . Further characterization of CD34-low/negative mouse hematopoietic stem cells. Ann.N.Y.Acad.Sci. 1999;872:57-66; discussion 66-70.:57-66. 188. Williams DA, Lemischka LR, Nathan DG, Mulligan RC. Introduction of new genetic material into pluripotent haematopoietic stem cells of the mouse. Nature. 1984;310:476-480. 189. Eglitis M A , Kantoff P, Gilboa E, Anderson WF. Gene expression in mice after high efficiency retroviral-mediated gene transfer. Science. 1985;230:1395-1398. 190. Keller G, Snodgrass R. Life span of multipotential hematopoietic stem cells in vivo. J.Exp.Med. 1990;171:1407-1418. 191. Capel B, Hawley R, Covarrubias L, Hawley T, Mintz B. Clonal contributions of small numbers of retrovirally marked hematopoietic stem cells engrafted in unirradiated neonatal W/Wv mice. Proc.Natl.Acad.Sci.U.S.A. 1989;86:4564-4568. 57 192. Capel B, Hawley RG, Mintz B. Long- and short-lived murine hematopoietic stem cell clones individually identified with retroviral integration markers. Blood. 1990;75:2267-2270. 193. Jordan CT, Lemischka IR. Clonal and systemic analysis of long-term hematopoiesis in the mouse. Genes Dev. 1990;4:220-232. 194. Van Zant G, Chen JJ, Scott-Micus K. Developmental potential of hematopoietic stem cells determined using retrovirally marked allophenic marrow. Blood. 1991;77:756-763. 195. Drize NJ, Keller JR, Chertkov JL. Local clonal analysis of the hematopoietic system shows that multiple small short-living clones maintain life-long hematopoiesis in reconstituted mice. Blood. 1996;88:2927-2938. 196. Zhong RK, Astle C M , Harrison DE. Distinct developmental patterns of short-term and long-term functioning lymphoid and myeloid precursors defined by competitive limiting dilution analysis in vivo. J.Immunol. 1996;157:138-145. 197. Smith L G , Weissman IL, Heimfeld S. Clonal analysis of hematopoietic stem-cell differentiation in vivo. Proc.Natl.Acad.Sci.U.S.A. 1991;88:2788-2792. 198. Randall TD, Weissman IL. Phenotypic and functional changes induced at the clonal level in hematopoietic stem cells after 5-fluorouracil treatment. Blood. 1997;89:3596-3606. 199. Morrison SJ, Weissman IL. The long-term repopulating subset of hematopoietic stem cells is deterministic and isolatable by phenotype. Immunity. 1994;1:661-673. 200. Lanzkron SM, Collector MI, Sharkis SJ. Hematopoietic stem cell tracking in vivo: a comparison of short-term and long-term repopulating cells. Blood. 1999;93:1916-1921. 201. Ramos CA, Bowman TA, Boles NC et al. Evidence for Diversity in Transcriptional Profiles of Single Hematopoietic Stem Cells. PLoS.Genet. 2006;2: 202. Wagers AJ, Sherwood RI, Christensen JL, Weissman IL. Little evidence for developmental plasticity of adult hematopoietic stem cells. Science. 2002;297:2256-2259. 203. Cao Y A , Wagers AJ, Beilhack A et al. Shifting foci of hematopoiesis during reconstitution from single stem cells. Proc.Natl.Acad.Sci.U.S.A. 2004;101:221-226. 58 204. Camargo FD, Chambers SM, Drew E, McNagny K M , Goodell M A . Hematopoietic stem cells do not engraft with absolute efficiencies. Blood. 2006;107:501-507. 205. Cho RH, Muller-Sieburg CE. High frequency of long-term culture-initiating cells retain in vivo repopulation and self-renewal capacity. Exp.Hematol. 2000;28:1080-1086. 206. Muller-Sieburg CE, Cho RH, Thoman M , Adkins B, Sieburg HB. Deterministic regulation of hematopoietic stem cell self-renewal and differentiation. Blood. 2002;100:1302-1309. 207. Sieburg HB, Cho RH, Dykstra B et al. The hematopoietic stem compartment consists of a limited number of discrete stem cell subsets. Blood. 2006;107:2311-2316. 208. Muller-Sieburg CE, Sieburg HB. The GOD of hematopoietic stem cells: a clonal diversity model of the stem cell compartment. Cell Cycle. 2006;5:394-398. 209. Kopp HG, Avecilla ST, Hooper AT, Rafii S. The bone marrow vascular niche: home of HSC differentiation and mobilization. Physiology (Bethesda.). 2005;20:349-56.:349-356. 210. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J.Pathol. 2006;169:338-346. 211. Kiel MJ, Iwashita T, Yilmaz OH, Morrison SJ. Spatial differences in hematopoiesis but not in stem cells indicate a lack of regional patterning in definitive hematopoietic stem cells. Dev.Biol. 2005;283:29-39. 212. Harrison DE, Zhong RK, Jordan CT, Lemischka IR, Astle C M . Relative to adult marrow, fetal liver repopulates nearly five times more effectively long-term than short-term. Exp.Hematol. 1997;25:293-297. 213. Rebel VI, Miller CL, Eaves CJ, Lansdorp PM. The repopulation potential of fetal liver hematopoietic stem cells in mice exceeds that of their liver adult bone marrow counterparts. Blood. 1996;87:3500-3507. 214. Jordan CT, McKearn JP, Lemischka LR. Cellular and developmental properties of fetal hematopoietic stem cells. Cell. 1990;61:953-963. 215. Morrison SJ, Hemmati HD, Wandycz A M , Weissman IL. The purification and characterization of fetal liver hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 1995;92:10302-10306. 216. Morrison SJ, Wandycz A M , Akashi K, Globerson A, Weissman IL. The aging of hematopoietic stem cells. Nat.Med. 1996;2:1011-1016. 59 217. de Haan G, Van Zant G. Dynamic changes in mouse hematopoietic stem cell numbers during aging. Blood. 1999;93:3294-3301. 218. Sudo K, Ema H, Morita Y, Nakauchi H. Age-associated characteristics of murine hematopoietic stem cells. J.Exp.Med. 2000;192:1273-1280. 219. Kim M , Moon HB, Spangrude GJ. Major age-related changes of mouse hematopoietic stem/progenitor cells. Ann.N.Y.Acad.Sci. 2003;996:195-208.:195-208. 220. Rossi DJ, Bryder D, Zahn JM et al. Cell intrinsic alterations underlie hematopoietic stem cell aging. Proc.Natl.Acad.Sci.U.S.A. 2005;102:9194-9199. 221. Liang Y, Van Zant G, Szilvassy SJ. Effects of aging on the homing and engraftment of murine hematopoietic stem and progenitor cells. Blood. 2005;106:1479-1487. 222. Liang Y, Van Zant G. Genetic control of stem-cell properties and stem cells in aging. Curr.Opin.Hematol. 2003;10:195-202. 223. Suda T, Suda J, Ogawa M . Proliferative kinetics and differentiation of murine blast cell colonies in culture: evidence for variable GO periods and constant doubling rates of early pluripotent hemopoietic progenitors. J.Cell Physiol. 1983;117:308-318. 224. Gothot A, Pyatt R, McMahel J, Rice S, Srour EF. Assessment of proliferative and colony-forming capacity after successive in vitro divisions of single human CD34+ cells initially isolated in GO. Exp.Hematol. 1998;26:562-570. 225. Punzel M , Liu D, Zhang T et al. The symmetry of initial divisions of human hematopoietic progenitors is altered only by the cellular microenvironment. Exp.Hematol. 2003;31:339-347. 226. Trentin JJ. Determination of bone marrow stem cell differentiation by stromal hemopoietic inductive microenvironments (HIM). Am.J.Pathol. 1971;65:621-628. 227. Metcalf D. Lineage commitment and maturation in hematopoietic cells: the case for extrinsic regulation. Blood. 1998;92:345-347. 228. Abkowitz JL, Persik MT, Shelton GH et al. Behavior of hematopoietic stem cells in a large animal. Proc.Natl.Acad.Sci.U.S.A. 1995;92:2031-2035. 229. Kirkland M A . A phase space model of hemopoiesis and the concept of stem cell renewal. Exp.Hematol. 2004;32:511-519. 230. Roeder I, Kamminga L M , Braesel K et al. Competitive clonal hematopoiesis in mouse chimeras explained by a stochastic model of stem cell organization. Blood. 2005;105:609-616. 60 231. Szilvassy SJ, Ragland PL, Miller CL, Eaves CJ. The marrow homing efficiency of murine hematopoietic stem cells remains constant during ontogeny. Exp.Hematol. 2003;31:331-338. 232. Dykstra B, Ramunas J, Kent D et al. High-resolution video monitoring of hematopoietic stem cells cultured in single-cell arrays identifies new features of self-renewal. Proc.Natl.Acad.Sci.U.S.A. 2006;103:8185-8190. 61 Chapter 2 Different In Vivo Repopulating Activities of Purified Hematopoietic Stem Cells Before And After Being Stimulated To Divide In Vitro With The Same Kinetics* Naoyuki Uchida*, Brad Dykstra*, Kristin J. Lyons, Frank Y. K. Leung, and Connie J. Eaves. Terry Fox Laboratory, British Columbia Cancer Agency and Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada. *N. Uchida and B. Dykstra contributed equally to this study. A version of this chapter has been published: Experimental Hematology 2003 Dec; 31(12):1338-47. 62 2.1. Introduction Life-long blood cell production is dependent on the continuous activation of a small population of multi-potent hematopoietic stem cells (HSCs) found primarily in the bone marrow (BM) 1' 2. This population is maintained by the ability of HSCs to divide and generate daughter cells that, at least half of the time, retain the same proliferative and differentiation potentialities as the original parent HSCs. Murine HSCs can be quantitated by limiting dilution assays that detect their longterm (>4 months) multi-lineage differentiation activity in competitively repopulated hosts3"5. The self-renewal activity of these cells in vivo has been well documented and, when individual HSCs are evaluated, remarkable heterogeneity in their sustained self-renewal responses is seen6. At a molecular level, very little is known about the regulation of the HSC self-renewal process. One approach has been to look for effects of various growth factor conditions on the preservation or loss in vitro of HSC function. However, this requires systems for tracking highly purified HSCs and their progeny in order to distinguish between changes in HSC numbers caused by their differentiation rather than their death. At the time the present studies were being pursued, several protocols for isolating murine HSCs at purities of 15-20% had been reported7"11 and recent studies using HSC-enriched cell populations had already yielded some evidence of growth factor-determined modulation of HSC self-renewal versus differentiation decisions in vitro12"14. Interestingly, these findings are paralleled by a lack of evidence for growth factor control of lineage choice in "normal" hematopoietic cells15"18. Taken together, these observations suggest a model of HSC regulation in which the stem cell state is maintained by intracellular regulators that 63 block the activation of multiple latent lineage-specific differentiation programs. Such regulators may be distinct from transcription factors that progressively restrict which differentiation program will ultimately be activated. Candidate genes regulating the maintenance of HSCs in an undifferentiated state would be Stat3]9 and HoxB420, whereas 21 22 • 23 candidate genes regulating differentiation would be Gata-1 ,PU.l and Pax5 . In other systems, extrinsic control of cell differentiation decisions is well established. In some of these, alterations in the duration of Gi have been implicated (reviewed in 2 4). Interestingly, a number of studies of primitive hematopoietic cells have suggested that their proliferative potential may be inversely correlated with their rate of cell cycle entry25"28. However, the link between control of cycling and differentiation in HSCs is complex, since the time taken for highly purified quiescent HSCs to complete a first cell division can be influenced by the combination of growth factors to which they are exposed13. In addition, the downstream checkpoint regulators involved appear to differ between HSCs and the early differentiated progenitor cells they generate29'30. During an initial series of experiments designed to analyze the properties of different fractions of side population (SP) cells isolated from adult mouse B M on the basis of their ability to efflux Hoechst 33342 (Ho)3 1, we noted that greater than 40% of the lineage marker-negative (lin") Rhodamine-123-negative (Rho") subset of SP cells have longterm multi-lineage in vivo repopulating ability when assessed using single cell transplants. We then used these highly purified HSCs to investigate further the relationship between the kinetics of HSC mitogenesis in vitro and the maintenance of HSC function. Our results show that 2 growth factor cocktails with the same potent 64 mitogenic effects on HSCs elicit marked differences in their execution of a self-renewal division. 2.2. Results 2.2.1. Characterisation of the in vivo repopulating activity of lin" Rho" SP cells from adult mouse BM Figure 2.1 A shows a typical FACS profile of the SP cells in suspensions of normal adult mouse B M (Pep3b-Ly5.i) after co-staining with Rho and antibodies against various lin markers. In this experiment, 0.2% of the starting B M population was contained within the SP fraction defined by a verapamil-sensitive ability to efflux Ho. Approximately 50% of the SP B M cells were lin but only 2 ± 1 % proved to be both lin" and Rho" (ie -0.004% of the starting B M cells, as shown in the lower left gate in Fig. 2. IB). In a preliminary experiment to assess the content of HSCs in the lin" Rho" subset of SP B M cells, 10 W 4 1 mice were each injected with -10 cells (10 FACS-sorted lin" Rho" SP events) from congenic Pep3b-Zv5.1 donors. Stable donor-derived repopulation (4-68%), including all major WBC lineages (B-, T-, and G/M cells) was seen in 9 of the recipients for at least 6 months post-transplant. The 10th mouse did not show any evidence of donor-derived (Ly5.1+) WBCs throughout this period. When these numbers (one negative mouse out of 10 injected with -10 test cells each) were used to derive a preliminary estimate of the HSC frequency in the lin" Rho" SP fraction using Poisson statistics and the method of 65 maximum likelihood, a value of 1 HSC per 4 lin" Rho" SP cells was obtained (67% CI = 1/3 to 1/7). A more definitive quantification of HSC frequency within the lin" Rho" SP fraction was then obtained directly by an analysis of 105 mice injected with single cells. As summarised in Table 2.1, 42 (40%) of these mice showed detectable levels of repopulation (37/94 W 4 1 and 5/11 B6 recipients). Forty of the 42 clones produced were multi-lineage (B + T + G/M) and 35 of these were maintained in vivo for at least 4 months. Two additional clones were classified as 'lineage restricted' because no T-cells were detected in them at any time point; both of these also had low levels of total repopulation (0.3 and 0.7%). An example of a long-lived multi-lineage clone is shown in Fig. 2.1C. As also shown in Table 2.1, retrospective analysis revealed a significantly reduced (2.4-fold, p=0.005, chi-square test) frequency of repopulation of the sublethally irradiated female W 4 1 recipients of single male cells in comparison to all other donor-recipient sex combinations (for which no host anti-graft effect would have been anticipated34). In the latter (sex antigen-compatible) recipients, 28/66 (42%) of the lin" Rho" SP cells met the most stringent criteria of longterm multi-lineage repopulation and 34/66 (52%) had some in vivo repopulating activity. The level of repopulation attained by individual multi-lineage clones at 4 months post-transplant varied widely, i.e., from 0.2% to 75% of the total number of circulating WBCs (Fig. 2.2 A, left panel), independent of the genotype (W 4 1 or B6) of the recipient (data not shown). Interestingly, unlike the frequency of mice that became repopulated, the level of repopulation subsequently achieved was independent of the donor-host sex 66 combination. Further follow-up on 9 of the in vivo clones for another year revealed 2 general patterns of clone development (Fig. 2.3 A). Approximately half (4/9) showed an initial large output of WBCs at 8 weeks that rapidly subsided to a detectable, but much lower output of WBCs over the next year. The other 5 clones were small or not yet detectable at 8 weeks and did not achieve maximum WBC outputs until at least 6 months post-transplant although they also showed a subsequent decline over the latter 8 months of follow-up. Interestingly, the distribution of the different lineages of mature WBCs contained within each of these clones was highly variable regardless of the initial kinetics of clonal expansion, and in some clones, marked fluctuations in these distributions were also seen over time. Three examples are shown in Fig. 2.3B. Femoral B M cells obtained by aspiration 6 months post-transplant from 3 highly repopulated (>50% Ly5.1 WBCs) mice that had each received a single lin" Rho" SP cell were transplanted into pairs of secondary recipients. All 6 of these secondary recipients showed sustained (>8 months) multi-lineage repopulation by the progeny of the initial cells injected into the 3 primary mice (Table 2.2). B M cells were aspirated from 4 of the secondary recipients (again at the 6-month post-transplant time point) and injected into 8 tertiary recipients. Six months later, 4 of these tertiary recipients showed high levels of multi-lineage repopulation by cells derived from the same original lin" Rho" SP cell (Fig. 2. ID). Another 2 of these tertiary mice died prior to analysis and the remaining 2 were negative. 67 2.2.2. Most of the HSCs are in the Rho" subset of the lin" SP fraction of adult BM To determine whether the properties of the HSCs present in the lin" Rho" SP fraction were representative of the majority of the HSCs in adult mouse B M , the remaining lin" SP cells were subdivided into a Rho* fraction (lower middle gate in Fig. 2.IB) and a Rho + fraction (lower right gate in Fig. 2.IB) as defined in the Methods. The frequency and hence total number of HSCs in each of these fractions was then determined by limiting dilution analysis. As shown in Table 2.3, no HSC activity was detected in the Rho+ subset of lin" SP cells and the frequency of HSCs in the Rho* fraction was -0.3%. Thus, although the lin" Rho" fraction contained only 2% of the SP cells, it was calculated to contain 88% of all HSCs present in the SP fraction (assuming all HSCs are lin"), with 11%) of the HSCs being found in the lin" Rho* fraction and <1% in the lin" Rho + fraction. Calculation of the average size of the clones produced by the minor population of Rho* HSCs and their content of B, T and G M cells showed no differences in any of these parameters with those obtained for the more prevalent Rho" HSCs: the average clone size at 4 months for the Rho* HSCs was 24% of the total WBC value versus 22 ± 4% for the Rho" HSCs and the average % B, T and G M cells within each clone was 33%, 49%, and 18% for the Rho* HSCs versus 37 ± 3 %, 46 ± 4 %, and 17 ± 3 % for the Rho" HSCs. 2.2.3. A minor fraction of lin" Rho" SP cells are LTC-ICs We next sought to determine the frequency of lin" Rho" SP cells that could be detected in vitro as LTC-ICs. In one experiment (A in Table 2.4), single lin" Rho" SP B M cells were sorted directly into a 96 well plate pre-loaded with irradiated feeder cells. In a second 68 experiment (B in Table 2.4), single cells of this phenotype were first sorted into a plate pre-loaded with medium only and then visually confirmed cells were transferred to wells with feeders. In both experiments, the frequency of LTC-ICs was less than the frequency ascertained for HSCs using the in vivo assay and the average LTC-IC frequency from analyses of a total of 123 single lin" Rho" SP cells was 24%. 2.2.4. Lin" Rho" SP cells cultured in SF and TPO or SF and IL-11 (± FL) demonstrate the same initial division kinetics but their progeny exhibit different in vivo repopulating activities We then took advantage of the high purity of HSCs in the lin" Rho" SP fraction to compare the mitogenic and self-renewal responses of HSCs to different growth factor cocktails. Single lin" Rho" SP cells were deposited using the FACS into separate wells of a 96-well plate containing either 10 ng/ml of SF and 100 ng/ml of TPO or 300 ng/ml of SF, 20 ng/ml of LL-11 ± 1 ng/ml of FL. Each well was first examined 4-16 hours later using an inverted microscope to identify those containing a single viable (refractile) cell. These cells were monitored every 4-8 hours thereafter to determine their kinetics of entry into a first division, and in some cases (with the SF+IL-11+FL and SF+TPO cocktails only), a second division. The first growth factor cocktail was chosen because the initial kinetics of activation and self-renewal behavior of HSCs purified by a different method had already been reported13. The other 2 cocktails were chosen because we had previously found that they stimulate optimal and equivalent HSC expansion in 10-day cultures of Sca-1+ c-kit+ lin" cells35. Fig. 2.4 shows the combined results of 6 experiments 69 in which the proliferation kinetics of a total of 424 single lin" Rho" SP cells cultured with either SF+IL-11+FL or SF+TPO were determined. No cell division (appearance of doublets) was seen in any of the cultures before 24 hours of incubation. However, over the next 24 hours -90% of the cells cultured in SF+IL-11+ FL and -75% of those cultured in SF+TPO had completed a first division with slight increases in both of these values over the following 12 to 16 hours. Both the first and second divisions were highly synchronous and occurred with similar kinetics in the 2 cocktails used for these studies. Thus -70% of the cells that divided within 60 hours did so between 29 and 41 hours after initiating the cultures and the second division followed -14 hours later. In 2 additional experiments, a total of 249 cells were cultured either in SF+IL-11+FL or in SF+IL-11 only. Similar assessment of the rate of doublet formation showed the results in these 2 conditions to be the same as those indicated in Fig. 2.4 (data not shown). Doublets that had been observed to appear from single cells in the previous 6 hours (candidate Gi cells) were injected as unseparated pairs of cells into irradiated recipient mice to determine whether at least one daughter cell had retained HSC activity. The timing of this selection was chosen to avoid the anticipated complication of a loss of in vivo engraftment ability that reversibly affects HSCs as they transit S/G2/M 3 6 ' 3 7 . Intriguingly, 14 (48%) of 29 recipients (11 of 20 W 4 1 and 3 of 7 B6 mice) injected with doublets generated in cultures containing SF+IL-11+FL showed longterm multi-lineage repopulation from these cells and an additional W 4 1 recipient showed longterm lineage-restricted (G/M plus B but no T cells) repopulation (Table 2.1). Moreover, the overall levels of repopulation (i.e. 0.6-55%, Fig. 2.2, center panel) spanned the same range seen 70 in recipients of the original lin" Rho" SP cells (Fig. 2.2, left panel). In marked contrast, only 4 (11%) of the 36 doublets generated in cultures containing SF+TPO showed HSC activity in injected mice (all W 4 1 ) and the WBC outputs that were obtained from these were all relatively low (i.e. 0.4-2.1%, Fig. 2.2, right panel). However, almost half (16/36) of the recipients in this second group were female and had been injected with male cells whereas there were no sex-incompatible combinations in the assays of doublets generated in response to SF+IL-11+FL. Therefore, the differential effect of these 2 growth factor cocktails would likely be overestimated unless the comparison was restricted to sex-matched combinations in the SF+TPO group. Nonetheless, when this was done, a significant difference was still seen (14/29 = 48%> of mice injected with doublets generated in SF+IL-11+FL vs 3/20 =15% of mice injected with doublets generated in SF+TPO, p = 0.008, chi-square test). 