@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en, "Medical Genetics, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Fraser, Christopher Charles"@en ; dcterms:issued "2008-12-23T22:24:52Z"@en, "1992"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The adult bone marrow contains stem cells capable of reconstituting all hemopoietic lineages for the life-time of a recipient following lethal irradiation and bone marrow transplantation. Despite more than two decades of research to characterize this population of pluripotent stem cells, mechanisms that regulate the survival, proliferation and differentiation in the initial stages of hemopoietic development in vivo remain poorly understood. Methods for maintaining, expanding and following the fate of the most primitive hemopoietic cells in vitro thus offers major opportunities for the investigation into these mechanisms. The major objectives of this work were therefore initially to determine if totipotent stem cells capable of repopulating hemopoiesis in lethally irradiated mice could be maintained and proliferate for extended times in vitro, and subsequently to quantitate stem cell numbers in these cultures and assess their in vivo repopulating potential in lethally irradiated hosts. Initial studies were aimed at determining if stem cells capable of repopulating all hemopoietic lineages in lethally irradiated mice could be maintained and proliferate when grown in a long-term culture (LTC) system. Marrow cells from 5-fluorouracil treated male mice were infected with a helper-free recombinant virus carrying the neomycin resistance gene and seeded onto irradiated adherent layers of pre-established, long-term marrow cultures of female origin. At 4 weeks, cells from individual cultures were transplanted into multiple recipients. Southern blot analysis of hemopoietic tissues 45 days post transplant demonstrated large clonal populations common to lymphoid and myeloid tissues as indicated by the presence of unique retroviral insertion fragments. In a number of cases it was found that multiple recipients of a single flask were repopulated with the same clonally marked totipotent cells indicating expansion of such cells during culture prior to their injection into irradiated recipients. These results demonstrated for the first time both maintenance and expansion in vitro of totipotent stem cells with in vivo repopulating ability."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/3279?expand=metadata"@en ; dcterms:extent "3457654 bytes"@en ; dc:format "application/pdf"@en ; skos:note "IN VITRO GROWTH AND GENETIC MANIPULATION OFHEMOPOIETIC REPOPULATING CELLSbyCHRISTOPHER CHARLES FRASERB.Sc.H., The University of British Columbia, 1986A THESIS SUBMITI’ED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESGenetics ProgrammeWe accept this thesis as conformingto the required standardTHE UNiVERSITY OF BRITISH COLUMBIADecember, 1991© Christopher Charles Fraser, 1991Signature(s) removed to protect privacyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.- (Signature)_______________________Department of 4The University of British ColumbiaVancouver, CanadaDate L i,DE-6 (2/88)Signature(s) removed to protect privacyIIABSTRACTThe adult bone marrow contains stem cells capable of reconstituting all hemopoieticlineages for the life-time of a recipient following lethal irradiation and bone marrowtransplantation. Despite more than two decades of research to characterize this population ofpluripotent stem cells, mechanisms that regulate the survival, proliferation and differentiation inthe initial stages of hemopoietic development in vivo remain poorly understood. Methods formaintaining, expanding and following the fate of the most primitive hemopoietic cells in vitrothus offers major opportunities for the investigation into these mechanisms. The majorobjectives of this work were therefore initially to determine if totipotent stem cells capable ofrepopulating hemopoiesis in lethally irradiated mice could be maintained and proliferate forextended times in vitro, and subsequently to quantitate stem cell numbers in these cultures andassess their in vivo repopulating potential in lethally irradiated hosts.Initial studies were aimed at determining if stem cells capable of repopulating allhemopoietic lineages in lethally irradiated mice could be maintained and proliferate when grownin a long-term culture (LTC) system. Marrow cells from 5-fluorouracil treated male mice wereinfected with a helper-free recombinant virus carrying the neomycin resistance gene and seededonto irradiated adherent layers of pre-established, long-term marrow cultures of female origin.At 4 weeks, cells from individual cultures were transplanted into multiple recipients. Southernblot analysis of hemopoietic tissues 45 days post transplant demonstrated large clonalpopulations common to lymphoid and myeloid tissues as indicated by the presence of uniqueretroviral insertion fragments. In a number of cases it was found that multiple recipients of asingle flask were repopulated with the same clonally marked totipotent cells indicatingexpansion of such cells during culture prior to their injection into irradiated recipients. Theseresults demonstrated for the first time both maintenance and expansion in vitro of totipotentstem cells with in vivo repopulating ability.mIn a subsequent and more extensive study, a quantitative in vivo assay for competitiverepopulating units (CRU) was combined with retroviral marking of the initial cell population inorder to quantitate stem cell numbers, and to determine if long-term repopulating ability hadbeen maintained. These studies included the analysis of 46 different clones obtained inrecipients 5 weeks to 7 months after transplantation of the cultured marrow. Half of theseclones (22 of 46) included both lymphoid and myeloid progeny. Eight of the 22 lymphomyeloid clones were represented in multiple recipients, in some cases following the injection oflimiting numbers of CRU, indicating repopulation from sibling totipotent stem cells with long-term repopulating potential. Quantitation using the CRU assay demonstrated that in spite of thesignificant expansion of some totipotent stem cells revealed by retroviral marking, there was anet decrease in total CRU numbers after 4 weeks in LTC. Such results suggest that thebehavior of individual CRU in LTC may be very heterogeneous, with some undergoingextensive amplification even in the face of concurrent mechanisms leading to a net loss ofCRU, presumably due to their differentiation and/or death. Serial analysis of cells released intothe nonadherent fraction of LTC for up to 7 weeks provided additional evidence for continuedproliferation of cells with long-term repopulating potential. LTC can thus clearly support themaintenance and amplification of totipotent hemopoietic stem cells for extensive periods of timewithout diminution of their long-term in vivo repopulating potential. These results set the stagefor future studies into manipulation of development and proliferation of cells with thesecapabilities.The final series of investigations explored an approach for manipulating earlyhemopoietic cell behavior based on reconstitution of bone marrow following retroviral transferand subsequent constitutive expression of a candidate regulator. As a model, a recombinantretrovirus carrying the interleukin-7 (IL-7) gene was used to infect bone marrow cells forreconstitution of lethally irradiated hosts. IL-7 has been implicated as a regulator in early T-cellivand early B-cell development. Constitutive expression of IL-7 might therefore be anticipated tocause an amplification of stages early in hemopoietic development in mice reconstituted withIL-7 producing clones of hemopoietic cells. IL-7 reconstituted recipients exhibited a severelymphoproliferative disorder with hyperplastic lymph nodes and splenomegaly, duepredominantly to expansion in extrathymic lymphoid tissues of an unusual T cell populationresembling immature thymocytes. Further experiments will be required to establish whetherIL-7 acts directly on this population of immature T-cells to induce their expansion or whetherthis cytokine affects the commitment or expansion of cells at developmentally earlier stages ofhemopoiesis.The results presented here provide for the first time evidence of in vitro amplification oftotipotent long-term repopulating hemopoietic stem cells, establishing the LTC as a startingpoint for delineating the regulatory factors that influence the maintenance, proliferation anddevelopmental decisions of these cells. Additional evidence for a role of IL-7 in early stages ofhemopoietic development and a murine model for these effects in vivo are described.VTABLE OF CONTENTSPageABSTRACT iiLIST OF TABLES ixLIST OF FIGURES xABBREVIATIONS xiiiACKNOWLEDGEMENTS xivChapter I INTRODUCTION1) The Hemopoietic System: Critical Issues in HemopoieticStem Cell Biology 1A) An Overview of Hemopoiesis 1B) Assays for Early Hemopoietic Cells 4a) In Vitro Colony Assays 4b) Spleen Colony Assays (CFU-S) 6c) Assays for Early Lymphocytes 7d) In Vivo Assays for Hemopoietic Repopulating Cells 92) Properties of Murine Hemopoietic Stem Cells 14A) Developmental Potential and Dynamics ofHemopoietic Stem Cells 14B) Phenotyping Murine Hemopoietic Stem Cells 18C) Heterogeneity Within the Hemopoietic Stem CellPopulation: Differing Self-Renewal and ProliferativePotentials 21a) Self-renewal of CFU-S and CFU-C 21b) Variability in Stem Cell Capacity for Long-TemHemopoiesis 223) Regulation of Primitive Hemopoietic Cells 25A) The Hemopoietic Microenvironment 25B) Hemopoietic Growth Factors 28C) Growth Factors in the Early Stages of Hemopoiesis 31a) Early Myeloid Progenitors 31b) Early Lymphocyte Precursors 32c) Growth Factor Effects in Vivo 334) Growth of Primitive Hemopoietic Cells in Vitro 34A) Liquid Culture Systems 34B) Dexter LTBMC 35vi5) Hemopoietic Manipulation Using Retroviruses .37A) Retroviral Features 37B) Design of Recombinant Reiroviruses 40a) Production of Helper Free ReplicationDefective Retroviruses 40b) Retroviral Vector Design 43C) Retroviral Gene Transfer to Hemopoietic Cells 456) Thesis Objectives and General Strategy 48References 51Chapter II MATERIALS AND METHODS1) Retrovirus Construction, Production and Assays 75A) Recombinant Retroviruses 75a) tkneol9 75b) JZen-neo 75c) JZenmlL7tkneo 77B) Viral Packaging Cell Lines 78C) Generation of Viral Producer Cell Lines 78D) Viral Titering and Helper Virus Assay 802) Hemopoietic Cell Culture and Assays 81A) Mice 81B) Retroviral Infection of Bone Marrow Cells 81C) Long-Term Bone Marrow Culture 82D) Clonal Analysis of Repopulating Cells Recoveredfrom Long-Term Cultures 83E) Limiting Dilution Analysis of CompetitivelyRepopulating Cells 85F) CFU-S Assays 86G) Methylcellulose Assays 863) Molecular Analysis 87A) Southern Blot Analysis 87B) Northern Blot Analysis 884) Immunological Analyses 89A) Antibodies 89B) FACS Analysis and Cell Sorting 89C) IL-7 Activity Assay 90D) White Blood Cell Counts 91References 92vuEXPANSION IN VITRO OF RETROVIRALLY MARKEDTOTIPOTENT HEMOPOIETIC STEM CELLS1) Introduction 952) Experimental Strategy 963) ResultsA) Maintenance of Lympho-Myeloid Stem Cells 98B) Proliferation of Totipotent Stem Cells 1004) Discussion 104References 107PROLIFERATION OF TOTIPOTENT HEMOPOIETIC STEM CELLSIN VITRO WITH RETENTION OF LONG-TERM COMPETITIVE INVIVO RECONSTITUTING ABILITY.1) Introduction 1092) Experimental Strategy 1103) ResultsA)B)C)D)Phenotypic Heterogeneity in Disease Presentationof Mice Reconstituted With mIL-7 Virus InfectedBone Marrow 131B) Clonal Analysis and IL-7 Expression 134C) Perturbation in Early T-Lymphopoiesis in IL-7 Mice 136Chapter ifiChapter IVIn Vitro Recovery of Long-Term Repopulating Cells . . . 111Quantitation of Competitive Repopulating Units(CRU)inLTC 113Evidence of Totipotent Stem Cell AmplificationinVitro 117Serial Studies of Repopulating Stem Cell ClonesDuring LTC 1214) Discussion 123References 126Chapter V ALTERATIONS IN LYMPHOPOIESIS FOLLOWING HEMOPOIETICRECONSTITUTION WITH INTERLEUKIN-7 VIRUS INFECTEDBONE MARROW.1) Introduction 1292) Experimental Strategy 1303) ResultsA)vifi4) Discussion.144References 147Chapter VI SUMMARY AND FUTURE DIRECTIONS 150References 154xLIST OF TABLESPageTABLE I. Hemopoietic Growth Factors and Their Major Activities. 29TABLE II. The Frequency of CRU in 4 Week Old Long-Term Bone MarrowCultures. 116TABLE III. Tissue Distribution of Marked Clones in Recipients of LTC Cells. 118TABLE IV. White Cell Count and Phenotype Observed in Mice Reconstituted WithJZen mlL7tkneo Infected Marrow. 133TABLE V. Summary of Cell Surface Phenotypes of Tissues From MiceReconstituted With JZenmL7tkneo and JZen-neo Infected BoneMarrow. 139xLIST OF FIGURESPageFIGURE 1. Schematic representation of the organization and regulatory factorsinvolved in the hemopoietic system showing developmentalcompartmentalization and maturation compared to the potential for selfrenewal and proliferation of myeloid cells, and antigen expressionduring lymphoid cell differentiation. 3FIGURE 2. Representitive scheme for use of viral integration patterns for the studyof hemopoietic differentiation patterns in vivo. Clonal analysis ofhemopoietic tissues is performed by Southern blot analysis followingtransplantation with retrovirally infected hemopoietic stem cells. Therestriction enzyme sites within the retrovirus and cellular genomedetermine the size of fragment generated with each integration event. 10FIGURE 3. Two models for lineage restricted hemopoietic development from atotipotent stem cell: (A.) by production and maintenance of restrictedstem cells that contribute continuously to a specific lineage; or (B.)generation of lineage committed cells with high proliferative potentialthat undergo sequential differentiation and expansion. 16FIGURE 4. The retroviral life cycle beginning with expression of viral RNA andproteins which then associate and bud from the cell surface. Infectiousparticles then recognise specific receptors, the viral RNA enters the hostcell, undergoes reverse transcription to DNA which then integrates intothe host genome and the cycle begins again. 38FIGURE 5. Production of Packaging cell lines by transfection of sequences thatpermit production of viral proteins but not packagable viral RNA due tolack of P packaging sequences. Transfection or infection with arecombinant virus deleted of viral protein coding sequence allowsproduction of infectious virus that can only undergo one cycle ofinfection. 42FIGURE 6. Schematic representation of tkneol9,as well as construction of JZen-neoand JZenmlL7tkneo. 76FIGURE 7. Schematic representation of the protocols used to demonstrate maintenanceof lymphoid-myeloid stem cells (A.), and to demonstrate proliferation oflymphoid-myeloid stem cells (B.) in long-term marrow cultures. 97xiFIGURE 8. Clonal analysis by Southern blot of hemopoietic tissues from three mice45 days after reconstitution with cells from the adherent or non-adherentfractions of separate 4 week old long-term marrow cultures. 99FIGURE 9. Clonal analysis of multiple mice transplanted 45 days previously witheither the adherent layer or non-adherent fraction of single long-termcultures A, B and C. 101FIGURE 10 Clonal analysis of long-term secondary reconstituted mice. 103FIGURE 11. Presence of unique retroviral insertion fragments in bone marrow (BM),spleen (Spl), separately isolated macrophage (0) and mast cellpopulations (mast), separated splenic T and B lymphocyte populations(Spl B, Spl T), thymus (Thy), and lymph node (LN) tissues of 2 mice(A and B) detected by Southern blot analysis of HindIll digested DNA.112FIGURE 12. Outline for the quantitation of CRU in single long-term marrow cultures.Retrovirally marked male LTC cells were injected at limiting dilutions intoirradiated recipients together with female compromised cells. The proportion ofanimals positive for male repopulation was determined 5 weeks and 7 monthspost transplant, and the clonal contributions from the LTC derived stem cellsdetermined by retroviral integration events. 114FIGURE 13. Proportion of mice (4-8 animals per group) negative for reconstitution(<5%) of marrow or thymus with male cells 5 weeks aftertransplantation with varying proportions of a 4-week old long-termculture initiated with Day 4 5-FU cells, together with 2 xcompromised female marrow cells. Circles represent results fromanalyses of three separate long term cultures. A straight line fit to thecombined data based on maximum likelthood analysis is shown by thesolid line. The broken line represents previously published data forsimilar analyses of fresh Day 4 5-FU marrow cells. 115FIGURE 14. Demonstration of unique retroviral insertion fragments in varioushemopoietic tissues of multiple mice sacrificed either 5 weeks or 7months after transplantation of cells from single 4-week old long-termcultures initiated with retrovirally infected Day 4 5-FU marrow cells. 119FIGURE 15. Serial analysis of repopulating cells in the nonadherent fraction ofsingle long-term cultures assessed after 3, 5, 6 and 7 weeks by injectioninto multiple recipients. 122FIGURE 16. Schematic representation of JZen-neo and JZenmlL7tkneo proviruses.The coding sequences for murine Interleukin-7 (mIL-7) are followed bya thymidine kinase promoted NEO coding sequence (tkneo). 132xliFIGURE 17. Presence of unique retroviral insertion fragments in bone marrow,spleen, thymus, and pooled or individual lymph node tissues of mouse1,2 3 and 5. 135FIGURE 18. Northern blot analysis of IL-7 mRNA expression in tissues of recipientsof JZenmlL7tkneo infected bone marrow and a recipient of JZen-neoinfected bone marrow 137FIGURE 19. Sample FACS analysis of CD4 (vertical axis) and CD8 (horizontal axis)expression in thymus, spleen, and lymph node of a lethally irradiatedmouse repopulated with JZen-neo infected bone marrow (NEO) or micereconstituted with JZenmlL7tkneo infected bone marrow (IL-7). 138FIGURE 20. Cell surface phenotypic analysis and proliferative responses of lymphnode cells from a mouse repopulated with JZen-neo infected bonemarrow or mouse 5 reconstituted with JZenmlL7tkneo infected bonemarrow. 142FIGURE 21. FACS analysis of CD4 and CD8 expression and proliferative responsesin lymph node and ascites cells of mouse 6, reconstituted withJZenmlL7tkneo infected bone marrow, and sacrificed 16 weeks posttransplant. 143FIGURE 22. Schematic representation for in vitro clonal expansion and manipulationof an individual stem cell, followed by in vivo assessment of progenystem cells repopulating capacity. 152xliiLIST OF ABBREVIATIONSCFC Colony forming cell.CFU-E Colony forming unit-Erythroid.CFU-G Colony forming unit-Granulocyte.CFU-M Colony forming unit-Macrophage.CFU-S Colony forming unit-Spleen.CFU-GM Colony forming unit-Granulocyte-Macrophage.CFU-GEMM Colony forming unit-Granulocyte-Erythrocyte-Monocyte-Megakaryocyte.CSF Colony stimulating factor.CRU Competitive repopulating unit.FACS Fluorescence activated cell sorter.5-FU 5-fluorouracil.G- CSF Granulocyte-colony stimulating factor.G418 Geneticin.GM-CSF Granulocyte-macrophage colony stimulating factor.G6PD Glucose-6-phosphate dehydrogenase.Hb Hemoglobin.HPP-CFC High proliferative potential colony forming cell.HPRT Hypoxanthine phosphoribosyl transferase.Ig Immunoglobulin.IL Interleukin.Kb Kiobase.LW Leukemia inhibitory factor.LTC Long-term culture.LTR Long terminal repeat.LTC-IC Long-term culture initiating cell.M-CSF Macrophage colony stimulating factor.Mo-MuLV Moloney murine leukemia virus.PGK Phosphoglycerate kinase.Rh Rhodamine.RFLP Restriction fragment length polymorphism.SCID Severe combined immune deficiency.SF Steel factor.TCR T-cell receptor.TGF-13 Transforming growth factor (.TNF-a Tumor necrosis factor-a.xivACKNOWLEDGMENTSI would like to thank and express my gratitude:to my supervisor, Dr. R. Keith Humphries for the opportunity to do graduate trainingat the Terry Fox Laboratory for Hematology/Oncology and for support and guidancethroughout this project;to Dr. C.J. Eaves and Dr. F. Takei for their discussions, guidance, collaboration andenthusiastic support;to my friends and collegues Dr. S. Szilvassy, Dr. G Dougherty, Dr. R. Kay, Dr. P.Hughes and fellow graduate student J.D. Thacker,who were always under any circumstancesno matter what, even late in the evening or on Saturdays, willing to instruct, collaborate ordiscuss science, or anything else.to Dr. D. Juriloff, Dr. W. Jeffries and Dr. J. Emerman for serving on my graduatecommittee in the Department of Genetics; and Dr. J. E. Barker (The Jackson Laboratory, BarHarbor, ME) for serving on my thesis examining committee.to Patty Rosten and Dolores Fatur-Saunders, and Grace Lima for the excellent technicalassistance , and Fred Jensen for taking care of the mice in the animal facility.To Don Henkelman. for statistical analysis, and Karen Windham for help in manuscriptpreparation.to the National Cancer Institute of Canada for fmancial support.To my parents for their years of support and understanding, and above all to Yvonnewho has worked by my side from start to finish together with James, Casimar and Alexei.1CHAPTER IINTRODUCTION1) THE HEMOPOIETIC SYSTEM.A) AN OVERVIEW OF HEMOPOIESIS: CRiTICAL ISSUES IN HEMOPOIETICSTEM CELL BILOLOGY.Hemopoiesis is a continuous process of cell turnover, involving the regulateddifferentiation and amplification of blood cell precursors to allow the replacement ofapproximately 200 billion erythrocytes (1) and 60 billion neutrophilic leukocytes in the adultnormal human every day (2). This is achieved in part by the continued regulated contributionof earlier less specialized precursors and progenitors that have some proliferative ability but arenot self-maintaining. These too must be supplied from a developmentally earlier source. Thisprocess ultimately depends on the continued turnover of cells termed “hemopoietic stem cells??that are both self-maintaining with the capacity for extensive proliferation, and aremultipotential with the ability to produce daughter cells that can contribute along multiplelineages. In terms of studying hemopoietic development, it is vital to understand the molecularmechanisms regulating these initial stages of development in order to distinguish extrinsic andintrinsically determined control points. This introduction examines the current state ofknowledge of hemopoietic stem cells and assays that can be used to identify their physical andfunctional properties. The overall objective of this thesis is then to improve the current state of2knowledge and increase the understanding of events occuring at the hemopoietic stem celllevel.Hemopoietic cells at different stages of development have been operationally defmed bytheir differentiative, proliferative, and self-maintenance capacity (Figure 1). The mature bloodcells that carry out specific hemopoietic functions can be classified into two lineages, myeloidand lymphoid. The myeloid cells are produced in the bone marrow (3) and consist ofgranulocytes/monocytes/macrophages, (granulocytic), erythrocytes (erythroid) and platelets(megakaryocytic) and their precursors. The lymphoid cells are produced to various degrees inthe bone marrow, spleen, thymus and lymph node and consist of T-lymphocytes and B-lymphocytes. Mature functional end cells and their immediate precursors are not self-maintaining, and undergo extensive maturation steps within relatively few divisions (4,5).Myeloid committed cells at earlier stages of development (progenitors), and earlier lymphoidcells are not easily definable by morphological criteria. Myeloid progenitors have beenrecognized retrospectively by their capacity to form specific colony types in vitro, or in vivo.Similar clonal assays for early lymphoid cells are less well developed, but recently havebecome better characterized by recognition of specific antigen expression during development.Myeloid progenitors and early lymphoid cells can be generated from a commonprecursor, the totipotent hemopoietic stem cell. Hemopoietic stem cells can be operationallydefined by their ability to regenerate and sustain both myeloid and lymphoid blood cellproduction for extensive times in vivo. Many important questions remain unansweredconcerning the nature and regulation of the hemopoietic stem cell compartment. These includebasic understanding of their numbers, usage over time and both intrinsic and extrinsicregulatory processes. These issues are the major focus of this thesis, which has the main3POTENTIAL FORFigure 1. Schematic representation of the organization and regulatory factorsinvolved in the hemopoietic system showing developmentalcompartmentalization and maturation compared to the potential for selfrenewal and proliferation of myeloid cells, and antigen expression duringlymphoid cell differentiation.