2.3. Discussion In this study we demonstrate the ease with which an essentially pure population of HSCs (>40% with longterm multi-lineage in vivo repopulating activity, -25% with LTC-IC activity) can be reproducibly obtained from normal adult mouse B M in a single sorting step. This is achieved by the selection of cells that actively efflux both Rho and Ho using the powerful double wave-length emission analysis of Ho-effluxing cells after stringent removal of all cells expressing B220, L y l , Macl, Grl and the antigen recognized by Terl 19. These are all properties assigned to HSCs more than a decade ago7'32 and used since by many investigators but not previously combined in this particular strategy. A 71 very similar approach involving the additional use of Sca-1 and pyroninY staining was recently reported38 but with a reported 3-fold lower HSC frequency in the recovered population. Goodell et al reported that the majority of HSCs in normal adult B6 B M have an SP phenotype31 and here we show that 88% of these have a Rho" phenotype as previously shown for HSCs that express Sca-133. Therefore HSCs isolated from adult mouse B M on the basis of their lin" Rho" Sp characteristics would be assumed to be representative of the majority of HSCs and display the same CD34", Thy-11 0, Sca-1\ c-+ • 8 11 kit phenotype used variably by others to also obtain highly enriched HSC populations " . This assumption is further supported by the finding that the frequency of lin" Rho" SP cells in unfractionated adult mouse B M (0.004%, of which half appear to be HSCs) corresponds closely to the reported frequency of HSCs (0.001%) measured functionally using the same transplantation procedure as that employed here39. P-glycoprotein, the product of the mdr-la/lb genes in the mouse, can efflux both Rho and H o 4 0 and is exclusively responsible for the Rho" phenotype of HSCs as shown by studies of mdr-la/lb'^ mice 4 1' 4 2. Bcrp/Abcg2 is another verapamil-sensitive A B C transporter that can also efflux Ho and appears to be responsible for the SP (Ho") phenotype of HSCs 4 1 ' 4 3 . Thus the differential expression of only 7 cell surface molecules appears sufficient to allow a functionally defined homogeneous population of HSCs to be isolated from normal adult mouse B M . Analysis of other subsets of the lin" SP population showed that -90% of HSCs were contained in the Rho" fraction with the remaining 10%) displaying a reduced ability to efflux Rho. Thus, the expression (or function) of p-glycoprotein appears to be less 72 stable than that of Bcrp/Abcg2 within the HSC population present in adult mouse B M . Furthermore, we could not discern any differences between the Rho" and Rho* HSCs in terms of the types or numbers of progeny they produced over a period of at least 4 months. These findings are consistent with the possibility that changes in the ability of HSCs to efflux Rho may be reversible and not necessarily indicative of a change in HSC potential. Reversible changes in other phenotypic properties of HSCs including expression of Mac 1, CD34 and CD38 have also been reported44"46. Evidence of biologic heterogeneity even within the lin" Rho" SP population of HSCs was seen when the kinetics of their output of mature WBCs was compared in single cell reconstituted mice monitored over a period of greater than 1 year. As noted by g others in a related study of mice repopulated with limiting numbers of HSCs , we observed 2 patterns of reconstitution. These patterns were distinguished by a rapid versus a delayed onset of clonal expansion and accompanying differences in the stability of the clone size initially attained. Interestingly, stability of the overall clone size did not extend to specific WBC lineages (B, T and G/M) whose numbers fluctuated in a highly clone-specific fashion. Thus, clonal stability can be a deceiving parameter when documented at the level of total mature WBCs. Further experiments will be required to determine if and how clone size may be related to HSC self-renewal activities post-transplant. Additional evidence of heterogeneity within the lin" Rho" SP fraction of B M cells is provided by the finding that HSCs were twice as numerous as cells detectable as LTC-ICs. Thus, although previous studies have formally demonstrated that some LTC-ICs can generate HSCs 4 7 ' 4 8 , indicating a degree of overlap between cells detectable by both assays, this 73 overlap is clearly not complete. Unfortunately, since neither of these functionally defined cell types accounted for greater than 50% of the lin" Rho" SP cells, the present data do not clarify whether all lin" Rho" SP LTC-ICs are also HSCs or whether they represent only a partially overlapping population. An unexpected finding that emerged from a retrospective analysis of sublethally irradiated female W 4 1 mice transplanted with single male lin" Rho" SP cells was a 2.4-fold decrease in the HSC frequency measured as compared to experiments in which myeloablated and/or sex-antigen-compatible hosts were used. A role of the weak Y-antigen has been described34 but not previously shown to affect HSC frequency determinations. The effect seen here may be attributable to the combined use of single cell transplants and sublethally irradiated W 4 1 recipients whose residual immunocompetence is likely to have been greater than in +/+ mice given a myeloablative dose of radiation plus a radioprotective graft of 105 normal B M cells. It is interesting to note that the distribution of clone sizes generated in the female mice repopulated with single male HSCs was not altered suggesting that the effect on HSC detection was restricted to an early period post-transplant. The finding that in sex antigen-compatible donor-host combinations, more than 50%) of the lin" Rho" SP cells could repopulate transplanted mice and more than 40% produced durable clones is remarkable in terms of the extraordinary HSC detection efficiency these numbers imply. This is of particular interest in view of the value of 10% recently documented as the 24-hour BM homing efficiency of unpurified HSCs measured using a second transplant to detect the fraction of HSCs that can be recovered from the 74 B M of primary recipients 24 hours after being transplanted . One possible explanation of the 4-fold discrepancy between the 2 values is that many BM-derived HSCs that re-home to the B M within 24 hours of being injected into the bloodstream are transiently compromised in their ability to do so again. Alternatively, some of the original HSCs injected may home initially to other tissue sites and only later return to the B M to proliferate and differentiate. Evidence that longterm repopulating HSCs have a high detection efficiency (at least 40%) in irradiated mice has also been recently reported using different methods of HSC purification and detection, providing additional support for a significant disparity between the proportion of injected HSCs detectable in the B M 24 hours later and the proportion that can produce a detectable clone over the subsequent 4 months49. This finding also indicates that freshly isolated populations in which only 10-20% of the cells demonstrate longterm in vivo HSC activity are not likely to be more than 20-40% pure. The actual biologic purity of phenotypically defined cell isolates has important implications for their use in gene expression analyses to identify candidate gene products responsible for maintaining the stem cell state of HSCs 5 0" 5 4 . These analyses will be confounded by the significant contamination of such populations with cells that do not possess HSC activity, particularly since the contamination is likely to consist of downstream, but closely related, cell types. Another comparative approach is now suggested by the observation here that HSC function appears to remain unaltered for periods of up to 16 hours in vitro in the presence of potently mitogenic growth factors. Given the likelihood that many changes in gene expression would be initiated during this 75 time, comparative analyses of such briefly cultured HSCs may provide a strategy to select for "stem cell" genes whose continued expression would not be perturbed. Finally, our studies provide evidence of a dissociation between the growth factor regulation of HSC mitogenesis and their self-renewal behavior since differences in the latter could be demonstrated in spite of identical cell division kinetics predicted by and confirmatory of those described by others13'55. Specifically, we found the frequency of first division doublets with HSC activity to be the same as the frequency of HSCs in the original lin- Rho- SP cells when they had been stimulated to divide in the presence of SF+IL-11+FL. Thus every HSC in mouse B M that has a lin" Rho" SP phenotype can be stimulated to execute a self-renewal division in the first cell cycle it completes in vitro when stimulated with this growth factor cocktail. In contrast, the second growth factor cocktail, consisting of a lower concentration of SF and a relatively high concentration of TPO produced first division doublets that showed a pronounced loss of HSC activity in spite of a similar HSC mitogenic response. This loss of HSC activity cannot be simply attributed to a decreased proportion of HSCs stimulated to divide since it was already shown by Ema et al 1 3 that only 25% of the cells that remain quiescent for the first 3 days (<15% of the initial total) retain HSC activity when cultured under these conditions. Therefore, it can be concluded that a significant proportion (>40%>) of the lin" Rho" SP cells that divide in the first 48 hours in SF+TPO were HSCs prior to culture and that more than half of them lose HSC activity by the time they complete a first division. Taken together, these findings confirm the ability of growth factor receptor-mediated signals to alter HSC self-renewal decisions12'14'56"58 and demonstrate that these 76 can be differently enacted within a single cell cycle without affecting the rate at which the cells progress through that cycle. Additional exploitation of this experimental model should allow further dissection of the mechanisms involved in regulating the stem cell state of primitive hematopoietic cells. 2.4. Materials and methods 2.4.1. Mice C57Bl/6J:Pep3b-Zy5.7 (Pep3b) mice were used as B M donors and C57B1/6J V/'/W41-Ly5.2 (W 4 1) or occasionally C57B1/6J-Z.v5.2 (B6) mice as recipients. All mice were bred and maintained in the animal facility at the BC Cancer Research Centre in microisolators and were provided with sterilized food and water. W 4 1 mice were irradiated with 400 cGy and B6 mice with 900 cGy of 1 3 7 Cs y-rays prior to being injected intravenously with one or more test cells as indicated. All B6 recipients were also injected with a radioprotective transplant of 105 normal B6 B M cells. Both types of recipients were then given water acidified with hydrochloric acid for the following 8 weeks. It has been previously shown that the frequency of repopulating cells detected in sublethally irradiated (400 cGy) W 4 1 mice and radioprotected B6 mice given 900 cGy is the same, as are the repopulation levels and lineage ratios obtained5'14'59. Therefore, we pooled the results from these two recipient types. To obtain B M cells for transfer to secondary and tertiary recipients, femoral B M aspirates were performed on anaesthetized mice as described60 and 3-10 x 106 cells were then injected per recipient. 77 2.4.2. Cell preparation and flow cytometry B M cells were suspended in Hank's balanced salt solution (HBSS, StemCell Technologies, Vancouver, BC, Canada) containing 2% fetal bovine serum (HF, StemCell Technologies) and, in most cases, the majority of the lin + cells were then removed immunomagnetically on a column (StemSep, StemCell Technologies) as described by the supplier. Cells were then suspended at 106 cells/ml in pre-warmed Iscove's modified Dulbecco's medium (LMDM) supplemented with 10 mg/ml bovine serum albumin , 10 p,g/ml insulin, and 200 p.g/ml transferrin (BIT, StemCell Technologies), 100 units/mL penicillin, 100 ug/mL streptomycin (both from StemCell Technologies), 10"4 M 2-p1-mercaptoethanol (2-ME, Sigma) and 0.1 (xg/mL of Rho (Molecular Probes Inc., Eugene, OR). After 30 minutes incubation at 37°C, the cells were washed with HF and resuspended at 106 cells/ml in the same medium without Rho. The cells were then immediately incubated with 0.1 ug/ml of Ho (Molecular Probes) for 90 minutes at 37°C, while a parallel aliquot was also incubated with 50 uM verapamil (Sigma Chemicals, St.Louis, MO). Cells were then washed, resuspended at 107 cells/ml in ice-cold HF plus 5% rat serum (Sigma) and 3 ug/ml of a Fc receptor blocking antibody (2.4G2) 6 1 followed by monoclonal antibody staining for 30 minutes on ice. Antibodies used for analysis or sorting were biotinylated anti-Macl (Ml/70: monocytes), anti-Grl (RB6-8C5: granulocytes), anti-B220 (RA3-6B2: B-lymphocytes), anti-Lyl (53-7.3: T-lymphocytes), and fluorescein isothiocyanate (FITC)-conjugated anti-CD45.1 (anti-Ly5.1, from clone A20, originally obtained from Dr. G. Spangrude, University of Utah and subsequently maintained in the Terry Fox Laboratory) and biotin-conjugated TER-119 and 78 allophycocyanin (APC)-conjugated anti-CD45 obtained from Becton Dickinson (BD, San Jose, CA). The lineage cocktail consisted of B220, Gr l , Macl, L y l , and TER119 antibodies. Cells stained with biotinylated antibodies were washed and incubated for 20 minutes on ice with Streptavidin-phycoerythrin (SA-PE, BD), then washed once with HF and once with HF plus 2 ul/ml propidium iodide (PI, Sigma). Cells were analysed using a FACSort or FACSCalibur (BD) with CellQuest software (BD) and sorted on a triple laser FACS Vantage equipped with UV, argon, and helium-neon lasers (BD). Within the lin" SP fraction, Rho" cells were defined as those showing less fluorescence in the FL-1 channel than that exhibited by >99.9 % of unstained cells, Rho* as those comprising -60% of the next most Rho + cells and Rho+ cells as the remaining lin" SP cells. In 2 experiments, CD45" cells were excluded using anti-CD45-APC staining to increase the clarity of the SP profile (data not shown). 2.4.3. Single cell liquid cultures Single lin" Rho" SP cells were sorted using the single cell deposition unit of the sorter into the individual wells of round-bottom 96 well plates preloaded with 200 u.1 of IMDM containing BIT, antibiotics, 2-ME and 40 (J-g/ml human low density lipoproteins (LDLs, Sigma). Just prior to transfer to 37°C either 300 ng/ml murine Steel factor (SF, StemCell Technologies) plus 20 ng/ml human interleukin (IL) -11 (Genetics Institute, Cambridge, MA) with or without 1 ng/ml human Flt3-ligand (FL, Immunex, Seattle, WA) as indicated, or 10 ng/ml SF plus 100 ng/ml human thrombopoietin (TPO, Genentech, Inc., 79 South San Francisco, CA) were added. After 4-16 hours incubation, each well was checked using an inverted microscope and all those that contained one and only one refractile (viable) cell were identified. These wells were subsequently re-examined at 4-8 hour intervals to determine the kinetics of a first and, in some experiments, also a second division (i.e., by tracking the appearance first of a doublet and then of 3 or 4 cells). 2.4.4. Assays for hematopoietic activity Long-term culture-initiating cells (LTC-ICs) were measured by assessing the in vitro colony forming cell (CFC) content of 4-week co-cultures of single sorted cells seeded onto previously established, irradiated mouse marrow feeder layers62. To assess single cells (or doublets) for their in vivo repopulating activity, single lin" Rho" SP cells were deposited using the single cell deposition unit of the sorter directly into the individual wells of a 96-well plate and then 2-16 hours later, the presence of a single cell in each well (and later doublets) was verified using an inverted microscope. In initial experiments, cells were harvested after the first 2-4 hours of incubation in wells pre-loaded with EVIDM plus BIT plus LDLs to allow those containing viable single cells to be identified. Subsequent results obtained with cells incubated for up to 16 hours with added growth factors showed the same frequency of repopulating cells as detected after 2 hours (data not shown) and hence all data for single cells incubated for 2-16 hours were pooled. After cells to be injected were identified, the entire plate was placed on ice until injection to prevent any cell cycle progression during that interval. For cells assayed in W 4 1 recipients, the entire volume of each well was harvested into a 1 ml syringe pre-80 loaded with 200 u.1 of saline and then taken up and down several times into the syringe before being injected intravenously. For cells to be assayed in B6 recipients, a 1 ml syringe was first loaded with 200 u.1 of medium containing 105 freshly isolated B6 B M cells. These were then combined with the contents of the well prior to being injected. To assess donor-derived repopulation, peripheral blood samples were collected 2 and 4 months post-transplant, and in some cases at later intervals up to 15 months post-transplant. Following ammonium chloride (NH4CI) treatment to lyse the red blood cells (RBCs), the white blood cells (WBCs) were stained with FITC-conjugated anti-CD45.1 for the detection of donor-derived (Pep3b) cells, together with biotinylated anti-B220, anti-Lyl, and anti-Grl plus anti-Macl, to detect B-cells, T-cells, and granulocytic/monocytic (G/M) cells, respectively. SA-PE was used as a secondary reagent to detect the biotin-labelled populations. A minimum of 5,000 viable cells were analyzed on a FACSCalibur (BD). Unless otherwise indicated, recipients were considered to be multi-lineage repopulated if their WBCs contained greater than 0.05% CD45.1+ events of each lineage (B220, Ly-1, and Grl/Macl). The gates that were used excluded >99.99%> of the events obtained when the WBCs from unmanipulated W 4 1 or B6 mice were stained with the anti-CD45.1, B220, Ly-1 and Grl/Macl , antibodies. The frequencies of HSCs in the limiting dilution experiments were calculated with L-calc™ software (StemCell Technologies) which uses Poisson statistics and the method of maximum likelihood3. The frequency of HSCs in the single-cell transplant experiments were calculated directly from the proportion of positive mice. In both cases, only mice 81 showing longterm (>1 months) multi-lineage repopulation were defined as positive and all other recipients were classified as negative. 82 Table 2.1. Frequency and type of repopulation in irradiated recipients of single lin" Rho" SP BM cells or their immediate progeny No. of repopulated mice* Sex of donors Total no. of Multi-lineage, Lin-restricted, Multi-lineage, Cells injected And recipients mice injected longterm longterm short term Single lin'Rho" SP cells M to M, F to F, F to M 66 28 1 5 M to F 39 7 1 0 SF + IL-l 1 + FL doublets M to M, F to F, F to M 29 14 1 0 M to F 0 0 0 0 SF + TPO doublets M to M, F to F, F to M 20 3 0 0 M to F 16 1 0 0 M = male, female = female *Multi-lineage longterm repopulation = 3 lineages (B220+, Grl/Macl + and Lyl +) of donor-derived WBCs at >16 weeks. Lin-restricted longterm repopulation = <3 lineages of donor-derived WBCs at >16 weeks. Multi-lineage short term repopulatiom= 3 lineages of donor-derived WBCs at 8 but not 16 weeks. Table 2.2. Single L i n - Rho- SP cells can generate sufficient daughter HSCs to repopulate secondary and tertiary recipients % Ly5.1 + donor repopulation Lineage distribution of Ly5,l+ cells, % in PB, 6 mo post-transplant B220 Grl/Macl Lyl Recipient Lin-Rho-SP cell A: Primary recipient A Secondary recipient Al Tertiary recipient Ala Tertiary recipient Alb Secondary recipient A2 Tertiary recipient A2a Tertiary recipient A2b* Lin-Rho-SP cell B: Primary recipient B Secondary recipient B1 Tertiary recipient B1 a Tertiary recipient Bib Secondary recipient B2 Tertiary recipient B2a Tertiary recipient B2b Lin-Rho-SP cell C: Primary recipient C Secondary recipient Cl Secondary recipient C2 62 83 21 47 81 30 16 54 3 0 died 10 0 died 68 1 1 43 20 58 42 18 34 43 17 47 0 42 0 48 17 23 22 35 11 9 48 13 16 82 15 0 3 26 40 35 45 31 50 34 53 41 1 38 0 40 0 49 60 37 Secondary recipients Al and A2 were transplanted with BM cells from primary recipient A, Bl and B2 from B, Cl and C2 from C. Tertiary recipients Ala and Alb were transplanted with BM cells from secondary recipient Al, A2a and A2b from A2, Bla and Bib from Bl, B2a and B2b fromB2. *Mouse died 5.5 months post-transplant, thus data shown for A2b is 5 months post-transplant. 84 Table 2.3. Frequency of HSCs in the Rho± and Rho+ subsets of lin-SP adult mouse BM cells Lin-SP Subset* % o f S P f Cells/mouse Positive mice/total HSC frequency! Rho+/- 29±9 1000 4/4 1/270 (1/200-1/380) 300 4/6 100 2/10 30 2/8 10 0/8 Rho+ 15±5 300 0/5 <l/2000 (<l/740-< 1/5400) § 100 0/6 * Defined by the gates shown in Figure 2. IB f Values represent ± standard error of the mean (SEM) from 3 experiments. J HSC frequencies were calculated from the proportion of repopulated mice measured at 6 months post-transplant. Values shown in brackets are the upper and lower limits defined by ± S EM (67% CI). § The value shown was calculated assuming one mouse given the maximum number of cells tested had been positive and serves to indicate the sensitivity of the assay in this instance. 85 Table 2.4. LTC-IC assay of single Lin- Rho- SP cells Experiment LTC-IC Frequency Colonies per L T C - I C * A 20/66 (30%) 44 ± 15 B 10/57 (18%) 29 ± 2 0 A+B 30/123 (24%) 39 ± 12 * Values represent the mean ± S E M from 3 replicates for each of exp'ts A and B. 86 Hoechst-red Rhodamine C D 4 5 . 1 Figure 2.1. FACS profiles of the lin" Rho subsets of freshly isolated mouse B M SP cells and their first and third generation clonal progeny produced in serially transplanted mice. (A) FACS profile of unseparated adult mouse B M cells after staining with Ho and analysis of the fluorescence at two wavelengths as described 3 1 . The gate indicates the SP fraction of cells whose high Ho efflux ability is blocked by verapamil (0.2% of total). (B) FACS profile of adult mouse B M SP cells concurrently stained with Rho and lin-PE. The gates indicate Rho" (2%, left gate) Rho* (-60%, middle gate), and Rho + (-38%, right gate) populations of lin" SP cells defined as described in the Methods. (C) FACS profiles of a highly repopulated recipient of a single lin" Rho" SP cell. (D) FACS profiles of a tertiary recipient of ~107 B M cells from a secondary recipient of ~10 7 BM cells aspirated from the mouse shown in (C). 87 100 c o 1 1 0 H Q. O Q. a> i_ TJ > "iZ a> • o c o a 0.1 * A A A A A A A A A A A • • A A * A • A A A A • A A • A A A * A A • • • • • A A • o A S i n g l e ce l l P o s t - d i v i s i o n in P o s t - d i v i s i o n (pre -d iv is ion) S F + IL-11 + F3L in S F + T P O Figure 2.2. Repopulation levels in mice transplanted with single lin" Rho" SP B M cells and doublets derived from them in vitro. This figure shows the proportions of donor-derived W B C s measured 4-6 months post-transplant in multilineage repopulated recipient mice injected with single lin"Rho" SP B M cells (triangles), or doublets originating from single lin'Rho" SP B M cells after being cultured in 300 ng/ml SF, 20 ng/ml IL-11, and 1 ng/ml F L (squares) or 10 ng/ml SF and 100 ng/ml T P O (circles). Open symbols represent the repopulation levels seen in female W 4 1 recipients of male cells; solid symbols are all other donor-recipient combinations. 88 20 40 60 20 40 60 20 40 60 20 40 60 20 40 60 Time post-transplant (wk) Figure 2.3. Longterm repopulation kinetics of single cell-transplanted mice. (A) Total donor-derived (Ly5.1 +) W B C s detected in 9 mice transplanted with a single lin" Rho" SP B M cell and followed for 15 months. On the left are shown the clones that exhibited a pattern of initial rapid expansion (peak size at 8 weeks post-transplant) followed by a decline to low but detectable levels throughout the subsequent period of observation. On the right are shown the remaining clones whose size did not reach a peak until >24 weeks post-transplant. (B) Distribution of B220 +(circles), Grl /Macl + (squares) , and Lyl +(triangles) W B C s in 3 randomly selected clones (indicated in panel A ) . 89 Incubation time (hours) Figure 2.4. Cell division kinetics of lin" Rho" SP B M cells cultured in 2 different cytokine cocktails. Single lin" Rho" SP B M cells were deposited into the individual wells of 96 well plates (using the cell sorter) in a total of 6 experiments and then the cells were cultured in serum-free medium containing either 300 ng/ml SF plus 20 ng/ml IL-l 1 and 1 ng/ml FL (circles: 4 experiments, 27 to 154 cells per experiment, 248 cells total), or 10 ng/ml SF and 100 ng/ml TPO (triangles: 4 experiments, 21 to 71 cells per experiment, 176 cells total). At 4-8 hour intervals, each well was examined and cells scored as having completed a first division (solid symbols) when 2 or more cells were first seen, and a second division (open symbols) when 3 or 4 cells were first seen. Each symbol shows the proportion of single viable cells initially identified in a given experiment that had divided (once or twice) by the time point shown. All data from all 6 experiments are shown. 90 2 . 5 . References 1. Bradford GB, Williams B, Rossi R, Bertoncello I. Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment. Exp.Hematol. 1997;25:445-453. 2. Cheshier SH, Morrison SJ, Liao X, Weissman IL. In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells. Proc.Natl.Acad.Sci.USA 1999;96:3120-3125. 3. Szilvassy SJ, Humphries RK, Lansdorp PM, Eaves A C , Eaves CJ. Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy. Proc.Natl.Acad.Sci.USA 1990;87:8736-8740. 4. Zhong R-K, Astle C M , Harrison DE. Distinct developmental patterns of short-term and long-term functioning lymphoid and myeloid precursors defined by competitive limiting dilution analysis in vivo. J.Immunol. 1996;157:138-145. 5. Miller CL, Eaves CJ. Expansion in vitro of adult murine hematopoietic stem cells with transplantable lympho-myeloid reconstituting ability. Proc.Natl.Acad.Sci.USA 1997;94:13648-13653. 6. Keller G, Snodgrass R. Life span of multipotential hematopoietic stem cells in vivo. J.Exp.Med. 1990;171:1407-1418. 7. Wolf NS, Kone A, Priestley GV, Bartelmez SH. In vivo and in vitro characterization of long-term repopulating primitive hematopoietic cells isolated by sequential Hoechst 33342-rhodaminel23 FACS selection. Exp.Hematol. 1993;21:614-622. 8. Spangrude GJ, Brooks D M , Tumas DB. Long-term repopulation of irradiated mice with limiting numbers of purified hematopoietic stem cells: in vivo expansion of stem cell phenotype but not function. Blood 1995;85:1006-1016. 9. Morrison SJ, Hemmati HD, Wandycz A M , Weissman IL. The purification and characterization of fetal liver hematopoietic stem cells. Proc.Natl.Acad.Sci.USA 1995;92:10302-10306. 10. Osawa M , Hanada KI, Hamada H, Nakauchi H. Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science 1996;273:242-245. 91 11. Wagers AJ, Sherwood PJ, Christensen JL, Weissman IL. Little evidence for developmental plasticity of adult hematopoietic stem cells. Science 2002;297:2256-2259. 12. Yonemura Y, Ku H, Hirayama F, Souza L M , Ogawa M . Interleukin 3 or interleukin 1 abrogates the reconstituting ability of hematopoietic stem cells. Proc.Natl.Acad.Sci.USA 1996;93:4040-4044. 13. Ema H, Takano H, Sudo K, Nakauchi H. In vitro self-renewal division of hematopoietic stem cells. J.Exp.Med. 2000;192:1281-1288. 14. Audet J, Miller CL, Rose-John S, Piret J, Eaves CJ. Distinct role of gpl30 activation in promoting self-renewal divisions by mitogenically stimulated murine hematopoietic stem cells. Proc.Natl.Acad.Sci.USA 2001;98:1757-1762. 15. Pharr PN, Ogawa M , Hofbauer A, Longmore GD. Expression of an activated erythropoietin or a colony-stimulating factor 1 receptor by pluripotent progenitors enhances colony formation but does not induce differentiation. Proc Natl Acad Sci U.S.A 1994;91:7482-7486. 16. Goldsmith M A , Mikami A, You Y et al. Absence of cytokine receptor-dependent specificity in red blood cell differentiation in vivo. Proc.Natl.Acad.Sci.USA 1998;95:7006-7011. 17. Metcalf D. Lineage commitment and maturation in hematopoietic cells: the case for extrinsic regulation. Blood 1998;92:345-348. 18. Stoffel R, Ziegler S, Ghilardi N et al. Permissive role of thrombopoietin and granulocyte colony-stimulating factor receptors in hematopoietic cell fate decisions in vivo. Proc.Natl.Acad.Sci.USA 1999;96:698-702. 19. Oh I-H, Eaves CJ. Overexpression of a dominant negative form of STAT3 selectively impairs hematopoietic stem cell activity. Oncogene 2002;21:4778-4787. 20. Sauvageau G, Thorsteinsdottir U, Eaves CJ et al. Overexpression of HOXB4 in hematopoietic cells causes the selective expansion of more primitive populations in vitro and in vivo. Genes Dev. 1995;9:1753-1765. 21. Pevny L, Simon M C , Robertson E et al. Erythroid differentiation in chimaeric mice blocked by a targeted mutation in the gene for transcription factor GATA-1. Nature 1991;349:257-260. 92 22. Scott EW, Simon M C , Anastasi J, Singh H. Requirement of transcription factor PU.l in the development of multiple hematopoietic lineages. Science 1994;265:1573-1577. 23. Urbanek P, Wang Z-Q, Fetka I, Wagner EF, Busslinger M . Complete block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking Pax5/BSAP. Cell 1994;79:901-912. 24. Zhu L, Skoultchi AL Coordinating cell proliferation and differentiation. Curr.Opin.Genet.Dev. 2001;11:91-97. 25. Ponchio L, Conneally E, Eaves C. Quantitation of the quiescent fraction of longterm culture-initiating cells (LTC-IC) in normal human blood and marrow and the kinetics of their growth factor-stimulated entry into S-phase in vitro. Blood 1995;86:3314-3321. 26. Berardi A C , Wang A, Levine JD, Lopez P, Scadden DT. Functional isolation and characterization of human hematopoietic stem cells. Science 1995;267:104-108. 27. Jordan CT, Yamasaki G, Minamoto D. High-resolution cell cycle analysis of defined phenotypic subsets within primitive human hematopoietic cell populations.' Exp.Hematol. 1996;24:1347-1355. 28. Brummendorf TH, Dragowska W, Zijlmans JMJM, Thornbury G, Lansdorp PM. Asymmetric cell divisions sustain long-term hematopoiesis from single-sorted human fetal liver cells. J.Exp.Med. 1998; 188:1117-1124. 29. Cheng T, Rodrigues N, Shen H et al. Hematopoietic stem cell quiescence maintained by p21 c i p l / w a f l . Science 2000;287:1804-1808. 30. Cheng T, Rodrigues N, Dombkowski D, Stier S, Scadden DT. Stem cell repopulation efficiency but not pool size is governed by p27kipl. Nature Medicine 2000;6:1235-1240. 31. Goodell M A , Brose K, Paradis G, Conner AS, Mulligan RC. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J.Exp.Med. 1996;183:1797-1806. 32. Spangrude GJ, Heimfeld S, Weissman LL. Purification and characterization of mouse hematopoietic stem cells. Science 1988;241:58-62. 33. Spangrude GJ, Johnson GR. Resting and activated subsets of mouse multipotent hematopoietic stem cells. Proc.Natl.Acad.Sci.USA 1990;87:7433-7437. 93 34. Simpson E. The role of H-Y as a minor transplantation antigen. Immunol.Today 1982;3:97-106. 35. Audet J, Miller CL, Eaves CJ, Piret JM. Common and distinct features of cytokine effects on hematopoietic stem and progenitor cells revealed by dose response surface analysis. Biotechnol.Bioeng. 2002;80:393-404. 36. Habibian HK, Peters SO, Hsieh CC et al. The fluctuating phenotype of the lympho-hematopoietic stem cell with cell cycle transit. J.Exp.Med. 1998;188:393-398. 37. Glimm H, Oh I, Eaves C. Human hematopoietic stem cells stimulated to proliferate in vitro lose engraftment potential during their S/G2/M transit and do not reenter G 0 . Blood 2000;96:4185-4193. 38. Huttman A, Liu SL, Boyd AW, Li CL. Functional heterogeneity within rhodamine 123l0 Hoechst33342lo/sp primitive hemopoietic stem cells revealed by pyronin Y. Exp Hematol. 2001;29:1109-1116. 39. Szilvassy SJ, Ragland PL, Miller CL, Eaves CJ. The marrow homing efficiency of murine hematopoietic stem cells remains constant during ontogeny. Exp.Hematol. 2003;31:331-338. 40. Shapiro AB, Ling V. Reconstitution of drug transport by purified P-glycoprotein. J.Biol.Chem. 1995;270:16167-16175. 41. Zhou S, Scheutz JD, Bunting KD et al. The A B C transporter Bcrpl/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat.Med. 2001;7:1028-1034. 42. Uchida N, Leung F Y K , Eaves CJ. Liver and marrow of adult mdr-la/lb'1' mice show normal generation, function and multi-tissue trafficking of primitive hematopoietic cells. Exp.Hematol. 2002;30:862-869. 43. Zhou S, Morris JJ, Barnes Y et al. Bcrpl gene expression is required for normal numbers of side population stem cells in mice, and confers relative protection to mitoxantrone in hematopoietic cells in vivo. Proc Natl Acad Sci U.S.A 2002;99:12339-12344. 44. Randall TD, Weissman IL. Phenotypic and functional changes induced at the clonal level in hematopoietic stem cells after 5-fluorouracil treatment. Blood 1997;89:3596-3606. 45. Sato T, Laver JH, Ogawa M . Reversible expression of CD34 by murine hematopoietic stem cells. Blood 1999;94:2548-2554. 94 46. Higuchi Y, Zeng H, Ogawa M . CD38 expression by hematopoietic stem cells of newborn and juvenile mice. Leukemia 2003;17:171-174. 47. Ploemacher RE, Van Der Sluijs JP, van Beurden CAJ, Baert M R M , Chan PL. Use of limiting-dilution type long-term marrow cultures in frequency analysis of marrow-repopulating and spleen colony- forming hematopoietic stem cells in the mouse. Blood 1991;78:2527-2533. 48. Fraser CC, Szilvassy SJ, Eaves CJ, Humphries RK. Proliferation of totipotent hematopoietic stem cells in vitro with retention of long-term competitive in vivo reconstituting ability. Proc.Natl.Acad.Sci.USA 1992;89:1968-1972. 49. Benveniste P, Cantin C, Hyam D, Iscove NN. Hematopoietic stem cells engraft in mice with absolute efficiency. Nat.Immunol. 2003;4:708-713. 50. Phillips RL, Ernst RE, Brunk B et al. The genetic program of hematopoietic stem cells. Science 2000;288:1635-1640. 51. Terskikh A V , Easterday M C , Li L et al. From hematopoiesis to neuropoiesis: evidence of overlapping genetic programs. Proc.Natl.Acad.Sci.USA 2001;98:7934-7939. 52. Ramalho-Santos M , Yoon S, Matsuzaki Y, Mulligan RC, Melton DA. "Sternness": transcriptional profiling of embryonic and adult stem cells. Science 2002;298:597-600. 53. Ivanova NB, Dimos JT, Schaniel C et al. A stem cell molecular signature. Science 2002;298:601-604. 54. Akashi K, He X, Chen J et al. Transcriptional accessibility for genes of multiple tissues and hematopoietic lineages is hierarchically controlled during early hematopoiesis. Blood 2003;101:383. 55. Reddy GP, Tiarks CY, Pang L et al. Cell cycle analysis and synchronization of pluripotent hematopoietic progenitor stem cells. Blood 1997;90:2293-2299. 56. Zandstra PW, Conneally E, Petzer A L , Piret JM, Eaves CJ. Cytokine manipulation of primitive human hematopoietic cell self-renewal. Proc.Natl.Acad.Sci.USA 1997;94:4698-4703. 57. Maguer-Satta V, Oostendorp R, Reid D, Eaves CJ. Evidence that ceramide mediates the ability of tumor necrosis factor to modulate primitive human hematopoietic cell fates. Blood 2000;96:4118-4123. 95 58. Bryder D, Ramsfjell V, Dybedal I et al. Self-renewal of multipotent long-term repopulating hematopoietic stem cells is negatively regulated by fas and tumor necrosis factor receptor activation. J.Exp.Med. 2001;194:941-952. 59. Trevisan M , Yan X-Q, Iscove NN. Cycle initiation and colony formation in culture by murine marrow cells with long-term reconstituting potential in vivo. Blood 1996;88:4149-4158. 60. Verlinden SF, van Es HH, van Bekkum DW. Serial bone marrow sampling for long-term follow up of human hematopoiesis in NOD/SCLD mice. Exp.Hematol. 1998;26:627-630. 61. Unkeless JC. Characterization of a monoclonal antibody directed against mouse macrophage and lymphocyte Fc receptors. J.Exp.Med. 1979;150:580-596. 62. Miller CL, Rebel VI, Lemieux M E et al. Studies of mutant mice provide evidence for alternate mechanisms capable of activating hematopoietic stem cells. Exp.Hematol. 1996;24:185-194. 96 Chapter 3 High Resolution Video Monitoring of Hematopoietic Stem Cells Cultured in Single-Cell Arrays Identifies New Features of Self-renewalf Brad Dykstra,1'2* John Ramunas,3* David Kent,1'2 Lindsay McCaffrey,1 Erin Szumsky,3 Liam Kelly, 3 Kristen Farn,3 April Blaylock,3 Connie Eaves1'2 and Eric Jervis3 'Terry Fox Laboratory, BC Cancer Agency, Vancouver, BC 2University of British Columbia, Vancouver, BC 3Department of Chemical Engineering, University of Waterloo, ON * B. Dykstra and J. Ramunas contributed equally to this study. A version of this chapter has been published: Proceedings of the National Academy of Sciences 2006 May 23; 103(21):8185-90 97 3.1. I n t r o d u c t i o n Time-lapse video imaging offers unique opportunities to determine how specific physical properties of individual living cells change with respect to one another over time and under different conditions. It has been used for more than half a century1"4 to study cell morphology during attachment and migration5'6, cell lifetimes7'8, growth9, death10'11, contact inhibition12, clonal heterogeneity13, and mitosis14. Software for extracting and analyzing cell lineage15 and morphology16 data from videos of cells also has an extensive history. Time-lapse studies of primitive hematopoietic cells have provided information about their cell membrane dynamics when co-cultured with stromal cells 1 7, 1 8 or fibronectin19, their kinetics of division20, their morphology and migration21, their localization in vivo22 and their simultaneous expression of different fluorescent proteins23. Here we asked whether time-lapse video imaging could be used to identify new behavioral traits of hematopoietic stem cells (HSC) with functionally validated long term multi-lineage repopulating activity in vivo. A number of groups have reported methods for obtaining highly purified (>20% pure) populations of HSCs from normal adult mouse bone marrow24"29. One of these methods involves isolating cells lacking surface markers characteristic of mature blood cells; (i.e., lineage marker-negative, or lin" cells) that efficiently efflux the fluorescent dyes, Rhodamine-123 (Rho" cells) and Hoechst 3334226. Efflux of Hoechst 33342 results in the appearance of a side population of cells (SP cells) in 2 dimensional plots of fluorescent events30. In mouse bone marrow (BM), the subset of lin"Rho"SP cells represents -0.004% of all the cells. Assessment of the blood cells generated in mice following injection of single lin"Rho"SP cells has shown that 40% of these cells can produce all blood cell types for many (>4) months26. Interestingly, most of the markers used to isolate HSC-enriched populations from steady-state mouse B M are not directly associated with HSC functional potential, since these phenotypes are 98 altered when HSC are activated or stimulated to divide31"35. In fact, very few stable properties of HSCs, apart from their defining developmental potential, have been identified. To search for new identifying properties of HSCs that are useful when they are proliferating, we developed a novel microwell array imaging system to visualize clones derived from individual HSCs over a 4-day period under conditions that support HSC self-renewal divisions ' ' . Each clone was then recovered and assayed for the presence of HSCs with long term multi-lineage in vivo repopulating activity. Video images of these assayed clones were then used to correlate visible characteristics of the cultured cells with those that had produced functionally defined daughter HSCs. 3.2. Results 3.2.2. Cell division kinetics of CD45midlin"Rho"SP cells determined by high resolution video tracking. In vivo transplantation assays of 83 freshly isolated CD45midlin"Rho"SP cells showed that 31% of these were functionally detectable HSCs (Figures 3.1A and 3.1C). Additional cells of this phenotype were shipped overnight from Vancouver, BC to Waterloo, ON and then 67 of these were loaded into the individual wells of 3 silicone microwell array chambers containing serum-free medium (SFM) and 300 ng/ml murine Steel factor (SF) plus 20 ng/ml human interleukin-11 (IL-l 1) and 1 ng/ml human flt3-ligand (FL). The arrays were incubated at 37°C for 4 days and imaged using a 5x objective at 3-minute intervals throughout this period to allow the morphology and behavior of each cell and its progeny to be recorded and tracked (Figures 3.2B and 3.2C). This sequence of images provided precise information about the timing of every cell division that occurred (n=679) and hence the duration of each intervening cell cycle. From these data we constructed pedigree diagrams for each of the 67 clones generated (Figure 3.2D). The average 99 times to the first, second and third division determined for all cells that completed these cycles were 39.5±7.6, 18.2±5.2 and 15.8±3.8 hours, respectively (Figure 3.2E). Sister cells (i.e., paired progeny derived from the same parental cell) divided with remarkable synchrony throughout the culture period (Figure 3.2F). Transplantation data were obtained on 61 individually harvested clones and the results of these showed that 17 of the 61 clones (28%) contained HSCs. This finding demonstrated that a high proportion of the input HSCs had executed at least one self-renewal division during imaging, in spite of the overnight transit of the cells before and after the 4-day culture period (Figures 3.IB and 3.ID). 3.2.3. Association of smaller clone sizes and longer cell cycle times with retention of H S C activity. Retrospective analysis showed that the 4-day clones containing HSCs were significantly smaller than those in which HSCs were not detected (log2 average size 8.8±1.1 cells, n=17 versus 17.6±1.2 cells, n=44, p<.005, Figure 3.3C). This difference in clone size corresponds to an average difference of one fewer cell generation (3.1 versus 4.1) in the clones in which HSC self-renewal divisions were subsequently shown to have occurred. The average cell cycle time of cells that completed 1, 2, and 3 divisions was also significantly longer for all 3 cycles (p<.005) in the HSC-containing clones as compared to those without detectable HSCs (Figure 3.3A). The best discrimination between these 2 types of clones was obtained by combining all 3 cell cycle times (Figure 3.3B). Note that for these calculations, we excluded clones in which 3 divisions or more did not occur although there were only 7 such clones in all. Interestingly, neither of the 2 cells that remained viable, but did not divide during the 4-day imaging period, displayed repopulating activity when subsequently injected into mice. Also, clones containing HSCs had significantly (PO.05, one tailed t-test) greater asymmetry between the cell cycle times of the 100 daughters of the clone founder than clones in which HSCs were not detected (see Supporting Information for details). 3.2.4. Association of a late prevalence of cells with uropodia with loss of HSC activity. We also looked for other features of cell behavior in clones that might be associated with their retention (or loss) of HSC activity including the acquisition and loss of different types of cellular projections. During the first 14-18 hours only 6 of the 67 wells (-9%) contained cells with lagging posterior projections (uropodia) although 45 (-67%) contained cells with other cytoplasmic extensions. At later times, uropodia became more prevalent, particularly in some clones (Figures 3.4A and 3.4B). Filopodia (long, thin projections) were observed with high resolution imaging (20X and 40X objectives) on most cells at the start and end of the period of monitoring, but these were not consistently visible in the lower resolution images collected every 3 minutes (using the 5X objective) and were therefore not included in the present analysis. When cells were scored for the presence or absence of uropodia during the final 12 hours of the 4-day culture period, the majority of the cells in 25 of the clones (-37%) contained uropodia. A significant association (p<0.05) with the presence or absence of HSC activity was only found in the latter case, where none of the clones with a late predominance of cells with uropodia were found to contain HSCs. 3.2.5. Identification of a combination of monitored parameters that are predictive of HSC self-renewal divisions. We then asked whether combining 2 different parameters of cell behavior in the clones (time to 3r d division and lack of uropodia on the fourth day of culture) would identify HSC-containing clones more efficiently than either of these parameters on its own. To apply the first parameter, 101 we chose a minimal cell cycle time that included all HSC-containing clones and excluded a maximum number of non-HSC-containing clones. To define such a cutoff in a way that could be applied to other data sets, we set it equal to the mean time to the 3rd-division measured on the entire data set minus 0.5 standard deviations (SD). For the data set shown in Figure 3, this value was 67.23 hours. This value was then used as a gate to subdivide clones into 2 groups; those clones in which the first cell to reach a 3r d mitosis did so in less than 67.23 hours, and those in which the first cell to reach a 3r d mitosis took longer than this threshold period (see Methods and Table 3.2 for additional information). We then further subdivided the clones into 2 groups based on whether or not the majority of the cells within the clone displayed uropodia at any point during the final 12 hours of culture. Selection of clones in which the time to the 3 r d division was >67.23 hours and >^0% of the cells exhibited uropodia in the last 12 hours of culture identified clones that contained HSCs at a 2.26-fold higher frequency than in the original 61 clones analyzed (Figure 3.4D, Table 3.1). The robustness of these criteria to identify HSC-containing clones was then tested by applying them to similar data acquired from 2 independently executed experiments of the same design. As in the first experiment, maintenance of HSC activity was evident in the clones analyzed after culturing single CD45midlin"Rho"SP cells for 4 days (Figure 3.2E and 3.2F). Importantly, application of the same criteria identified in the first experiment to the data obtained from the 2 later experiments allowed the HSC-containing clones to again be predicted with a 2 to 3-fold increased efficiency (Table 3.1). 3.3. Discussion Here we describe a novel time-lapse video monitoring system that allows high-resolution real-time tracking of cells in multiple expanding clones in vitro to be coupled with functional 102 assays of the individually harvested clones at the end of the monitoring period. These unique features have made it possible to address new questions about the biology of HSCs that previously have not been amenable to investigation. The objective of our study was to identify new parameters that might be associated with HSC self-renewal divisions in vitro. From a survey of numerous cell features (see Supporting Information for details of other features considered), we identified 2 that each showed a significant association with clones containing HSCs after 4 days of culture: a prolonged cell cycle time measured over 3 divisions and a reduced proportion of progeny with uropodia at any time between 84 and 96 hours of culture. In combination, these parameters identified all of the HSC-containing clones in each of the 3 experiments performed and consistently enhanced the identification of HSC-containing clones 2-to 3-fold independent of the starting purities of the HSCs tested (see controls in Table 3.1) or other inter-experimental variations likely to have occurred. This suggests that these biomarkers are indeed robust features of mouse B M HSCs. These findings extend the results of previous studies that correlated longer cell cycle times of primitive hematopoietic cells of both mouse38 and human3 9'4 0 origin with the retention of their primitive cell properties. The experiments described here have taken this line of investigation a step further through the use of a more highly purified HSC starting population, a higher spatial-temporal resolution monitoring system, and functional assessment of the HSC activity retained (or not) by each tracked clone. In this way, a link between HSC cell cycle time and their self-maintenance in culture could, for the first time, be definitively established. Schroeder41 has recently described a complementary computer-aided culture and time-lapse imaging system that he has used to describe the generation of HSC-derived clones on stromal cell feeder layers but without data for HSC activity in the clones produced. We anticipate that 103 further use of both systems will provide valuable new insights into how primitive hematopoietic cells interact with external cues to regulate their self-renewal and differentiation potential. Multiple studies have previously associated a variety of cell projections with primitive hematopoietic cells42"46. In particular, Frimberger et al. 4 7 observed several types of projections on the leading edge and periphery of cells in HSC-enriched populations using high-speed optical-sectioning microscopy and inverted fluorescent video microscopy. Giebel and colleagues48 have described the appearance of uropodia at the rear pole of human CD34 + cells. Here, we found that the late presence of uropodia was negatively associated with retained HSC activity. Clearly, attention to the criteria used to define different categories of projections as well as the particular culture conditions used and the time in culture at which they are assessed will be important to future investigations of whether these projections play a role in HSC biology. Although our approach is potentially applicable to any HSC-containing population, all candidate biomarkers would need to be screened again if a different isolation strategy were used, because the non-HSC component of such populations would likely be different. A strength of the approach used here is that it can be adapted to any source of HSCs or HSC isolation strategy because it makes no assumptions about the biological homogeneity of the cells being monitored. The most useful biomarkers are, however, those that can be directly linked to the defining developmental properties of HSCs. The technology and experimental design described here thus represent an important advance in the definitive identification of such features. In addition, the system we have described has the flexibility of allowing specific cells with tracked histories to be removed by micromanipulation at any time point and then assayed or analyzed. Cells containing reporter genes or labeled surface or internal components will further broaden the scope of cellular events that can be monitored. We thus anticipate increasing application of this powerful technology to many areas of cell biology. 104 3.4. Materials and Methods 3.4.1. Mice Bone marrow donors were 8-12 week-old C57B1/6J-Zy5.7 or -Ly5.2 mice. Transplant recipients were Zy5-congenic C 5 7 B 1 / 6 J - ^ V ^ / mice sublethally irradiated with 360 cGy X-rays at -350 cGy per minute. Peripheral blood (PB) was collected at 4, 8, 12, and 16 weeks post transplant and the leukocytes were then stained with antibodies for donor and recipient CD45 allotypes plus lymphoid and myeloid specific markers. Long term repopulation was defined as the detection of donor-derived leukocytes at >1% levels in the PB for at least 16 weeks. Multi-lineage repopulation was defined as the detection of >1% of both donor type lymphoid and myeloid cells at 4, 8, 12, and/or 16 weeks after transplantation. Evidence of both long term and multi-lineage repopulation in the same recipient was used to infer that ^ HSC had been injected. For further details, see Supporting Information. 3.4.2. Purification, culture and shipment of CD45midlin"Rho"SP cells Cell purification was performed as previously described26, with minor modifications. See Supporting Information for details. For controls, single CD45midlin"Rho"SP cells were sorted into the individual wells of a round-bottom 96-well plate containing 100-200 /xl of serum-free media (SFM), visually confirmed, and were then injected individually directly into sublethally irradiated C57Bl /6J-^ / / I^ / recipients. The, CD45m i dlmRho"SP cells to be imaged were sorted and collected into a 1.4 ml Eppendorf tube pre-filled with SFM in Vancouver, BC, and then shipped via overnight courier (18-22 hours) at 4°C to the University of Waterloo in Ontario. On arrival, the cells were then warmed to 25°C and 300 ng/ml murine SF (StemCell Technologies, Vancouver, BC), 20 ng/ml human IL-l 1 (Genetics Institute, Cambridge, MA), and 1 ng/ml human FL (Immunex, Seattle, WA) added to the medium. Single CD45midlin"Rho"SP cells were 105 then micromanipulated into the individual microwells of an array chamber (Figure 3.2A, prepared as described below) which was then placed at 37°C in a humidified, 5% CO2 atmosphere and imaged every 3 minutes using phase contrast optics. The time of cytokine addition was set as zero hours of culture time for all experiments. At the end of the 4 days of culture, the clones in the arrays were harvested individually, placed into separate 0.65 ml microcentrifuge tubes and shipped via overnight courier at 4°C to Vancouver, where the cells in each tube were then resuspended and injected into individual sublethally-irradiated C57B1/6J-vf'/W4' recipients. 