(Modified from Reference (190)).LYMPHOIDSTEM CELL?SELF- PROUFERARENEWAL flONTOTIPOTENTSTEM CELLCl-I1-3\\1-6G-CSFI HPP-CFCI BLAST-CFCd12 CFU-SiI CFU-GEMM0 [ood8 CFU.S MYELOIDSTEMCELLSI PROGENITORSaBFU.Mkpro-Bpre-BB220÷aBFU-E1-3orGM-CSF+CD8+CFU-GMCFU-M CFU-Gb‘CF U-eoblastIg+CFU-Mk CFU-E6 omega erythroblast blastPRECURSORS-CSFCSFmono myeloblast blast4objectives of developing approaches to determine regulators and mechanisms that play criticalroles in stem cell behavior.Assays that identify primitive hemopoietic cells as well as the current understanding oftheir properties and developmental regulation are briefly reviewed below.B) ASSAYS FOR EARLY HEMOPOIETIC CELLS.a) In Vitro Colony Assays.In vitro assays for cells with the ability to form colonies in semi-solid media have beenextremely useful in identifying cells at early stages in the hemopoietic hierarchy. It is nowpossible to recognize a wide range of cells termed progenitors that can give rise to colonies ofeither specific single lineages or multiple lineages of mature cells. Progenitors are theprecursors to differentiated end cells. Unlike end cells and their immediate precursors,progenitors are difficult to defme by morphological criteria, and are relatively rare (1-3% ofhemopoietic cells). Such cells were first identified in assays simultaneously developed in 1965by Pluznik and Sachs (6) and Bradley and Metcalf (7) that allowed the clonal growth ofhemopoietic cells (8). Proliferation of progenitors in vitro has been shown to be criticallydependent on specific growth factors that have in the past been supplied throughsupplementation with conditioned media derived from various sources (9). More recently, anumber of distinct gene products with hemopoietic colony stimulating activity have beenidentified and genetically cloned, so that these functional proteins can now be obtained in pureform (Reviewed in (10)).In vitro clonogenic cells can be defined within a hierarchical structure by theirproliferative and differentiative ability as well as for their capacity for generating progeny also5detectable as colony-forming cells (self-renewal). The more mature forms have limitedproliferative ability, yield mature progeny after shorter intervals in vitro and form coloniesrestricted to a single lineage (11,12). Cells capable of forming in vitro colonies have beentermed colony forming units (CFU). Some of these colonies are erythroid (CFU-E),granulocyte (CFU-G), macrophage (CFU-M), mast cell and megakaryocyte specific (13-16).Immediate precursors to these are bi-potential progenitors that form mixed colonies of two celltypes, for example granulocyte-macrophage progenitors (CFU-GM) (17). These progenitorsare also restricted in proliferative ability and form in vitro in a short time. Both uni and bipotential progenitors are killed in vivo by drugs such as 5-Fluorouracil (5-FU) which arecytotoxic to cycling cells, indicating these progenitors are in active cycle under steady-stateconditions (18,19).The criteria that separate more primitive types of clonogenic progenitors are a moreextensive proliferative capacity, more prolonged periods intervening before mature progeny areseen, and the ability to generate secondary colonies upon replating. Progenitors that fit withinthese categories are CFIJ-GEMM (or B-macro), HPP-CFC and blast colony forming cells. Alarge proportion of at least the latter two cases of earlier progenitors are unaffected by cycle-specific cytotoxic drugs such as 5-FU, suggesting many of them are in a non-cycling state invivo (20-22). CFU-GEMM is the name given to a clonogenic cell population that can formcolonies containing a mixture of granulocytes, erythrocytes, macrophages and megakaryocyteswithin 12-14 days after plating (14,23). These were termed macrosopic burst-forming cells(B-Macro) because it was found that large colonies(> i04 cells) of erythroid natureconsistently also contained mature cells of other lineages. Self-renewal potential of B-macrosin vitro has been documented, but this ability appears to be limited (24). Recently precursorsto B-Macros termed pre-CFCm1jhave been identified by their ability to expand underappropriate conditions in liquid culture and subsequently produce B-Macros in semi-solid6cultures (25,26). Likely this assay detects a cell equivalent or overlapping with blast colonyforming cells.Massive in vitro macrophage colonies (containing 10 cells) termed high proliferativecolony forming cells (HPP-CFC) can be identified when 5-FIJ treated murine bone marrow isplated at low density under appropriate conditions (20). Although these progenitors canproduce significant numbers of in vitro derived cells, their re-plating ability is low, suggestingthat these represent a more committed progenitor that has maintained a significant proliferativepotential. Despite this it has been suggested that there is a strong correlation of these cells withthose capable of repopulating bone marrow of irradiated mice (27,28). The human equivalentof HPP-CFC are found in low frequency in bone marrow, and generate large (5X10)macrophage colonies (29).Clonogenic cells giving rise to small “blast” colonies have also been described formouse and man (30-32). Characteristics of these blast colony forming cells are delayed onsetof colony formation, resistance to cell cycle specific chemotherapeutic agents and the ability togenerate a wide variety of cell types upon replating into secondary cultures including numbersof secondary blast colonies as well as B-Macros. These properties are suggestive of primitivestem cells, however demonstration of their in vivo repopulating ability has not to date beendocumented.b) Spleen Colony Assay (CFU-S’.The in vivo spleen colony assay first described by Till and McCulloch in 1961 (31) hasbeen extensively used as a measure of hemopoietic cells with stem cell properties. The assay isperformed by injecting lethally irradiated mice with syngeneic hemopoietic cell populations andscoring of macroscopic nodules visible on the spleen of the recipient 8-14 days post transplant.7Spleen colonies generated in this time frame are heterogeneous (34,35). Spleen coloniesobserved at days 12-14 are of single cell origin as demonstrated by studies using irradiationinduced chromosomal markers (36,37) and more recently using unique retroviral integrationevents as markers (38,39), and generally are of mixed myeloid cell types, and contain cellscapable of further spleen colony formation on transfer into secondary recipients. Day 12 CFUS thus manifest properties of pluripotent stem cells including an extensive proliferative ability,multi-potentiality (40,41), and self-renewal ability (42). Spleen colonies visible at days 8-9 arerelatively small, contain mainly erythroid cells, and do not contain cells capable of furtherspleen colony formation on transfer to secondary recipients (34).Although the CFU-S assay has been a valuable tool for generating concepts of stem celldifferentiation, the relationship of CFU-S to cells with totipotent long-term repopulating abilityis unclear. The ability to produce lymphoid cells within the colonies has not been demonstrated(43,44). Considerable evidence has accumulated indicating that most CFU-S in normal adultmice are separable from cells capable of sustained hemopoiesis in transplanted mice (reviewedin (45,46)) as will be discussed later in this text.c) Assays for Early Lymphocytes.Clonogenic assays for the earliest stages of lymphocyte development are not available.In vitro assays similar to those described for myeloid progenitors that allow growth of eithermurine B-cell colonies (CFU-B)(47), or T-cell colonies (CFU-T)(48) have been developed,but these are thought to represent outgrowth of activated mature cells. Other approaches havebeen taken to identify the earlier stages of lymphocyte differentiation, primarily through thedetection of developmentally specific cell surface markers using the fluorescence activated cell8sorter (FACS) to isolate and identify pure cell populations, and through determining thepotentials of these cells by various assays.Several stages of B-cell development can be discriminated by expression of cytoplasmicand surface immunoglobulin (Ig)(49,50), plus other non Ig antigens (reviewed in (51)). Pre-Bcells are at the earliest and best defmed stage of differentiation, recognized by having begun torearrange their Ig heavy chain genes (Figure 1) (52-54). When this process is complete andfunctional rearrangement has occured Ig light chain genes rearrange to eventually give rise tosurface 1gM expressing B cells. Pre-B cells and B cells can be recognized by a monoclonalantibody to the B lineage specific antigen B220 (55,56). Hemopoietic cells capable ofrepopulating lethally irradiated mice express low levels of Thy-i antigen (Thyil0) (57). Cellsexpressing both B220 and low levels of Thy-i do not form spleen colonies or rescue lethallyirradiated mice, but can repopulate lymphopoiesis when co-injected with normal bone marrow(58). It is not known however, to what extent these cells overlap with those capable ofconiributing to T-cell development.Early T-cell development begins with migration of early hemopoietic stem cells to thethymus. The most immature thymocytes do not appear to express many of the T-cell antigens(CD3,CD4, CD8, and the T-cell receptor (TCR)) at readily detectable levels except for lowlevels of Thy-i, similar to totipotent stem cells found in the bone marrow. This population ofthymocytes can be functionally defined by their ability to reconstitute the thymus of anirradiated mouse or a fetal thymus depleted of lymphoid cells (59-61), and to differentiate invitro into more mature T-cells (Figure 1) (62-64). Thymocytes begin to rearrange and thenexpress TCR and proceed through an intermediate CD3medCD4+CD8+ stage prior tobecoming3high and single CD4 or CD8 positive functional T-cells.9d) In Vivo Assays for Hemopoietic RepoDulating Cells.Evidence from chromosomal markers, isoenzyme analysis and retroviral marking havedemonstrated the presence in adult marrow of hemopoietic cells with both myeloid andlymphoid long-tem repopulating potential. The earliest experimental studies indicating thiswere those of Wu et. al.(36), who demonstrated repopulation of lymphoid and myeloidlineages in bone marrow transplanted recipient mice from contributions of a single stem cellthrough the use of irradiation induced chromosomal markers. Subsequent studies usingisoenzyme markers provided additional evidence for the existence of totipotent stem cells.Ivlintz et a! (65) demonstrated a monoclonal derivation of lymphoid and myeloid lineages fromthe inoculation of fetal wiwv mice with limiting numbers of fetal liver cells derived fromstrains that produce distinguishable hemoglobin and isoenzymes. Similar results have beenachieved in more recent studies using bone marrow cells from adult congenic mice as a sourceof allelic differences (66). Reiroviral marking studies have similarly demonstrated totipotentstem cell repopulation using irradiated (67) as well as W/WV (38) recipient mice, and usingdonor stem cells from either adult 5-fluorouracil treated bone marrow (67), or fetal liver cells(68). The scheme usually used to detect different integration patterns to demonstrate clonalhemopoietic populations in reconstituted mice is outlined in Figure 2. The demonstration oftotipotent stem cells is not limited to the murine system. Analysis of chromosomalabnormalities and X-linked G-6PD isoenzyme studies in human leukemias have pointed to anorigin from a multipotential cell (69). The existence of a normal totipotent stem cell was alsorecently demonstrated in studies of patients receiving marrow from donor femalesheterozygous for an X-linked restriction fragment length polymorphism (RFLP) whounderwent apparent monoclonal or oligoclonal hemopoietic reconstitution (70).RecombinantRetrovirus10He mop o let icStem CellIRRADIATEDHOSTSouthernblotanalysisRepresentative tissue distributionsfrom 1 or 2 stem cellsFigure 2. Representitive scheme for use of viral integration patterns for the study ofhemopoietic differentiation patterns in vivo. Clonal analysis of hemopoietictissues is performed by Southern blot analysis following transplantation withretrovirally infected hemopoietic stem cells. The restriction enzyme siteswithin the retrovirus and cellular genome determine the size of fragmentgenerated with each integration event.infection\\0transplantationHIND III HIND IIIceIluIarJ PF?DBE L_ce1Iu1argenome genomeRETROVIRAL INTEGRATION INTO STEM CELL GENOME4, à/AiAmarrow spleen thymus marrow spleen thymus marrow spleen thymus— — — — — -— — — — — -- — — — —11The studies described above have indicated totipotent differentiation capacity and long-term repopulating potential as properties of the most primitive types of hemopoietic cells.Assays for CFU-S as described above are inadequate for specifically quantitating numbers ofcells with these properties. The only valid method for detecting the most primitive type ofhemopoietic stem cell must measure its ability to reconstitute and sustain normal hemopoiesisfor extensive periods of time in lethally irradiated recipients. CFLJ-S are particularlysusceptible to radiation and have a small capacity to undergo sub-lethal damage repair (71).Death of these cells occurs in all hemopoietic tissues in a logarithmic time-dependent fashion asa result of total body irradiation (72). Since myeloablative treatments (e.g. doses of irradiationthat do not allow survival of mice) do not necessarily kill all residual hemopoietic stem cells,the presence of regenerated hemopoiesis by donor stem cells needs to be verified using areliable genetic marker. Some markers used in mouse studies are sex differences (detection ofthe Y-chromosome)(73), isoenzyme markers, or hemoglobin variants (74) and radiationinduced chromosomal markers (75) as mentioned above.A number of approaches have been taken to quantitate stem cells capable ofrepopulating hemopoiesis in irradiated mice. One such technique has been to provide bonemarrow cells at limiting dilution, and then to assay for survival of recipients 30 days posttransplant (72). There are difficulties with this assay however, because there are a number ofill-defined variables that can contribute to mortality under these conditions, and even after lethaldoses of irradiation, regeneration of host hemopoiesis may occur (76,77). Recent attemptshave been made to determine more effective dose rates and fractionated total body irradiationschemes to attempt to ensure endogenous stem cells are depleted (78).An alternative approach to lethal irradiation for hemopoietic stem cell ablation is the useof mice harboring mutations at the white-spotting (W) (79) locus, which result in severe12deficiencies in fertility and hemopoiesis (reviewed in (80)). The hemopoietic defects inaffected homozygous or compound heterozygous mice (e.g.W/W”) are severe, and stem cellsof W/W” mutant mice are unable to make macroscopic spleen colonies (CFU-S) whentransplanted into normal syngeneic mice (81). Normal (+1+) syngeneic bone marrow however,can form spleen colonies in, and permanently cure non-irradiated w,wv mice. The use ofW/W” mice for stem cell studies has the advantage of overcoming the unpredictable effects ofirradiation on survival and host stem cell recovery. This assay however has a potentialdisadvantage of a low seeding efficiency of donor stem cells to the marrow.Estimations based on the number of clones repopulating hemopoiesis in radiationchimera.s when reconstituted with graded numbers of adult bone marrow cells suggest afrequency of 1-3 per i0 (82-84). Other approaches, using rescue of lethally irradiated mice(72), curing W/W” mice (85), or in competitive repopulation assays (77) indicate a 10-foldhigher frequency. Using mixtures of genetically different bone marrows for transplantation ina competitive strategy has allowed stem cell numbers to be estimated using the binomial modelwith covariance (86). Levels of both lymphoid and erythrocyte types were found to be closelycorrelated when sampled over a period of several months suggesting most circulating cellswere derived from common precursors.More recently a quantitative competitive strategy has been designed that makes use oflimiting dilution of male donor cells and female recipients (77). The assay involvestransplantation of male test cells together with female cells that have been compromised by twoprevious cycles of transplantation. The compromised female cells ensure not only thedetection of a very primitive class of repopulating cells, but also ensure the survival of lethallyirradiated mice transplanted with very low numbers of test cells. The frequency of competitiverepopulating units (CRU) is then determined using Poisson statistics by the proportion ofrecipients positive for male cell repopulation.13The reasons for differences in estimations of stem cell numbers using various assays isunclear, but in the case of W/WV cure and competitive repopulation, survival is not dependenton donor stem cells. Discrepancies may also be accounted for by intrinsic differences in thestem cells of different strains of mice (87). Frequency of stem cell numbers has also beenassessed by extrapolation from numbers of purified repopulating cells, and fall within the rangeof these studies (84). Similar studies have also been used for fetal liver, which have estimatedstem cell content to be 1-10 per i0 cells (88,89).Efficient and consistent assays to determine the frequency of repopulating cellsrepresent an important step to address questions regarding factors that influence proliferationand differentiation of these cells. Recent attempts have been made to establish limiting dilutionassays in long-term marrow cultures as an alternative simple and routine procedure to quantitatemarrow repopulating cells (90,9 1). From these studies the numbers of cells responsible forinitiating murine long-term cultures are estimated to be in the range of estimates forrepopulating cells generated by in vivo studies. These results are of particular importance tohuman studies where equivalent in vivo models for stem cell development are not available. Ithas been suggested that the sustained output of clonogenic progenitors observed in humanlong-term cultures is the result of activation of very primitive hemopoietic cells termed longterm culture initiating cells (LTC-IC) (92). Limiting dilution analysis techniques have made itpossible to quantitate LTC-IC frequencies from various cell populations. (93). For murineand human cultures it has yet to be shown that the cells responsible for culture establishmentare equivalent to repopulating cells.142) PROPERTIES OF MURINE HEMOPOIETIC STEM CELLS.A) DEVELOPMENTAL POTENTIAL AND DYNAMICS OF HEMOPOIETIC STEMCELLS.Key questions regarding hemopoietic stem cells include: what is the developmental andproliferative potential of the most primitive stem cell; how many stem cells are required at anytime point to sustain hemopoiesis; what are the dynamics of stem cell contribution posttransplantation, is it a sustained long-lasting contribution of very few stem cells, or are stemcells a fuel for hemopoiesis, to be used sequentially from a pooi and burned-out as needed?As discussed above, the existence of totipotent hemopoietic stem cells are now welldocumented. The existence