3.4.3. Videotracking system Cells were cultured in custom-designed microwell chambers. Briefly, these were constructed by applying silicone gel to a glass coverslip to form a film approximately 20 jim thick and a 100-/xm wide glass scraper was then used to machine 2 sets of perpendicular rows to form the array wells before the gel set (Figure 3.2A). A glass tube was then affixed around the array to form a reservoir to contain the culture medium. To deposit the cells within the array, the entire reservoir was filled with 1 ml of medium containing approximately 50 cells which were then allowed to settle. Each of the 40 microwells was then loaded with a single cell by repositioning the settled cells using a glass micropipet guided by a 3-axis motorized micromanipulator. The micropipets were made from capillary tubes (Drummond, 3-000-203-G/X) using a vertical pipet puller (Kopf, Model 720) and cut with a single-crystal diamond-tipped glass etcher to give an opening 15-30 fim wide. Images were obtained on a Zeiss Axiovert 200 microscope equipped with phase contrast optics and a Sony XCD-SX900 digital camera. Cells were exposed to light only during imaging. Each cell in each image of the approximately 1850-image time courses was scored for morphological characteristics, location, and parentage using human-assisted custom cell tracking 106 software that generated pedigree diagrams with other data superimposed upon them for visulaization. Data from these diagrams were then imported into standard analysis programs (Excel, M A T L A B , Prism) to test correlations between candidate biomarkers and HSC activity . 3.4.4. Purification of CD45midlin"Rho"SP cells Bone marrow cells were collected by flushing femurs and tibias with Hank's balanced salt solution containing 2% fetal bovine serum (HF, StemCell Technologies, Inc, Vancouver BC, Canada). Following lysis of the red blood cells with a buffered ammonium chloride solution (StemCell), the majority of the cells expressing CD5, CD1 lb, CD45R, 7-4, Ly-6G, and glycophorin A were removed immunomagnetically (using the EasySep™ negative selection protocol and the biotinylated Mouse Hematopoietic Progenitor Cell Enrichment Cocktail from StemCell as described by the supplier). Cells were then washed, spun, and re-suspended at 106 cells/ml in pre-warmed Iscove's modified Dulbecco's medium (IMDM) supplemented with 10 mg/ml bovine serum albumin, 10 u.g/ml insulin, and 200 u.g/ml transferrin (BIT, StemCell) and 0.1 uM 2-mercaptoethanol (Sigma) (collectively called serum-free medium, SFM) as well as 0.1 (ig/ml of Rho (Molecular Probes Inc., Eugene, OR). After 30 minutes incubation at 37°C, the cells were washed with HF, resuspended at 106 cells/ml in the same medium minus Rho, and immediately incubated with 5 p:g/ml of Hoechst 33342 (Sigma) for 90 minutes at 37°C. Cells were then washed with ice-cold HF and resuspended at 107 cells/ml in ice-cold HF followed by staining for 30 minutes on ice with additional biotinylated EasySep™ Mouse Hematopoietic Progenitor Cell Enrichment Cocktail (StemCell) and anti-CD45-allophycocyanin (APC, BD Biosciences). After washing the cells once with ice-cold HF, they were incubated on ice with phycoerythrin (PE)-conjugated streptavidin (BD Biosciences) for visualization of residual lin+ cells. The cells were then washed once with ice-cold HF and resuspended in ice-cold HF plus 1 107 u l / m l p r o p i d i u m iodide (PI, S igma) . C e l l s were sorted i n V a n c o u v e r us ing a F A C S V a n t a g e equipped w i t h U V , argon, and he l ium-neon lasers ( B D Biosc iences ) . U s i n g gates set to exclude P F events, cel ls w i t h a C D 4 5 m i d l i n ~ R h o " S P phenotype were selected (Figure 3.5). 3.4.5. In vivo repopulation assay F o r in vivo assays o f freshly isolated single (control) cel ls , the entire v o l u m e o f each w e l l containing a v i sua l l y conf i rmed single c e l l was harvested into a 1 m l syringe preloaded w i t h 300 /xl o f saline and then taken up and d o w n several t imes into the syringe before be ing injected intravenously. F o r in vivo assays o f 4-day clones shipped i n microcentr i fuge tubes, the entire v o l u m e was loaded into a 1 m l syringe and injected intravenously. P B samples were col lected from the ta i l v e i n o f injected mice 4, 8, 12, and 16 weeks later. F o l l o w i n g lys is o f the red b lood cells w i t h a m m o n i u m chlor ide (S temCel l ) , the leukocytes were resuspended i n H F plus 3 ug /ml o f an ant i -mouse I g G receptor ant ibody for 20 minutes to m i n i m i z e nonspecif ic staining. C e l l s were then stained w i t h antibodies for donor and recipient C D 4 5 al lotypes ( a n t i - C D 4 5 . 1 - A P C and ant i -CD45.2- f luoresce in isothiocyanate [F ITC] ) plus a n t i - L y 6 g - P E / a n t i - M a c l - P E for m y e l o i d cel ls , or a n t i - B 2 2 0 - P E for B - c e l l s , or a n t i - C D 5 - P E for T-ce l l s ( C D 4 5 . 2 - F I T C pur i f ied and conjugated i n the Ter ry F o x Laboratory, V a n c o u v e r B C ; C D 4 5 . 1 - A P C f rom eBiosciences , San D i e g o C A ; a l l others f rom B D Biosc iences , San Jose C A ) . T o calculate levels o f donor-derived leukocytes i n the P B o f the recipient mice , events negative for C D 4 5 . 1 and C D 4 5 . 2 or posi t ive for both C D 4 5 . 1 and C D 4 5 . 2 were excluded, and the % o f cel ls co-s ta ining w i t h each lineage marker and the donor C D 4 5 allotype was determined (Figure 3.6). L o n g term repopulat ion was defined as the persistence o f donor leukocytes at >1% for at least 16 weeks post-transplantation. Mul t i - l i neage repopulat ion was defined as the detection o f both donor type l y m p h o i d and m y e l o i d cel ls at levels >1% o f a l l l y m p h o i d or m y e l o i d cells at any point after transplantation. 108 Evidence of both long term and multi-lineage repopulation in the same recipient was used to infer that at least one HSC had been injected. 3.4.6. Calculation of the time to a 3 r d division exclusion parameter We chose a cutoff time that in the first experiment included all HSC-containing clones and excluded a maximum number of non-HSC-containing clones. To make this cutoff time applicable to future data sets we defined it as 0.5 SD less than the mean time to 3 r d division for the entire data set. Application of this criterion to the clones in the first experiment allowed all HSC-containing clones to be selected (Figure 3.7). For clones in which one or more cells did not complete a 3 r d division, estimation of the missing values was performed as follows: In each clone where all 4 possible 3 r d divisions occurred, the ratio of the average 3 r d cycle time to the average 2 n d cycle time was calculated. The average of these ratios was then multiplied by the observed 2 n d cycle time to obtain estimated 3 r d cycle times for clones where all 4 3 r d divisions did not occur. For these clones, estimated times to 3 r d division were then calculated by adding the estimated time to 3 r d division to the observed times to 1s t and 2 n d division. In clones that completed a 2 n d division but not a 3 r d division, either the estimated time to 3 r d division or the total culture time, whichever was greater, was assigned to each cell. Note that this estimation procedure cannot be applied to clones containing less than 3 cells. For 2-cell clones, the estimated time to 3 r d division was calculated as follows. In each clone where all 4 possible 3 r d divisions occurred, the ratio of the average 2 n d plus 3 r d cycle time to the 1s t cycle time was calculated. The average of these ratios was then multiplied by the observed 1s t cycle time and added to the observed 1 s t cycle time to obtain an estimate of the time to 3 r d division for the 2-cell clones. 109 3.4 .7 . Image tracking methodology and candidate biomarker testing The location, morphological characteristics, and parentage of each cell in each image of the ~1850-image time course sequences were documented and tracked using custom software (available from the authors on a collaborative basis). This software presents a table for each image in the time course that allows multiple characteristics to be assigned a coded score. Cells were scored as having a projection i f one was visible for at least 3 consecutive images (spanning at least 6 minutes) in the video. The software can then derive values such as cell speed and cell-cell distance from the scored characteristics and superimpose the results on pedigree diagrams. These data were used to identify candidate biomarkers (see Table 3.2 for examples of biomarkers considered). Additional biomarkers are still being investigated. Data were imported from the custom software into standard analysis programs (Excel, M A T L A B , Prism) to test correlations between candidate biomarkers and clones identified from the in vivo assays as HSC-containing or not. Candidate biomarkers with significantly different scores between H S C - and non-HSC-containing clones were used to develop criteria that would exclude as many non-HSC clones as possible (high selectivity) while retaining most or all HSC-containing clones (high sensitivity). Each criterion had a parameter that could be adjusted to achieve these goals. In the case of the uropod criterion, the maximum % of cells with uropodia that a clone at late times contained was the most discriminating criterion. In the case of the time to a 3 r d division criterion, it was the factor by which we multiplied the SD before subtracting it from the mean that proved most useful. A 3D surface plot in which the fold-enrichment was plotted as a function of both criteria was then generated and the peak in this plot used to identify the optimal parameter values for the two criteria (>50% of cells with uropodia, and a factor of 0.5, respectively). Regions of the plot in which not all H S C wells passed were excluded. 110 Table 3.1 - Application of selection criteria developed from the 'training set' of data to results from 2 additional experiments. TRAINING SET Fresh Cultured (control) Non-gated Gate T Gate 2T Gates 1,2 Number of HSC containing clones 6/11 17/61 17/36 17/40 17/27 Percentage 55 28 47 43 63 Fold Enrichment 1.00 1.69 1.53 2.26 TEST SET #1 Fresh Cultured (control) Non-gated Gate 1 Gate 2 Gates 1,2 Number of HSC containing clones 3/16 4/24 4/13 4/18 4/12 Percentage 19 17 31 22 33 Fold Enrichment 1.00 1.85 1.33 2.00 TEST SET #2 Fresh Cultured (control) Non-gated Gatel Gate 2 Gates 1,2 Number of HSC containing clones 4/18 5/73 5/42 5/35 5/27 Percentage 24 6.8 12 14 19 Fold Enrichment 1.00 1.74 2.09 2.72 * Excluding clones containing one or more cell with a cumulative time to a 3r d division faster than the mean minus 0.5 SD. * Excluding clones containing >50% of cells with uropodia during the final 12 hours of culture. I l l T a b l e 3.2: C a n d i d a t e b i o m a r k e r s cons ide r ed in the search for those with the most power to distinguish HSC-containing from non HSC-containing clones. Score % that pass gat Criterion (per well unless noted) Sub-criterion HSC wells Non-HSC wells Significant? (PO.05) Gate1 HSC Non HSC Mean length of cell cycle (hours) 2 n d 21.1+/-1.4 16.9+/-0.6 Yes Well mean > experiment mean - (0.45 x SD) 100 57 - ^ r d 17.9+/-0.9 15.2+/-0.5 Yes Well mean > experiment mean - (0.9 x SD) 100 75 Mean cumulative time to division (hours) 1s t 44.0+/-1.9 37.9+/-1.0 Yes Well mean > experiment mean - SD 100 84 2 n d 65.1+/-2.4 53.3+/-1.9 Yes Well mean > experiment mean - (0.75 x SD) 100 73 ^ r d 82.7+/-2.2 70.4+/-1.7 Yes Well mean > experiment mean - (0.5 x SD) 100 43 Clone size (cells) 48 h 1.6+/-0.2 2.2+/-0.2 Yes Clone size < 3 100 75 72 h 4.1+/-0.3 7.7+/-0.7 Yes Clone size < 7 100 41 96 h 9.5+/-1.0 24.9+/-2.9 Yes Clone size < 17 100 39 Founder divided - 17/17 (100%) 42/44 (95%) No 2 Clone founder divided 100 95 50% or more cells with uropodia First 12 h 0/17 (0%) 6/44 (14%) No 2 - - -Last 12 h 0/17 (0%) 30/44 (68%) Yes 50% or fewer cells with uropodia 100 32 Any pseudopodia3 on > 50% of cells in first 12 h - 11/17 (65%) 31/44 (70%) No - - -1 or more deaths - 1/17 (6%) 1/44 (2%) No - - -Difference between daughters of founder (i.e. asymmetry in first generation) Cycle time (hours)4 2.4+/-0.6 1.2+/-0.2 Yes Difference > 0.25 x SD 82 66 Mean migration speed (microns/hour) 2.8+/-0.9 2.8+/-0.8 No Mean migration speed of founder in first 12 h (microns/hour) 11.0+/-2.4 11.6+/-1.1 No The gates were designed to pass all HSC wells while excluding as many non-HSC wells as possible, except in the case of the difference between cell cycle times of the daughters of the founder, in which case this was not possible due to a few very symmetric sisters in the HSC well population. In this case the gate was designed to maximize the fold-enrichment. The data suggest a correlation but n may not be sufficiently large. 3 Includes lamellipodia, uropodia, and processes that generally radiated from the cell bodies in directions that were not consistently from the leading nor trailing edges. Resolution at 5x magnification was too low to detect filopodia and so they were not scored, but were visible at higher magnification. 4 Similar results were obtained by taking the ratio of the cell cycle times instead of the difference. Mean values are ± SEM, n=61. 112 A B C D E F Figure 3.1 - In vivo repopulation characteristics of single CD45midlin"Rho"SP cells or their clonal progeny. A. Representative FACS profiles from a mouse repopulated with a single CD45midlin~Rho~SP cell. B. Representative FACS profiles from a mouse repopulated with the in vitro progeny of a single CD45m,dlin"Rho"SP cell cultured for 4 days in an array chamber. C. Proportion of PB leukocytes produced from a single freshly isolated CD45rnidlin"Rho"SP cell transplanted 16 weeks previously. Filled circles identify mice in which the level of donor-type leukocytes indicated that at least one HSC was present in the clone injected (>1% donor-type leukocytes at 16 weeks and >1% of both lymphoid and myeloid cells present at some point during the period the mice were serially monitored). Open circles represent mice in which some donor leukocytes could be detected at 16 weeks (>0.1%) but these were either <1% of the total and/or had not shown all lineages to have been included in the cells produced. Mice showing no (<0.1%) repopulation by donor-type cells are not shown. Horizontal bars show the geometric mean size of the clones produced in vivo from the injected HSCs of HSC-containing clones. D, E, and F. Proportion of donor-type leukocytes seen in the PB of mice injected 16 weeks previously with a 4-day clone derived from a single CD45m,dlin"Rho"SP cell in the first imaging experiment (D) and in the second 2 experiments (E and F). 113 Figure 3.2 - Description of the high resolution time lapse array system and representative culture results. A. A digital image of an array showing 40 silicone microwells each capable of holding up to -150 cells that can be tracked simultaneously. B. Higher power view of a representative well containing one CD45midlin"Rho~SP cell suspended in SFM plus 300 ng/ml SF, 20 ng/ml IL-11 and 1 ng/ml FL. C. Close-up of the well shown in B after 4 days at 37°C. D. The pedigree diagram of the clone that developed in the well shown in C, illustrating the precision with which sequential cell divisions could be timed. E . Cell cycle time histogram of 67 individually cultured CD45m,dlin"Rho"SP cells. A delayed initial cell cycle was observed followed by synchronously maintained subsequent divisions. Cells that did not complete the corresponding cell cycle were excluded from this histogram. F. Comparison of the cell cycle times of individual progeny pairs demonstrating the pronounced synchrony retained between such 'sister' cells in spite of the wide range of cycle times observed. Cells whose sisters did not complete the corresponding cell cycle were not included in the plot. G. Example of part of a clone in which many cells have large trailing projections (uropodia). Arrows indicate cells with uropodia. H . Example of part of a clone in which very few cells have uropodia. 114 A C Figure 3.3 - HSC activity is associated with smaller clone sizes and longer cell cycle times. A. Duration of the 1st, 2 n d , and 3r d cell cycles was significantly longer in clones containing HSC than in clones without HSCs. Cells that did not complete a 1st, 2n d, or 3 r d cell cycle were excluded from this analysis. B. The cumulative time to a 3 r d division of cells in HSC-containing clones was significantly longer than the corresponding value for clones without HSCs. Clones in which there were <3 cell divisions but the cells remained viable until the end of culture were assigned a time to 3r d division equal to the total culture time. Error bars represent SEM, n=67. C. Comparison of 4-day clone size distributions for those that contained HSCs and those that did not. Horizontal bars indicate the geometric mean values which are significantly different (p<.005) On average, clones with HSCs executed one less division over the 4 days than clones that did not contain HSCs. 115 100 "i 90 A .2 80 H 70 H Excluded by t t + uropod gate ] T 60 H 50 HSC frequency of 17/27 (63%) Ml I IS Excluded by time to 3 r d division gate Figure 3.4 - Use of behavioral parameters defined by cell tracking to predict HSC-containing clones. Circles indicate the mean time to a 3 r d division in each clone. Bars indicate the ranges of these times. Arrows indicate that one or more of the cells did not complete a 3 r d division by the end of the culture period. Asterisks indicate wells in which the original cell had not yet divided at 96 hours when the cultures were terminated. Fi l led circles represent clones that contained a detectable H S C , open circles represent clones that did not. Grey symbols represent clones that were excluded by one or both of the 2 criteria applied (i.e., the average time to a 3 r division was <67.23 hours and/or >50% of cells within the clone displayed uropodia during the final 12 hours of culture). Black symbols identify the 34 clones that were not excluded by either criteria (i.e., the fastest time to a 3 r d division was >67.23 hours and ^ 0 % of cells within the clone displayed uropodia during the final 12 hours of culture). The latter allowed the frequency of HSC-containing clones in the remainder to be increased from 28% to 63%, a 2.26-fold increase. 116 10 5 0 1 0 0 1 5 0 2 0 0 2 5 0 Fo rwa rd S c a t t e r 5 0 1 0 0 1 5 0 2 0 0 2 5 0 U V r*4 ( FL4 ) 0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 Forward Sca t t o r 0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 F o r w a r d S c a t t e r 0 50 1 0 0 1 5 0 2 0 0 2 5 0 U V r a d (F 14) Figure 3.5. Purification of CD45midlin"Rho"SP cells. Stained bone marrow cells were first gated using FSC/SSC (A) and PI (B) to exclude debris, erythrocytes, dead cells, and cell clumps. Gates were then set around the CD45 m i d (C), SP (D), and lin'Rho" (E) populations. (F) A l l 5 gates in combination resulted in the isolation of -0.004% of the original bone marrow cells. 117 io° i t 1 io z ic 1 i V i»° tt1 it1 it3 ii* io° I O 1 ic* J » 3 ie 4 C M S : ( l i t C04S2-FITC C W S 2- FITC Figure 3.6. Sample calculation of the donor-derived leukocyte level in the P B of a transplanted mouse. A s a control, a non-transplanted C57BU6J-W41/W"-Ly5.] recipient stained with both CD45.1 and CD45.2 showed no events in the CD45.2 single positive (donor) gate (A) . To calculate the level of donor-derived leukocytes present in the P B of a transplanted mouse, events negative for CD45.1 and CD45.2 or positive for both CD45.1 and CD45.2 were first excluded ( B ) , and the overall level of donor-derived leukocytes was then calculated within the single positive events (C). Mice whose P B contained >1% donor cells at 16 weeks were considered to be repopulated long term. To calculate the donor contribution to the myeloid lineages, the donor contribution of the Ly6g /Mac l positive compartment was calculated ( D ) . Similarly, to calculate the donor contribution to the lymphoid lineages, the donor contribution of the B cells (E) and T cells (F) was calculated. If >1% of the myeloid and >1% of the lymphoid cells were of donor origin, mice were considered to have shown multi-lineage repopulation by the injected cell(s). Evidence of both long term and multi-lineage repopulation in the same recipient was used to infer ^ H S C had been injected. 118 o o ° o o 0 ° o ° o 0 o l * f | * i» i l ' l i i l f tt t f l I o g O Q O o ° o o O ° 67.23 h 0 r-ALL HSC non-HSC Figure 3.7 - Calculation of time to a 3 r division exclusion parameter. Construction of the time to a 3r d division exclusion gate as the average of the mean time to a 3 r d division for all clones minus one half the SD, in this case 67.23 hours. Dotted symbols indicate an estimated time to a 3r d division. 119 References: 1. Schwobel W. Description of a simple apparatus for time-lapse microphotography. Mikroskopie. 1952;7:115-120. 2. Schrek R, Ott IN, Jr. Study of the death of irradiated and non-irradiated cells by time-lapse cinemicrography. AMA.Arch.Pathol 1952;53:363-378. 3. Kramis NJ. Time-lapse, phase contrast cine photomicrography of tissue culture cells. J Biol.Photogr.Assoc 1956;24:27-29. 4. Siskin J. Analyses of variations in intermitotic time. In: Rose GG, ed. Cinemicrography cell biology. New York: Academic Press; 1963:143-168. 5. Allen RD, Zacharski LR, Widirstky ST et al. Transformation and motility of human platelets: details of the shape change and release reaction observed by optical and electron microscopy. J Cell Biol. 1979;83:126-142. 6. DiMilla PA, Stone JA, Quinn JA, Albelda SM, Lauffenburger DA. Maximal migration of human smooth muscle cells on fibronectin and type IV collagen occurs at an intermediate attachment strength. J Cell Biol. 1993;122:729-737. 7. Hsu TC. Generation time of HeLa cells determined from cine records. Tex.Rep.Biol.Med 1960;18:31-33. 8. Froese G. The distribution and interdependence of generation times of hela cells. Exp Cell Res 1964;35:415-419. 9. Zetterberg A, Killander D. Quantitative cytochemical studies on interphase growth. II. Derivation of synthesis curves from the distribution of DNA, RNA and mass values of individual mouse fibroblasts in vitro. Exp Cell Res 1965;39:22-32. 10. Schrek R, Ott JN, Jr. Study of the death of irradiated and non-irradiated cells by time-lapse cinemicrography. AMA.Arch.Pathol 1952;53:363-378. 11. Marin G, Bender M A . Radiation-induced mammalian cell death: lapse-time cinemicrographic observations. Exp Cell Res 1966;43:413-423. 12. Martz E, Steinberg MS. The role of cell-cell contact in "contact" inhibition of cell division: a review and new evidence. J Cell Physiol 1972;79:189-210. 13. Absher PM, Absher RG. Clonal variation and aging of diploid fibroblasts. Cinematographic studies of cell pedigrees. Exp Cell Res 1976;103:247-255. 14. Concha M L , Adams RJ. Oriented cell divisions and cellular morphogenesis in the zebrafish gastrula and neurula: a time-lapse analysis. Development 1998;125:983-994. 120 15. Sylwester D, Dennis SM, Absher M . Computer processing of cell lineage data from time lapse cinematography studies. Comput.Biol.Med 1980;10:103-108. 16. Potel MJ, Sayre RE, Robertson A. A system for interactive film analysis. Comput.Biol.Med 1979;9:237-256. 17. Frimberger A E , McAuliffe CI, Werme K A et al. The fleet feet of haematopoietic stem cells: rapid motility, interaction and proteopodia. Br.J.Haematol. 2001 ;112:644-654. 18. Wagner W, Saffrich R, Wirkner U et al. Hematopoietic progenitor cells and cellular microenvironment: behavioral and molecular changes upon interaction. Stem Cells 2005;23:1180-1191. 19. Fruehauf S, Srbic K, Seggewiss R, Topaly J, Ho AD. Functional characterization of podia formation in normal and malignant hematopoietic cells. J Leukoc.Biol. 2002;71:425-432. 20. Punzel M , Liu D, Zhang T et al. The symmetry of initial divisions of human hematopoietic progenitors is altered only by the cellular microenvironment. Exp Hematol. 2003;31:339-347. 21. Francis K, Palsson B, Donahue J, Fong S, Carrier E. Murine Sca-l(+)/Lin(-) cells and human K G l a cells exhibit multiple pseudopod morphologies during migration. Exp Hematol. 2002;30:460-463. 22. Suzuki N, Ohneda O, Minegishi N et al. Combinatorial Gata2 and Seal expression define hematopoietic stem cells in bone marrow niche. Proc Natl Acad Sci U.S A 2006 23. Stadtfeld M , Varas F, Graf T. Fluorescent protein-cell labeling and its application in time-lapse analysis of hematopoietic differentiation. Methods Mol Med 2005;105:395-412. 24. Osawa M , Hanada KI, Hamada H, Nakauchi H. Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science 1996;273:242-245. 25. Wagers AJ, Sherwood RI, Christensen JL, Weissman IL. Little evidence for developmental plasticity of adult hematopoietic stem cells. Science 2002;297:2256-2259. 26. Uchida N, Dykstra B, Lyons KJ, Leung F Y K , Eaves CJ. Different in vivo repopulating activities of purified hematopoietic stem cells before and after being stimulated to divide in vitro with the same kinetics. Exp.Hematol. 2003;31:1338-1347. 27. Benveniste P, Cantin C, Hyam D, Iscove NN. Hematopoietic stem cells engraft in mice with absolute efficiency. Nat.Immunol. 2003;4:708-713. 28. Chen C-Z, Li L, Li M , Lodish H. The Endoglin P o s i t i v e Sea-l P o s i t i v eRhodamine L o w phenotype defines a near-homogeneous population of long-term repopulating hematopoietic stem cells. Immunity 2003;19:525-533. 121 29. Matsuzaki Y, Kinjo K, Mulligan RC, Okano H. Unexpectedly efficient homing capacity of purified murine hematopoietic stem cells. Immunity 2004;20:87-93. 30. Goodell M A , Brose K, Paradis G, Conner AS, Mulligan RC. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J.Exp.Med. 1996;183:1797-1806. 31. Sato T, Laver JH, Ogawa M . Reversible expression of CD34 by murine hematopoietic stem cells. Blood 1999;94:2548-2554. 32. Huygen S, Giet O, Artisien V et al. Adhesion of synchronized human hematopoietic progenitor cells to fibronectin and vascular cell adhesion molecule-1 fluctuates reversibly during cell cycle transit in ex vivo culture. Blood 2002;100:2744-2752. 33. Uchida N, Dykstra B, Lyons K et al. A B C transporter activities of murine hematopoietic stem cells vary according to their developmental and activation status. Blood 2004;103:4487-4495. 34. Habibian HK, Peters SO, Hsieh CC et al. The fluctuating phenotype of the lympho-hematopoietic stem cell with cell cycle transit. J.Exp.Med. 1998;188:393-398. 35. Zhang CC, Lodish HF. Murine hematopoietic stem cells change their surface phenotype during ex vivo expansion. Blood 2005;105:4314-4320. 36. Miller CL, Eaves CJ. Expansion in vitro of adult murine hematopoietic stem cells with transplantable lympho-myeloid reconstituting ability. Proc.Natl.Acad.Sci.USA 1997;94:13648-13653. 37. Audet J, Miller CL, Eaves CJ, Piret JM. Common and distinct features of cytokine effects on hematopoietic stem and progenitor cells revealed by dose response surface analysis. Biotechnol.Bioeng. 2002;80:393-404. 38. Suda T, Suda J, Ogawa M . Proliferative kinetics and differentiation of murine blast cell colonies in culture: Evidence for variable G 0 periods and constant doubling rates of early pluripotent hemopoietic progenitors. J.Cell Physiol. 1983;117:308-318. 39. Brummendorf TH, Dragowska W, Zijlmans JMJM, Thornbury G, Lansdorp PM. Asymmetric cell divisions sustain long-term hematopoiesis from single-sorted human fetal liver cells. J Exp Med 1998;188:1117-1124. 40. Srour EF, Tong X, Sung K W et al. Modulation of in vitro proliferation kinetics and primitive hematopoietic potential of individual human CD34+CD38-/lo cells in GO. Blood 2005;105:3109-3116. 41. Schroeder T. Tracking hematopoiesis at the single cell level. Ann.N.Y.Acad Sci 2005;1044:201-209. 122 42. Francis K, Palsson B, Donahue J, Fong S, Carrier E. Murine Sca-l(+)/Lin(-) cells and human K G l a cells exhibit multiple pseudopod morphologies during migration. Exp Hematol. 2002;30:460-463. 43. Frimberger A E , Stering A l , Quesenberry PJ. An in vitro model of hematopoietic stem cell homing demonstrates rapid homing and maintenance of engraftable stem cells. Blood 2001;98:1012-1018. 44. Fruehauf S, Srbic K, Seggewiss R, Topaly J, Ho AD. Functional characterization of podia formation in normal and malignant hematopoietic cells. J Leukoc.Biol. 2002;71:425-432. 45. Wagner W, Ansorge A, Wirkner U et al. Molecular evidence for stem cell function of the slow-dividing fraction among human hematopoietic progenitor cells by genome-wide analysis. Blood 2004;104:675-686. 46. Giebel B, Corbeil D, Beckmann J et al. Segregation of lipid raft markers including CD 133 in polarized human hematopoietic stem and progenitor cells. Blood 2004;104:2332-2338. 47. Frimberger A E , McAuliffe CI, Werme K A et al. The fleet feet of haematopoietic stem cells: rapid motility, interaction and proteopodia. Br.J.Haematol. 2001;112:644-654. 48. Giebel B, Corbeil D, Beckmann J et al. Segregation of lipid raft markers including CD133 in polarized human hematopoietic stem and progenitor cells. Blood 2004;104:2332-2338. 123 Chapter 4 Longterm symmetric propagation in vivo of functionally distinct subtypes of hematopoietic stem cells* Brad Dykstra1'2, David Kent1'2, Michelle Bowie1 , 2, Lindsay McCaffrey1, Melisa Hamilton1, Kristin Lyons and Connie Eaves ' . 'Terry Fox Laboratory, BC Cancer Agency, Vancouver, British Columbia, V5Z 1L3, Canada department of Medical Genetics, Medicine, and Pathology and Laboratory Medicine, University of British Columbia, Vancouver, British Columbia, V6T 1Z3, Canada A version of this chapter has been submitted for publication 124 4.1. Introduction The mammalian blood-forming system is a complex dynamic cellular system designed to sustain the required levels of multiple lineages of short-lived mature blood cells throughout life. It is now well established that all mature blood cell types are continuously generated from a relatively small population of undifferentiated self-sustaining pluripotent hematopoietic stem cells (HSCs). In the adult, this process usually spans many cell divisions during which lineage choices are first progressively restricted and then executed in a well co-ordinated sequence (reviewed in '). The properties of pluripotent hematopoietic cells and the intrinsic and extrinsic molecular mechanisms that regulate their initial specification and subsequent maintenance or differentiation are amongst the most intriguing, but still poorly understood aspects of hematopoiesis. The investigation of these questions has been confounded by the biological heterogeneity now recognized among hematopoietic cells with the same apparent breadth of pluripotency. One aspect of this heterogeneity is manifested as differences in the durability of mature blood cell production displayed both in vivo2'6 and in vitro1. The fact that differences in the types of white blood cells (WBCs) produced and the longevity of their production can be associated with phenotypically distinct and prospectively separable subsets of hematopoietic cells has led to the widely held view that the earliest stages of hematopoiesis can be modelled as a linear branching hierarchy8. The demonstration of parallel changes in HSC gene expression programs has provided further support for this model 9 - 1 1 . These investigations have been greatly facilitated by the introduction of in vivo repopulation assays that use prolonged outputs of mature blood cell types as endpoints to allow input HSCs with putative lifelong self-sustaining ability to be specifically discriminated from cells that display more limited growth or differentiation activity4;12-14. 125 Adult HSCs with long term reconstituting activity (>1% contribution to the circulating WBCs at 4-6 months post-transplant4'5'15"17) have been conceptualized as a relatively homogeneous population distinct from multi-lineage but short term reconstituting cells (STRCs, also sometimes referred to as ST-HSCs, and MPPs 1 8' 1 9) or more restricted myeloid or lymphoid differentiation potential (CMPs and CLPs 2 0). Nevertheless, even amongst HSCs with long term reconstituting activity, significant variation in progeny outputs is a well recognized hallmark of their behaviour when examined at a clonal level. Historically, this was first noted at the level of HSCs in experiments that tracked the different numbers and types of cells produced in mice transplanted with retrovirally-marked HSCs 2 1" 2 5 . Later, when limiting dilution HSC assays were established, the functional heterogeneity of individual HSCs was confirmed26"31. More recently, additional evidence of heterogeneous HSC behavior has been obtained from analyses of mice repopulated with purified populations of HSCs 3 2" 3 4. Nevertheless, the extent to which this 35 heterogeneity in HSC behavior may reflect a predetermined intrinsic diversity , or chance exposure to different environments36"38, or stochastic events affecting intrinsic pathways that regulate HSC behavior39"42 remains unresolved. To address these possibilities, we have taken advantage of the recent development of a robust method for isolating HSCs at a high purity (~30%)17'33. This made it practical to transplant a large number of mice with single HSCs (or their immediate in vitro generated HSC progeny). Their individual clonal WBC outputs were then tracked in vivo over a 6-month period and their individual self-renewal activities evaluated in secondary and tertiary recipients (Figure 4.1). Analysis of these data revealed 4 distinct patterns of WBC output in the primary recipients, 2 of which were robustly and faithfully transmitted to secondary and tertiary recipients. However, when the initial cells were first stimulated to divide in vitro, the reconstitution pattern 126 of their progeny rapidly changed, although with remarkable symmetry within clones. These findings suggest the existence of adult HSC subsets that have intrinsically determined differentiation programs that are stably perpetuated through many self-renewal divisions in vivo, but are also subject to rapid alteration in vitro. 4.2. Results 4.2.1. Recipients of a single HSC (or HSC-derived in vitro clone) display one of 4 distinct patterns of WBC reconstitution The power of single cell transplants is that all progeny subsequently detected at any time can be ascribed to the same original starting cell. Here we used multi-parameter fluorescence-activated cell sorting (FACS) to isolate the CD45 m i d , lineage marker-negative, Rhodamine-123du", Hoechst 33342-excluding side population (CD45midlin"Rho"SP) fraction of normal adult ] 7'33 mouse bone marrow (BM) cells, that we have previously shown to contain -30% HSCs ' . We then injected 352 sublethally irradiated Ly5-congenic W4''/W4' mice with visually confirmed, single cells of this phenotype, or with in vitro clones generated from such cells after 4 days in serum-free medium containing a growth factor cocktail that supports a modest expansion of HSCs 4 3 . Sublethally irradiated W4''/W41 mice were used as recipients because they provide an environment for the detection of individual HSCs that is equivalent to lethally irradiated mice in conventional limiting dilution assays while avoiding the complication of having to transplant additional cells to ensure survival of the recipients14'15. The number and proportion of myeloid (granulocytes/monocytes, GM) and lymphoid (B-cell and T-cell) WBCs of the donor Ly5 genotype present in the blood of each of the 352 mice transplanted was then determined 4, 8, 12, 127 16 and 24 weeks later (Figure 4.1 and Figure 4.8). Significant longterm reconstitution (i.e., >1% of the total circulating WBCs present 16-24 weeks post-transplant) from the injected cell or clone was obtained in 93 of the 352 mice (26%). Another 26 of the 352 recipients (7%) showed a detectable but unsustained transplant-derived contribution to the circulating WBCs (>0.5% prior to 16 weeks, but <1% at 16-24 weeks). The small fraction of the CD45midlin~Rho~SP cells responsible for this latter activity may be viewed as contaminating STRCs according to current functional definitions4'5'18'44'45. Accordingly, we eliminated these latter mice from all subsequent analyses in order to ensure the exclusion of STRCs as well as later cell types with more restricted differentiation potentials (e.g., CMPs and CLPs) that are not present in the SP fraction46. In the remaining 233 primary recipients (66%), donor-derived WBCs were not detected in the blood at any time (<0.5% at all time points). The specific levels of donor-derived WBCs in each of the 93 mice that showed significant longterm repopulation varied considerably between mice throughout the 6 month period of follow-up (Figure 4.2A), as expected. We then calculated the separate contribution of donor cells at 16 weeks post-transplant to each lineage (GM, B or T). When these 16-week lineage-specific contributions were plotted as normalized ratios for each mouse on a ternary graph (see Figure 4.8 for details), the ratios segregated into 4 distinct clusters (Figure 4.3). We have designated these as a, P, y and 8. The a and P clusters were defined, respectively, by donor-derived GM:(B + T) ratios of >2 and 0.25 to 2. Mice that had donor-derived GM:(B + T) ratios of <0.25 were further subdivided into 2 clusters according to whether there was a continuing >1%> donor contribution to the myeloid lineage as well as to both lineages of lymphoid cells at 16 weeks (y-type), or whether they contributed exclusively to the lymphoid lineages at this time (5-type). Values for transplants of freshly isolated single cells (n=49) and 4-128 day clones (n=44) clustered in the same ways (Figure 4.9, panels A and B) and were therefore pooled for subsequent analyses. Interestingly, the contribution of donor-derived cells to the total number of WBCs 16 weeks post-transplant was not a highly discriminating feature of each cluster (Figure 4.2B), although each cluster did display a distinct kinetic pattern of reconstitution when the average values were plotted as a function of time post-transplant (Figure 4.2C). Specific examples of the 4 reconstitution patterns are shown in Figures 2D and 2E. In the upper panels of these figures, the data are displayed as additive contributions of the donor-derived G M , B and T cell contributions to the total WBC pool as a function of time after the initial cells were injected. In the lower panels of these figures, the results are displayed as separate specific contributions to each of the 3 types of WBCs monitored. The second method of data presentation is useful because it is not confounded by variations in total G M , B or T values that may change during the course of the experiment. Note that when the average lineage-specific contributions were plotted over time, 4 distinct patterns were again evident (Figure 4.2F). Based on these findings, we defined the cells responsible for each WBC reconstitution pattern operationally as a-, (3-, y- and §- longterm reconstituting cells (LTRCs). 4.2.2. a- and P-LTRCs display extensive self-renewal activity in vivo with long term preservation of the original pattern of WBC reconstitution Given the different patterns of WBC reconstitution observed in primary recipients of single LTRCs (or their immediate LTRC progeny generated in vitro), it was of interest to determine whether these would be equally or differentially associated with long-term in vivo self-renewal activity. To test this, we transplanted pairs of sublethally irradiated Ly5-congenic 129 W4''/W41 mice with the regenerated B M cells harvested individually from 46 of the 93 clonally repopulated mice 6-7 months after the initial cell(s) had been injected. The pattern of WBC reconstitution in each of the secondary recipients was then monitored and analyzed as in the primary mice. The results for the 16-week time point showed that cells from 21 of the 46 primary clones contained detectable secondary LTRCs in one or both secondary recipients (i.e., >1% of the circulating WBCs were of the same Ly5 genotype as the initially transplanted cell(s)). Strikingly, all 21 of the primary clones containing secondary LTRCs originated from cc-or p-LTRCs, and this included almost every a- or P-LTRC tested (10/11 a-LTRCs and 11/12 P-LTRCs) (Figure 4.4A and Figure 4.1 OA). In contrast, no y- (0/6) or 5-LTRCs (0/17) demonstrated in vivo self-renewal in these experiments. Interestingly, the L T R C subtype appeared more predictive of secondary reconstituting activity than either the total donor-derived WBC levels in the primary recipients or the extent of the individual donor contribution to the G M , B or T lineages (Figure 4.1 OB). Paired tertiary transplants were also performed on B M cells harvested from 21 of the successfully repopulated secondary mice and the reconstituted G M , B and T cells were then monitored in the tertiary recipients for up to 24 weeks post-transplant. These assays showed that most of the a- and P-LTRCs had undergone further expansion in the secondary mice (Figure 4.4A). Interestingly, the lineage-specific WBC contributions seen in the secondary and tertiary recipients at 16 weeks (see examples in Figure 4.4B), as well as the overall reconstitution patterns obtained (see examples in Figure 4.5A), were both highly reminiscent of the 4 patterns identified in the primary mice. Moreover, comparison of each cohort of serial transplants revealed a striking preservation of each particular reconstitution pattern obtained in multiple recipients of cells from the same original LTRC (see examples 1-3 in Figure 4.4B for a- and P-130 LTRCs, respectively). The similarity in the reconstitution patterns displayed by the progeny of individual a- or P-LTRCs is even more obvious when the paired recipients of cells from the same primary or secondary donors are compared (Figure 4.5A). Taken together, these findings point to an extensive self-renewal ability of a- and P-LTRCs and very high stability of their WBC reconstituting properties over many divisions (given that in most cases, only a 10% sampling of the progeny of 1 cell repopulated 2 of 2 secondary recipients and 10% of their progeny usually repopulated further pairs of tertiary recipients). Evidence of a change from one reconstitution pattern to another was, however, noted in approximately half of the secondary recipients of cells from a- and P-LTRC-repopulated primary mice (Figure 4.4A). Specifically, in 5 of the 10 a-LTRC-repopulated primary mice tested, one or both of the secondary recipients showed a more balanced donor-derived GM:B + T ratio at 16 weeks, typical of P-LTRCs (examples 4 and 5 in Figure 4.4B for a-LTRCs). Similarly, in 5 of the 11 P-LTRC-repopulated primary mice tested, one or both of the secondary recipients showed a low donor-derived GM:B + T ratio at 16 weeks, typical of y- or 5-LTRCs (examples 4 and 5 in Figure 4.4B for P-LTRCs). These findings indicate that some a-LTRCs can also produce P-LTRCs (and/or y- and 5-LTRCs) and some P-LTRCs can also produce y- and 8-LTRCs. However, these pattern switches are clearly not common events in vivo since many a- and P-LTRCs sustained the production of the same type of daughter LTRCs for over a year. It is important to note that 11 of the 16 a-LTRCs identified in the original set of transplants would not have been classified as HSCs by conventional endpoints (i.e., >1% donor contribution to the total WBCs and >1% donor contribution to the G M , B and T lineages at 16 weeks) due to their failure to produce mature B- and/or T-cells until after 16 weeks post-transplant. However, of the 6 whose progeny were analyzed in secondary transplants, 3 131 produced lymphoid cells in numbers consistent with traditionally defined HSCs (examples 1 and 4 in Figure 4.4B for a-LTRCs). Thus a-LTRCs can be precursors of conventionally defined HSCs even though they might not, themselves, be recognized as HSCs in a primary recipient followed for up to 6 months. 4.2.3. L T R C subtypes are rapidly but symmetrically altered in vitro We next asked whether HSCs stimulated to proliferate in vitro generate the same or a different distribution of LTRC subtypes. As noted above, the transplanted 4-day in vitro clones produced the same 4 patterns of WBC reconstitution at 16 weeks post-transplant as seen in recipients of freshly isolated cells (Figure 4.9). However, after 4 days in vitro, during which time >95% of the input cells divide at least once17'33, the cells harvested produced a different ratio of WBC reconstitution patterns. The proportion of primary mice displaying y and 8 patterns increased from 17/49 (35% for freshly isolated cells) to 36/44 (82% for the 4-day clones) and the proportion of those with a and P patterns decreased correspondingly from 32/49 (65%) to 8/44 (18%) (Figure 4.6). Since the cloning efficiency in vitro under these conditions is so high 1 7' 3 3, the increased number of y and § patterns seen must be attributable either to the generation of clones of y- and/or 6-LTRCs that arise either from non-LTRCs or to cells that initially were a- or p-LTRCs. After 10 days in culture, no a- or p-LTRCs were detected and all mice repopulated with any LTRCs remaining in the 10-day clones appeared to have been transplanted with y- and/or 5-LTRCs (Figure 4.6 and Figure 4.9C). To determine whether the altered reconstitution patterns obtained by the cultured cells reflected the properties of a single persisting LTRC in each clone or whether they might be shared by multiple LTRCs present in the same clone, we transplanted equal portions of 18 132 different 10-day clones into 2-3 mice each and then again followed the W B C output patterns obtained in each mouse (Figure 4.1). From these analyses, 8 subdivided 10-day in vitro clones produced >1% of the circulating W B C s at 16 weeks later in 2 or more recipients. This result demonstrates that multiple L T R C s had been generated in at least 8 of the 18 clones tested (44%). Moreover, recipients repopulated with portions of the same clone displayed remarkably similar kinetics of W B C reconstitution and lineage-specific contributions from the injected cells (see examples in Figure 4.5B). Thus, marked symmetry was also evident in the generation of L T R C s in vitro, in spite of a rapid change in the types of L T R C s being produced. 4.3. Discussion For more than 50 years it has been appreciated that the B M of mice contains cells with an impressive ability to produce all types of mature blood cells for many months. The subsequent development of methods to quantify such cells in limiting dilution transplantation assays and prospectively isolate them as unique populations separate from cells with more restricted proliferative or differentiation potential has led to the widespread adoption of a linear branching model of hematopoiesis 4 7. Recent studies of various gene knockout mice have provided further support for the concept that different stages of hematopoietic cell development in this complex hierarchy are differentially dependent on the functional contributions of specific key genes, and that the intrinsic regulation of short and long term renewal of pluripotentiality may involve mechanistically distinct processes 4 8" 5 0. Historically, data suggesting that the vast majority of H S C s in the adult were quiescent and, once activated, could only be propagated for 2 or 3 serial transplants were used to infer a model in which lifelong blood production had to be sustained by the sequential recruitment and removal of HSCs from an ultimate reservoir of cells with similar 133 finite self-renewal potentialities22'40. More recent experiments have altered this view with the demonstration that all adult mouse HSCs are likely to divide approximately once a month51'52 and that, when serially transplanted, can expand their numbers at least 8000-fold53. Similarly, analyses of the average telomere length in WBCs from humans of different ages suggests that these WBCs derive from a pool of HSCs that are slowly but continuously entering the cell cycle throughout adult life54. Nevertheless, key questions about the durability and intrinsic control of self-renewal potential of individual HSCs in vivo remain largely unanswered. To address these questions, we took advantage of recently developed methods for obtaining a population of B M that is highly enriched for HSCs (-30% purity) and is also phenotypically separate from less primitive cell types with more restricted reconstituting and/or differentiation ability (including ST-HSCs, MPPs, CMPs and CLPs). Measurements of WBC outputs in the present study showed that ST-HSCs accounted for less than 10% of the total number of cells/clones assayed. The purity of HSCs in the starting populations tested and their minimal contamination with less primitive cells thus made it practical to analyze, over long periods of time (up to 18 months), the progeny of a large number of single initial HSCs. Using this approach, we showed that the kinetics of WBC reconstitution could be subdivided into 4 distinct patterns. The fact that 2 of the patterns were also consistently associated with robust self-renewal along with stable perpetuation or symmetric conversion of the reconstituting pattern in their progeny strongly argues that these 4 reconstituting patterns are intrinsically pre-determined in HSCs prior to transplant. Accordingly, we have assigned them the operational designations of a-, (3, y- and 8-LTRCs. Unfortunately, as yet there is no known method for subdividing the rare CD45m i dlinTlho~SP B M cells to allow these functionally recognized subsets 134 to be prospectively isolated. Hence candidate molecular differences remain inaccessible to definition, although this would clearly be a future interesting avenue of investigation. The cells defined here as P-LTRCs correspond to what has been traditionally envisaged as highly competitive HSCs because of their rapid and sustained pluripotent differentiation activity in primary recipients and extensive self-renewal activity in secondary and tertiary mice. y-LTRCs meet the expectation of a class of less competitive HSCs with rapidly apparent and strong initial pluripotent differentiation activity, but declining outputs of all lineages at 16 weeks and insufficient self-renewal activity to produce daughter LTRCs detectable in secondary mice transplanted with cells harvested from primary mice after 6 to 7 months. 