of populations of stem cells that are restricted to either myeloid orlymphoid development has been difficult to verify. Restricted developmental potential of stemcells was initially suggested by analysis of mice reconstituted with marrow cells containingchromosomal markers (75). This observation was strengthened by the ability to transfer thesame differentiation pattern to secondary transplant recipients. However, these studies arelimited by the difficulties in analyzing a large population of cells from a particular lineage wherea small contribution from a particular clone may go undetected. Restricted contributions toeither lymphoid or myeloid lineages has also been observed in retroviral marking studies,which have the technical advantage of allowing more sensitive and quantitative assessments ofdifferent cell populations (39,67). In addition to these observations, the existence of a highlyproliferative stem cell that is responsible for long-term maintenance of erythropoiesis has beenpostulated (66,94). This class of stem cells was proposed when preferential long-termmonoclonal erythropoietic reconstitution was detected by injecting limiting numbers of +1+15cells into W/W” mice and following reconstitution using cellular markers (Hbb, Gpi, Pgk). Itwas subsequently demonstrated using similar procedures that the stem cells responsible forerythropoiesis may be multipotential (95), and may also contribute to lymphopoiesis (96).Detection of lineage-restricted reconstitution by genetic or induced markers in all casesdiscussed above, may have arisen through the self-renewal and contribution from restrictedstem cells (Figure 3 A), but it is also possible that stochastic models (97,98) can explain theseoccurrences. In this case a single totipotent stem cell may self-renew, plus generate a non-selfrenewing daughter cell with extensive proliferative capacity restricted to a particular lineage.The subsequent proliferation and further committment would then provide large numbers ofdifferentiated progeny appearing along one pathway (Figure 3B). Such a model wouldpreclude the need to maintain self-renewing lineage restricted stem cells. Standard markingtechniques in transplanted recipients cannot distinguish between the two possibilities.The number of stem cell clones and their usage in steady state hemopoiesis and postBMT in the murine system remains unclear. Theoretical possibilities include maintenance ofhemopoiesis by continual proliferation of a few, or many stem cells versus the sequentialactivation of stem cell clones that proliferate, differentiate, then eventually decline. The formerpossibility was supported by Harrison et al (99), who determined by analysis of allophenicmice harboring distinct stem cell populations which produce different hemoglobin types, thathemopoiesis in these mice resulted from the simultaneous contribution from most if not all oftheir stem cells. Similar conclusions have been drawn from the analysis of radiation chimeras(82).The latter model of “clonal succession” was initially proposed by Kay in 1965 (100).A large number of transplantation studies have demonstrated results consistent with the modelof16B.restricted stem cell proliferation andself-renewal commitmenttotipotent stem cell() myeloid restricted stem cell() lymphoid restricted stem cellnon self renewing committed cellFigure 3. Two models for lineage restricted hemopoietic development from a totipotent stem cell:(A.) by production and maintenance of restricted stem cells that contribute continuouslyto a specific lineage; or (B.) generation of lineage committed cells with highproliferative potential that undergo sequential differentiation and expansion.A.17stem cell succession, with polyclonal reconstitution and clonal fluctuations over extendedperiods of time. Inconsistent with this model however, has been the observation of mice thatmaintain detectable clones for extensive periods of time in excess of 1 year suggesting longlasting hemopoiesis from single clones was possible. This was also seen in isoenzymestudies, where major shifts in the contributions of donor stem cells to hemopoiesis wereobserved including both gains and losses, plus in some cases long lasting (1 year) monoclonalreconstitution. (65). The generation of clonal markers by retroviral integration has provided aparticularly powerful method for tracking clonal contribution of hemopoietic repopulating cells(Fig. 2). Analysis of mice reconstituted with retrovirally marked marrow has revealedpolyclonality in most cases, with major fluctuations of stem cell contribution in the first fewmonths post transplant at first indicating hemopoietic development was consistent with clonalsuccession (39,101). More recent studies have indicated that after this time frame hemopoiesismay be dominated by fewer long-lasting totipotent stem cell clones (102,103). Stable long-lasting clones have also been demonstrated to contribute to the majority of hemopoiesis for acombined life-span of over 2 years when transferred into secondary recipients (104).Jordan and Lemischka (102) have derived a model for stem cell behavior based onsequential analysis of clonal contributions in mice transplanted with retrovirally marked cells.The model proposes that stem cells undergo an initial disequilibrium post transplant, reflectedby dramatic clonal fluctuations in vivo. These initial months post-transplant may be mistakenlyinterpreted as clonal succession. The authors suggest that the initial disequilibrium is not dueto intrinsically distinct classes of stem cells with different developmental and proliferativeproperties, but instead represent an expanding pool of totipotent stem cells undergoingstochastic comniittment versus self-renewal events. Eventually a stable hemopoietic systememerges in which one or few stem cell clones dominate for the life-time of the recipient.Although this model can explain some scenarios, stem cell clones do occasionally emerge at18long times post transplant (102), and in secondary transplant recipients (39), suggesting thepresence of dormant stem cells.In summary, relatively few, and perhaps single totipotent stem cells are sufficient tosustain long-lasting hemopoiesis in the mouse. Stem cell utilization has not been completelyresolved, and results appear to be dependent on the system used. Long term hemopoiesis intransplanted murine recipients appears to favor the maintained contribution of relatively fewstem cell clones after an initial disequilibrium. This however may be a function of the numberof stem cells given to a host at the time of transplantation, and a large dose of stem cells mayfavor polyclonal reconstitution. Despite the model proposed by Jordan and Lemiscka (102),the possibility of clonal fluctuation due to the decline of short-lived, less competitive stem cells,and eventual dominance by the progeny of more primitive stem cells cannot be discounted.Heterogeneity with regards to both physical and functional properties within stem cells havebeen well documented and will be discussed below. In addition, after the initial expansion of astem cell clone, it is not known how the stem cell progeny are regulated for extended periods.It may be that the progeny are utilized successively, making it necessary to introduce markersin vivo in order to address this possibility.B) PHENOTYPING MIJRTNE HEMOPOIETIC STEM CELLS.Numerous purification strategies for hemopoietic stem cells have been developed overthe last two decades based on attempts to discriminate cells by size and density, cell surfacecharacteristics, and staining by supravital dyes (reviewed in (45,46)), and assay forenrichment of CFU-S. Initial studies separated bone marrow cells by size and density oncentrifugation gradients and showed enrichment of CFU-S in fractions of smaller and lessdense cells when compared to whole marrow (105,106). These procedures are routinely used19as starting steps for stem cell enrichment. Other physical characteristics have been defined bythe fluorescence-activated cell sorter (FACS), which can not only measure fluorescence, butalso light scatter. When compared to whole bone marrow, CFU-S were found to have mediumforward (a measure of cell size) and medium to low perpendicular (a measure of cell granularityand complexity) light scatter intensities (107).Further enrichment for CFU-S and/or repopulating cells has been possible based onlectin binding (108,109), use of monoclonal antibodies to antigens on the cell membranes(110,111), or uptake of supravital dyes such as rhodamine-123 (112,113). The highestenrichments for CFU-S using these procedures range from 100- to 1000-fold compared tonormal bone marrow (109,111,114).A combination of positive selection using monoclonal antibodies against Thy-i andSca-i as well as negative selection using a cocktail of antibodies which allows the removal ofthe majority of mature end cells (Lin-) (110) resulting in a population of Thy1l0Scai+Lincells has provided the highest purity of CFU-S (1/10 cells) to date (iii). Thy1l0Scai+Lincells contain a high frequency of cells (1/10) capable of proliferating when injected into thethymus (CFU-T) suggesting lymphoid potential (115). Even at this level of purity thispopulation of cells is heterogeneous based on rhodamine dye uptake (116). Analysis ofcongenic radiation chimeras injected with limiting numbers of Thy1l0Sca1+Lin cellsestimated that 1/40 of these cells were capable of contributing to lymphoid and myeloidreconstitution for greater than 9 weeks (84). However in most of these cases contribution ofdonor cells was low (< 10 % donor derived), and conclusive evidence that these representtotipotent contributions from individual cells remains to be demonstrated.In contrast to the results demonstrating co-purification of CFU-S and repopulatingcells, some investigators have demonstrated qualitative and physical separation of CFU-S fromthose cells capable of contributing radioprotective ability to irradiated mice. Treatment of mice20with a dose of 5-fluorouracil (which selectively eliminates proliferating cells) causes athousand-fold reduction in day- 10 CFU-S without affecting the repopulating ability of the bonemarrow (117). Similarly, after repeated serial transplantations, the bone marrow mayregenerate normal CFU-S numbers, but loses the ability to contribute to long term repopulationupon transfer to subsequent lethally irradiated mice (118). Some reports have demonstratedphysical separation of most CFU-S from those cells capable of long-term reconstitution.Rhodamine-123 (Rh) fluorescence intensity has been used to separate CFU-S (Rh-bright) fromcells capable of both raclioprotective ability (Rh-dull) and long-term culture initiation, althoughsome overlap occurs (90,119). A more striking separation was reported by Jones et. al. (120)using counterfiow centrifugal elutriation, which separates cells according to size and density.Recent approaches combining retroviral marking and stem cell purification haveallowed isolation and subsequent in vivo characterization of the developmental potential ofrepopulating cells (121,122). Using this combined procedure a highly enriched population offetal liver stem cells (500-1000 fold) was isolated based on recognition with monoclonalantibody AA4. 1, cell density, and lack of fibronectin (FNA) and UN markers (AA4. 1 FNADen1°65107 Linlo), and subsequent developmental potential of these cells assessed by clonalanalysis in repopulated mice (122). A similar procedure was employed to assess cells in a 5-FU resistant Thy1l0H2K population in a competitive assay. Retroviral marking of thispopulation demonstrated that at least some proportion of these cells were capable ofrepopulating both myeloid and lymphoid lineages for extended times (121).21C) HETEROGENEITY WITFIIN THE HEMOPOIETIC STEM CELL POPULATION:DWFERING SELF-RENEWAL AN]) PROLIFERATIVE POTENTIALS.a) Self-renewal of CPU-S and CFU-C.In the process of examining spleen colonies for the presence of secondary CFU-S (anindicator of self-renewal), Siminovitch et. al. (123) noted the distribution per colony wasextremely heterogeneous, with some colonies containing many CFU-S (200-300), while otherscontained very few (< 10). Subsequent analysis revealed that the distribution of CFU-Scontent of individual spleen colonies could be described by a probabalistic model in whichCFU-S have a probability of self-renewal versus differentiation or death (97). This modelpredicts that the heterogeneity observed can be due to developmental decisions from ahomogeneous starting population. The model does not however predict to what extent thesedecisions are determined intrinsic to the cell or by extrinsic factors such as growth factors orthe microenvironment. Transplantaton studies have demonstrated a rapid progressive declinein self-renewal ability of CPU-S with each serial passage (124,125). Models to explain thisdecline argue either in favor of environmental influences, in which under the appropriateconditions CPU-S can be indefinitely maintained (124), or in favor of differences due tomitotic history, in which the quality of the CPU-S declines with each successive division(126).The heterogeneity observed in CPU-S generation in vivo hasalso been observed invitro by analysis of generation of CPU-S in macroscopic mixed colonies (24). Similarheterogeneity has also been seen for self-renewal in vitro of blast colony forming cells (127).These results cast doubt on the role of extrinsic factors in determining stem cell differentiationin vivo (24,127,128). To date difficulties in studying repopulating cells at the single cell level22have precluded analogous assessment of the heterogeneity in proliferative behavior of thesecells.b) Variability in Stem Cell Capacity for Long-Term Hemopoiesis.Nevertheless, considerable evidence indicates that stem cells differ in repopulating/selfrenewal capacity. In 1965 Hayflick proposed that normal diploid fibroblasts are capable of afinite number of cell doublings (129). This led investigators studying hemopoieticdevelopment to question the theory of self renewal of stem cells capable of repopulatingirradiated hosts. True self renewal implies that at least one daughter cell after each divisionremains exactly the same as the parent cell. If a stem cell’s potential is based on its mitotichistory, then any bone marrow population may be heterogeneous unless all stem cells hadundergone equal doublings. This concept of a continuum of maturing totipotent stem cells wasoriginally termed the generation-age hypothesis (130), and predicts a maturation step coincidentwith each cell division. Hemopoiesis would then be maintained from a population of stem cellswith differing potentials. Depending on the rigor of the assay employed, cell division couldthen give the appearance of self-renewal, when in fact the daughter cells had undergone aminor differentiation step.Marking studies have revealed that progeny of single stem cells can contribute tohemopoiesis in primary and secondary recipients for the equivalent of the lifetime of a mouse(104). How these progeny are maintained and regulated is not known. It has beendemonstrated that clonal progeny in a transplanted recipient can contribute to totipotent longlasting hemopoiesis in multiple secondary recipient mice, suggesting self-renewal from aparental cell some time post transplant (103). Early times post transplant indicate a number offluctuating short-lived clones (102). This phase of engraftment coincides with similar23observations found by serial transplantation studies (118), and in reconstitution of W/W” miceat limiting dilution where erythrocyte replacement was found to precede leukocyte replacement(94). These may represent stem cells with intrinsically less potential than the long term clones;they may alternatively reflect normal environmental or stochastic mechanisms influencing ahomogeneous start population of stem cells.An experimental approach to explore stem cell heterogeneity and self renewal has beenthrough the use of sequential bone marrow transplantation. Initial studies demonstrated thatCFU-S content (126), and the competence to induce recovery of irradiated mice (131)decreased with short (14-35 day) sequential bone marrow transplants. The number of serialtransplants is limited, with little hemopoietic recovery after the third transfer, even with largeinoculums of cells (132). An analogous decline is observed in long-term bone marrow culturesthat were originally described by Dexter (133), which have been shown to maintain CFU-S forprolonged times. These cultures demonstrate a rapid decline in CFU-S self-renewal ability andrepopulating ability within the first three weeks (134,135). These data support the model ofstem cell decline with proliferative history. Additional support for this model has come fromobservations that circulating stem cells have decreased ability to form secondary CFU-S, and itwas speculated that these represent waste products at the bottom end of the stem cell hierarchy(136). In addition, stem cells that have survived alkylating agents have a reduced capacity tomaintain sequential transplantations (132,137).Although intrinsic stem cell differences may explain the observed decline, it does notdiscount the possibility of exthnsically mediated changes to otherwise self-renewing stem cells.Stem cell repopulating ability as defined by potential for serial transplantation or ability tooutcompete other donor cells does not decline with age (138- 141), an indication that these cellsremain completely functional under normal physiological conditions. Young and old bonemarrows were equally depleted of competitive ability against fresh marrow after one cycle of24transplantation. Under different conditions for serial transplantation than those describedabove, the rapid decline in the ability to maintain hemopoiesis is not as acute. Sequentialtransplantation of normal bone marrow into WIW” mice with longer time intervals (12-14months) could maintain erythropoiesis continuously for over 6 years (142,143). These micedid exhibit a slightly decreased cure rate and number of CFU-S with each initialtransplantation, but after six successive transplantations have not lost their ability to repopulate.It is not known what the role of prolonged times between transplant plays in the recovery ofreconstituting ability, but this observation has been confirmed and disputed by others(118,144) who have used marking to ensure stem cells were donor derived. The fact that thedecline in repopulating potential is observed in serial transplants in unirradiated WIW” micediscredits the possibility that the damage results from irradiation of the carriermicroenvironment.( 145)More recent investigations have begun to address the nature of the decline observed insequential transplantation. If stem cell proliferation is the cause of the decline, it washypothesized that transplantation with limiting numbers would induce a greater proliferativestress to reconstitute a host, and this would be reflected in subsequent transplantations. Thiscombined with observations of CFU-S numbers led the authors to suggest both a decrease instem cell content and self-renewal capacity were responsible for the decline (144). Thisconclusion was confirmed by another report in which estimates of stem cell numbers anddepletion were based on competitive repopulation studies (86). It was suggested from theseresults that transplantation caused a two-fold decrease in stem cell numbers, and a sevenfoldreduction in marrow repopulating ability. In contrast, similar results have led otherinvestigators to conclude that the decline was solely due to dilution of repopulating cells withmore mature progenitors (118).25These results point to the importance of developing systems where repopulating stemcells can proliferate within a controlled environment (ie in vitro). Such a system coupled with arigorous quantitative assay for repopulating cells may resolve questions regarding intrinsicdifferences between clonal progeny of proliferating stem cells, and regarding the effects ofextrinsic factors on proliferation and developmental capacity. Such answers are crucial tofuture goals for both bone marrow transplantation and genetic engineering.3) REGULATION OF PRIMITIVE HEMOPOJETIC CELLS.A) THE HEMOPOIETIC MICROENVIRONMENT.Several lines of evidence suggest that regulation of hemopoietic stem cell proliferationoccurs in distinct localized microenvironments within the bone marrow (Reviewed in (146)).Strong evidence for a localized control of proliferation comes from shielded leg experiments(147,148). In these experiments CFU-S are induced to proliferate in response to total bodyirradiation. The CFU-S in lead-shielded femurs of irradiated mice, however remain in thesteady-state “G0phase” of the cell cycle, indicating the existence of local proliferation controlmechanisms. It has been suggested that there exists specialized compartments within the bonemarrow that regulate the pathway of stem cell proliferation and development (149), but there isno clear evidence suggesting such pathways exist, and it may be equally possible thatdevelopmental decisions may be determined by pathways that are intrinsic to a stem cell, withthe microenvironment only providing non-specific activation signals (97).The bone marrow microenvironment contains both hemopoietic cells and nonhemopoieitic cells, either of which may be involved in regulating hemopoiesis. Macrophages(150,151) and T-lymphocytes (152,153) are present within the bone marrow, and both are26capable of secreting multiple growth factors that affect hemopoietic development. A mainfocus of attention for potential cells responsible for this local control of blood productionwithin the bone marrow are reticular cells, endothelial cells, fibroblasts, and fat cells (154)operationally defined as the supportive or “stromal” environment. Morphological studies havedemonstrated a close association between these fixed tissue elements within the bone marrowand blood cells (155-157). A strong indication that these cells are at