8-LTRCs also lack this degree of self-renewal activity and, although pluripotent and capable of long-term repopulation, do not produce myeloid cells beyond the first 4 months post-transplant. 5-LTRCs thus appear to share the functional properties of other cells described as STRCs 5 5 . a-LTRCs are the most interesting and novel type of LTRCs identified here because they typically produced detectable levels of mature WBCs only after many weeks and those eventually detected were largely myeloid for many months. Nevertheless, a-LTRCs also possessed extensive self-renewal activity and in vivo could occasionally regenerate LTRCs that were able to produce significant numbers of lymphoid progeny in secondary recipients. The presence in adult mouse B M of phenotypically distinct subsets of HSCs that produce mature WBCs after a long delay has been noted by others32'56, although it was not appreciated that they might display a unique lineage output program. Evidence of sustained lineage biases in the repopulation patterns of HSCs has been previously suggested from analyses of the reconstituting activity of cells present in 4-week stromal cell-containing cultures initiated with limiting dilutions of unseparated B M cells57. Interestingly, in those experiments, a subtype of 135 self-renewing lymphoid-biased HSCs was also proposed, which was not observed in the current study. This may be explained by the different histories of the cells used in the 2 studies, which we show here to be a critical parameter. Figure 4.7 summarizes the relationships found between the 4 types of LTRCs we describe. P-, y- and 5-LTRCs can be readily accommodated within the classical hierarchical CQ scheme but a-LTRCs do not fit readily into this paradigm. Many a-LTRCs generated myeloid cells throughout 3 cycles of hematopoietic reconstitution over a period of 18 months with low or negligible contributions to the lymphoid lineages, thus allowing their distinction and independent maintenance to be convincingly documented. Serial transplants of the in vivo generated progeny of P-LTRCs also suggested stable self-renewal of this program for similarly extended periods, although it should be recognized that the output of multiple LTRC subtypes in secondary and tertiary transplants might be indistinguishable from that of p-LTRCs. Interestingly, occasional production of p-LTRCs (and/or y- and 5-LTRCs) from a-LTRCs was seen in vivo and may also have occurred in vitro. In vivo production of exclusively y- and/or 5 -LTRCs from P-LTRCs was also sometimes noted. On the other hand, evidence of the generation of a-LTRCs from P-LTRCs was not obtained, although this would not have been detectable if any P-LTRCs (and/or y- or 5-LTRCs) had also been present since the lymphoid cells they produced would override the pattern characteristic of co-transplanted a-LTRCs. Future experiments utilizing limiting dilution transplants into secondary recipients would circumvent this problem and will thus be of interest to clarify the full spectrum of inter-program conversions that a- and P-LTRCs can undertake. Perhaps the most striking and unexpected result of the present study is the symmetry of reconstitution behavior sustained by clonally amplified LTRCs both in vivo, where the input program was stably maintained over many self-renewal divisions, and in vitro, where there was a 136 rapid shift to less competitive repopulation programs. Thus, even when many daughter cells from the same in vivo or in vitro clone were injected and the repopulation patterns were different from those characteristic of the parental cells, the patterns produced by different aliquots of the same clone mimicked one another with extraordinary similarity. Collectively, our findings suggest that the repopulation patterns later displayed by primitive pluripotent hematopoietic cells are intrinsically pre-set, possibly at a much earlier point in their development. This would suggest the possibility that genes involved in setting the pace and ease of activation of particular WBC differentiation lineages may be epigenetically modified in different ways in different HSC subsets. Such a possibility would also fit with current evidence of lineage priming of pluripotent LTRCs and their subsequent differentiation by progressive suppression of lineage options (reviewed in 5 9). However, regardless of the underlying mechanism(s), it now seems clear that self-renewal divisions of HSCs in vivo do not randomly reset the differentiation options of daughter HSCs. Preliminary repopulation studies suggest that type a-LTRCs may be less prevalent in embryonic day 14.5 fetal liver. Among 10 mice that were each repopulated with single fetal liver HSCs, the numbers repopulated by a-, P-, y- and 5-LTRCs were 0, 30%, 30% and 40%, respectively (compared to the distribution in adult B M of 27%, 39%, 12% and 22%, see Figure 4.9). Interestingly, several studies suggest that cells with type a characteristics may increase as mice age 6 0 - 6 3 Such a shift is consistent with our data suggesting a reduced or absent frequency of a-LTRCs in fetal mice and might contribute to the decrease in lymphoid cells characteristic of aging mice (reviewed in 6 4). Understanding the mechanisms that allow a- and P-LTRCs to sustain their characteristic WBC output patterns should give new insights into the pathways 137 involved in HSC differentiation. The findings reported here may also help to clarify the heterogeneity seen in genetically similar leukemias. 4.4. Materials and Methods 4.4.1. Mice B M donors were 8-12 week-old C57B1/6J-Zv5.i or -Ly5.2 mice. All transplant recipients were ZyJ-congenic C51BV61-W4'/W4' (W^'/W41) mice given a sublethal dose of irradiation (360 cGy X-rays at -350 cGy per minute). 4.4.2. Purification, culture and transplantation of CD45midlin~Rho~SP cells Single viable (propidium iodide-negative) CD45midlin"Rho'SP cells were sorted by FACS as previously described17 into the individual wells of a round-bottom 96-well plate containing 100-200 /xl of serum-free medium, centrifuged at 700 rpm, and visually confirmed. Some single cells were then injected intravenously into sublethally irradiated W41 /W41 recipients, as described17'33. Others were cultured for 4 or 10 days with 300 ng/ml murine Steel factor (StemCell Technologies, Vancouver, BC), 20 ng/ml human IL-11 (Genetics Institute, Cambridge, MA) and 1 ng/ml human Flt3-ligand (Immunex, Seattle, WA). Clones present in 4-day culture plates were then individually harvested and injected intravenously into sublethally irradiated W4''/W41 recipients. Clones present in 10-day cultures were subdivided into equal aliquots and then injected intravenously into groups of 2-3 sublethally irradiated W4''/W4' recipients. 24 weeks post-transplant, B M was harvested from selected repopulated primary recipients of single 138 CD45midlin~Rho~SP cells or 4-day clones, and the cell content equivalent of one femur (-10% of the total BM) was injected into each of 2 secondary sublethally irradiated W4']/W41 recipients. B M was similarly harvested from selected secondary recipients 24 weeks post-transplant and the equivalent of one femur from each secondary mouse was injected into a pair of tertiary irradiated FT'Vlf^recipients. 4.4.3. A n a l y s i s o f in vivo r e p o p u l a t i o n Peripheral blood samples were collected from the tail vein of mice 4, 8, 12, 16 and 24 weeks after transplantation. Following lysis of the red blood cells with ammonium chloride (StemCell), the WBCs were stained with antibodies for donor and recipient CD45 allotypes (anti-CD45.1-allophycocyanin [APC] and anti-CD45.2-fluorescein isothiocyanate [FITC]) plus anti-Ly6g-phycoerythrin [PE]/anti-Macl-PE for myeloid (GM) cells, or anti-B220-PE for B-cells, or anti-CD5-PE for T-cells (CD45.2-FITC purified and conjugated in the Terry Fox Laboratory, Vancouver BC; CD45.1-APC from eBiosciences, San Diego CA; all others from BD Biosciences, San Jose CA). To calculate repopulation levels, events negative for CD45.1 and CD45.2 or positive for both CD45.1 and CD45.2 were excluded, and the contributions of the injected (donor) cells to the populations of circulating G M , B, T, and total WBCs were calculated. Recipients with >1% donor-derived WBCs at 16 and/or 24 weeks post-transplant were considered to be repopulated with LTRCs. Different LTRC subtypes (a, /?, y and 5) were discriminated as described in the Results. Further details are given in Figure 4.8. 139 Purified BM cells (CD45m,dlirrRho-SP) • • 1 **J Single cells or 4-day dones injected individually 4 8 12 16 24 weeks , 4-day clonal cultures 10-day clonal cultures 10-day dones split between 2-3 mice I Measure WBC output I Harvest BM | ' » . . . 8 1 6 24 weeks 4 4 4 Measure WBC output~| Pairs of secondary transplants 8 16 24 weeks 4 4 4 Measure WBC output Harvest BM Pairs of tertiary transplants 8 16 24 week f 1 1 ' Measure WBC output Figure 4.1. Schematic Representation of the Overa l l Exper imenta l Design 158 freshly isolated CD45.1 CD45midlin~Rho"SP cells and 194 in vitro clones (generated from single CD45m i dlin Rho'SP cells cultured for 4 days in 300 ng/ml Steel factor, 20 ng/ml IL-l 1 and 1 ng/ml Flt3-ligand) were transplanted into sublethally irradiated W4''/W4' recipients. At regular intervals post-transplant, WBC samples were analyzed for the presence of donor cells of B-cell, T-cell and myeloid (GM) lineages. 24 weeks post-transplant, B M was harvested from 46 positive recipients and 1 femur equivalent was injected into each of 2 secondary sublethally irradiated W41 /W41 recipients (total of 96, of which 90 survived). Similarly, B M was harvested from 21 selected secondary recipients at 24 weeks post-transplant and injected into pairs of tertiary recipients (total of 42, of which 40 survived). 10-day in vitro clones were also generated from single CD45midlin"Rho"SP cells, and 18 such clones were injected into groups of 2-4 sublethally irradiated W41/W41 mice each (total of 48 mice). 140 Single cells 4-day clones 4 8 12 16 20 24 Weeks post-transplant (X-LTRC O +3 §1 51 = 1 ° £ 4 8 12 16 cc-LTRC B *:* o so * 40 O § 30 •a IF 20 ctLTRC^_--# a-LTRC (1-LTRC j-LTRC 5-LTRC B-LTRC y-LTRC 4 8 12 16 20 24 Weeks post-transplant 8-LTRC 4 8 12 16 4 8 12 16 4 8 12 16 Weeks post-transplant P-LTRC y-LTRC 8-LTRC 4 8 12 16 4 8 12 16 24 4 8 12 16 Weeks post-transplant i Total WBC I Myeloid (GM) ] B-cells • T-cells Figure 4.2. W B C Outputs in Recipients of Single L T R C s or Their Clonal Progeny Generated In Vitro (A) Donor contributions to the total circulating WBCs for each of the 93 reconstituted mice shown individually. 141 (B) Variations in the donor contributions to the total circulating WBCs at 16 weeks post-transplant of all 4 LTRC subtypes. Each point represents an individual mouse. Horizontal bars indicate mean values. (C) Distinct patterns of WBC reconstitution by each of the 4 LTRC subtypes. Values are the mean ±SEM of all mice in each LTRC group as defined in the text. (D) and (E) Examples of individual mice repopulated with freshly isolated LTRCs (D), or LTRCs from 4-day clones generated in vitro (E). Top panels: Colored areas represent donor WBC of G M (red), B-cell (blue) and T-cell (yellow) lineages as a percentage of all WBCs over time post-transplant. Data are stacked such that the sum of each lineage contribution represents the percentage donor contribution to all WBCs. Bottom panels: For each time point, the separate donor contributions to the G M (red), B-cell (blue) and T-cell (yellow) lineages are shown as bars, and the percentage donor contribution to the total WBCs is shown as a grey area. (F) Distinct patterns of donor contributions to the G M (red), B-cell (blue) and T-cell (black/yellow) lineages over time are shown. Values are the mean ±SEM of data from all mice in each LTRC group. 142 Figure 4.3. Identification of L T R C Subtypes in Ternary Plots of Their Lineage-specific Contributions at 16 Weeks Post-transplant (A) Comparison of the ratio of donor clone contributions to the G M (up), B-lineage (lower left) and T-lineage (lower right) WBCs. If a clone were contributing only G M cells, the value would lie at the upper vertex; if only B cells, the value would lie at the left vertex; and if only T-cells, at the right vertex. If a clone were contributing equally towards all 3 lineages, the value would lie in the centre of the triangle. Greater relative contributions to the G M , B or T lineages shift the values to the top, lower left or lower right, respectively. The graph can be divided into 3 sections based on specific myeloid to lymphoid contribution ratios: High GM:(B+T) = ratio > 2:1; low GM:(B+T) = ratio < 1:4; balanced GM:(B+T) = ratio between 1:4 and 2:1. (B) Subdivision of LTRCs according to the ratio of their contributions to the myeloid and lymphoid lineages at 16 weeks post-transplant. a-LTRCs display a high GM:(B+T) contribution, p-LTRCs a balanced GM:(B+T) contribution, y-LTRCs a low GM:(B+T) contribution, 5-LTRCs contribute B-cells and/or T-cells but no (<1%) G M cells at 16 weeks. 143 1" LTRC type a-LTRC |) LTRC y-LTRC 6-LTRC Proportion of 1° LTRCs that repopulated 2° recipients 10/11 11/12 0/6 0/17 WBC output patterns observed in 2° recipient pairs 5 a only 1 a+ |! 4 only 6 r only 1|i+y 1 (5+8 1 Y + 5 2 8 only n/a n/a Proportion of 1° LTRCs that repopulated 3° recipients 4/6 6/8 n/a n/a WBC output patterns observed in 3° recipient pairs 2 a only 1 a + |) 1y only 5 |' only 1 (5 + 5 n/a n/a B 1° LTRC type: Example #1: Example #2: Example #3: Example #4: Example #5: 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 a-LTRC L ia III Li I J nd 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 \ B - L T R C y-LTRC 8-LTRC * * 400 BO CC 48 M « J 100 80 60 40 20 0 J 100 80 60 40 20 0 J 100 80 60 40 20 0 J 100 80 40 20 0 Jl * * Recipients: 3° I donor % of GM 1° 2° • donor % of B 1° • donor % of T Figure 4.4. Clonal Propagation of Repopulation Patterns in Secondary and Tertiary Recipients. (A) Summary of the repopulation patterns seen in pairs of secondary and tertiary recipients transplanted with the progeny of each of the 4 LTRC subtypes. 144 (B) Five examples of serial transplants originating from each of the 4 L T R C types are shown. Bars represent the percent donor contribution to the G M (red), B-cell (blue) and T-cell (yellow) lineages at 16 weeks post-transplant in primary, secondary and tertiary recipients (as indicated). Negative recipients (<1% donor WBCs at 16 weeks post-transplant) are indicated with an asterisk. nd = not done. f indicates the recipient mouse died before 16 weeks post-transplant. 145 Recipient pair #1 Recipient pair #2 A Recipient pair #3 l\ ll 1 M Ii I Recipient pair #4 iTi. ii mil I r ri Weeks post-transplant Subdivided clone #1 Subdivided clone #2 Subdivided clone #3 Subdivided clone #4 J Jl -75 50 n 25 0 75 50 _ ru.... . n 25 0 rl rfl :::: ^ Tl a 16 24 . ...n n n 8 18 21 Weeks post-transplant m donor % of WBC I donor % of GM • donor % of B • donor % of T Figure 4.5. Intra-clonal Comparisons of L T R C Progeny (A) Results in paired secondary recipients. For each time point, the donor contributions to the G M (red), B-cell (blue) and T-cell (yellow) lineages are shown as bars and the donor contribution to the total WBCs is shown as a grey area. (B) Results for subdivided 10-day clones. For each time point, the donor contributions to the G M (red), B-cell (blue) and T-cell (yellow) lineages are shown as bars, and the donor contribution to the total WBCs is shown as a grey area. 146 B Fresh (n = 49) 4-day clones (n = 44) Subdivided 10-day clones (n = 21) Figure 4 . 6 . Rapid Alteration of L T R C Distributions In Vitro (A) The distribution of a-, p-, y-, or 5-LTRCs identified in freshly isolated CD45m,dlin"Rho"SP B M cells. (B) The distribution of a-, P-, y-, or 8-LTRCs identified in 4-day clones generated in vitro from freshly isolated CD45m , dlmRho~SP B M cells. (C) The distribution of a-, P-, y-, or 8-LTRCs identified in subdivided 10-day clones generated in vitro from freshly isolated CD45midlin"Rho"SP B M cells. Portions of 8 clones were injected into a total of 21 recipients. 147 r \ [5-LTRC * y - L T R C 5 - L T R C L y m p h o p o i e s i s Figure 4.7. Schematic Diagram of the Relationships Between L T R C Subtypes and the Mature W B C Types They Generate. (3-, y-, or 5 - L T R C s appear to form a hierarchy that corresponds to conventional models of H S C differentiation where the propensity for generating mature myeloid progeny diminishes progressively before pluripotentiality is lost. a - L T R C s represent a novel cell type, in which a very strong propensity for myeloid cell output can be independently and exclusively sustained over multiple cycles o f hematopoietic reconstitution, although crossover to the P-, y-, or 5 - L T R C stream is not precluded. 148 • JW.TRC O o 64.TRC I a-LTRC ' |!-LTRC A y-LTRC r O i l ' UJ S . r Ul r S '4 r 9.22 5.86 1.41 0° 101 102 103 1C CD45.2-FITC (donor) 4.52 ^^ ^^  11.1 m •^ J-MMI 60 0° 101 102 103 1C CD45.2-FITC (donor) |9.71 f||l. 0.9 1 fwiAsgluQ I.- 'HI. *r* 10° 101 102 103 10' CD45.2-FITC (donor) 10u 10' 10' 10 10* CD45.2-FITC (donor) % donor of W B C : (5.86+1.41) + (0.54+6.23) + (0.21+6.8) 13 -7 .0% % d o n o r o r G M : 5.86/(3.86+9.22)= 58.86% % donor o r B : 0.54/(0.54+42.9)- 1.24% % donor of T : 0.21/(0.21+40.7)= 0.51% Normalized G M : B : T ratio Tor ternary graph: 36.86 : 1.24 : 0.51- 0.955 : 0.032 : 0.013 GM:(B+T) ratio: 36.86/(1.24+0.51)= 21.06 35.4 116 W 0° 101 102 103 10 p 114 23.8 % d o n o r o r W B C : (11.1+60) + (35.4+34.5) + (23.8+47.5)/3 = 70.8% % d o n o r o r G M : 11.1/(11.1+4.52)= 71.1% % donor o r B : 35.4/(35.4+13.6)= 72.2% % donor o f T : 23.8/(23.8+11.4) = 67.6% Normalized G M : B : T ratio for ternary graph: 71.1 : 72.2 : 67.6= 0.337 : 0.342 : 0321 17.3g|fc . .LMTIM " ' CD45.2-FITC (donor) 10 10 10* 10 10 CD45.2-FITC (donor) 10 10 10' 10 10 CD45.2-FITC (donor) 10* 10' CD45.2-FITC (donor) 10" 10" 10' 10 CD45.2-FITC (donor) GM:(B+T) ratio: 71.1/(72.2+67.6) - 0.51 % donor of W B C : (0.9+20.2) + (4.7+16) + (14.4+5.82)/3-20.7% % donor of G M : 0.9/(0.9+9.74) = 8.5% % donor oTB: 4.7/(4.7+39) • 10.8% % donoroTT: 14.4/(14.4+32.4)= 50.8% Normalized G M : B : T ratio for ternary graph: 8.5 : 10.8 : 30.8 - 0.169 : 0.216 : 0.615 G M :(B+T) ratio: 8.3/(10.8+30.8)= 0.20 oS-LTRC O10 3 J19.6 0.01 ..22 1.83 CD45.2-FITC (donor) 10u 10 10 10" 10 CD45.2-FITC (donor) 10 10' CD45.2-FITC (donor) % donor of W B C : (0.01+8.22) + (1.83+6.59) + (6.47+1.8) ti - 8.31% % donor of B: 1.83/(1.83+32.4)= 5.3% % donor of T : 6.47/(6.47+39.4)= 14.1% Normalized G M : B : T ratio for ternary graph: 0.03 :3.3 : 14.1 =0.000 : 0.273 : 0.727 G M :(B+T) ratio: 0 05/(3.3+14.1) = 0.003 Figure 4.8. Representative Examples of Mice Repopulated with Each L T R C Subtype Illustrating the Gating Strategies and Relevant Calculations. (A) Viable WBCs are initially identified on the basis of their forward/side light scattering characteristics and lack of staining with PI, followed by the exclusion of events negative for CD45.1 and CD45.2 or positive for both CD45.1 and CD45.2. 149 (B) The overall donor repopulation level is then calculated as a proportion of the single positive events. (C) As a control, a non-transplanted C57BV6J-W41/W41-Ly5.1 recipient stained with both anti-CD45.1 and anti-CD45.2 antibodies shows no events in the CD45.2 single positive (donor) gate. (D) Representative examples of 4 mice injected 16 weeks previously with one of each of the 4 LTRC subtypes are shown plotted on a ternary graph. (E) Here the corresponding 16-week FACS plots, gating strategy, and relevant calculations are shown for the same 4 examples as in (D) after gating for PI" WBCs as described in the first 2 panels. 150 E Fresh BM Number % of total 4-day clones Number % of total Subdivided 10-day clones Number % of total Fresh FL Number % of total a-LTRC 13 27% 3 7% 0 0% 0 0 P-LTRC 19 39% 5 11% 0 0% 3 30% y-LTRC 6 12% 6 14% 5 24% 3 30% 6-LTRC 11 22% 30 68% 12 76% 4 40% Total 49 100% 44 100% 21 100% 10 100% Figure 4.9. Ternary Plots of the Lineage-Specific Contributions Show the Same LTRC Subtypes in Fresh and Cultured Cells but in Different Proportions (A) Ternary plot of 16-week lineage-specific contributions for individual mice transplanted with single freshly isolated CD45midlin"Rho"SP LTRCs. (B) Ternary plot of 16-week lineage-specific contributions for individual mice transplanted with entire 4-day LTRC-containing clones generated in vitro from single freshly isolated CD45midlin" Rho'SP cells. 151 (C) Ternary plot of 16-week lineage-specific contributions for individual mice transplanted with subdivided 10-day clones that were generated in vitro from single freshly isolated CD45midlin~ RhoSP cells and that contained multiple LTRCs. (D) Ternary plot of 16-week lineage-specific contributions for individual mice transplanted with single freshly isolated lin"Sca-l+CD43+Macl+ LTRCs from E14.5 fetal liver (FL). (E) Table summarizing the representation of the different L T R C subtypes in the different sources of cells tested. FL = fetal liver. 152 A B donor % of GM donor % of B donor % of T Relative donor contribution to B cells of WBC • 2° LTRCs detected o 2° LTRCs not detected • 2" LTRCs detected o 2° LTRCs not detected Figure 4.10. Differential Association of Secondary L T R C Act iv i ty with Different Measures of P r imary W B C Output (A) a- and P-LTRCs, but not y- or 5-LTRCs, demonstrate long-term self-renewal in vivo. Each point represents a primary recipient whose B M cells were transplanted into 2 secondary recipients 6-7 months post-transplant. The black points indicate that one or both secondary recipients were repopulated (>1% donor WBCs at 16 weeks post-transplant). The white points indicate that neither of the 2 secondary recipients was repopulated. (B) Percent donor contribution to the myeloid lineage predicts for secondary LTRC activity in vivo better than the percent donor contribution to either the total WBCs or the B- or T-cell lineages. Each point represents a primary recipient whose B M cells were transplanted into 2 secondary recipients at 6-7 months post-transplant. The black points indicate that one or both secondary recipients were repopulated (>1% donor WBCs at 16 weeks post-transplant). The white points indicate that neither of the 2 secondary recipients was repopulated. 153 4.5. References 1. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 2. Spangrude GJ. Characteristics of the hematopoietic stem cell compartment in adult mice. Int.J Cell Cloning 1992;10:277-285. 3. Jones RJ, Wagner JE, Celano P, Zicha MS, Sharkis SJ. Separation of pluripotent haematopoietic stem cells from spleen colony-forming cells. Nature 1990;347:188-189. 4. Morrison SJ, Weissman IL. The long-term repopulating subset of hematopoietic stem cells is deterministic and isolatable by phenotype. Immunity 1994;1:661-673. 5. Adolfsson J, Borge OJ, Bryder D et al. Upregulation of flt3 expression within the bone marrow Lin"Scal+c-kit+ stem cell compartment is accompanied by loss of self-renewal capacity. Immunity 2001;15:659-669. 6. Passegue E, Wagers AJ, Giuriato S, Anderson WC, Weissman IL. Global analysis of proliferation and cell cycle gene expression in the regulation of hematopoietic stem and progenitor cell fates. J.Exp.Med. 2005;202:1599-1611. 7. Ploemacher RE, Van Der Sluijs JP, Voerman JSA, Brons NHC. An in vitro limiting-dilution assay of long-term repopulating hematopoietic stem cells in the mouse. Blood 1989;74:2755-2763. 8. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 9. Ivanova NB, Dimos JT, Schaniel C et al. A stem cell molecular signature. Science 2002;298:601-604. 10. Park IK, He Y, Lin F et al. Differential gene expression profiling of adult murine hematopoietic stem cells. Blood 2002;99:488-498. 11. Zhong JF, Zhao Y, Sutton S et al. Gene expression profile of murine long-term reconstituting vs. short-term reconstituting hematopoietic stem cells. Proc Natl Acad Sci U.S.A 2005;102:2448-2453. 12. Szilvassy SJ, Humphries RK, Lansdorp PM, Eaves A C , Eaves CJ. Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy. Proc.Natl.Acad.Sci.USA 1990;87:8736-8740. 13. Harrison DE, Jordan CT, Zhong RK, Astle C M . Primitive hemopoietic stem cells: direct assay of most productive populations by competitive repopulation with simple binomial, correlation and covariance calculations. Exp.Hematol. 1993;21:206-219. 154 14. Trevisan M , Iscove N. Phenotypic analysis of murine long-term hemotopoietic reconstituting cells quantitated competitively in vivo and comparsion with more advanced colony-forming progeny. J.Exp.Med. 1995;181:93-103. 15. Miller CL, Eaves CJ. Expansion in vitro of adult murine hematopoietic stem cells with transplantable lympho-myeloid reconstituting ability. Proc.Natl.Acad.Sci.USA 1997;94:13648-13653. 16. Kiel MJ, Yilmaz OH, Iwashita T et al. S L A M family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 2005; 121:1109-1121. 17. Dykstra B, Ramunas J, Kent D et al. High-resolution video monitoring of hematopoietic stem cells cultured in single-cell arrays identifies new features of self-renewal. Proc.Natl.Acad.Sci.USA 2006; 103:8185-8190. 