least partially responsiblefor support is the observation that marrow-derived stromal cells transplanted into ectopic sitescan initiate hemopoiesis (158,159). Stromal cells within the bone marrow are generallyconsidered to be of non-hemopoietic origin and mesenchymally derived (160,161). Thepossibility of a stem cell common to both hemopoietic and at least some stromal cells persistingin adult marrow remains controversial (162).The importance of stromal-cell stem-cell interactions are highlighted by the discovery ofcomplementary phenotypes in mice harboring mutations at the dominant steel (Sl) (163) and thewhite-spotting (W) (79) loci. Both strains express similar phenotypes, with defectivepigmentation and severe deficiencies in fertility and hemopoiesis (reviewed in (80)). Thehemopoietic defects in affected homozygous or compound heterozygous mice (e.g. Sl/Sidl orW/W”) are macrocytic anemia (164), defective megakaryocytopoiesis (165,166), reducedgranulocytopoiesis (167) and a deficiency in mast cells (168,169). The defects were found tobe complementary when it was discovered that stem cells of 51151(1 mutant mice are able tomake macroscopic spleen colonies (CFU-S) when transplanted into normal syngeneic mice(170), whereas wiwv mice were not (81). Normal syngeneic bone marrow, however, couldform CFU-S i wiwv mice but not in S1/Sl hosts. In addition irradiated Sl/Sld stroma couldnot serve to maintain in vitro hemopoiesis when supplied with syngeneic or W/W” bonemarrow. Conversely however,WIW” stroma could support normal and Sl/Sld bone marrow(171). These observations were interpreted as a defect in S1/Sl” microenvironment, and a27complementary defect intrinsic to W/W” stem cells. Recently the genes at the Si and W iocihave been cloned and their molecular defects characterized (172-175). The gene for the W lociwas first identified to be allelic with the c-kit proto-oncogene, a member of the tyrosine kinasereceptor family. It was speculated that the defect in the Si loci would be loss of the ability ofthe bone marrow microenvironment to produce the ligand for c-kit , which is expressed onhemopoietic stem cells. This assumption was shown to be correct when the Si locus wascloned and its product subsequently termed Steel Factor (SF), and found to be a potenthemopoietic growth factor.Further evidence that stromal cells are responsible for locally mediated blood cellproduction is supported by in vitro studies using a long-term culture system that maintains stemcells and progenitors for extended periods developed by Dexter (133), and through establishingbone marrow stromal cell lines. Maintenance and differentiation of early hemopoietic cells inboth murine and human long-term cultures depends on the development of a supportivepopulation of adherent cells. The adherent cell layer contains adipocytes, macrophages,fibroblasts and endothelial cells although the precise role and importance of each of these celltypes in supporting early hemopoiesis in such cultures is just beginning to be clarified(176,177).A number of stromal cell lines have been isolated from the bone marrow and long-termcultures and characterized with respect to morphology, phenotype, growth factor production,and ability to support hemopoiesis. Most of these cell lines constitutively express mRNA forseveral cytokines including M-CSF , IL-6, and TGF-13 (178-18 1). In addition some of thesecell lines can be induced to express other growth factors by stimulation with inflammatorymediators such as IL-i (182), TNF-o (183) and lipopolysaccharide (LPS) (184). Theseobservations and others have suggested a model in which the microenvironment may play arole in the regulation of hemopoiesis either by constitutive, or induced expression of growth28factors by stromal cells (185,186). This is supported by the observations that some stromalderived cell lines can maintain both myelopoiesis and B-lymphopoiesis in vitro (187), althoughthe ability to maintain totipotent stem cells has not yet been demonstrated.B) HEMOPOIETIC GROWTH FACTORS.A large number of growth factors and receptors involved in the regulation ofhemopoiesis have been identified and their molecular and functional characterisation areongoing processes (reviewed in (188-190). A list of these factors and some of their knownactivities are summarized in Table I, and some of their targets displayed in Figure 1. The firstfactors were initially identified by their ability to support in vitro colony growth, and termedcolony stimulating factors (CSF) due to their ability to stimulate granulocyte, macrophage andmixed colonies in vitro and are termed G-CSF, M-CSF, GM-CSF and multi-CSF (also termedinterleukin-3) respectively. Since that time a large number of other factors involved inregulation of hemopoiesis have been identified and their genes cloned. Twelve of thesemolecules have been termed interleukins (IL-la and 3 to IL-i 1).Classification and nomenclature of factors involved in hemopoietic regulation are nowbecoming increasingly confusing since the original concepts of lineage-restricted activity,and/or a single cellular source of a given factor is found not to be the case in most if not allinstances. For example, monocytes can produce M-CSF and GM-CSF, IL-i and TNF(reviewed in (190)) but IL-i and TNF can also induce fibroblasts and endothelial cells toproduce GM-CSF, G-CSF, M-CSF IL-6 and IL-7. T-cells are known to produce IL-2, 3,4,and 5 and GM-CSF. Different cell types are also continuously being identified that express orreact to a number of these factors alone or in combination (for example IL-7 can inducemonocytes to secrete cytokines (191). Classification becomes increasingly difficult as multiple29Table I. Hemopoietic Growth Factors (CSFs and Interleukins)Name Abbreviations Major ActivitiesErythropoietin Epo ErythroidMacrophage CSF M-CSF, CSF- 1 Monocyte/macrophageGranulocytic CSF G-CSF NeutrophilGranulocytic-macrophageCSF GM-CSF NeutrophWmacrophageSteel Factor SF, SCF, MCF,c-kit ligand C-S EC, Mast cellsInterleukin-1 alpha IL-la C-S EC, T-cellsInterlekin-1 beta IL-1{ C-S EC, T-cellsInterleukin 2 IL-2 T-cellsInterleukin-3 IL-3 Most myeloid lineagesInterleukin-4 IL-4 B-cells, T cellsInterleukin-5 IL-5 eosinophil, B cellsInterleukin-6 IL-6 C-S EC, B cellsInterleukin-7 IL-7 Pre-B cells, T cellsInterleukin-8 IL-8 Neutrophils, T-cellsInterleukin-9 IL-9 Early erythroidInterleukin- 10 IL- 10 T-cells, mast cellsInterleukin-1 1 IL-il c-S EC, B-cellsC-S EC is an abbreviation for co-stimulator of early cells.30targets for the activity of specific growth factors, and the ability of multiple factors to synergisein stimulating a cellular response are being identified. In addition some growth factors aremultifunctional, and during myeloid differentiation a response can change from beingproliferative in progenitors to inducing functional priming in end cells (192). The ability toinduce differentiation has been difficult to segregate from proliferation, but data from cell linessuggests such a role exists for growth factors (193). Although some interleukins appear to bemore or less lymphoid or myeloid specific, many appear to be co-stimulatory in combinationwith other growth factors at various stages in hemopoietic development. For example 11-5induced differentiation of bone marrow cells into eosinophils is enhanced by IL-i and IL-3,and IL-5 also acts on B lymphocytes (194,195).In addition to the complexity generated by the interaction of multiple factors , a numberof factors have recently been identified and cloned that can reversibly inhibit primitivehemopoietic cell proliferation either directly, or by antagonist actions. Transforming growthfactor-{3 can be a potent inhibitor of proliferation of some hemopoietic cells (196), is a potentinhibitor of IL-i receptor expression, and can be produced by marrow stromal cells (197).Recently an antagonist to the IL-i receptor proteins (IL-ira) has been cloned (198), as well as aprotein involved in inhibition of early progenitor proliferation, MIP-la (199).To what extent these and other factors play a role in the very earliest stages ofhemopoietic development are unknown. In the next section some of the evidence implicating anumber of growth factors in these events will be reviewed.31C) GROWTH FACTORS IN THE EARLY STAGES OF HEMOPOIESIS.a) Early Myeloid Progenitors.In vitro colony assays for those progenitors that are capable of generating secondarycolonies remains the major source of information for determining growth factor effects on earlyhemopoiesis. The most widely implicated growth factors in this stage of development areinterleukins 1, 3 and 6, G-CSF, and more recently IL-il, SF, and LW. IL-3 has beendemonstrated to have mast cell growth factor activity, and granulocyte-macrophage colonystimulating activity (200). IL-3 can also act alone in the proliferation and differentiation ofblast cell colonies and multilineage colonies (201), as well as induce colony formation fromemiched CFU-S (202). IL-i however cannot function as a a colony stimulating factor alone,but was identified as a factor that synergised with M-CSF in macrophage colony formation(203). IL-i is a potent synergistic factor in the production of earlier multiineage colonies incombination with M-CSF or IL-3 (204). This combination is also effective, and found to beoptimal for the production of HPP-CFC from 5-PU treated marrow (205), and from stem cellpopulations enriched by FACS (206,207). Both TGF-f (208) and MIP-ict (199) have beendemonstrated to inhibit murine HPP-CFC formation in the presence of the above growthfactors or conditioned media respectively.Combinations of IL-i and IL-3 act synergistically in the production of blast cellcolonies (209). IL-6, originally identified as a B cell stimulatory factor (210), and as a growthfactor for murine GM progenitors (211,212), enhances the proliferation and time of emergenceof blast cell colonies in combination with IL-3 (213). This synergistic effect with IL-3 ofshortening the dormancy period of quiescent progenitors has also been described for G-CSF(214), and for IL-il (215). One study has demonstrated that survival of blast cell progenitors32in culture does not seem to require IL-i, IL-3, IL-6, or G-CSF (216), it may be however, thatother growth factors are responsible for stem cell maintenance.b) Early Lymphocyte Precursors.A number of factors have been implicated as regulators in the early stages oflymphopoiesis. Some evidence from pre-B cell lines indicates that IL-3 influences replicationof B-lymphocyte precursors (217,218), but the physiological significance of these findings isnot clear since long-term cultures that support B- lymphopoiesis do not express detectable IL-3. Interleukin 7 was recently cloned and identified as the primary candidate responsible forproliferation of the later stages of bone marrow derived pre-B cells in Whitlock-Witte long termcultures (219). It was subsequently shown to be a potent mitogen for pro- and pre-B cells,but does not induce immunoglobulin-positive B cells to proliferate (220,221; reviewed in(222)). Equally important molecules in early lymphopoiesis are factors that inhibitdifferentiation, providing an additional regulatory mechanism (reviewed in (223)). While IL-4has been shown to have a functional role in terminal maturation of B- cells (224), it can alsoinhibit maturation of B-cell precursors produced in long-term cultures (225). TGF-f3 has alsobeen indicated as an antagonist of B lymphopoiesis (226).Combinations of growth factors have been demonstrated to function as proliferativesignals to early T-cell precursors. IL-4 in the presence of the mitogen PMA induces theproliferation of fetal thymocytes (227,228) and early adult thymocytes that don’t express CD3,CD4, or CD8 (TN, triple negative thymocytes) (229). Fetal thymus (day 15 gestation) canalso be stimulated with combinations of TNF-cz + IL-2, IL-9 + IL-2, or IL-lO + IL-2 (229).Adult TN thymocytes can be stimulated with IL-lO + IL-2 + IL-4 (230), or TNF-a + IL-2(229). These combinations however, do not maintain the T-cells in an undifferentiated state,33demonstrated in part by the loss of ability to repopulate fetal thymus organ cultures. IL-7,initially described as a pre-B cell growth factor has also been shown to stimulate proliferationof fetal thymocytes (229),and adult thymocytes that are negative for expression of CD4 andCD8 (231) in the absence of co-stimulation. TN thymocytes do not proliferate in response toIL-7, but can be maintained in an undifferentiated state in liquid culture, suggestingmaintenance but not self renewal as a function of IL-7 at this early stage of development (232).c) Growth Factor Effects in Vivo.The recent surge in cloning and expression of recombinant hemopoietic growth factorshas made available large quantities of these products for in vivo studies, which have primarilyfocussed on their potential use for clinical enhancement of hemopoietic recovery afterchemotherapy, radiotherapy and bone marrow transplantation (reviewed in (233)). Themurine model has provided the means to study growth factor effects on hemopoieticstimulation in vivo with or without previous irradiation. In general, recombinant growthfactors have a limited life span in vivo, and experiments are usually performed by repeatedinjections over a prolonged period or by delivery via a continuous perfusion pump.Comparisons between the effectiveness of different growth factors is difficult due to thevariability between published studies with regards to cytokine concentrations and half-life,routes of administration, administration schedules, and duration of treatment.A number of factors have been found to enhance hemopoietic recovery inmyelosuppressed mice with or without transplantation of marrow cells. These include G-CSF(234), GM-CSF (235,236), IL-3 (237), IL-i (238,239), IL-6 ( 240,241), or withcombinations of growth factors (242,243). Some studies have analysed the effect of in vivoadministration on early hemopoietic progenitors (CFU-S). Significant amplification of CFU-S34numbers have been observed after treatment with G-CSF (244), IL-3 (245) and IL-6 (241).The difficulty with determining the specific direct effects of in vivo administration ofrecombinant growth factors on stem cells or progenitors is the potential for secondary effectsby stimulation of other growth factors. For example a high level of TL-6 can be detected inserum of mice treated with IL-i, and may play a major role in the observed response (246).4) GROWTH OF PRIMITIVE HEMOPOIETIC CELLS IN VITRO.A) LIOUID CULTURE SYSTEMS.Enriched hemopoietic CFU-S populations have been tested for their ability to proliferatein short-term liquid suspensions, with the readouts being either visualizing cell numbers or byestablishing in vitro progenitor content, and found that IL-3 had the most significant effect(207,201). An assay that detects precursors to multilineage in vitro colonies termed preCFCmuiij has been developed. The presence of these cells was inferred by the ability of IL-iand IL-3 in suspension culture to produce a net increase in multi lineage colonies, suggestingproliferation of an earlier cell (25). It was later shown that expansion of these cells could bequantitated, and the observed amplification was dependent upon IL-i and IL-3, andcombinations of a number of other growth factors could not replace the effect (26). Similarattempts have been made to amplify CFU-S in suspension culture initially without success(247,248). More recent attempts have been successful however, demonstrating an increase inCFU-S in culture in response to combined IL-3 and IL-6 (249). This amplification has alsobeen observed using purified stem cells as a starting population, and culturing in the presenceof mast cell growth factor (MGF, also known as stem cell factor) with either IL-i or IL-3.35These investigators observed a 2 to 12 fold increase in CPU-S depending on the combinationof growth factor and stem cell source (250).While these experiments define factors capable of maintaining or initiating proliferationof progenitors and CFU-S, they do not necessarily identify factors that regulate the control ofproliferation of cells at the earliest stages of hemopoiesis. Assays capable of detecting factorsthat regulate maintenance, proliferation and commitment of totipotent repopulating stem cellshave yet to be defined. Initial attempts to address some of these questions have been made.Bone marrow cultured in IL-3 in combination with IL-6 appeared to outcompete cells culturedwith each of these factors alone when transplanted into W/W” recipients, and the totipotentnature of these cells was determined by retroviral marking (249). However, such experimentscannot distinguish between quantitative and qualitative changes and therefore do not providedefinitive evidence of proliferation.B) DEXTER LONG TERM BONE MARROW CULTURES.A culture system for maintaining CPU-S in vitro for prolonged periods was initiallydescribed by Dexter et. al. in 1977 (133). Dexter long-term cultures consist of a non-adherentpopulation of cells containing myeloid cells andprogenitors (25 1-253), but not maturelymphocytes (254), and an adherent layer that provides a microenvironment of macrophages,adipocytes and fibroblasts. Both CFU-S and CFU-C can be maintained at high levels for atleast 12 weeks. Although CPU-S numbers are maintained at significantly high levels, theability for serial transplantation, and CPU-S self-renewal ability (as measured by numbersproduced in a single 14-day serial transplantation of bone marrow ) declines rapidly within thefirst few weeks of culture (255). Both adherent and non-adherent fractions of the cultures may36contain CFU-S, but the non-adherent cells present after 1-2 weeks have a greatly reducedpotential for regenerating CFU-S in irradiated recipients.Maintenance of CFU-S production requires constant close range cell-cell interactions ofstem cells with the adherent cells (256). frradiation of the adherent layer eliminates CFU-Sproduction, but the culture can be returned to normal with a fresh inoculum of bone marrowcells (257,258). Attempts to serially transfer CFU-S from the non-adherent or adherentfractions onto irradiated adherent layers results in both a loss in CFU-S numbers and acomplete loss of CFU-S regenerating potential within one transfer (258). This decreasedrepopulating ability is analogous to the loss observed in sequential bone marrow transplantationafter the second or third transfer. A number of suggestions have been proposed to explain thisphenomenon (259).Early studies on maintenance of repopulating cells in long-term culture demonstratedusing donor marrow from mice with a hereditary chromosomal marker (CBA/HT6T6), thatsubstantial lymphoid and to a lesser extent myeloid reconstitution could be detected in mice 2-4months post transplant after receiving cells that had been in culture for up to seven weeks(260). Erythropoietic repopulating ability can be maintained in LTC, but it is substantiallyreduced when assayed by competitive repopulation (261). Reconstitution of immunodeficientmice that lack mature B-cells (CBAIN) showed a delayed B-cell recovery with long-termcultured cells compared to normal marrow (262). This observation and the observation thatAbelson virus transformable pre-B cells are not present in these cultures suggests that either avery early lymphoid restricted stem cell, or a totipotent stem cell are responsible for thereconstitution. SCID mice (severe combined immune deficient), which carry a mutation thatresults in an absence of B and T lymphocytes (263) have been used to quantitate cells fromlong-term cultures capable of reconstituting lymphopoiesis. Compared to normal bonemarrow, long-term culture derived cells were suggested to be 4-fold enriched for cells capable37of reconstituting lymphopoiesis, which was attributed to preferential maintenance of lymphoidrestricted cells (264). The overall results have led investigators to suggest that long-termcultures may maintain relatively few, if any totipotent stem cells, but the cultures do maintainCFU-S, myeloid restricted, and lymphoid restricted stem cells (261,264).Some attempts have been made to isolate and characterize the cells responsible forinitiating LTC. Initial studies using stem cells purified due to lectin-binding propertiesas thesource to initiate LTC were unsuccessful (265), probably because the purification was basedon isolation of CFU-S. Purified cells capable of initiating LTC however, can be separatedfrom the majority of CPU-S based on low Rhodamine-123 uptake which is also a property ofcells capable of long-term reconstitution of lethally irradiated mice (119).5) HEMOPOIETIC MANIPULATION USING RETROVIRUSES.A) RETROVIRAL FEATURES.Retroviruses at present provide the most efficient means to transfer foreign genes intomammalian somatic cells. Gene transfer using retroviruses is non-toxic, and integrations intothe host genome are stable and occur as single unique events. These features make retroviralgene transfer a powerful tool for both genetically marking cells for clonal analysis and forexpressing foreign genes allowing functional tests of putative regulatory molecules. Keyfeatures of the retroviral life cycle are summarized in Figure 4. Critical steps in this processare: packaging of the virion RNA into structural viral proteins and budding of the intact virionfrom the cell surface; infection of a host cell via specific receptors; reverse transcription of theviral RNA into cDNA; integration of the proviral cDNA into the host genome; and expression38integrated provirus viral mRNAviral proteinsFigure 4. The retroviral life cycle beginning with expression of viral RNA andproteins which then associate and bud from the cell surface. Infectiousparticles then recognise specific receptors, the viral RNA enters the hostcell, undergoes reverse transcription to DNA which then integrates into thehost genome and the cycle begins again.REPLICATIONbudding ofinfectious virionINFECTIONrecognition ofspecific cellularreceptorviral DNA39of the viral RNA from the integrated DNA. Events occurring at each of these stages are nowknown in considerable detail.Retroviruses in general have three stuctural genes (gag, pol and env) which encodeproducts that function as internal viral proteins, an RNA dependant DNA polymerase termedreverse transcriptase, and external envelope proteins respectively (reviewed in (266,267)).The proviral DNA is flanked by repeats termed LTR’s (long-terminal repeats) which containregulatory elements , and polyadenylation signals. The host RNA polymerase II transcribesthe retroviral RNA from the proviral DNA, and two molecules associate with the viralstructural proteins to make a complete virion. Association of the viral RNA molecule withpackaging protein is dependent on the presence of a specific viral sequence termed the ‘Ppackaging signal region (between MoMuLV bases 215 and 355) adjacent to the 5’ LTR(268,269) This sequence is necessary but additional sequences extending into the gag regionare now also recognized as essential for efficient RNA packaging (270,271). The intactretroviral particle buds from the cell surface and searches for a new host to infect whichrequires recognition of specific cell-surface receptors. Several cell surface molecules that canserve as retroviral receptors have now been identified. These include CD4 for HTV (272) andWi which has recently been cloned and identified as the receptor for murine type C ecotropicretroviruses (273). The viral particle fuses with the cell membrane either at the surface, orwithin an acidified endosome after receptor mediated endocytosis, releasing the virion core intothe cytoplasm.Once inside the host cell the viral RNA is used to generate DNA for integration into thehost genome. Reverse transcription uses both strands of RNA which are used as the templatesby reverse transcriptase to generate a double stranded DNA provirus (274,275). Retroviralintegration requires that the cells are undergoing DNA synthesis, but partially synthesized viralDNA may persist in quiescent cells for some time until activated to proliferate (276). Retroviral40integration is completed totally due to the function of a reverse transcriptase proteolyticcleavage product IN, which produces a 4 base pair staggered cut in the cellular genome, andjoins the ends of the linear viral DNA to the target DNA resulting in a flanking 4 base pairduplication (277-280; reviewed in (281)).Most retroviral integrations occur randomly within the genome, but it has beendemonstrated that approximately 2 out of every 8000 integrations will occur at the same site.These preferential sites of integration have been estimated to be present at 500-1000 pergenome. Molecular isolation of preferred sites of integration revealed no obvious sequencesimilarity between them (282). Once integrated, subsequent expression of the viral genome isdue to enhancer and promoter sequences within the LTR. The absolute level of expression isdependent upon both chromosomal position and the type of enhancer sequences within theLTR (283-285).B) DESIGN OF RECOMBINANT RETROVIRUSES.a) Production of Helper Free Replication Defective Reirovirus.High efficiency gene transfer is a prerequisite for experiments whose goal is studyinghemopoietic development by marking stem cells or by introduction of putative regulatorygenes. Infection with recombinant reiroviruses currently provides the most effective approachfor gene transfer to primary hemopoietic cells. The utility of this method has been greatlyfacilitated by the development of a system for packaging infectious replication defectiveretrovirus free of helper (replication competent) virus. Such helper-free packaging systemsminimize potential problems of multiple integrations into the genome, which may complicateexperiments in which retroviruses are used as markers for clonal analysis or increase the41chance of the activation of harmful genes such as oncogenes, or inactivation of importantregulatory genes (286-287). This has been achieved by generation of packaging ccli lines thatcan provide functional viral proteins in trans. but cannot alone produce infectious virus, due tothe absence of the critical packaging sequence in the retroviral sequence used to construct theline. Transfecting a packaging cell line with a plasmid containing the replication incompetentrecombinant virus that contains the ‘I’ sequence can subsequently produce a particle that caninfect but not replicate, and therefore is termed “replication defective” or “helper free” (Figure5).The concept of a helper free reiroviral packaging cell line was introduced in 1983 byMann et al (288) who developed the P-2 packaging cell line by deleting the sequence in acloned Moloney murine leukemia virus (M-MuLV). Transfer of the mutant packaging RNA insuch a line is not completely blocked. Infrequently both packaging RNA and recombinantvirus RNA are packaged together allowing recombination to take place during reversetranscription (275) resulting in a functional competent helper virus. Newer generations ofpackaging cell lines have therefore been developed that greatly minimize the risk of helper virusproduction. These involve removing sequences homologous to those in recombinant viruses todecrease the probability of recombination. For example in the packaging cell line PA3 17produced by Dusty Miller (289), the 3’ LTR was replaced with the polyadenylation signal fromthe simian virus SV4O DNA. This approach has greatly reduced the chance of helper virusproduction, but it still occurs at low frequency. The most recent series of cell lines (eg GP-E86(290) and ‘P-CRE(291) ) contain physically separated packaging functions by incorporatinggag and poi genes into one plasmid and the env gene into another, and then separatelyIransfecting them into a cell line. Physically separating the viral protein encoding genesdecreases the likelihood of recombination events leading to replication competent virus.packaging sequencesviral proteinsfrom packagingsequencesFigure 5. Packaging cell lines are produced by transfection of sequences that permitproduction of viral proteins but not packagable viral RNA due to lack of I’packaging sequences. Transfection or infection with a recombinant virusdeleted of viral protein coding sequence allows production of infectious virusthat can only undergo one cycle of infection.LTR poly(A)-LTR env poiy(A)packaginglinetransfection of viralpackaging sequencesnon packaging viralmRNAproteinsretroviral vectorinserted genetransfection orinfection withretroviral vectorRNAbudding ofreplication defectiveinfectious retrovirus43By altering the source of the env gene which encodes the protein responsible forrecognizing the cellular receptor, the host range can be altered. To date packaging cell lineswith a host range limited to mice (ecotropic) have used the env gene from M-MuLV, and thosewhich have an extended host range including human cells (amphotropic) are derived frommurine virus 4070A (292-294).Two main procedures can be employed to introduce the recombinant retrovirussequence into the packaging cell line. Transfection procedures are simple and generally arefollowed by selection of producer clones generating high litre virus. High titres have beenfound to be more routinely possible by infecting the packaging cell line with helper free virusgenerated from a packaging line of opposite env host range (to overcome receptorinterference), for example by infection of an ecotropic viral producer cell line with anamphotropic virus (295). Virus production from packaging cell lines transfected with a viralDNA construct in general are at least 10-fold lower than those observed from cell lines infectedwith the same virus (296,297). In addition, transfected viral DNA is less stable and has ahigher chance of rearranging and producing altered virus than if virus is introduced to thepackaging cell line by infection procedures (298). Various strategies have been effectivelyemployed to increase viral litres including “ping-ponging” in which amphotropic and ecotropicviral producer lines can be co-cultivated or undergo repeated infections to facilitate multipleintegration events.b) Retroviral Vector Design.A number of approaches have been taken to design retroviral vectors, depending on theinvestigators’ desired application (reviewed in (299)). Approximately 80% of the viral genome44(gag, poi, env) encodes for functional protein, most of which can be removed and replaced byexogenous DNA while retaining the ability to package the viral RNA and form an infectiousparticle. With each construct generated the investigator must deal with two essential concerns;viral titre and hence gene transfer efficiency; and gene expression in transduced target cells.Often it is a trade off between manipulation of the viral genome to produce sufficient levels ofregulated expression at the expense of viral titres, each of which must be determined by testingindividual constructs separately. At present there are no rules governing the outcome of thesevariables and it is difficult to predict which will function most effectively.The amount of virus expressed from a viral producer cell line (viral titre) is a crucialfactor for infection efficiency into bone marrow cells, particularly due to the low frequency ofhemopoietic stem cells in this population, and can be a major concern for both marking studiesand for the introduction of expressable foreign genes. All the factors that influence viral titresare not known, but investigators are now aware that inclusion of sequences extending into theviral gag region are important (270,271). Low viral titres have been a major drawback to someretroviral designs, particularly those vectors that have a “crippled” 3’ LTR designed so thatafter infection of the target cell the LTR promoter and enhancer are non-functional (300).The two main options for obtaining expression of the transfected gene contained withinthe recombinant retrovirus are the use of the viral LTR promoter itself, or linkage of the gene toits own or exogenous promoter. To date, LTR driven expression has proven equal or betterthan that obtained from a broad range of internal promoters including those of viral oreukaryotic gene origin (301). Levels of expression and tissue restriction of expression fromretroviral LTRs depends on the nature of the viral LTR used (302). Mutations within LTRshave allowed expression of retroviral genes in embryonal carcinoma cells, where viralinactivation normally occurs possibly due to negative regulation by cellular factors (283,285).Isolation of these mutants have provided retroviruses with a larger host range capable of45expressing in primitive cells, which are valuable tools for construction of recombinantretroviruses carrying foreign genes.In some instances it is desirable to have more than one gene expressing in a vector, forexample one gene that can be used for positive selection of infected cells by inducing drugresistance, and a second that expresses a gene with a particular cellular function. This may beobtained by using the LTR promoter and producing a polycistronic message, or by placing onegene in the naturally occurring splice junctions within the virus (303), or by allowing one geneto be expressed from the LTR, and the second gene placed under control of an internalpromoter (304). In general the introduction of more than one exogenous gene into theretroviral genome causes a decrease in expression of those genes compared to expression ifonly a single gene were present (305-307).Despite these difficulties, long term expression of foreign genes such as humanadenonosine deaminase (308), x and f3-globin (309,310), as well as some hemopoietic growthfactors discussed below have been documented in the hemopoietic systems of micetransplanted with retrovirally infected bone marrow.C) RETROVIRAL GENE TRANSFER TO HEMOPOIETIC CELLS.Retrovirally mediated gene transfer has proven a powerful tool for the study ofhemopoiesis. The main applications have been for marking and fate mapping of hemopoieticstem cells as described above, and for the introduction of expressable genes either for genereplacement as a model for human genetic diseases or as a means to test regulatory moleculessuch as putative oncogenes and growth factors.Introduction of foreign genes into primitive hemopoietic cells using retrovirusesrequires optimal gene transfer efficiencies that depend on both retroviral titre, and the state of46the target cells. Low efficiencies are a particular concern for clonal analysis, in studies whereonly a limited number of target cells are available, or in investigations of gene function whereperturbation may be masked by uniiifected cells. Attempts to maximize gene transfer toprimitive murine hemopoietic cells has empirically identified several important factors. Theseinclude high viral titre, presumably based on increasing viral to target ratios, and pre-treatmentof donors with 5-FU and the use of suitable growth factors prior to and during retroviralinfection.A common source of cells for retroviral infection is 5-FIJ treated bone marrow, whichhas been shown to be much more accessible to retroviral integration than normal bone marrow(311). This effect is presumably due to the fact that 5-FU treated stem cells are cycling (312),and retroviruses infect or integrate more efficiently into cycling (as opposed to non-cycling)cells. Comparison of marrow for retroviral gene transfer efficiency 2-8 days after 5-FUtreatment indicated that day 4 was the best source of infectable stem cells for long-termreconstitution in irradiated hosts (313). Not only the source of stem cells, but also theconditions under which they are infected appear to be important. Inappropriate infectionconditions may lead to loss of reconstitutive capacity of stem cells (314). Initially conditionedmedias were used as a source of growth factors (39), but more recently the use of recombinantgrowth factors including murine leukemia inhibitory factor (LIF) (315), and a combination ofIL-3 and IL-6 (249) have been shown to increase gene transfer efficiencies to CFU-S and long-term repopulating cells respectively. Optimal growth factor conditions for the maintenance ofstem cells in vitro during the infection period, and for enhanced gene transfer efficiencies tothese cells have not yet been well defined. Alternative approaches to improving efficiencies areavailable. The infection procedure may vary over a period of 1 to 4 days in which bonemarrow cells are either cultured in the presence of viral supernatants, or co-cultivated with viralproducer cells. The latter technique results in improved gene transfer efficiencies to long-term47repopulating cells, but appears to result in an overall loss of repopulating ability, perhapsthrough loss due to adherence of bone marrow to the viral producer cell lines (311). Theproportion of cells expressing from the integrated provirus can be enhanced by including adrug resistance gene within the virus allowing subsequent 24-48 hour positive selection ofinfected stem cells; this technique however results in a significant reduction in the number ofviable cells (>90%) (307). When the infection protocol is complete, bone marrow cells arethen collected and introduced into lethally irradiated hosts for long-term donor reconstitution.These protocols will reproducibly give 80 to 100 % gene transfer to CFU-S.Retroviral gene transfer has now been used to introduce a large number of genes intohemopoietic repopulating cells. Of major interest in context of hemopoietic regulation arestudies of the effect of over or ectopic expression of several hemopoietic growth factor genes.Reconstitution of mice with marrow infected with a reirovirus carrying the GM-CSF generesulted in a lethal myeloproliferative syndrome primarily due to expansion of neutrophils andmacrophages which was not paralleled by an expansion in progenitors (316). Similarobservations were made in mice overexpressing the IL-3 gene (317,318), except for a moredramatic increase in hemopoietic cellularity, and an increase in progenitor content in the spleen.Cells from primary recipients in both cases were unable to develop tumors in non-irradiatedsecondary recipients, suggesting that secondary genetic events were required to develop a fullleukemic phenotype. The most recent growth factor to be tested in this way is IL-6 which hasbeen implicated in the development of multiple myeloma (319,320), myeloid leukemia (321),and lymphoma (322). Dysregulated IL-6 expression in retrovirally transduced bone marrowrecipients resulted in a non neoplastic disorder with the most pronounced hematologicphenotype being extensive expansion of B-cell derived plasma cells, and a transient increase ingranulocyte numbers (323). The progenitor content in these mice was not reported.48These results have in common the amplification of end cell stages of hemopoiesis astheir major observed phenotype. With the large numbers of growth factor genes that haverecently been cloned, it will be interesting to determine using retroviral gene transfer whichones will induce preferential proliferation and expansion of cells in the early stages ofhemopoiesis. Recently the CPU-S content has been measured in retrovirally infected miceexpressing constitutively high levels of G-CSF, GM-CSF and IL-3 (324). An increase in thetotal CPU-S content in the spleen was observed in all cases. Individual spleen colonieshowever, did not contain an elevated number of secondary CPU-S suggesting the increase wasdue to recruitment from more primitive stem cells rather than CPU-S self-renewal. Thesestudies did not look at the effects on cells earlier in hemopoiesis than CPU-S. However, thisapproach coupled with improved analytical techniques, should prove extremely powerful forresolving factors involved in the regulation of totipotent stem cells.6) THESIS OBJECTIVES AND GENERAL STRATEGY.As reviewed above, a large body of data now supports the existence of a class of veryprimitive hemopoietic cells with the capacity to differentiate down both myeloid and lymphoidpathways and to contribute to blood cell production for a lifetime. The mechanisms thatregulate the proliferation, maintenance and clifferentiative decisions of these key cells in thehemopoietic system however, remain poorly understood. Some of the key questions that haveyet to be answered include: what factors initiate proliferation of these cells from a quiescentstate; is there a mandatory loss of potential with each cellular division as suggested by thegeneration-age hypothesis; what factors influence developmental decisions?The research in this thesis was initiated in an effort to develop new approaches forstudying stem cell behavior and to begin to use these to obtain insight into these fundamental49questions. The experimental design incorporated three key elements. First an in vitro systemwas chosen that had the potential to support the maintenance and expansion of hemopoieticstem cells with long-term repopulating ability. Second a retroviral marking technique wasemployed to enable Iracking the in vivo proliferative and developmental potential of individualstem cells. Third, the recently developed CRU assay based on limiting dilution of stem cellswas used to allow quantitation and specific assessment of stem cells with long-termrepopulating ability.As a starting point for these studies, the Dexter long-term culture was chosen as acandidate in vitro model (133) to support the earliest hemopoietic cells. Previous studies hadshown that long-term marrow cultures could support cells capable of reconstituting erythroid(260) and lymphoid lineages (263) in lethally irradiated recipients. The loss of CFU-S selfrenewal ability (254) and loss of competitive repopulating ability however, led investigators topredict that totipotent stem cells were not proliferating in these cultures, and probably were notbeing well maintained.My first objective was to determine if totipotent stem cells could be detected inestablished (4 week old) Dexter long-term cultures and second to determine if evidence of theirself-renewal in the cultures could be obtained. In order to do this, hemopoietic stem cells from5-FU treated bone marrow were genetically marked by infection with a recombinant retrovirus,then placed in long-term cultures. The ability of long-term cultures to maintain stem cells wasthen tested by repopulation of lethally irradiated mice with culture derived cells. Recipientmice were analysed for clonal reconsitution from culture derived retrovirally marked stem cells.It was projected that if amplification were occurring under similar conditions in vitro, thenrepopulation of multiple mice from the progeny of a single stem cell could be detected.The next step was to delineate the extent of stem cell maintenance and amplification andto determine whether a capacity for long-lasting hemopoiesis had been sustained. By50combining retroviral marking of stem cells with long-term culture and the CRU assay, it washoped these objectives could be reached. The results of this approach are outlined in ChapterIv.The ability to quantitate and qualitate stem cell development in vitro provides a means toassess growth factor effects on proliferation and differentiation decisions of those cells. As afirst step towards such an investigation, the work presented in Chapter V was initiated toassess the possible role of one such candidate regulator, Interleukin-7 (IL-7) on earlyhemopoietic cell behavior. These studies began by analyzing the effects of reconstituting micewith marrow infected with an IL-7 recombinant retrovirus, which yielded a number ofunexpected findings. Subsequent in vivo and in vitro studies would clearly be useful todetermine the possible role of IL-7 as a stem cell regulatory factor.The ability to determine the developmental potential of individual stem cells derivedfrom an in vitro clonal expansion provides an important starting point for delineation of themolecular mechanisms that control the processes of self-renewal and/or developmentalrestriction of totipotent stem cells.51REFERENCES1. Erslev AJ. Production of erythrocytes. In: Williams WJ, Beutler E, Erslev AJ,Lichtman MA. Hematology. New-York: McGraw-Hill (1983).2. Dancey if, Deubelbeiss KA, Harker LA, Finch CA. Neutrophil kinetics in man. I ClinInvest. 