18. Christensen JL, Weissman IL. Flk-2 is a marker in hematopoietic stem cell differentiation: A simple method to isolate long-term stem cells. Proc.Natl.Acad.Sci.USA 2001 ;98:14541-14546. 19. Yang L, Bryder D, Adolfsson J et al. Identification of Lin(-)Scal(+)kit(+)CD34(+)Flt3-short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients. Blood 2005;105:2717-2723. 20. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 21. Dick JE, Magli M C , Huszar D, Phillips RA, Bernstein A. Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/Wv mice. Cell 1985;42:71-79. 22. Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell 1986;45:917-927: 23. Jordan CT, Lemischka IR. Clonal and systemic analysis of long-term hematopoiesis in the mouse. Genes Dev. 1990;4:220-232. 24. Keller G, Snodgrass R. Life span of multipotential hematopoietic stem cells in vivo. J.Exp.Med. 1990;171:1407-1418. 25. Pawliuk R, Eaves C, Humphries RK. Evidence of both ontogeny and transplant dose-regulated expansion of hematopoietic stem cells in vivo. Blood 1996;88:2852-2858. 26. Smith L G , Weissman IL, Heimfeld S. Clonal analysis of hematopoietic stem-cell differentiation in vivo. Proc.Natl.Acad.Sci.USA 1991;88:2788-2792. 155 27. Spangrude GJ, Brooks D M , Tumas DB. Long-term repopulation of irradiated mice with limiting numbers of purified hematopoietic stem cells: in vivo expansion of stem cell phenotype but not function. Blood 1995;85:1006-1016. 28. Zhong R-K, Astle C M , Harrison DE. Distinct developmental patterns of short-term and long-term functioning lymphoid and myeloid precursors defined by competitive limiting dilution analysis in vivo. J.Immunol., 1996;157:138-145. 29. Randall TD, Weissman IL. Phenotypic and functional changes induced at the clonal level in hematopoietic stem cells after 5-fluorouracil treatment. Blood 1997;89:3596-3606. 30. Szilvassy SJ, Ragland PL, Miller CL, Eaves CJ. The marrow homing efficiency of murine hematopoietic stem cells remains constant during ontogeny. Exp.Hematol. 2003;31:331-338. 31. Sieburg HB, Cho RH, Dykstra B et al. The hematopoietic stem compartment consists of a limited number of discrete stem cell subsets. Blood 2006;107:2311-2316. 32. Jones RJ, Collector MI, Barber JP et al. Characterization of mouse lymphohematopoietic stem cells lacking spleen colony-forming activity. Blood 1996;88:487-491. 33. Uchida N, Dykstra B, Lyons KJ, Leung F Y K , Eaves CJ. Different in vivo repopulating activities of purified hematopoietic stem cells before and after being stimulated to divide in vitro with the same kinetics. Exp.Hematol. 2003;31:1338-1347. 34. Ema H, Sudo K, Seita J et al. Quantification of self-renewal capacity in single hematopoietic stem cells from normal and Lnk-deficient mice. Dev.Cell 2005;8:907-914. 35. Muller-Sieburg CE, Sieburg HB. Clonal diversity of the stem cell compartment. Curr.Opin.Hematol. 2006; 13:243-248. 36. Trentin JJ. Determination of bone marrow stem cell differentiation by stromal hemopoietic inductive microenvironments (HIM). Am.J.Pathol. 1971;65:621-628. 37. Metcalf D. Lineage commitment and maturation in hematopoietic cells: the case for extrinsic regulation. Blood 1998;92:345-348. 38. Moore K A , Lemischka IR. Stem cells and their niches. Science 2006;311:1880-1885. 39. Till JE, McCulloch EA, Siminovitch L. A stochastic model of stem cell proliferation, based on the growth of spleen colony-forming cells. Proc.Natl.Acad.Sci.USA 1964;51:29-36. 40. Abkowitz JL, Persik MT, Shelton GH et al. Behavior of hematopoietic stem cells in a large animal. Proc.Natl.Acad.Sci.USA 1995;92:2031-2035. 41. Kirkland M A . A phase space model of hemopoiesis and the concept of stem cell renewal. Exp.Hematol. 2004;32:511-519. 156 42. Roeder I, Kamminga L M , Braesel K et al. Competitive clonal hematopoiesis in mouse chimeras explained by a stochastic model of stem cell organization. Blood 2005; 105:609-616. 43. Audet J, Miller CL, Eaves CJ, Piret JM. Common and distinct features of cytokine effects on hematopoietic stem and progenitor cells revealed by dose response surface analysis. Biotechnol.Bioeng. 2002;80:393-404. 44. Kiel MJ, Yilmaz OH, Iwashita T et al. S L A M family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 2005; 121:1109-1121. 45. Yang L, Bryder D, Adolfsson J et al. Identification of Lin(-)Scal(+)kit(+)CD34(+)Flt3-short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients. Blood 2005;105:2717-2723. 46. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 47. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 48. Lessard J, Faubert A, Sauvageau G. Genetic programs regulating HSC specification, maintenance and expansion. Oncogene 2004;23:7199-7209. 49. Park IK, Qian D, Kiel M et al. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 2003;423:302-305. 50. Cheng T, Rodrigues N, Shen H et al. Hematopoietic stem cell quiescence maintained by p 2 1 c i P i / w a f i S c i e n c e 2000;287:1804-1808. 51. Bradford GB, Williams B, Rossi R, Bertoncello I. Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment. Exp.Hematol. 1997;25:445-453. 52. Cheshier SH, Morrison SJ, Liao X, Weissman IL. In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells. Proc.Natl.Acad.Sci.USA 1999;96:3120-3125. 53. Iscove NN, Nawa K. Hematopoietic stem cells expand during serial transplantation in vivo without apparent exhaustion. Curr.Biol. 1997;7:805-808. 54. Rufer N, Brummendorf TH, Kolvraa S et al. Telomere fluorescence measurements in granulocytes and T lymphocyte subsets point to a high turnover of hematopoietic stem cells and memory T cells in early childhood. J.Exp.Med. 1999;190:157-167. 55. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 157 56. Ortiz M , Wine JW, Lohrey N et al. Functional characterization of a novel hematopoietic stem cell and its place in the c-kit maturation pathway in bone marrow cell development. Immunity 1999;10:173-182. 57. Muller-Sieburg CE, Cho RH, Thoman M , Adkins B, Sieburg HB. Deterministic regulation of hematopoietic stem cell self-renewal and differentiation. Blood 2002;100:1302-1309. 58. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J Pathol 2006;169:338-346. 59. Miyamoto T, Akashi K. Lineage promiscuous expression of transcription factors in normal hematopoiesis. Int.J Hematol. 2005;81:361-367. 60. Sudo K, Ema H, Morita Y, Nakauchi H. Age-associated characteristics of murine hematopoietic stem cells. J Exp Med 2000;192:1273-1280. 61. Kim M , Moon HB, Spangrude GJ. Major age-related changes of mouse hematopoietic stem/progenitor cells. Ann.N.Y.Acad Sci 2003;996:195-208. 62. Rossi DJ, Bryder D, Zahn JM et al. Cell intrinsic alterations underlie hematopoietic stem cell aging. Proc Natl Acad Sci U.S.A 2005;102:9194-9199. 63. Liang Y, Van Zant G, Szilvassy SJ. Effects of aging on the homing and engraftment of murine hematopoietic stem and progenitor cells. Blood 2005;106:1479-1487. 64. Linton PJ, Dorshkind K. Age-related changes in lymphocyte development and function. Nat.Immunol 2004;5:133-139. 158 Chapter 5 Discussion and Future Directions 5.1 Major contributions In the studies presented in this thesis, we analyzed the functional properties of a large number of individually purified HSCs to obtain new information about the regulation of HSC self-renewal both in vitro and in vivo. This approach allows the clonal outputs of individual HSCs to be unambiguously assessed over time. It thus provides a powerful strategy for dissecting the unique characteristics of individual HSCs within a large and heterogeneous HSC population. In recipients of single HSCs, variable WBC repopulation levels were documented at 16 weeks post-transplant (Ch 2-4). When the contribution to the WBC pool was tracked at multiple time points, distinct reconstitution patterns were discerned (Ch 2 and 4). Differences in cell cycle times (Ch 2 and 3) and clone size in vitro (Ch 3) were also observed. Diverse ratios of differentiated WBC types produced in vivo from single transplanted cells allowed the identification of distinct HSC subtypes (Ch 4), and evidence that these subtypes were intrinsically pre-determined was obtained from the observation that their defining reconstitution patterns were faithfully transmitted through many self-renewal divisions in vivo. In Chapter 2,1 provided definitive evidence that two equally mitogenic growth factor cocktails can have remarkably disparate effects on HSC self-renewal, even within the span of a single cell cycle. Specifically, the frequency of first-division doublets with HSC activity was found to be the same as the frequency of HSCs in the original lin~Rho~ 159 SP cells when they had been stimulated to divide in the presence of 300 ng/ml SF plus 20 ng/ml IL-11 plus 1 ng/ml Flt3L, a combination described by Audet et al. 1' 2 to maximize HSC self-renewal over culture periods of up to 10 days. Thus every HSC in mouse B M that had a lin~Rho~ SP phenotype could be stimulated to execute a self-renewal division in the first cell cycle it completed in vitro when stimulated with this growth factor cocktail. In contrast, the second growth factor cocktail, consisting of a lower concentration of SF (10 ng/ml) and a relatively high concentration of TPO (100 ng/ml), produced first-division doublets that showed a pronounced loss of HSC activity in spite of a similar HSC mitogenic response. Therefore, it could be concluded that a significant proportion (>40%) of the lin"Rho" SP cells that divide in the first 48 hours in SF + TPO were HSCs prior to culture and that more than half of them lose HSC activity by the time they complete a first division. These findings extend previous results with less purified cells suggesting that growth factor receptor-mediated signals can alter HSC self-renewal decisions (see Chapter 1, section 1.2.4.3 for details) and demonstrate for the first time that these can be differently enacted within a single cell cycle without affecting the rate at which the cells progress through that cycle. These findings further demonstrate that growth factor regulation of HSC mitogenesis and self-renewal are dissociated control mechanisms, inviting interest in the future elucidation of the molecular basis of this dissociation. In Chapter 3,1 describe the use of a novel time-lapse video microscopy system to link data from high-resolution real-time monitoring of expanding HSC-derived clones in vitro with the results of in vivo reconstituting assays of the same clones evaluated individually. The scale of these experiments made it possible to search for distinct 160 features of in vitro cell behaviour that are associated with HSC self-renewal divisions. From a survey of numerous candidate features, we identified two that showed a significant association with 4-day HSC-containing clones. One feature was a prolonged cell-cycle time measured over three successive divisions. The second was the presence of a reduced proportion of progeny with uropodia at any time between 84 and 96 hours after initiation of the cultures. In combination, these parameters identified all of the HSC-containing clones in each of the three experiments performed and consistently enhanced the identification of HSC-containing clones 2- to 3-fold independent of the starting purities of the HSCs tested, these findings build on the results of previous studies that correlated longer cell-cycle times of self-renewing pluripotent cells with clonogenic activity in methyl cellulose cultures3 and extend them to HSCs. Through the use of a highly purified HSC starting population, increased spatial-temporal-resolution of the assessment of cell cycle times, and in vivo assessment of whether or not each tracked clone retained functional HSC activity, a link between HSC cell-cycle time and their self-maintenance in culture was definitively established. In Chapter 4,1 presented the results of longitudinal analyses of WBC outputs performed on a large number of recipients of single purified HSCs (or in vitro clones derived from HSCs). These analyses revealed that the adult HSC compartment can be resolved into 4 subtypes. Two of these stably and autonomously propagate their unique patterns of WBC reconstitution through many self-renewal divisions in vivo. Conversely, when stimulated to proliferate in vitro, HSCs generated within the same clone were shown to rapidly shift to less competitive patterns of WBC output, although remarkable similarity was again seen in the reconstitution patterns displayed by cells within the same 161 in vitro clone. Thus, even when many daughter cells from the same in vivo or in vitro clone were injected and the repopulation patterns were different from those characteristic of the parental cells, the patterns produced by different aliquots of the same clone mimicked one another with extraordinary similarity. Collectively, these findings suggest that one of a small number of predetermined, but not irrevocable programs of differentiation behavior are imposed on HSCs at an early point in their development and that these are then symmetrically propagated even when they are altered. The most likely mechanism for establishing and maintaining such symmetry during many HSC self-renewal divisions and yet allow rapid alterations in the face of appropriate external cues would seem to be via the epigenetic modification(s) of genes involved in setting the pace and ease of activation of particular WBC differentiation lineages. Regardless of the underlying mechanisms, it now seems clear that self-renewal divisions of particular LTRCs in vivo do not reset the differentiation options of their LTRC progeny. Together, these results show that individual members of the HSC compartment show previously unappreciated pre-existing heterogeneity, and that external influences on HSC self-renewal and differentiation in vivo remain delimited by these intrinsic properties. 162 5.2 Implications and Future Directions 5.2.1 Basis of Functional Heterogeneity in HSCs Over the years there has been a continuing debate regarding the basis of observed heterogeneity in the HSC compartment. While stochastic modelling may provide useful descriptions and hypotheses regarding the potential range of behaviours that HSC may display, the data presented in this thesis indicate that individual HSCs have predetermined self-renewal and subsequent differentiation properties although these can also be modified by exposure to different environmental cues. This concept has been reinforced by observations of inbred mouse strain-specific variation in the self-renewal and aging properties of HSCs4"7. Further studies have identified some of the genetic loci responsible for these differences8"1and have revealed that these are complex traits that involve multiple genes12. In addition, even in identical genetic backgrounds, intrinsic differences in HSC properties have been described when comparing HSCs from mice of different ages13'14 or developmental stages15. Furthermore, using in vitro limiting dilution techniques, Muller-Sieburg has described heritable intrinsic heterogeneity between HSCs from mice of the same age16. The observations described in this thesis have extended these latter findings using single cell transplants to directly examine the functional properties of a large number of individually purified HSCs, including assessments of their self-renewal abilities by secondary and tertiary transplantations. This permitted the identification of four distinct HSC subtypes based on the ratios of differentiated WBC types that they produced in vivo. 163 The two subtypes capable of self-renewal in vivo autonomously propagated their unique patterns of WBC production through many self-renewal divisions in vivo. If the observed heterogeneity could be explained by random internal or external events, we would have predicted that a similar degree of heterogeneity would be seen in secondary and tertiary recipients. However, the serial transplant results described in Chapter 4 suggest that this is not the case, and thus provide support for the alternative view; i.e., that this heterogeneity is largely intrinsically determined and not due to microenvironmental influences. These results also affirm that the control of lineage output and the control of self-renewal are two distinct processes. For example, y- and 8 LTRCs lack extensive self-renewal ability but can produce significant numbers of lymphoid and myeloid lineages. Conversely, a-LTRCs can self-renew extensively but produce primarily myeloid progeny. Based on our results, we have proposed the model shown in figure 5.1a to explain the relationships between the 4 LTRC subtypes identified. P-, y-, or 5-LTRCs appear to form a hierarchy that corresponds to conventional models of HSC differentiation (see section 1.1.3 for details) where the propensity for generating mature myeloid progeny diminishes progressively before pluripotentiality is lost. a-LTRCs represent a novel cell type, in which a strong propensity for myeloid cell output can be independently and exclusively sustained over multiple cycles of hematopoietic reconstitution, although direct or sequential transitioning into P-, y-, and 5-LTRCs is not precluded. An alternate model is shown in figure 5.1b. In this model, the P -LTRC is at the top of the hierarchy. P-LTRC can self-renew to produce additional P-LTRCs. Upon 164 division, if a progeny LTRC loses multi-lineage capacity, it becomes an a-LTRC, whereas if it loses extensive self-renewal ability, it becomes a y-LTRC. In general, these are one-way transitions, but in some cases, partial reversion may be seen. It should be noted that data from Muller-Sieburg et al16also suggests the existence of a self-renewing LTRC that has lost the majority of its myeloid differentiation potential. However no example of such a cell was observed in the present studies. While these models describing the relationships between L T R C types are both consistent with the observations described, further experiments are required to verify to what extent these models correct. From the data presented in Chapter 4, it is not possible to determine whether beta-LTRCs can produce alpha-LTRCs and if so, at what frequency. This is because the secondary and tertiary transplants were not performed using limiting LTRC numbers, and thus any alpha-LTRCs produced would be masked if co-transplanted with any other LTRC types due to their production of lymphoid cells. Another unresolved question regards the alpha-LTRCs that gained lymphoid differentiation ability upon serial transplantation. Through the use of serial transplants of limiting numbers of HSCs, it could be determined whether one, some, or all of the regenerated HSCs switched this functional property. By performing an additional round of transplants, it would be possible to determine if this gain of lymphoid differentiation ability was due to the production of beta-, gamma-, or delta-LTRCs. These issues are impossible to resolve with the current data, and would require performing the secondary and tertiary transplants at the single LTRC level. Ideally, this would be accomplished using single re-purified HSCs from the primary recipients. However, the CD45m i dLin" Rho'SP purification strategy is no longer as effective for regenerated HSCs, and so an 165 alternate purification strategy would be required. The S L A M receptors have recently been identified as promising candidates for the isolation of HSCs from previously transplanted mice17. An alternative would be to perform the transplants using limiting numbers of cells such that each recipient would be very likely to have received either one or zero HSCs. The latter model (Fig. 5.1b) might be explained by invoking both extrinsic and intrinsic control parameters. For example, alternate niches might exist that influence the progeny outcome of the division of P -LTRCs 1 8 . One niche might promote the production of additional P-LTRCs, another might promote the transition to a-LTRCs, and yet another might promote the transition to y-LTRCs. However, to explain the nearly exclusive regeneration of a-LTRCs over many self-renewal divisions in vivo, this model requires that the changes induced by these alternate niches are then fixed and largely irreversible even when cells are transferred to secondary or tertiary hosts. Changes in the relative sizes of these putative niches might also be invoked to explain why the HSC compartments of fetal liver, young adult B M , and aged adult B M differ with respect to their proportional content of the four LTRC subtypes. For example, during embryogenesis and early development, conditions may encourage the formation of additional P- and y-LTRCs, but not a-LTRCs. During adulthood and especially as aging progresses, the conditions that promote the formation of additional a-LTRCs would thus increase. Alternatively, the formation of a-LTRCs from P-LTRCs might be a low-frequency stochastic event and these cells simply accumulate with age. To determine whether the different HSC characteristics of fetal liver, young adult B M , and aged adult B M are due to exclusively intrinsic events or whether the 166 environment is p r i m a r i l y responsible, it w o u l d be appealing to transplant H S C s f rom different stages o f development into each o f these environments and measure the regeneration o f the different L T R C types produced by l i m i t i n g d i lu t ion assays i n secondary recipients. In addi t ion, it w o u l d be interesting to alter the expression o f genes * * 19 20' 21 shown to be i n v o l v e d i n H S C aging, such as Ezh-2 or pl 6 ' , and determine whether the aging-related functional changes are mit igated. S i m i l a r l y , i f the expression o f genes associated w i t h fetal l i ve r H S C s such as AMLl/Runxl22 or sci23 are manipulated, it w o u l d be interested to determine i f the numbers or types o f L T R C s are altered. A close examinat ion o f the data i n Chapter 4 also reVeals "microheterogenei ty" w i t h i n the 4 L T R C subtypes defined here. W h e n L T R C s were c l o n a l l y ampl i f i ed in vivo or in vitro, the W B C output over t ime is remarkably s imi la r i n "sister" L T R C s . That this occurs both in vivo and in vitro suggests that there is a degree o f int r ins ic p rog ramming i n each i nd iv idua l L T R C that specifies the behaviour o f a l l o f its progeny i n response to external s t imul i . Th i s provides further support for the intr insic nature o f H S C heterogeneity. T o investigate the mechanisms i n v o l v e d and possible invo lvement o f epigenetic changes to specific genes, it w o u l d be o f part icular interest to develop new methods for isola t ing enriched populat ions o f different types o f L T R C s and compare their chromat in status. 