58:705 (1976).3. Lichtman MA. The ultrastructure of the hemopoietic environment of the marrow: Areview. Exp Hematol 9:391 (1981)4. Lajtha LG, Oliver R. Studies on the kinetics of erythropoiesis: A model of erythron. InWolstenholme GEW and O’Conner M. Haemopoiesis Ciba Found Symp.London:Churchill (1960).5. Warner HR, Athens JW. An analysis of granulocyte kinetics in blood and bonemarrow. Ann NY Acad Sci 113:523 (1964).6. Pluznik DH, Sachs L. The cloning of normal ‘mast’ cells in tissue culture. I Cell CompPhysiol 66:3 19 (1966).7. Bradley TR, Metcalf D. The growth of mouse bone marrow cells in vitro. Aust J ExpBiol Med Sci 44:287 (1966).8. Till JE, McCulloch EA. Hemopoietic stem cell differentiation. Biochem Biophys Acta605:43 1 (1980).9. Metcalf D. Hemopoietic colonies. In vitro cloning of normal and leukemic cells. BerlinHeidelberg: Springer-Verlag (1977).10. Metcalf D. The hemopoietic colony stimulating factors. Amsterdam: Elsevier (1984).11. Gregory Ci, Henkelman RM. Relationships between early hemopoietic progenitor cellsdetermined by correlation analysis of their numbers in individual spleen colonies. In:Baum SJ, Ledney GD. Experimental Hematology Today. New York: Springer-Verlag(1977).12. Eaves CJ, Humphries RK, Eaves AC. In vitro characterization of erythroid precursorcells and the erythropoietic differentiation process. In: Stamatoyannopoulos G,Neinhuis AW. Cellular and Molecular Regulation of Hemoglobin switching. NewYork: Grune and Stratton (1979).13. Gregory Ci, Eaves AC. Human marrow cells capable of erythropoietic differentiationin vitro: Definition of three erythroid colony responces. Blood 49:855 (1977).14. Fauser AA, Messner HA. Identification of megakaryocytes, macrophages andeosinophils in colonies of human bone marrow containing neutrophilic granulocytesand erythroblasts. Blood 53:1023 (1979).5215. Long MW, Gragawski LL, Hefner CH, Boxer LA. Phorbol diesters stimulate thedevelopment of an early murine progenitor cell. The burst-forming unit megakaryocyte.J Clin Invest 76:43 1 (1985).16. Metcalf D, MacDonald HR, Odartchenko N, Sordat B. Growth of mousemegakaryocyte colonies in vitro. Proc Nati Acad Sci USA 72:1744 (1975).17. Metcalf D, Merchav 5, Wagemaker G. Commitment by GM-CSF or M-CSF of bipotential GM progenitor cells to granulocyte or macrophage formation. Jn:Baum,Ledney, Theirfelder. Experimental Hematology Today. Basel: Karger (1982).18. Van Zant G. Studies of hematopoietic stem cells spared by 5-fluorouradil. J Exp Med159:679 (1984).19. Gregory CJ, Eaves AC. Three stages of erythropoietic progenitor cell differentiationdistinguished by a number of physical and biological properties. Blood 5 1:527 (1978).20. Bradley TR, Hodgson GS. Detection of primitive macrophage progenitor cells inmouse bone marrow. Blood 54: 1446 (1979).21. Boswell HS, Wade Jr PM, Quesenberry PJ. thy-i antigen expression by murine high-proliferative capacity hematopoietic progenitor cells. I. relation between sensitivity todepletion by Thy-i antibody and stem cell generation potential. J Immunol 133:2940(1984).22. Suda T, Suda J, Ogawa M. Proliferative kinetics and differentiation of murine blast cellcolonies in culture: evidence for variable G0 periods and constant doubling rates ofearly pluripotent hemopoietic progenitors. J Cell Physiol 117:308 (1983).23. Johnson GR, Metcalf D. Pure and mixed erythroid colony formation in vitro stimulatedby spleen conditioned medium with no detectable erythropoietin. Proc Nail Acad SciUSA 74:3879 (1977).24. Humphries RK, Eaves AC, Eaves CJ. Self-renewal of hemopoietic stem cells duringmixed colony formation in vitro. Proc Natl Acad Sci USA 78:3629 (1981).25. Iscove NN, Shaw AR, Keller G. Net increase of pluripotential hematopoieticprecursors in suspension culture in response to IL-i and IL-3.J Immunol 142:2332(1989).26. Iscove NN, Yan X-Q. Precursors (pre-CFCmuljj) of multilineage hemopoietic colony-forming cells quantitated in vitro. J Immunol 145:190 (1990).27. Hodgson GS, Bradley TR, Radley JM. The organization of hemopoietic tissue asinferred from the effects of 5-fluorouradil. Exp Hematol 10:26 (1982).28. Hodgson GS, Bradley TR, Radley JM. In vitro production of CFU-S and cells witherythropoiesis repopulating ability by 5-fluorouracil treated bone marrow. mt J CellCloning 1:49 (1983).5329. McNiece 1K, Stewart FM, Deacon DM, Temeles DS, Zsebo KM, Clark Sc,Quesenberry PJ. Detection of human CFC with a high proliferative potential. Blood74:609 (1989).30. Nakahata T, Ogawa M. Identification in culture of a class of hemopoietic colony-forming units with extensive capability to self-renew and generate mukipotentialcolonies. Proc Nail Acad Sci USA 79:3843 (1982).31. Leary AG Ogawa M. Blast cell colony assay for umbilical cord blood and adult bonemarrow progenitors. Blood 69:953 (1987).32. Rowley SD, Sharkis SJ, Hattenburg C, Sensenbrenner LL. Culture from human bonemarrow of blast progenitor cells with an extensive proliferative capacity. Blood 69:804(1987).33. Till JE, McCulloch EA. A direct measurement of the radiation sensitivity of normalmouse bone marrow cells. Radiat Res 14:213 (1961).34. Magli MC, Iscove NN, Odartchenko N. Transient nature of early haematopoietic spleencolonies. Nature 295:527 (1982).35. Wolf NS, Preistley GV. Kinetics of early and late spleen colony development. ExpHematol 14:676 (1986).36. Wu AM, SiminovitchL, Till JE, McCulloch EA. Evidence for a relationship betweenmouse hemopoietic stem cells and cells forming colonies in culture. Proc Nail Acad SciUSA. 59:1209 (1968a).37. Becker AJ, McCulloch EA, Till JE. Cytological demonstration of the clonal nature ofspleen colonies derived from transplanted mouse bone marrow cells. Nature 197:452(1963).38. Dick JE, Magli MC, Huszar D, Phillips RA Bernstein A, Introduction of a selectablegene into primitive stem cells capable of long-term reconstitution of the hemopoieticsystem of W/W” mice. Cell 42:7 1 (1985).39. Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamicbehavior of hematopoietic stem cells. Cell 45:917 (1986).40. Curry IL, Trentin JJ. Hemopoietic spleen colony studies: I. Growth anddifferentiation. Dev Biol 15:395 (1967).41. Fowler JH, Wu AM, Till JE, McCulloch EA, Siminovitch L. The cellular compositionof hemopoietic spleen colonies. J Cell Physiol 69:327 (1967).42. Wu AM, Siminovitch L, Till JE, McCulloch EA. Cytological evidence for arelationship between normal hemopoietic colony-forming cells and cells of thelymphoid system. J Exp Med 127:455 (1968).5443. Lala PK, Johnson GR. Monoclonal origin of B-lymphocyte colony forming cells inspleen colonies formed by multipotential hemopoietic stem cells. J Exp Med 148:1486(1978).44. Paige CJ, Kincade PW, Moore MAS, Lee G. The fate of fetal and adult B-cellprogenitors grafted into immunodeficient CBAIN mice. 3 Exp Med 150:548 (1979).45. Spangrude GJ. Enrichment of murine haemopoietic stem cells: diverging roads.Immunol Today 10:344 (1989).46. Visser JWM, Van Bekkum DW. Purification of pluripotent hemopoeitic stem cells: pastand present. Exp Hematol 18:248 (1990).47. Metcalf D, Nossal SJV, Warner NL, Miller JFP, Mandel TE, Layton JE, Gutman GA.Growth of B-lymphocyte colonies in vitro. 3 Exp Med 142:1534 (1975).48. Rozenszajn LA, Shoham D, Kalechman I. Clonal proliferation of PHA-stimulatedhuman lymphocytes in soft agar culture. Immunology 29:1041 (1975).49. Landreth KS, Rosse C, Clagett 3. Myelogenous production and maturation of Blymphocytes in the mouse. J Immunol 127:2207 (1981).50. Abney ER, Cooper MD, Kearney JF, Lawton AR, Parkhouse RME. Sequentialexpression of immunoglobulin on developing B lymphocytes: a systematic survey thatsuggests a model for the generation of im noglobulin isotype diversity. J Immunol120:2041 (1978).51. Kincade PW, Lee G, Watanabe T, Sum I, Scheid MP. Antigens displayed on murine Blymphocyte precursors. J Immunol 127:2726 (1981).52. Alt FW, Yancopoulus GD, Blackwell TK, Wood C, Thomas E,Boss M, Coffman R,Rosenberg N, Tonegawa 5, Baltimore D. Ordered rearrangements of immunoglobulinheavy chain variable region segments.EMBO J 3:1209 (1984).53. Coffman RL, Surface antigen expression and immunoglobulin gene rearrangementduring mouse pre-B cell development. Immunol Rev 69:5 (1983)54. Coffman RL, Weissman IL. Immunoglobulin gene rearrangement during pre-B celldifferentiation. J Mol Cell Immunol 1:31 (1983).55. Coffman RL, Weissman IL. B220, a B-cell specific member of the T200 glycoproteinfamily. Nature 289:68 1 (1981).56. Scheid MP, Landreth KS, Tung LS, Kincade PW. Preferential but nonexclusiveexpression of macromolecular antigens on B lineage cells. Immunol Rev69: 141.(1983).57. Muller-Sieburg CE, Tidmarsh GF, Weissman IL, Spangrude GJ. Maturation ofhematolymphoid cells that express Thy-i. In: Reif A, Schlesinger M. Thy-i:Immunology, Neurology and Therapeutic Applications. Marcel Dekker, New York(1989).5558. Tidmarsh GF, Heimfeld S. Whitlock CA, Weissman IL, Muller-Seiburg CE.Identification of a novel bone marrow derived B-cell progenitor population thatcoexpresses B220 and Thy-i and is highly enriched for Abelson leukemia virus targets.Mol Cell Biol 9:2665 (1989).59. Fowlkes BJ, Edison L, BJ Matheison, Cused TM. Early T lymphocytes: differentiationin vivo of adult intrathymic precursor cells. 3 Exp Med 162:802 (1985).60. Crispe IN, Moore MW, Husmann LA, Smith L, Bevan MJ, Shimonkevitz RP.Differentiation potentials of subsets of CD4-CD8- thymocytes. Nature 329:336 (1987).61. Pearse M, Wu L, Egerton M, Wilson A, Shortman K, Scollay R. A early murinethymocyte developmental sequence is marked by transient expression of the interleukin2 receptor. Proc Natl Acad Sci USA 86:1614 (1989).62. Suda T, Ziotnik A. LL-7 maintains the T-cell precursor potential of CD3-CD4-CD8-Thymocytes. J Immunol 146:3068 (1991).63. Takei F, Secher DS, Milstein C, Springer T. Use of a monoclonal antibody specificallynon-reactive with T cells to delineate lymphocyte subpopulations. Immunology 42:37 1(1981).64. Crispe IN, Bevan MJ. Expression and functional significance of the 31 id marker onmouse thymocytes. J Immunl. 138:2013. (1987).65. Mintz B, Anthony K, Litwin S. Monoclonal derivation of mouse myeloid andlymphoid lineages from totipotent hematopoietic stem cells experimentally engrafted infetal hosts. Proc Natl Acad Sci USA 8 1:7835 (1984).66. Nakano T, Waki N, Asai Hidekazu, Kitamura Y. Long-term monoclonal reconstitutionof erythropoiesis in genetically anemic W/WV mice by injection of 5-fluorouracil-treatedbone marrow cells of Pgk1afPgk1b mice. Blood 70: 1758 (1987).67. Keller G., Paige C, Gilboa E, Wagner EF. Expression of a foreign gene in myeloidand lymphoid cells derived from multipotent hemopoietic precursors. Nature 318:149(1985).68. Capel B, Hawley R, Covarrubias L, Hawley T, Mintz B. Clonal contributions of smallnumbers of hematopoietic stem cells engrafted in unirradiated neonatal W/W’ mice.Proc Nati Acad SciUSA 86:4564 (1989).69. Fialkow P3, Gartler SM, Yoshida A. Clonal origin of chronic myelocytic leukemia inman. Proc Natl Acad Sci USA 58:1468 (1967).70. Turhan AG, Humphries RK, Phillips GL, Eaves AS, Eaves CJ. Clonal hematopoiesisdemostrated by X-linked DNA polymorphisms after allogeneic bone marrowtransplantation. N Engl J Med 320:1655 (1989).5671. Glasgow GP, Beetham KL, Mill WB. Dose rate effect on the survival of normalhaematopoietic stem cells of BALBIc mice. mt j Radiat Oncol Biol Phys 9:557 (1983).72. McCulloch EA Till JE. The radiation sensitivity of normal mouse bone marrow cells,determined by quantitative marrow transplantation into irradiated mice. Radiat Res13:115 (1960).73. Hampson IN, Spooncer E, Dexter MT. Evaluation of a mouse Y chromosome probefor assassing marrow transplantation. Exp Hematol 17:313 (1989).74. Russell ES, McFarland EC. Hemoglobin, comparative molecular biology models forthe study of disease. Ann NY Acad Sci 241:25 (1974).75. Abramson S, Miller RG, Phillips RA. The identification in adult bone marrow ofpluripotent and restricted stem cells of the myeloid and lymphoid systems. 3 Exp Med145:1567 (1977).76. Brecher G, Neben 5, Yee M, Bullis J, Cronkite EP. Pluripotential stem cells withnormal or reduced self renewal survive lethal irradiation. Exp Hematol 16:627 (1988).77. Szilvassy SJ, Humphries RK, Lansdorp PM, Eaves AC, Eaves CJ. Quantitative assayfor totipotent reconstituting hematopoietic stem cells by a competitive repopulationstrategy. Proc Natl Acad Sci USA 87:8736 (1990).7. Down ID, Tarbell NJ Thames HI),. Mauch PM. Syngeneic and allogeneic bonemarrow engraftment after total body irradiation: Dependence on dose, dose rate andfractionation. Blood 77:661 (1991).79. Russell ES. Analysis of peiotropism at the W locus in the mouse. Relationship betweenthe effects of W and W” substitution on hair,pigmentation,and on erythrocytes.Genetics 34:708 (1949).80. Russell ES. Hereditary anemias of the mouse: a review for geneticists. Adv Genet20:357 (1979).81. McCulloch EA, Siminovitch L, Till JE. Spleen colony formation in anemic mice ofgenotype WIW”. Science 144:844 (1964).82. Micklem HS, Lennon JE, Ansell ID, Gray RA. Numbers and dispersion ofrepopulating hematopoietic cell clones in radiation chimeras as functions of injecteddose. Exp Hematol 15:25 1 (1987).83. Harrison DE, Astle CM, Lemer C. Number and continuous proliferative pattern oftransplanted primitive immunohematopoietic stem cells. Proc Natl Acad Sci 85:822(1988).84. Smith LG, Weissman IL, Heimfeld S. Clonal analysis of hematopoietic stem celldifferentiation in vivo. Proc Nail Acad Sci USA 88:2788 (1991).5785. Boggs DR, Boggs SS, Saxe DS, Gress RA, Confield DR. Hematopoietic stem cellswith high proliferative potential. Assay of their concentration in marrow by thefrequency and duration of cure of W/Wv mice. J Clin Invest 70:242 (1982).86. Harrison DE, Stone M, Astle CM. Effects of transplantation on the primitiveimmunohematopoietic stem cell. J Exp Med 172:43 1 (1990).87. Van Zant G, Chen JJ, Scott-Micus K. Developmental potential of hematopoietic stemcells determined using retrovirally marked allophenic marrow. Blood 77 :756 (1991).88. Metcalf D, Moore MAS.Haemopoietic cells. Amsterdam: Elsevier/North Holland(1971).89. Fleischman RA, Mintz B. Prevention of genetic anemias in mice by microinjection ofnormal hematopoietic stem cells into the fetal placenta. Proc Nati Acad Sci USA76:5736 (1979).90. Ploemacher RE, van der Sluijs IP, Voerman JSA, Brons NHC. An in vitro limiting-dilution assay of long-term repopulating hematopoietic stem cells in the mouse. Blood74:2755 (1989).91. Ploemacher RE, van der Sluijs JP, van Beurden CJM, Baert MRM, Chan PL. Use oflimiting-dilution type long-term marrow cultures in frequency analysis of marrowrepopulating and spleen colony-forming hemopoietic stem cells in the mouse. (inpress).92. Eaves CJ, Sutherland HJ, Cashman ID, Otsuka T, Lansdorp PM, Humphries RK,Eaves AC, Hogge DE. Regulation of primitive human hematopoietic cells in long-termmarrow cultures. Sem in Hematol 28:126 (1991).93. Sutherland HJ, Lansdorp PM, Henkelman DH. Functional characterization ofindividual human hematopoietic stem cells cultured at limiting dilution on supportivemarrow stromal layers. Proc Natl Acad Sci 873584 1990.94. Barker JE, Braun J, McFarland-Starr. Erythrocyte replacement precedes leukocytereplacement during repopulation of W/W” mice with limiting dilutions of +1+ donormarrow cells. Proc Natl Acad Sci USA 85:7332 (1988).95. Nakano T, Waki N, Asai Hidekazu, Kitamura Y. Different repopulation profilebetween erythroid and nonerythroid progenitor cells in genetically anemic W/W” miceafter bone marrow transplantation. Blood 74:1552 (1989).96. Nakano T, Waki N, Asai Hidekazu, Kitamura Y. Lymphoid differentiation of thehematopoietic stem cell that reconstitutes total erythropoiesis of a genetically anemicwiwv mouse. Blood 73: 1175 (1989).97. Till JE, McCulloch EA, Siminovitch L. A stochastic model of stem cell proliferationbased on the growth of spleen colony forming cells. Proc Nail Acad Sci USA 51:26(1964).5898. Vogel H, Niewisch H, Matioli G. The self renewal probability of hemopoietic stemcells. J Cell Physiol 72:221 (1968).99. Harrison DE, Lemer C, Hoppe PC, Carlson GA, Ailing D. Large numbers of primitivestem cells are active simultaneously in aggregated embryo chimeric mice. Blood 69:773(1987).100. Kay, HEM. How many cell generations? Lancet 2:4 18 (1965).101. Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of micereconstituted with retrovirus-infected stem cells. EMBO J 6:3955 (1987).102. Jordan CT, Lemischka JR. Clonal and systemic analysis of long-term hematopoiesis inthe mouse. Genes & Dev 4:220 (1990).103. Keiler G. Snodgrass R. Life span of multipotential hematopoietic stem ceils in vivo. JExp Med 171:1407 (1990).104. Cape! B, Hawley RG, Mintz B. Long- and short- lived murine hematopoietic stem cellclones individually identified with retroviral integration markers. Blood 75 :2267(1990).105. Worton RG, McCuiloch EA, Tiil JE. Physical separation of hemopoietic stem cellsdiffering in their capacity for self-renewal. J Exp Med 130:91 (1969).106. Miiler RG, Philips RA. Separation of cells by velocity sedimentation. J Cell Physiol73:191 (1969).107. van den Engh GJ, Visser JWM, Bol SJL, Trask B. Concentration of hemopoietic stemcells using a light-activated cell sorter. Blood Cells 6:609 (1980).108. Nicola NA, Burgess AW, Staber FG, Johnson GR, Metcalf D, Battye FL. Differentialexpression of lectin receptors during hemopoietic differentiation: enrichment forgranulocyte-macrophage progenitor cells. J Cell Physiol 103:217 (1980).109. Visser JWM, Bauman JGJ, Mulder AH, Eliason JF, de Leeuw AM. Isolation ofmurine pluripotential hemopoietic stem ceils. J Exp Med 159:1576 (1984).110. Muller-Seiburg CE, Whitlock CA, Weissman IL. Isolation of two early B-lymphocyteprogenitors from mouse marrow: a committed pre-pre B ceil and a clonogenic Thy-i-bhematopoietic stem cell. Cell 44:653 (1986).111. Spangrude GJ, Heimfeld 5, Weissman IL. Purification and characterization of mousehematopoietic stem cells. Science 241:58 (1988).112. Bertoncello I, Hodgson GS, Bradley TR. Multiparameter analysis of transplantablehemopoietic stem cells. I. The separation and enrichment of stem ceils homing tomarrow and spleen on the basis of rhodamine- 123 fluorescence. Exp HematOl 13:999(1985).59113. Ploemacher RE, Brons RHC. Separation of CFIJ-S from primitive cells responsible forreconstitution of the bone marrow hemopoietic stem cell compartment followingirradiation: evidence for a pre-CFU-S cell. Exp Hematol 17:263 (1989).114. Szilvassy SJ, Lansdorp PM, Humphries RK, Eaves AC, Eaves CJ. Isolation in asingle step of a highly enriched murine hematopoietic stem cell population withcompetitive long-term repopulating ability. Blood 74:930 (1989).115. Spangrude GJ, Scollay R. Differentiation of hematopoietic stem cells in irradiatedmouse thymis lobes: kinetics and phenotype of progeny. J Immunol 145:3661 (1990).116. Spangrude GJ, Johnson GR. Resting and activated subsets of mouse multipotenthematopoietic stem cells. Proc Nail Acad Sci USA 87:7433 (1990c).117. Hodgson GS, Bradley TR. Properties of haematopoietic stem cells surviving 5-fluorouracil treatment: evidence for a pre-CFLJS cell? Nature 281:381 (1979).118. Jones RJ, Celano P, Sharkis SJ, Sensenbrenner LL. Two phases of engraftmentestablished by serial bone marrow transplantation in mice. Blood 73:397 (1989).119. van der Sluijs JP, deJong JP, Brons RHC, Ploemacher RE. Marrow repopulatingcells, but not CFU-S establish long-term in vitro hemopoiesis on a marrow derivedstromal layer Exp Hematol 18:893 (1990).120. Jones RJ, Wagner JE, Celano P, Zicha MS, Sharkis SJ. Separation of pluripotenthaematopoietic stem cells from spleen colony-forming cells. Nature 347:188 (1990).121. Szilvassy SJ, Fraser CC, Eaves CJ, Lansdorp PM, Eaves AC, Humphries RK.Retrovirus -mediated gene transfer to purified hemopoietic stem cells with long-termlympho-myelopoietic repopulating ability. Proc Nail Acad Sci USA 86:8798 (1989).122. Jordan CT, McKearn JP, Lemischka JR. Cellular and developmental properties of fetalhematopoietic stem cells. Cell 61 :953 (1990).123. Siminovitch L, McCulloch EA, Till JE. The distribution of colony-forming cellsamong spleen colonies. J Cell Comp Physiol 62:327 (1963).124. Schofield R. The relationship between the spleen colony-forming cell and thehaemopoietic stem cell. Blood Cells 4:7 (1978).125. Schofield R, Dexter TM. Studies on the self-renewal ability of CFU-S which have beenserially transferred in long-term culture or in vivo. Leuk Res 9:305 (1985).126. Siminovitch L, Till JE, McCulloch EA. Decline in colony-forming ability of marrowcells subjected to serial transplantation into irradiated mice. J Cell Comp Physiol 64:23(1964).127. Nakahata T, Gross AJ, Ogawa M. A stochastic model of self-renewal and commitmentto differentiation of the primitive hemopoietic stem cells in culture. J Cell Physiol113:455 (1982).60128. Suda T, Suda J, Ogawa M. Single cell origin of mouse hemopoietic coloniesexpressing multiple lineages in variable combinations. Proc Nati Acad Sci USA80:6689 (1983).129. Hayflick L. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res37:614 (1965).130. Rosendaal M, Hodgson GS, Bradley TR. Haemopoietic stem cells are organized foruse on basis of their generation-age. Nature 264:68 (1976).131. Cudkowicz G, Upton AC, Shearer GM Hughes WL. Lymphocyte content andproliferative capacity of serially transplanted mouse bone marrow. Nature 201:165(1964).132. Heilman 5, Botnick LE, Hannon EC, Vignuelle RM. Proliferative capacity of murinehematopoietic stem cells. Proc Nati Acad Sci USA 75:490 (1978).133. Dexter TM, Allen TI), Lajtha LG. Conditions controlling the proliferation ofhemopoietic stem cells in vitro. 