5.2.2 HSC self-renewal The H S C compartment is maintained b y the process o f self-renewal, w h i c h refers to a ce l l d i v i s i o n i n w h i c h one or both o f the daughters o f an H S C retains its differentiation and proliferat ive potentialities. The dec is ion o f H S C s to self-renew or 167 become more restricted is governed by a complex interplay between intrinsically set properties and stimuli from the surrounding microenvironment (external cues). Several of the known mechanisms regulating self-renewal of HSCs are discussed in detail in section 1.2.4 of this thesis. Experimentally, self-renewal is difficult to define. In fact, it is fundamentally impossible to determine whether a daughter HSC shares the same properties as its parent since once the daughter cell is produced, the parent no longer exists in its original form. In addition, it is difficult to measure every possible key parameter to determine whether two cells are functionally identical. Thus, the criteria used to define functional equivalency are at best crude approximations of molecular events being implicated but not yet clarified. In the case of HSCs, it is generally assumed that long-term differentiated cell output is a useful indicator of self-renewal, based on the assumption that cells without self-renewal ability (i.e., non-HSCs) will exhaust their ability to produce mature progeny prior to the termination of the assay. This is one of the basic premises behind long-term in vitro assays such as CAFCs or LTC-ICs, as well as in vivo long-term repopulation assays24. Since HSCs are often defined by their readout in a particular assay, in many situations the most rigorous and relevant definition of self-renewal is to test daughter cells for HSC activity using the same assay as applied to the starting population. In this thesis, I have explored aspects of HSC self-renewal by analyzing in vitro and in vivo clones initiated with single highly purified HSCs. In Chapter 2,1 show that HSC self-renewal in vitro can be altered within a single cell cycle, in response to differential extrinsic factors. In Chapter 3,1 have described that in vitro self-renewal (or lack thereof) associates with cell cycle cycle length, clone size and lack of presence of 168 uropods. In Chapter 4,1 show that self-renewal in vivo, as determined by secondary transplantability, does not associate particularly well with total repopulation level, lymphoid cell output, or even the presence or absence of multi-lineage repopulation. Rather, in vivo self-renewal can be best predicted by the absolute contribution to myeloid WBCs produced at 16-24 weeks post-transplant, as well as a moderate to high proportion of myeloid versus lymphoid cell output. However, it should be noted that in the experiments just described, quantification of HSC self-renewal was not performed; rather, a simple "yes or no" answer was obtained in each case. An added layer of information could be obtained by the execution of rigorous limiting dilution transplants to determine the extent of self-renewal in each case. This would contribute a valuable additional parameter with which to examine the functional heterogeneity of HSC. It is interesting to note that when HSCs are stimulated to divide, the daughter cells produced tend to share similar functional properties with each other, suggesting the existence of intrinsic mechanisms in the original HSCs that delimit the properties of their progeny. Interestingly, in vivo, the progeny HSCs tend to retain the functional characteristics of the parent HSC, even after many self-renewal divisions. In contrast, I have demonstrated that HSC characteristics can be rapidly altered when they are stimulated to divide in vitro, although the daughters tend to be altered in a fashion that is predetermined by the parent HSC. This leads to the interesting question of how and why the in vivo and in vitro environments tested here have such discrepant effects on HSC self-renewal. There are many extrinsic and intrinsic factors that co-operate to allow HSC self-renewal to occur in vivo. Some of the factors involved in this process are discussed in section 1.2.4.1 and 1.2.4.2 of Chapter 1 of this thesis. When HSCs are stimulated in vitro, 169 however, the conditions are more defined, and there are likely one or more components that are lacking by comparison to the environments in vivo where HSCs self-renew. One component that seems to be of particular importance to HSC self-renewal is Steel factor (SF) as described in section 1.2.4.3 of Chapter 1. Indeed, recent data generated in our lab suggest self-renewal ability can be differentially altered in vivo even prior to a first cell division, in cultures with sufficiently reduced levels of SF but otherwise identical conditions (David Kent, personal communication). This effect was not due to alterations in apoptosis, since apoptosis was not seen in either condition, nor was HSC entry into and rate of progression through the cell cycle altered. It would also be of interest to modify the culture conditions and determine whether the culture-induced shift of LTRC types described in Chapter 4 can be altered. For example, if conditions previously demonstrated to increase HSC self-renewal were used, might the culture-induced shift be mitigated? Examples of such strategies might include the addition of Tat-HOXB4 2 5 , activation of the Wnt/Frizzled signalling pathwajr ' , the use of stromal feeder layers ' , or alternative cytokine conditions such as FGF-1 3 0 . Additionally, transgenic systems could be envisaged with inducible expression of intrinsic regulators known to increase HSC self-renewal, such as Bmi-1 31 32 overexpression and/or Pbx knockdown . It has also become apparent that differences in self-renewal ability exist between HSCs from different stages of development. In particular, HSCs from day 14.5 fetal liver have been shown to exhibit faster HSC regeneration rates in vivo than their adult B M counterparts15'33'34. It has recently been determined that these different HSC regeneration properties are largely intrinsically determined, and that a transition from fetal-like to 170 adult-like occurs rapidly, between 3 and 4 weeks post-birth15. Differences in self-renewal ability between HSCs from young and old mice have also been suggested14,35, but this has not yet been as rigorously addressed. However, it is also interesting to consider the extent to which different environments may play a role in altering the self-renewal behaviour of HSCs from different developmental stages. In recent studies of fetal liver and young adult HSCs, it has been suggested that several characteristics, including their self-renewal potential, are intrinsically modulated independently of the environment in which the HSCs are expanding15. However, this data is not complete and will require additional experiments to be conclusive. With respect to the differences between young and old HSCs, environment does seem to play a partial role, as determined by experiments where young and old HSCs are transplanted into young and old recipients14,36. An exciting next step would be to transplant fetal liver, young adult B M , and old adult B M HSCs into prenatal, young, and old mice, and thus resolve the extent of intrinsic versus extrinsic determination of behaviours seen in HSCs from different stages of development. 5.2.3 Intrinsic Determination of Lineage Output Patterns by Individual HSCs. As discussed in Chapter 1, primitive hematopoietic cells express low levels of many genes normally associated with differentiating or differentiated cell types (reviewed in 3 7). It has thus been hypothesized that open chromatin structure is maintained in 38 primitive cells, with progressive epigenetic changes occurring during differentiation . 171 Changes in transcription factor levels coincide with these epigenetic changes, resulting in the up-regulation and down-regulation of appropriate lineage-specific genes in a step-wise fashion39. Along with earlier work by Muller-Sieburg16, the data presented in Chapter 4 demonstrate the existence of differences in lineage output in mice repopulated with single HSCs which is transferred stably through many self-renewal divisions. Because these differences are present within (genetically identical) cells from the same individuals and yet can be so stably propagated, an epigenetic mechanism appears likely. HSCs with different regions of active or inactive chromatin could influence the production or differentiation of certain downstream cell types. For example, Muller-Sieburg has shown that lymphoid progenitors produced by a "myeloid-biased" (a-LTRC-like) HSC demonstrate a blunted response to IL-6 4 0. Presumably, in the HSC, the self-renewing program and myeloid differentiation program are active, but the lymphoid differentiation program has been dampened. It would thus be of particular interest to attempt to modify the epigenome of HSCs and determine whether different patterns of reconstitution could be obtained, and if the distribution or symmetry of LTRC types present were altered. One strategy would be to alter the expression in HSCs of one or more of the Polycomb group genes, which have well-known roles in epigenetic regulation (reviewed in 4 1' 4 2). Another approach would be to remodel the chromatin in a more random fashion by treatment with drugs that interfere with histone deacetylation or DNA methylation. Indeed, several examples of such drugs have been suggested to modify HSC properties in culture, including 5-azo2'deoxycytidine43'44, valproic acid4 5, trichostatin A 4 3 ' 4 4 ' 4 6 , trapoxin46, and chlamydocin46. 172 Interestingly, Sudo et al 1 3 and Rossi et al 1 4 suggest that self-renewing cells with a-LTRC characteristics increase with mouse age. A similar phenomenon has also been described by other groups, albeit less rigorously3 6'4 7 , 4 8. This shift in the HSC compartment with age might partly explain the well-documented decrease in lymphopoiesis with age (reviewed in 4 9). As a result, the a type is seen as a 'defective H S C ' 1 3 ' 4 0 that has retained self-renewal ability, lost lymphoid ability, and accumulates with age. While I did not examine aged mouse HSCs in this study, I did show using a serial transplantation model that a single a-LTRC can produce L T R C progeny that have a significant capacity for lymphopoiesis, suggesting that this 'defect' is not necessarily a permanent one. Nonetheless, it would be intriguing to test the LTRCs present in old mouse B M and determine whether the same LTRC types are seen but in different proportions, or whether the LTRCs themselves have changed their properties. In particular, it would be interesting to compare the a-LTRCs from young B M with the similar cells in old B M . The successful prospective isolation of the LTRC types identified in Chapter 4 would be an important accomplishment to provide additional support for their intrinsic nature, and would enable multiple avenues of further study. A first step towards identifying a phenotype would be to search for markers that subdivide the CD45m , dLin" Rho'SP population, followed by functional analysis to determine whether the LTRC types are segregated between the subpopulations. Potential candidates to test include CD34 5 0 , CD27 5 1 , Flk2/Flt3 5 2 ; 5 3, endoglin/CD105 5 4 , S L A M family receptors CD150 and CD244 5 5, EPCR/CD20 1 5 6 , or a-2 integrin/CD49b57. 173 If a phenotype could be identified that would allow distinct LTRC types to be separately isolated, this would permit comparative studies of their gene expression profiles to identify genes that are differentially expressed and hence might be considered candidates responsible for the different functional properties of these cells. Indeed, techniques enabling the global transcriptional profiling of single hematopoietic stem cells are now available58. Purified LTRCs of the various types would also enable the analysis of the epigenomic differences between them. In light of the hypothesized role of chromatin remodelling in maintaining their distinctive characteristics, this would be a particularly applicable strategy. Recent advances in chromatin immunoprecipitation technology have allowed this technique to be performed on samples as small as 100 cells59, which might enable the chromatin status of purified LTRCs to be analyzed. In addition, in vitro tracking of prospectively isolated cells would allow for further characterization of LTRC properties in vitro. Similarly, analysis of dissociated doublets and quadruplets could provide new insights into specific LTRC relationships. A complementary approach would be to attempt to prospectively identify L T R C types by carefully observing their behaviour with the time lapse video microscopy system described in Chapter 3. If certain characteristics could be associated with particular LTRC subtypes, it might allow for a higher throughput screening of candidate markers for prospective isolation. In addition, it might enable more rapid testing for various culture conditions that are better able to maintain LTRCs in vitro. It may also be informative to assay for and compare LTRC types in different mouse strains. If the LTRC proportions are sufficiently different between different mouse 174 strains, it would demonstrate a distinct genetic component, and further may present an opportunity to single out genes that contribute to the difference between the strains. 5.3. Concluding Comments For decades, the H S C compartment has been seen as a black box - early experiments clearly revealed that it exists, but what exactly was inside was not fully appreciated. Retroviral marking experiments, along with purification strategies that revealed functionally distinct subsets have provided a glimpse into the variety of cell potentialities within this primitive subset, but the exact details were still not elucidated. The advent of advanced purification strategies has ushered in an era of new possibilities, permitting a range of existing techniques to be modified to directly study single HSCs . Recent examples include global transcription analysis 5 8, l ipid raft clustering 6 0, localization in vivo6], and the analysis of lineage potential in vitro62. To this impressive list, we can now add the examples described in this thesis, including cytokine effects during initial cell cycling in vitro (Chapter 2), time-lapse photography of cells in culture (Chapter 3) and the analysis of H S C potentials using single cell transplantations in vivo (Chapter 4). Collectively, this work has opened up the black box a little bit, and the mysteries of the H S C compartment have begun to be unravelled. Certainly, it is now clear that H S C s are not functionally homogeneous. The studies described here have demonstrated that this diversity is primarily instrinsic yet can be influenced by external factors. This is an exciting realization, since it suggests that 175 different heritable molecular states exist for HSCs that define their functional properties, leading to the possibility of prospective identification of HSCs with different properties and the elucidation of the molecular mechanisms that distinguish them. Undoubtedly, some of the genes that are uniquely regulated in HSCs with different functional properties, and the mechanisms by which they are regulated, would offer insights into the control of the fundamental HSC properties of self-renewal and lineage output. If the pathways affecting these decisions could then be modulated, it would have far-reaching implications for many aspects of hematopoietic stem cell biology. For example, HSCs of particular types could conceivably be amplified ex vivo for transplantation purposes, for treatment specific to a particular blood disorder. Moreover, these pathways might be informative in the development of methods for generating hematopoietic cells from embryonic stem cells, and the subsequent production of differentiated blood cells of specific types. In addition, a greater understanding of these mechanisms could improve our understandings of leukemogenesis, as well as identify possible molecular targets for cancer therapy. 176 B Multipotent Myeloid restricted Figure 5.1. Hypothesized relationships between L T R C subtypes P-, y-, or 8-LTRCs appear to form a hierarchy that corresponds to conventional models of HSC differentiation where the propensity for generating mature myeloid progeny diminishes progressively before pluripotentiality is lost. a-LTRCs represent a novel cell type, in which a very strong propensity for myeloid cell output can be independently and exclusively sustained over multiple cycles of hematopoietic reconstitution, although crossover to the P-, y-, or 8-LTRC stream is not precluded. In panel A, a-LTRCs and P-LTRCs are seen as two discrete HSC types with unique lineage output properties, as described in chapter 4. In panel B, an alternate hypothesis is shown. P-LTRCs are seen as the only "true" HSCs, with both multipotentiality and self-renewal ability. Through subsequent intrinsic changes, the P-LTRCs lose either self-renewal ability and become y-LTRCs, or lose multipotency and become a-LTRCs. In mice repopulated with multiple LTRCs, these changes would not be seen since they would be masked by any remaining p-LTRCs, but their unique characteristics become apparent through single cell transplantations. 177 5.4. References 1. Audet J, Miller CL, Eaves CJ, Piret JM. Common and distinct features of cytokine effects on hematopoietic stem and progenitor cells revealed by dose-response surface analysis. Biotechnol.Bioeng. 2002;80:393-404. 2. Audet J, Miller CL, Rose-John S, Piret JM, Eaves CJ. Distinct role of gpl30 activation in promoting self-renewal divisions by mitogenically stimulated murine hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 2001;98:1757-1762. 3. Suda T, Suda J, Ogawa M . Proliferative kinetics and differentiation of murine blast cell colonies in culture: evidence for variable GO periods and constant doubling rates of early pluripotent hemopoietic progenitors. J.Cell Physiol. 1983; 117:308-318. 4. Muller-Sieburg CE, Riblet R. Genetic control of the frequency of hematopoietic stem cells in mice: mapping of a candidate locus to chromosome 1. J.Exp.Med. 1996;183:1141-1150. 5. de HG, Nijhof W, Van ZG. Mouse strain-dependent changes in frequency and proliferation of hematopoietic stem cells during aging: correlation between lifespan and cycling activity. Blood. 1997;89:1543-1550. 6. Chen J, Astle C M , Harrison DE. Development and aging of primitive hematopoietic stem cells in BALB/cBy mice. Exp.Hematol. 1999;27:928-935. 7. Kamminga L M , Akkerman I, Weersing E et al. Autonomous behavior of hematopoietic stem cells. Exp.Hematol. 2000;28:1451-1459. 8. Chen J, Astle C M , Muller-Sieburg CE, Harrison DE. Primitive hematopoietic stem cell function in vivo is uniquely high in the CXB-12 mouse strain. Blood. 2000;96:4124-4131. 9. Morrison SJ, Qian D, Jerabek L et al. A genetic determinant that specifically regulates the frequency of hematopoietic stem cells. J.Immunol. 2002;168:635-642. 10. Yuan R, Astle C M , Chen J, Harrison DE. Genetic regulation of hematopoietic stem cell exhaustion during development and growth. Exp.Hematol. 2005;33:243-250. 11. Muller-Sieburg CE, Cho RH, Sieburg HB, Kupriyanov S, Riblet R. Genetic control of hematopoietic stem cell frequency in mice is mostly cell autonomous. Blood. 2000;95:2446-2448. 12. Henckaerts E, Langer JC, Snoeck HW. Quantitative genetic variation in the hematopoietic stem cell and progenitor cell compartment and in lifespan are closely linked at multiple loci in BXD recombinant inbred mice. Blood. 2004;104:374-379. 178 13. Sudo K, Ema H, Morita Y, Nakauchi H. Age-associated characteristics of murine hematopoietic stem cells. J.Exp.Med. 2000;192:1273-1280. 14. Rossi DJ, Bryder D, Zahn JM et al. Cell intrinsic alterations underlie hematopoietic stem cell aging. Proc.Natl.Acad.Sci.U.S.A. 2005;102:9194-9199. 15. Bowie MB, Kent D, Dykstra B et al. Identification of a New Intrinsically Timed Developmental Checkpoint That Reprograms Key Hematopoietic Stem Cell Properties. Proc.Natl.Acad.Sci.U.S.A. 3006 16. Muller-Sieburg CE, Cho RH, Thoman M , Adkins B, Sieburg HB. Deterministic regulation of hematopoietic stem cell self-renewal and differentiation. Blood. 2002;100:1302-1309. 17. Yilmaz OH, Kiel MJ, Morrison SJ. S L A M family markers are conserved among hematopoietic stem cells from old and reconstituted mice and markedly increase their purity. Blood. 2006;107:924-930. 18. Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am.J.Pathol. 2006;169:338-346. 19. Kamminga L M , Bystrykh LV, de Boer A et al. The Polycomb group gene Ezh2 prevents hematopoietic stem cell exhaustion. Blood. 2006;107:2170-2179. 20. Zhang J, Grindley JC, Yin T et al. PTEN maintains haematopoietic stem cells and acts in lineage choice and leukaemia prevention. Nature. 2006;441:518-522. 21. Yilmaz OH, Valdez R, Theisen BK et al. Pten dependence distinguishes haematopoietic stem cells from leukaemia-initiating cells. Nature. 2006;441:475-482. 22. Okuda T, van Deursen J, Hiebert SW, Grosveld G, Downing JR. AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell. 1996;84:321-330. 23. Lecuyer E, Hoang T. SCL: from the origin of hematopoiesis to stem cells and leukemia. Exp.Hematol. 2004;32:11-24. 24. Coulombel L. Identification of hematopoietic stem/progenitor cells: strength and drawbacks of functional assays. Oncogene. 2004;%20;23:7210-7222. 25. Krosl J, Austin P, Beslu N et al. In vitro expansion of hematopoietic stem cells by recombinant TAT-HOXB4 protein. Nat.Med. 2003;9:1428-1432. 26. Willert K, Brown JD, Danenberg E et al. Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature. 2003;423:448-452. 179 27. Reya T, Duncan AW, Ailles L et al. A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature. 2003;423:409-414. 28. Wineman J, Moore K, Lemischka I, Muller-Sieburg C. Functional heterogeneity of the hematopoietic microenvironment: rare stromal elements maintain long-term repopulating stem cells. Blood. 1996;87:4082-4090. 29. Moore K A , Ema H, Lemischka IR. In vitro maintenance of highly purified, transplantable hematopoietic stem cells. Blood. 1997;89:4337-4347. 30. de HG, Weersing E, Dontje B et al. In vitro generation of long-term repopulating hematopoietic stem cells by fibroblast growth factor-1. Dev.Cell. 2003;4:241-251. 31. Iwama A, Oguro H, Negishi M et al. Enhanced self-renewal of hematopoietic stem cells mediated by the polycomb gene product Bmi-1. Immunity. 2004;21:843-851. 32. Krosl J, Beslu N, Mayotte N, Humphries RK, Sauvageau G. The competitive nature of HOXB4-transduced HSC is limited by PBX1: the generation of ultra-competitive stem cells retaining full differentiation potential. Immunity. 2003;18:561-571. 33. Harrison DE, Zhong RK, Jordan CT, Lemischka IR, Astle C M . Relative to adult marrow, fetal liver repopulates nearly five times more effectively long-term than short-term. Exp.Hematol. 1997;25:293-297. 34. Rebel VI, Miller CL, Eaves CJ, Lansdorp PM. The repopulation potential of fetal liver hematopoietic stem cells in mice exceeds that of their liver adult bone marrow counterparts. Blood. 1996;87:3500-3507. 35. Mauch P, Botnick L E , Harmon EC, Obbagy J, Hellman S. Decline in bone marrow proliferative capacity as a function of age. Blood. 1982;60:245-252. 36. Liang Y, Van Zant G, Szilvassy SJ. Effects of aging on the homing and engraftment of murine hematopoietic stem and progenitor cells. Blood. 2005;106:1479-1487. 37. Akashi K. Lineage promiscuity and plasticity in hematopoietic development. Ann.N.Y.Acad.Sci. 2005;1044:125-31.: 125-131. 38. Cross M A , Enver T. The lineage commitment of haemopoietic progenitor cells. Curr.Opin.Genet.Dev. 1997;7:609-613. 39. Bonifer C. Epigenetic plasticity of hematopoietic cells. Cell Cycle. 2005;4:211-214. 40. Muller-Sieburg CE, Cho RH, Karlsson L, Huang JF, Sieburg HB. Myeloid-biased hematopoietic stem cells have extensive self-renewal capacity but generate diminished lymphoid progeny with impaired IL-7 responsiveness. Blood. 2004;103:4111-4118. 180 41. Lessard J, Sauvageau G. Polycomb group genes as epigenetic regulators of normal and leukemic hemopoiesis. Exp.Hematol. 2003;31:567-585. 42. Iwama A, Oguro H, Negishi M , Kato Y, Nakauchia H. Epigenetic regulation of hematopoietic stem cell self-renewal by polycomb group genes. Int.J.Hematol. 2005;81:294-300. 43. Milhem M , Mahmud N, Lavelle D et al. Modification of hematopoietic stem cell fate by 5aza 2*deoxycytidine and trichostatin A. Blood. 2004;103:4102-4110. 44. Araki H, Mahmud N, Milhem M et al. Expansion of human umbilical cord blood SCED-repopulating cells using chromatin-modifying agents. Exp.Hematol. 2006;34:140-149. 45. Bug G, Gul H, Schwarz K et al. Valproic acid stimulates proliferation and self-renewal of hematopoietic stem cells. Cancer Res. 2005;65:2537-2541. 46. Young JC, Wu S, Hansteen G et al. Inhibitors of histone deacetylases promote hematopoietic stem cell self-renewal. Cytotherapy. 2004;6:328-336. 47. Kim M , Moon HB, Spangrude GJ. Major age-related changes of mouse hematopoietic stem/progenitor cells. Ann.N.Y.Acad.Sci. 2003;996:195-208.:195-208. 48. Spangrude GJ, Brooks D M , Tumas DB. Long-term repopulation of irradiated mice with limiting numbers of purified hematopoietic stem cells: in vivo expansion of stem cell phenotype but not function. Blood. 1995;85:1006-1016. 49. Linton PJ, Dorshkind K. Age-related changes in lymphocyte development and function. Nat.Immunol. 2004;5:133-139. 50. Osawa M , Hanada K, Hamada H, Nakauchi H. Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science. 1996;273:242-245. 51. Wiesmann A, Phillips RL, Mojica M et al. Expression of CD27 on murine hematopoietic stem and progenitor cells. Immunity. 2000;12:193-199. 52. Adolfsson J, Borge OJ, Bryder D et al. Upregulation of Flt3 expression within the bone marrow Lin(-)Scal(+)c-kit(+) stem cell compartment is accompanied by loss of self-renewal capacity. Immunity. 2001;15:659-669. 53. Christensen JL, Weissman IL. Flk-2 is a marker in hematopoietic stem cell differentiation: a simple method to isolate long-term stem cells. Proc.Natl.Acad.Sci.U.S.A. 2001;98:14541-14546. 181 54. Chen CZ, Li M , de Graaf D et al. Identification of endoglin as a functional marker that defines long-term repopulating hematopoietic stem cells. Proc.Natl.Acad.Sci.U.S.A. 2002;99:15468-15473. 55. Kiel MJ, Yilmaz OH, Iwashita T et al. S L A M family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell. 2005;121:1109-1121. 56. Balazs AB, Fabian AJ, Esmon CT, Mulligan RC. Endothelial protein C receptor (CD201) explicitly identifies hematopoietic stem cells in murine bone marrow. Blood. 2006;107:2317-2321. 57. Wagers AJ, Weissman IL. Differential expression of alpha2 integrin separates long-term and short-term reconstituting Lin-/loThyl.l(lo)c-kit+ Sca-1+ hematopoietic stem cells. Stem Cells. 2006;24:1087-1094. 58. Ramos CA, Bowman TA, Boles NC et al. Evidence for Diversity in Transcriptional Profiles of Single Hematopoietic Stem Cells. PLoS.Genet. 2006;2: 59. O'Neill LP, VerMilyea MD, Turner B M . Epigenetic characterization of the early embryo with a chromatin immunoprecipitation protocol applicable to small cell populations. Nat.Genet. 2006;38:835-841. 60. Yamazaki S, Iwama A, Takayanagi S et al. Cytokine signals modulated via lipid rafts mimic niche signals and induce hibernation in hematopoietic stem cells. EMBO J. 2006;25:3515-3523. 61. Suzuki N, Ohneda O, Minegishi N et al. Combinatorial Gata2 and Seal expression defines hematopoietic stem cells in the bone marrow niche. Proc.Natl.Acad.Sci.U.S.A. 2006;103:2202-2207. 62. Takano H, Ema H, Sudo K, Nakauchi H. Asymmetric division and lineage commitment at the level of hematopoietic stem cells: inference from differentiation in daughter cell and granddaughter cell pairs. J.Exp.Med. 2004;199:295-302. 182 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0100570/manifest

Comment

Related Items