3 Cell Physiol 91:335 (1977).134. Mauch P, Greenberger JS, Botnick L, Hannon E, Heilman S. Evidence for structuredvariation in self-renewal capacity within long-term bone marrow cultures. Proc NatlAcad Sci USA 77:2927 (1980).135. Harrison DE, Lerner CP, Spooncer E. Erythropoietic repopulating ability of stem cellsfrom long-term marrow culture. Blood 69:1021 (1987).136. Micklem HS, Anderson N, Ross E. Limited potential of circulating haemopoietic stemcells. Nature 256:41 (1975).137. Botnick LE, Hannon EC, Heilman S. Limited proliferation of stem cells survivingalkylating agents. Nature 262:68 (1976).138. Harrison DE, Astle CM, Doubleday JW. Stem cell lines from old immunodeficientdonors give normal responses in young recipients. J Immunol 118:1223 (1977).139. Ogden DA, Micklem HS. The fate of serially transplanted bone marrow cellpopulations from young and old donors. Transplantation 22:287 (1976).140. Harrison DE, Astle CM, Delaittre JA. Loss of proliferative capacity inimmunohemopoietic stem cells caused by serial transplantation rather than aging. 3 ExpMed 147:1526 (1978).141. Harrison DE, Astle CM. Loss of stem cell repopulating ability upon transplantation:Effects of donor age, cell number, and transplantation procedure. J Exp Med 156:1767(1982).142. Harrison DE. Normal function of transplanted mouse erythrocyte precursors for 21months beyond normal life spans. Nature 237:220 (1972).61143. Harrison DE. Normal production of erythrocytes by mouse marrow continuous for 73months. Proc Natl Acad sci USA 70:3 184 (1973).144. Mauch P, Heliman S. Loss of hematopoietic stem cell self-renewal after bone marrowtransplantation. Blood 74:872 (1989).145. Gardner RV, Astle CM, Harrison DE. The decrease in long-term repopulating capacityseen after transplantation is not the result of irradiation-induced stromal injury. ExpHematol 16:49 (1988).146. Dorshkind K. Regulation of hemopoiesis by bone marrow stromal cells and theirproducts. Ann Rev Immunol 8:111(1990).147. Croizat H, Frindel E, Tubiana M. Proliferative activity of the stem cells in the bonemarrow of mice after single and multiple irradiations (total or whole body exposure). JRadliat Biol 18:347 (1970).148. Gidali J, Lajtha LG. Regulation of haematopoietic stem cell turnover in partiallyirradiated mice. Cell Tissue Kinet 5:147 (1972).149. Trentin JJ . Determination of bone marrow stem cell differentiation by stromalhemopoietic inductive microenvironments (HIM). Am J Pathol 65:621(1971).150. Metcalf D, Nicola NA. Synthesis by mouse pertitoneal cells of G-CSF, thedifferentiation inducer for myeloid leukemia cells: Stimulation by endotoxin, M-CSF,and multi-CSF. Leuk Res 9:35 (1985)151. Suttles J, Gin JG, Mizel SB. IL-i secretion by macrophages: enhancement of IL-isecretion and processing by calcium ionophores. J Immunol 144: 175 (1990).152. Kishimoto T, Hirano T. Molecular regulation of B-lymphocyte response. Ann RevImmunol 6:485 (1988).153. Niemeyer CM, Sieff CA, Mathey-Prevot B, Wimperis JZ, Bierer BE, Clarke S,Nathan, DG. Interleukin-3 (IL-3) is produced only by activated human T-lymphocytes.Blood 73:945 (1989).154. Weiss L. The structure of bone marrow. J Morphol 117:467 (1965).155. Tavassoli M, Shaklai M. Absence of tight junctions in endothelium of marrow sinuses:possible significance for marrow cell egress. Br J Haematol 41:303 (1979).156. Weiss L. Hematopoietic microenvironment of the bone marrow: An ukrastructuralstudy of the stroma in rats. Mat Rec 186:161 (1976).157. Lichtman MA. The ultrastructure of the hemopoietic environment of the marrow: areview. Exp Hematol 9:39 1 (1981)158. Tavassoli M, Crosby WH. Transplantation of marrow to extramedullary sites. Science161:541 (1968).62159. Friedenstein AJ, Chailakhjan RK,Latsinik NV, Panasynk AF, Keliss-Borok IV.Stromal cells responsible for transferring the microenvironment of hemopoietic tissues.Transplantation 17:33 1 (1974).160. Ham AW, Cormack DH. The hematopoietic tissues: myeloid tissues. In: Histology.Philadelphia :Lipincott (1977).161. Metcalf D, Moore MAS. Embryonic aspects of haemopoiesis. In: Haemopoietic Cells.Amsterdam: North-Holland (1971).162. Singer JW, Keating A, Cuttner J, Gown AM, Jacobson R, Killen PD, Moohr JW,Najfeld V, Powell J, Sanders J, Striker GE, Fialkow PJ. Evidence for a stem cellcommon to hematopoiesis and its in vitro microenvironment: Studies of patients withclinical hematopoietic neoplasia. Leuk Res 8:535 (1984).163. Sarvella PA, Russell LB. Steel, a new dominant gene in the mouse. J Hered 47:123(1956).164. Bernstein SE, Russell ES Keighley G. Two hereditary mouse anemias (Sl/Sld and(W/W”) deficient in response to erythropoietin. Ann NY Acad Sci 149:475 (1968).165. Ebbe 5, Phalen E, Stohiman FJ. Abnormal megakaryopoiesis in WJW” mice. Blood42:857 (1973).166. Ebbe 5, Phalen E, Stohiman FJ. Abnormal megakaryopoiesis in 51jdmice. Blood42:857 (1973).167. Ruscetti FN, Boggs DR, Torok BJ, Boggs SS. Reduced blood and marrowneutrophils and granulocytic colony forming ceTls in SIJS1d mice. Proc Soc Exp BiolMed 152:398 (1976).168. Kitamura Y, Go S. Decreased production of mast cells in Sl/Sld anemic mice. Blood53:492 (1979).169. Kitamura Y, Go 5, Hatanaka S. Decreased production of mast cells in W/W” mice andtheir increase by bone marrow transplantation. Blood 52:447 (1978).170. McCulloch EA, SiminovitchL, Till JE. Russell ES, Bernstein SE. The cellular basis ofgenetically determined hemopoietic defect in anemic mice of genotype 5115ld Blood26:399 (1965).171. Dexter TM, Moore MAS. In vitro duplication and ‘cure’ of haemopoietic defects ingenetically anemic mice. Nature 269:412 (1977).172. Geissler EN, Ryan MA, Housman DE. the dominant-white spotting (W) locus of themouse encodes the c-kit proto-oncogene. Cell 55:185 (1988).63173. Chabot B, Stephenson DA, Chapman VM, Besmer P, Bernstein A. The protooncogene c-kit encoding a transmembrane tyrosine kinase receptor maps to the mouseW locus. Nature 335:88 (1988).174. Nocka K, Tan JC, Chiu E, Chu E, Ray TY, Traktma P. Bessmer P. Molecular basis ofdominant negative and loss of function mutations at the murine c-kit/white-spottinglocus: W37, W”, W41 and W. EMBO J 9:1810 (1990).175. Zsebo KM, Williams DA, Geissler EN, Broudy VC, Martin FH, Atkins HL, Hsu RY,Birkett NC, Okino KH, Mudock DC, Jacobsen FW, Langley KE, Smith KA, TakeishiT, Cattanach BM, Galli SJ, Suggs SV. Stem cell factor is encoded at the Sl locus of themouse and is the ligand for the c-kit tyrosine kinase receptor. Cell 63:2 13 (1990).176. Dexter TM. Stromal cell associated haemopoiesis. I Cell Physiol suppi 1:87 (1982).177. Eaves C, Coulombel L, Eaves A. Analysis of hemopoiesis in long-term human marrowcultures. In Killam AA, Cronkite EP, Muller-Berat CN. Haematopoietic stem cells.Characterization, proliferation, regulation. Copenhagen: Munksgaard (1983).178. Whitlock CA, Timardish GF, Muller-Sieburg C, Weissman IL. Bone marrow stromacell lines with lymphopoietic activity express high levels of a pre-B neoplasia associatedmolecule. Ce1l48:997 (1987).179. Gimble JM, Pietrangeli C, Henley A Dormein MA, Silver J, Namen A, Takeichi M,- Goridis C, Kincade PW. Characterization of murine bone marrow and spleen-derivedstromal cells: Analysis of leukocyte marker and growth factor mRNA transcript levels.Blood 74:303 (1989).180. Williams DA, Rosenblatt MF, Beirer DR, Cone RD. Generation of murine stromal celllines supporting hematopoietic stem cell proliferation by use of recombinant retrovirusvectors encoding simian virus 40 large T antigen. Mol Cell Biol 8:3864 (1988).181. Tsai S, Emerson SG, Sieff CA, Nathan DG. Isolation of a human stromal cell strainsecreting hemopoietic growth factors. J Cell Physiol 127:137 (1986).182. Yang YC, Tsai S, Wong GG, Clark SC. Interleukin-1 regulation of hematopoieticgrowth factor production by human stromal fibroblasts. J Cell Physiol 134:292 (1988).183. Seelentag WK, Mermod JJ, Montesano R, Vassalli P. Additive effects of interleukin-1and tumor necrosis factor-a on the accumulation of the three granulocyte andmacrophage colony-stimulating factor mRNAs in human endothelial cells. EMBO J6:2261 (1987).184. Rennick D, Yang G, Gemmell L, Lee F. Control of hemopoiesis by a bone marrowstromal cell clone: Lipopolysaccharide and interleukin-l inducible production ofcolony-stimulating factors. Blood 69:682 (1987).64185. Fibbe WE, van Damme J, Bilau A, Goselink HM, Voogt PJ, van Eeden G, Ralph P,Altrock BW, Falkenburg JHF. Interleukin-l induces human marrow stromal cells inlong-term marrow culture to produce granulocyte colony stimulating factor andmacrophage colony stimulating vactor. Blood 7 1:430 (1988).186. Zsebo KM, Yuschenkoff VN, Schiffer S, Chang D, McCall E, Dinarello CA, BrownMA, Altrock B, Bagby GJ. Vascular endothelial cells and granulopoiesis: Interleukin-1stimulates release of G-CSF and GM-CSF. Blood 7 1:99 (1988).187. Collins LS, Dorshkind K. A stromal cell line from myeloid long-term bone marrowcultures can support myelopoiesis and B lymphopoiesis. J Immunol 138:1082 (1987).188. Gabrilove JL. Introduction and overview of hematopoietic growth factors. Sem inHematol 26:1 (1989).189. Metcalf D. The molecular control of cell division, differentiation commitment andmaturation in haemopoietic cells. Nature 339:27 (1989).190. Williams D, Nathan DG. Introduction: The molecular biology of hematopoiesis. Sem inHematol 28:114 (1991).191. Alderson MR, Tough TW, Zeigler SF, Grabstein KH. Interleukin-7 induces cytokinesecretion and tumoricidal activity by human peripheral blood monocytes. J Exp Med173:923 (1991).192. Whetton AD, Dexter TM. Myeloid haemopoietic growth factors. Biochim BiophysActa 989:111(1989).193. Just U, Stocking C, Spooncer E, Dexter TM Ostertag W. Expression of the GM-CSFgene after retroviral transfer in hematopoietic stem cell lines induces synchronousgranulocye-macrophage differentiation. Cell 64:1163 (1991).194. Sanderson CJ, O’Garra A, Warren DJ, Klaus B. Eosinophil differentiation factor alsohas B cell growth factor activity: proposed name Interleukin 5. Proc Nat! Acad SciUSA 83:437 (1986).195. Warren DJ, Moore MAS. Synergism among interleukin 1, interleukin 3, andinterleukin 5 in the production of eosinophils from primitive hemopoietic stem cells. J Immunol 140:94 (1988).196. Lee G, Ellingsworth LR, Gillis S, Wall R, Kincade PW. f3-transforming growthfactors are potential regulators of B-lymphopoiesis. J Exp Med 166:1290 (1988).197. Dubois CM, Ruscetti FW, Palaszynski EW, Falk LA, Oppenheim JJ, Keller JR.Transforming growth factor f is a potent inhibitor of interleukin 1 (IL-i) receptorexpression: proposed mechanism of inhibition of IL-i action. 3 Exp Med 172:737(1990).65198. Eisenberg SP, Evans RJ, Arend WP, Verderber E, Brewer MT, Hannum CII,Thompson RC. Primary structure and functional expression from complementary DNAof a human interleukin-1 receptor antagonist. Nature 343:34 1 (1990).199. Graham GJ, Wright EG, Hewick R, Wolpe SD, Wilkie NM, Donaldson D, Lorimore5, Pragnell TB. Identification and characterization of an inhibitor of haemopoietic stemcell proliferation. Nature 344:442 (1990).200. Ihie IN, Keller J, Oroszlan S, Henderson LE, Copeland TD, Fitch F, Prystowski MB,Goidwasser E, Schrader JW, Palaszinski E, Dy M, Lebel B. Biological properties ofhomogeneous interleukin 31. Demostration of WEHI-3 growth factor activity, mast cellgrowth factor activity, P cell-stimulating factor activity, colony-stimulating factoractivity, and histamine-producing cell-stimulating factor activiy. J Immunol 131:282(1983).201. Suda T, Suda J, Ogawa M, Ihle JN. Permissive role of interleukin-3 (IL-3) inproliferation and differentiation of multipotential hemopoietic progenitors in culture. JCell Physiol 124:182 (1985).202. Muller-Sieburg CA, Townsend K, Weissman IL, Rennick D. Proliferation anddifferentiation of highly enriched mouse hematopoietic stem cells and progenitor cells inresponse to defined growth factors. J Exp Med 167:1825 (1988).203. Bartelmez SH, Stanley ER. Synergism between hemopoietic growth factors (HGFs)detected by their effectson cells bearing receptors for a lineage specific HGF: Assay ofhemopoietin- 1. J Cell Physiol 122:370 (1985).204. Stanley ER, Bartocci A, Patinkin D, Rosendaal M, Bradley TR. Regulation of veryprimitive, multipotent hemopoietic cells by hemopoietin-1. Cell 45:667 (1986).205. Zsebo KM. Wypych 3, Yuschenkoff VN, Lu H, Hunt P, Dukes PP, Langley KE.Effects of Hematopoietin-l and interleukin-1 activities on early hematopoietic cells ofthe bone marrow. Blood 7 1:962 (1988).206. Bartelmez SH, Bradley TR, Bertoncello I, Mochizuki DY, Tushinski RJ, Stanley ER,Hapel AJ, Young 1G. Kreigler AB, Hodgson GS. Interleukin 1 plus interleukin 3 pluscolony-stimulating factor 1 are essential for clonal proliferation of primitive myeloidbone marrow cells. Exp Hematol 17:240 (1989).207. Visser JWM, deVries P, Hogeweg-Platenburg MGC, Bayer 3, Schoeters G, Van DenHeuvel R, Mulder DH. Culture of hematopoietic stem cells purified from murine bonemarrow. SeminHematol28:117 (1991).208. Keller 1K, McNeice 1K, Ellingsworth LR, Quesenberry P3, Sing GK, Ruscetti FW.Transforming growth factor 13 directly regulates primitive murine hematopoietic cellproliferation. Blood 75:596 (1990).209. Ikebuchi K, Wong GG, Clark SC, Ihle JN, Hirai Y, Ogawa M. Interleukin-6enhancement of interleukin-3-dependant proliferation of multipotential hemopoieticprogenitors. Proc Natl Acad Sci USA 84:9035 (1987).66210. Hirano T, Yasukawa K, Harada H, Taga T, Watanabe Y,Matsuda T, Kashiwamura S,Naksjima K, Koyama K, Iwamatsu A, Tsunasawa S. Sakiyama F, Matsui H, TakaharaY, Taniguchi T, Kishimoto T. Complimentary DNA for a novel human interleukin(BSF-2) that induces B-lymphocytes to produce immunoglobulin. Nature 324:73,(1986).211. Wong GG, Witek-Giannotti JS, Temple TA, Kriz R, Ferenz C, Hewick RM, ClarkSC, Ogawa M. Stimulation of murine hematopoietic colony formation by humaninterleukin-6. J Immunol 140:3040 (1988)212. Suda T, Yamaguchi Y, Suda J, Miura Y, Okana A, Akiyama Y. Effect of interleukin-6(IL-6) on the differentiation and proliferation of murine and human hemopoieticprogenitors. Exp Hematol 16:3040 (1988).213. Ikebuchi K, Clark SC, Ihie JN, Souza LM, Ogawa M. Granulocyte colony-stimulatingfactor enhances interleukin-3 dependent proliferation of multipotential hemopoieticprogenitors. Proc Nati Acad Sci USA 85:3445 (1988).214. Ikebuchi K, Ihie IN, Hirai Y, Wong GG, Clark SC, Ogawa M. Synergistic factors forstem cell proliferation: Further studies of the target stem cells and the mechanism ofstimulation by interleukin-1, interleukin-6, and granulocyte colony-stimulating factor.Blood 72:2007 (1988).215. Musashi M, Yang Y-C Paul SR C1arkSC, Sudo T, Ogawa M. Direct and synergisticeffects of interleukin 11 on murine .hernopoiesis in culture. Proc NatlAcad Sci USA88:765 (1991).216. Leary AG, Hirai Y, Kishimoto T, Clark SC, Ogawa M. Survival of hemopoieticprogenitors in the Go period of the cell cycle does not require early hemopoieticregulators. Proc Nati Acad Sci USA 86:4535 (1989).217. Kinashi T, Inaba K, Tsubata T, Tashiro K, Palacios R, Honjo T. Differentiation of aninterleukin 3-dependent precursor B-cell clone into immunoglobulin-producing cells invitro. Proc Nati Acad Sci USA 85:4473 (1988).218. Palacios R, Stuber 5, Rolink A. The epigenetic influences of bone marrow and fetalliver stroma cells on the developmental potential of Ly-1+ pro-B lymphocyte clones.Eur J Immunol 19:347 (1989).219. Namen AE, Lupton 5, Hjerrild K, Wignall J, Mochizuki DY, Schmierer A, Mosely B,March CJ, Urdal D, Gillis 5, Cosman D, Goodwin RG. Stimulation of B-cellprogenitors by cloned murine interleukin-7. Nature 333:57 1 (1988).220. Namen AE, Schmierer A, March CJ, Overell RW, Park LS, Urdal D, Mochizuki DY.B cell precursor growth-promoting activity. Purification and characterization of agrowth factor active on lymphocyte precursors. J Exp Med 167:988 (1988).221. Takeda 5, Gillis 5, Palacios R. In vitro effects of recombinant interleukin 7 on growthand differentiation of bone marrow pro-B and pro-T lymphocyte clones and fetalthymocyte clones. Proc Nati Acad Sci USA 86:1634 (1989).67222. Henny CS. Interleukin 7: effects on early events in lymphopoiesis. Immunol Today10: 170 (1989).223. Kincade PW, Lee G, Pietrangeli CE, Hayashi S-I, Gimble JM. Cells and moleculesthat regulate B lymphopoiesis in bone marrow. In: Annual Review of Immunology,vol7. (1989).224. Hofman FH, Brock M, Taylor CR, Lyons B. 11-4 regulates differentiation andproliferation of human precursor B cells. J Immunol 141:1185 (1988).225. Peschel C, Green I, Paul WE. Preferential proliferation of immature B lineage cells inlong-term stromal cell-dependent cultures with IL-4. J Immunol 142:1558 (1989).226. Lee G, Ellingsworth LR, Gillis S, Wall R, Kincade PW. transforming growthfactors are potential regulators of B lymphopoiesis. J Exp Med 166: 1290 (1987).227. Suda TR, Murray R, Fischer M, Yokota T, Ziotnik A. Tumor necrosis factor x andP40 induce day 15 murine fetal thymocytes proliferation in combination with IL-2. JImmunol 144: 1783 (1990).228. Ziotnik A, Ransom J, Frank G, Fischer M, Howard M. Interleukin 4 is a growth factorfor activated thymocytes: possible role in T-cell ontogeny. Proc Natl Acad Sci USA84:3856 (1987).229. Suda TR, Murray R, Guidos C, Ziotnik A. Growth promoting activity of IL-la, IL-6,and tumor necrosis factor-a in combination with IL-2, IL-4, or IL-7 on murinethymocytes. J Immunol 144:3039 (1990).230. MacNeil IA, Suda TR, Moore KW, Zlotnilc A.Interleukin 10: a novel growth cofactorfor mature and immature T cells. J Immunol 145:4 167 (1990).231. Okazaki H, Ito M, Sudo T, Hattori M, Kamo S. Katsura Y, Minato N. 11-7 promotesthymocyte proliferation and maintains immunocompetent thymocytes bearing (43 orT-cell receptor in vitro: synergism with IL-2. J Immunol 143:29 17 (1989).232. Suda TR, Ziotnik A. 11-7 maintains the T cell precursor potential of CD3-CD4-CD8-thymocytes. J Immunol 146:3068 (1991).233. Glaspy JA, Golde DW. Clinical applications of the myeloid growth factors. Sem inHematol 26:14 (1989).234. Shimamura M, Kobayashi Y, Yuo A, Urabe A, Okabe T, Komatsu Y, Itoh 5, TakakuF. Effect of human recombinant granulocyte colony-stimulating factor on hematopoieticinjury in mince induced by 5-fluorouracil. Blood 69:353 (1987).235. Blazar BR, Widmer MB, Soderling CC, Urdal DL, Gillis 5, Robinson LL, ValleraDA. Augmentation of donor bone marrow engraftment in histocompatable murine68recipients by granulocyte/macrophage colony-stimulating factors. Blood 71:320(1988).236. Neta R, Oppenheim JJ, Douches SD. Interdependence of the radioprotective effects ofhuman recombinant interleukin alpha, tumor necrosis factor alpha, granulocyte colony-stimulating factor, and murine recombinant granulocyte-macrophage colony-stimulatingfactor. J Immunol 140: 108 (1988).237. Kindler V. Thorens B, deKossodo 5, Allet B, Eliason JF, Thatcher D, Farber N,Vassalli P. Stimulation of hematopoiesis in vivo by recombinant bacterial murineinterleukin 3. Proc Natl Acad Sci USA 83:1001 (1986).238. Morrissey P, Charrier K, Bressler L, Alpert A. The influence of IL-i treatment on thereconstitution of the hemopoietic and immune systems after sublethal radiation. JImmunol 140:4204 (1988).239. Benjamin WR, Tare NS, Hayes TJ, Becker JM, Anderson TD. Regulation ofhemopoiesis in myelosuppressed mice by human recombinant IL-ia. J Immunol142:792 (1989).240. Okano A, Suzuki C, Takatsuki F, Akiyama Y, Koilce K, Nakahata T, Hirano T,Kishimoto T, Ozawa K, Asano S. Effects of interleukin-6 on hemopoiesis in bonemarrow-transplanted mice. Transplantation 47:738 (1989).241. Patchen ML, MacVitte TJ, Williams JL, Schwartz GN, Souza LM. Administration ofinterleukin-6 stimulates mukilineage hematopoiesis and accelerates recovery fromradiation-induced hematopoietic depression. Blood 77:472 (1991).242. Moore MAS, Warren DJ. Synergy of interleukin-i and granulocyte colony stimulatingfactor: In vivo stimulation of stem cell recovery and hematopoietic regenerationfollowing 5-fluorouracil treatment of mice. Proc Natl Acad Sci USA 84:7134 (1987).243. Atkinson K, Matias C, Guiffre A, Seymour R, Cooley M, Biggs J, Munro V, Gihis S.In vivo administration of granulocyte colony-stimulating factor (G-CSF), granulocytemacrophage CSF, Interleukin-1 (IL-i), and IL-4, alone or in combination, afterallogeneic murine hematopoietic stem cell transplantation. Blood 77:1376(1991).244. Tamura M, Hattri K, Nomura H, Oheda M, Kubota N, Imazeki I, Ono M, Ueyama Y,Nagata 5, Sirafuji N, Asano S. Induction of neutrophilic granulocytosis in mice byadministration of purified human native granulocyte colony-stimulating factor (GCSF). Biochem Biophys Res Commun 142:454 (1987).245. Lord BI, Molineus B, Testa NG, Kelly M, Spooncer E, Dexter TM. the kineticresponse of haemopoietic precursor cells in vivo to highly purified recombinantinterleukin-3. Lymphokine Res 5:97 (1986).246. Kimura H, Ishibashi T, Shikama Y, Okano A, Akiyama Y, Uchida T, Maruyama Y.Interleukin- 1 3 (IL-i ) induces thrombocytosis in mice: possible implication of IL-6.Blood 76:2493 (1990).69247. Migliaccio AR, Visser 1MW. Proliferation of purified murine hemopoietic stem cells inserum-free cultures stimulated with purified stem-cell-activating factor. Exp Hematol14: 1043 (1986).248. Mulder AH, Visser JMW, van den Engh GJ. Thymus regeneration by bone marrowcell suspensions differing in the potential to form early and late spleen colonies. ExpHematol 13:768 (1985).249. Bodine DM, Karisson S, Neinhuis AW. Combinations of interleukins 3 and 6preserves stem cell function in culture and enhances retrovirus-mediated gene transferinto hematopoietic stem cells. Proc Nati Acad Sci USA 86:8897 (1989).250. de Vries P, Brasel KA, Eisenman JR, Alpert AR, Williams DE. The effect ofrecombinant mast cell growth factor on purified murine hematopoietic stem cells. J ExpMed 173: (1991).251. Testa NG, Dexter TM. Long-term production of erythroid precursor cells (BFU) inbone marrow cultures. Differentiation 9:193 (1977).252. Williams NH, Jackson H, Rabellino EM. Proliferation and differentiation of normalgranulopoietic cells in continuous bone marrow cultures. J Cell Physiol 93:435 (1977).253. Williams NH, Jackson H, Sheridan APC, Murphy MJ, Elste A, Moore MAS.Regulation of megakaryopoiesis in long term murine bone marrow cultures. Blood51:245 (1978).254. Dexter TM, Spooncer E. Loss of immunoreactivity in long-term bone marrow culture.Nature 275:135 (1978).255. Mauch P, Greenberger JS, Botnick L, Hannon E, Heilman S. Evidence for structuredvariation in self-renewal capacity within long-term bone marrow cultures. Proc NatlAcad Sci USA 77:2927 (1980).256. Bently SA. Close range cell:cell interaction required far stem cell maintenance incontinuous bone marrow culture. Exp Hematol 9:308 (1981).257. Mon KJ, Izumi H, Seto A. Stimulation and support of hemopoietic stem cellproliferation by irradiated siroma cell colonies in bone marrow cell culture in vitro. JRadiat Res 22: 109 (1981).258. Schofield R, Dexter TM. Studies on the self-renewal ability of CFU-S which have beenserially transferred in long-term culture or in vivo. Leuk Res 9:305 (1985)..259. Toksoz D, Dexter TM, Lord BI, Wright EG, Lajtha LG. The regulation of hemopoiesisin long-term bone marrow cultures. II. Stimulation and inhibition of stem cellproliferation. Blood 55:931(1980).260. Jones-Villeneuve E, Phillips RA. Potentials for lymphoid differentiation by cells fromlong-term cultures of bone marrow. Exp Hematol 8:65 (1980).70261. Harrison DE, Lemer CP, Spooncer E. Erythropoietic repopulating ability of stem cellsfrom long-term marrow culture. Blood 69:1021 (1987).262. Dorshldnd K, Phillips RA. Characterization of early B lymphocyte precursors presentin long-term bone marrow cultures. 3 Immunol 131:2240 (1983).263. Bosma GC, Custer RP, Bosma MJ. A severe combined immunodeficiency mutation inthe mouse. Nature 301:527 (1983).264. Fulop GM, Phillips RA. Use of scid mice to identify and quantitate lymphoid-restrictedstem cells in long-term bone marrow cultures. Blood 74:1537 (1989).265. Spooncer E, Lord BI, Dexter TM. Defective ability to self-renew in vitro of highlypurified primitive haematopoietic cells. Nature 3 16:62 (1985).266. Varmus HE, Swanstrom R. Replication of retroviruses. In: RNA Tumor Viruses. ColdSpring Harbor, New York: Cold Spring Harbor Laboratory (1984).267. Coffin J. Genome structure.In: RNA Tumor Viruses, vol 2. Cold Spring Harbor, NewYork: Cold Spring Harbor Laboratory (1985).268. Mann R, Baltimore D. Varying position of a retrovirus packaging sequence results inencapsidation of both unspliced and spliced RNAs. J Virol 54:401 (1985).269. Schwartzenberg P, Colicelli I, Goff SP. Deletion mutants of Moloney murine leukemiavirus which lack glycosylated gag protein are replication competent. Virology 91:481(1983).270. Bender MA, Palmer TD, Gelinas RE, Miller AD. Evidence that the packaging signal ofmolony murine leukemia virus extends into the gag region. 3 Virol 61:1639 (1987).271. Armentano D, Yu S-F, Kantoff PW, Von R T, Anderson WF, Gilboa E. Effect ofinternal viral sequences on the utility of retroviral vectors. J Virol 61:1647 (1987).272. Maddon PJ, Dangleish AG, McDougal IS, Clapham PR, Weiss RA, Axel R. The T4gene encodes the AIDS virus receptor and is expressed in the immune system and thebrain. Cell 47:333 (1986).273. Albritton LM, Tseng L, Scadden D, Cunningham JM. A putative murine ecotropicretrovirus receptor gene that encodes a multiple membrane spanning protein and conferssusceptibility to virus infection. Cell 57:659 (1989).274. Panganiban AT, Fiore D. Ordered interstrand and inirastrand DNA transfer duringreverse transcription. Science 241:1046 (1988).275. Hu W-S , Temin HM. Retroviral recombination and reverse transcription. Science250: 1227 (1990)276. Zack JA, Arrigo SJ, Weitsman SR, Go AS, Allyson H, Chen ISY. HIV-1 entry intoquiescent primary lymphocytes: molecular analysis reveals a labile, latent viralstructure. Cell 61:213 (1990).71277. Fujiwara T, Mizuuchi K. Retroviral DNA integration: structure of an integrationintermediate. Cell 54:497 (1988).278. Brown P0, Bowerman B, Varmus HE, Bishop JM. Retroviral integration: structure ofthe initial covalent product and its precursor, and role for the viral IN protein. Proc NailAcad Sci USA 86:2525 (1989).279. Fujiwara T, Craigie R. Integration of mini-retroviral DNA: A cell-free reaction forbiochemical analysis of retroviral integration. Proc Nati Acad Sci USA 86:3065 (1989).280. Craigie R, Fujiwara T, Bushman F. The IN protein of Moloney murine leukemia virusprocesses the viral DNA ends and accomplishes thier integration in vitro. Cell 62: 829(1990).281. Grandgenett DP, Mumm SR. Unravelling retrovirus integration. Cell 60:3 (1990).282. Shih C-C, Stoye JP, Coffin JM. Highly preferred targets for retrovirus integration.Cell 53:53 1 (1988).283. Barklis E, Mulligan RC, Jaenisch R. Chromosomal position or virus mutation permitsretrovirus expression in embryonal carcinoma cells. Cell 47:39 1 (1986).284. Taketo M, Tanaka M. A cellular enhancer of retrovirus gene expression in embryonalcarcinoma cells. Proc Natl Aca1 Sci USA 84:3748 (1987). -285. Hilberg F, Stocking C, Ostertag W, Grez M. Functional analysis of a retroviral host-range mutant: Altered long terminal repeat sequences allow expression in embryonalcarcinoma cells. Proc Nail Acad Sci USA (1987).286. Varmus HE, Quintrell N, Oritz S. Retroviruses a mutagens: insertion and excision of anontransforming provirus alter expression of a resident transforming provirus. Cell25:23 (1981).287. Neel BG, Hayward WS, Robinson HL, Fang J, Astrin SM. Avian leukosis virus-induced tumors have common proviral integration sites and synthesize discrete newRNAs: oncogenesis by promoter insertion. Cell 23:323 (1981).288. Mann R, Mulligan RC, Baltimore D. Construction of a retrovirus packaging mutantand its use to produce helper-free defective retrovirus. Cell 33:153 (1983).289. Miller AD, Buttimore C. Redesign of retrovirus packaging cell lines to avoidrecombination leading to helper virus production. Mol Cell Biol 6:2895 (1986).290. Danos 0, Mulligan RC. Safe and efficient generation of recombinant retroviruses withamphotropic and ecotropic host ranges. Proc Natl Acad Sci USA 85:6460 (1988).291. Markowitz D, Goff S, Bank A. A safe packaging line for gene transfer: separating viralgenes on two different plasmids. J Virol 62:1120 (1988).72292. Rein A, Schultz A. Different recombinant murine leukemia viruses use different cellsurface receptors. Virology 136:144 (1984).293. Cone RD, Mulligan RC. High-efficiency gene transfer into mammalian cells: generationof helper-free recombinant retroviruses with broad mammalian host range. Proc Nat!Acad Sci USA 8 1:6349 (1984).294. Miller DA, Law M-F, Verma IM. Generation of helper-free amphotropic retrovirusesthat iransduce a dominant-acting, methoirexate-resistant dihydrofolate reductase gene.Mol Cell Biol 5:43 1 (1985).295. Bodine DM, McDonagh KT, Brandt SJ, Ney PA, Agricola B, Byrne E, Nienhuis AW.Development of a high titer retrovirus producer cell line capable of gene transfer intorheusus monkey hematopoietic stem cells. Proc Natl Acad Sci USA 87:3738 (1990).296. Hwang LS, Gilboa E. Expression of genes introduced into cells by retroviral infectionis more efficient than that of genes introduced into cells by DNA transfection. J Virol50:417 (1984).297. Miller AD, Jolly DJ, Freidmann T, Verma TM. A transmissible retrovirus expressinghuman hypoxanthine phosphoribosyl transferas (HPRT): gene transfer into cellsobtained from humans deficient in HPRT. Proc Nati Acad Sci USA 80:4709 (1983).298. Miller AD, Ong ES, Rosenfeld MG, Verma TM, Evans RM. Infectious and selectableretrovirus containing an inducible rat growth hormone minigene. Science 225:993(1984).299. Chang JM, Johnson GR. Gene transfer into hemopoietic stem cells using retroviralvectors. Tnt J Cell Cloning 7:264 (1989).300. Hawley RG, Covarrubias L, Hawley T, Mintz B. Handicapped retroviral vectorsefficiently transduce foreign genes into hematopoietic stem cells. Proc Natl Acad SciUSA 84:2406 (1987).301. Belmont JW, MacGregor GR, Wagner-Smith K, Fletcher FA, Moore KA, Hawkins D,Vilalon D, Chang SMW, Caskey CT. Expression of human adenosine deaminase inmurine hematopoietic cells. Mol Cell Biol 8:5116(1988).302. Holland CA, Anidesaria P, Sakakeeny MA, Greenberger JS. Enhancer sequences of aretroviral vector determine expression of a gene in multipotent hematopoieticprogenitors and committed erythroid cells. Proc Natl Acad Sci USA 84:8662 (1987).303. Cepko CL, Roberts BE, Mulligan RC. Construction and applications of a highlytransmissible murine retrovirus shuttle vector, cell 37:1053 (1984).304. Magli M-C, Dick TE, Huszar D, Bernstein A, Phillips RA. Modulation of geneexpression in multiple heatopoietic cell lineages following retroviral vector genetransfer. Proc Natl Acad Sci USA 84:789 (1987).305. Kaleko M, Garcia JV, Osborne WRA. Expression of human adenosine deaminase inmice after transplantation of genetically modified bone marrow. Blood 75:1733 (1990).73306. Lim B, Apperly JF, Orkin SH, Williams DA. Long-term expression of humanadenosine deaminase in mice transplanted with retrovirus infected hemopoietic stemcells. Proc Nati Acad Sci USA 86:8892 (1989).307. Wilson JM, Danos 0, Grossman M, Raulet DH, Mulligan RC. Expression of humanadenosine deaminase in mice reconstituted with retrovirus-transduced hematopoieticstem cells. Proc Nati Acad Sci USA 87:439 (1990).308. Apperley JF, Luskey BD, Williams DA. Retroviral-mediated gene transfer of humanadenosine deaminase into murine hematopoietic cells. Sem Hematol 28:170(1991).309. Li CL, Dwarki VJ, Verma IM. Expression of human cz-globin and mouse/humanhybrid [-globin genes in murine hemopoietic stem cells transduced by recombinantretroviruses. Proc Natl Acad Sci USA 87:4349 (1990).310. Bender MA, Gelinas RE, Miller AD. A majority of mice show long-term expression ofa human f3-globin gene after retrovirus transfer into hematopoietic stem cells. Mol CellBiol 9: 1426 (1989).311. Bodine DM, McDonagh KT, Seidel NE, Neinhuis AW. Survival and retrovirusinfection of murine hematopoietic stem cells in vitro: effects of 5-FU and method ofinfection. Exp Hematol 19:206 (1991).312. Harrison DE, Lerner CP. Most primitive hempoietic stem cells (PHSC) are stimulatedto cycle rapidly after treatment with 5-FU. (in press).313. Botwell DDL, Johnson GR, Kelso A, Cory S. Expression of genes transferred tohaemopoietic stem cells by recombinant retroviruses. Mol Biol Med 4:229 (1987).314. Dumenil D, Jacquemin-Sablon H, Neel H, Frindel E, Dautry F. Mock retroviralinfection alters the developmental potential of murine bone marrow stem cells. Mol CellBiol 9:4541 (1989).315. Fletcher FA, Williams DE, Maliszewski C, Anderson D, Rives M, Belmont 3W.Murine leukemia inhibitory factor enhances retroviral-vector infection efficiency ofhematopoietic progenitors. Blood 76:1098 (1990).316. Johnson GR, Gonda TJ, Metcalf D, Harihan 1K, Cory S. A lethal myeloproliferativesyndrome in mice transplanted with bone marrow cells infected with a retrovirusexpressing granulocyte-macrophage colony stimulating factor. EMBO J 8:441 (1988).317. Wong PMC, Chung 5, Dunbar CE, Bodine DM, Ruscetti SR, Neinhuis AW.Retrovirus-mediated gene transfer and expression of the interleukin-3 gene in mousehemopoietic cells results in a myeloproliferative disorder. Mol Cell Biol 9:797 (1989).318. Chang JM, Metcalf D, Lang RA, Gonda RJ, Johnson GR. Nonneoplastichematopoietic myeloproliferative syndrome induced by dysregulated multi-CSF (IL-3)expression. Blood 73:1487 (1989).74319. Kawano M, Hirano T, Matsuda T, Taga T, Horii Y, Iwato K, Asaoku H, Tang B,Tanabe 0, Tanaka H, Kuramoto A, Kishimoto T. Autocrine generation andrequirement of BSF-211L-6 for human multiple myelomas. Nature 332:83 (1988).320. Klein B, Zhang X-G, Jourdan M, Content J, Houssiau F., Aarden L, Piechaczyk M,Bataille R. Paracrine rather than autocrine regulation of myeloma-cell growth anddifferentiation by interleukin-6. Blood 73:5 17 (1989).321. Hoang T, Haman A, Goncalves 0, Wong GG, Clark SC. Interleukin-6 enhancesgrowth factor-dependent proliferation of blast cells of acute myeloblastic leukemia.Blood 72:823 (1988).322. Yee C, Biondi A, Wang XH, Iscove NN, deSousa J, Asrden LA, Wong GG, ClarkSC, Messner HA, Minden MD. A possible autocrine role for interleukin-6 in twolymphoma cell lines. Blood 74:798 (1989).323. Brandt SI, Bodine DM, Dunbar CE, Neinhuis AW. Dysregulated interleukin 6expression produces a syndrome resembling Castleman’s disease in mice. J Clin Invest86:592 (1990).324. Chang JM, Johnson GR. Effects on spleen colony-forming unit self renewal afterretroviral-mediated gene transfer of multi-colony-stimulating factor, granulocytemacrophage colony stimulating factor, or granulocyte colony-stimulating factor. ExpHematol 19:602 (1991).75CHAPTER IIMATERIALS AND METHODS1) RETROVIRUS CONSTRUCTION. PRODUCTION.AND ASSAYS.A) Recombinant Retroviruses.a) Tkneol9.Tkneol9 is a replication defective murine-leukemia virus based vector containing thebacterial gene for neomycin phosphotransferase (neoT)linked to the herpes simplex virusthymidine kinase (Tk) promoter in the reverse orientation relative to the hybrid Moloney andHarvey murine leukemia virus long terminal repeat (LTR) region (Figure 6). This vector wasderived by Dr. Donna Hogge of the Terry Fox Lab (TFL) from MMCV-neo (1,2) by removalof the myc and Tkneo sequences and re-insertion of the latter in the 3’-S orientation relative toretroviral transcription (3). A helper-free viral producer cell line was generated in the ‘P-2packaging cell line by Dr. Phil Hughes of the TFL as described below.b) JZen-neoJZen-neo and JZenmiL7tkneo viruses (Figure 6) were constructed and produced in acollaboration with J.D. Thacker, a graduate student in the Terry Fox Laboratory. Both wereconstructed from the JZen. 1 viral backbone provided by Gregory R. Johnson (Walter Elisa76MoMuLV 5’ LTRXho I (Hpa I /Smal)rnIL-7JZenmlL7tkneondtkpromoterHybridMPSV/MoMuLV3’ LTRFigure 6. Schematic representation of (A.) tkneol9, as well as construction of (B.) JZenneoand (C.) JZenmlL7tkneo. Plasmid pMClneo and pTZl8Rtkneo were used in theproduction of JZenneo and JZenmlL7tkneo. The derivation of each is described indetail in Materials and Methods.HybridHaSVJMoMuLVHaSV 3’ LTRA.B.C.IJZen-neoMlu I / Hinc II bluntinto JZen HpaII)MPSV/MoMuLV3’ LTRpTZl8Rtkneoblunt mIL-7fragment intoblunt Xba I sitemIL-7Sma I / Hind III IL7tkneo fragment intoHpa I / Hind III of JZen77Hall, Melbourne Australia) (4). JZen. 1 construction has been previously described, and wasderived by replacing the 3’ LTR enhancer from a Moloney virus based vector ZipNeoSV(X)(5) with an enhancer from the LTR of the myeloproliferative sarcoma virus. To constructJZen-neo (Figure 6), an 854 base pair MluJJHincII fragment encompassing only the codingregion of the neor gene was isolated from pMClneo (6) blunted and inserted into the HpaI siteof JZen.1 using standard procedures (7). High-titre viral producing cell lines used were thengenerated using the packaging cell lines GP+envAM- 12 and GP+E-86 as described in this text.c) JZenmlL7tkneo.To construct JZenmlL7tkneo (Figure 6), a XhoJ/SalI 1092 bp fragment from pMClneo(6) containing the neo1 gene linked to the promoter region of the HSV virus and polyomaenhancer was inserted into the SalT site of pTZ18R (Pharmacia, Bale d’Urfe Quebec) toproduce pTZl8RTkneo. A 524 base pair BamHl/XmnT fragment encompassing the completeIL-7 coding sequence was isolated from plasmid 1046B (8) (a gift from A. Namen,Immunex.Corp.). This plasmid contains an IL-7 cDNA engineered to remove all upstreamstart codons (ATG’s) found in the mature TL-7 cDNA. This fragment was inserted afterblunting (7) into the blunted Xba I site of PTZl8Rtkneo. The mIL7-tkneo cassette was thenremoved by digestion with Small HindIll and subcloned into the HpaTlHindIll site of theJZen. 1 polylinker. Generation of high-titre GP+E-86 JZenmlL-7tkneo clones were carried outas described for JZen-neo.78B) Viral Packain Cell Lines.The viral producer cell lines used for production of recombinant retroviruses were theamphotropic packaging line PA3 17 (from D. Miller, Fred Hutchinson Research Center,Seattle, WA )(9,1O), the ecotropic packaging line psi-2 (from R.C. Mulligan, Whitehead Inst.,Boston MA)(11); and amphotropic and ecotropic packaging lines GP+env-AM12 and GP+E86 respectively (from Arthur Bank, Columbia University, New York, NY) (12). All cellswere maintained at 37°C in a 5% C02 incubator in Dulbecco’s modified Eagle’s medium(DMEM) with high glucose (4.5 g/L) with the following additives. For PA317 cell linesDMEM contained 10% fetal calf serum (FCS), for the psi-2 cell line DMEM contained 10%heat inactivated (60°C, 2 hours) newborn calf-serum (CS) (Johns Scientific). GP+env-AM12and GP+E-86 were maintained in HXM media (DMEM containing 10% calf serum, 15tgfmLhypoxanthine, 0.25 mg/mL xanthine, and 2.5 j.tg/mL mycophenolic acid (Sigma ChemicalCo., St. Louis, MO).C) Generation of Viral Producer Cell Lines.The following protocol was used for production of infectious recombinant TKneol9retroviruses from ‘P-2 viral producer cell lines. The proviral plasmid DNA was introduced intothe packaging PA3 17 cells by the calcium phosphate (CaPO4) transfection technique. DNAprecipitate was formed by combining 10-20 jig of plasmid DNA in 0.05 mL of 2.5 M CaC12with 0.5 mL of 2 X HBS (50mM Hepes, 3M NaC1, 1.5 mM Na2HPO4, pH 7.12) whilegently bubbling N2 through the solution to mix. The solution was allowed to stand at roomtemperature for 30 minutes, gently mixed, then 0.5 mL added to 4.5 mL of media on thePA3 17 cells plated 24 hours previously at a density of 2 x cells per 60mm tissue culture79dish. After 18 hours the media was changed and 24 hours later media was removed to harvesttransiently expressed virus. This supernatant was filtered (O.22p.m filter, Millipore, Bedford,MA) and overlayed on ‘P-2 cells ( 1x105 cells/6Omm dish), in the presence of 4 .ig/mLpolybrene (Sigma Chemical CO, St Louis, MO). After 48 hours the media was changed, andG418 was added to a final concentration of 1 mg/mL active substance (Geneticin, 540.tg/mgactive wt., Gibco, N.Y.). Media was changed every 3 days, and large macroscopic coloniesisolated 2-3 weeks later. Macroscopic G418 resistant colonies were isolated using cloningcylinders, expanded and individually titred (see below) for viral production on NIH3T3 cells(ATCC, Rockville, MD). Titres of selected producer clones ranged from 1 x 106 to 1 xviruses! mL of supematant.Production of recombinant ecotropic retroviruses JZen-neo and JZenmIL-7tkneo usingthe GP+E-86 cell line was performed as described above with the following modifications.GP+AM-12 was used as the amphotropic cell line for CaPO4 transfection, and individualG418 resistant clones isolated and titred. A GP÷AM-12 clone producing a high viral titre (>lx i06 cfu/mL)was selected. Supematant from this cell line was then used to infect GP+E-86cells. Multiple infections of GP+E-86 were performed using 24 hour conditioned media fromsubconfluent GP+AM-12, which was filtered (0.22p.m filter, Millipore, Bedford MA) andoverlayed on GP+E-86 cells ( 1x105 cells!6Omm dish), in the presence of 4 p.g/mL polybrene(Sigma Chemical Co, St Louis, MO ), once every 24 hours for 7-10 days. G418 resistantGP+E-86 colonies were selected and titred. A high dire clone was selected (>lx i06 cfu/mL),and the multiple infection and selection process repeated using GP-i-AM- 12 supematants.Clones of> lx 106 cfu! mL producing GP+E-86 cells were routinely isolated.80D) Viral Titering and Helper Virus Assay.Viral titres were determined by assays of medium conditioned by viral producer celllines for transfer of the neo’ gene to 3T3 cells. The day before assaying media was changed onsub-confluent viral producer cell lines, and 3T3 cells were plated (2 x i05 cells /60 mm tissueculture dish). Supematants were harvested from producer cells 20 hours later, filtered(0.22.t,Millipore) and various dilutions placed in final 2 mL volumes and added to thepreviously established dishes of 3T3 cells in the presence of polybrene (4 pg/mL). Freshmedia was placed on the 3T3 cells 24 hours later, then G418 (1 mg/mL) added after another 24hours. G4 1 8r colonies were scored after staining with methylene blue to derive the number ofinfectious particles carrying neor generated by the viral producers (colony forming-units / mL).Helper virus assays were performed by attempts to serially passage neor carryingretroviruses on 3T3 cells as described by R. Mann (11). Confluent dishes of G418 resistant3T3 cells were obtained after retroviral infection with 4 mLs of filtered (0.22i.t, Millipore)supernatants from viral producer cell lines followed by selection in G418 as described above.Media was changed on subconfluent 60 mm dishes of G418 resistant 3T3 cells and replacedwith 4 mLs of fresh media and then 20 hours later supernatant (4 mLs) was collected, filteredand then placed on pre-established dishes of normal 3T3 cells in the presence of polybrene.After 24 hours the media on the normal 3T3 cells was then changed and cells were selected inG418 and colonies scored 2-3 weeks later as described above. As a positive controlsubconfluent G418 resistant 3T3 cells were inoculated with of 1-10 .tL of supematant fromMoC12 (from A. Bernstein, Toronto), a cell line producing high levels of infectious competentMo-MuLV (13), and passaged for 1 week in the presence of polybrene prior to testing forhelper activity. Supematant from G418 resistant 3T3 cells treated with MoC12 supematantsroutinely produced> i06 cfu/ml.812) HEMOPOIETIC CELL CULTURE AND ASSAYS.A) Mice.Mice were 6 to 12 week old C3H/HeJ or B6C3F1 (C57BL/6J X C3H/HeJ Fl), eitherobtained directly from the Jackson Laboratories (Bar Harbor, ME) or bred and maintained inthe animal facility of the British Columbia Cancer Research Centre (BCCRC) from parentalbreeders obtained from Jackson Laboratories. Mice were maintained as prescribed by theCanadian Council for Animal Care and care monitored by the BCCRC and the University ofBritish Columbia.B) Retroviral Infection of Bone Marrow CellsSupematant infection protocol: Bone marrow cells were collected from adult maleB6C3F1 mice treated 4 days earlier by i.v. injection of 5-fluorouracil (5-FU, 150mg/kg bodyweight (Hoffman-LaRoche Ltd, Mississauga, Ont). Viral containing supernatants wereobtained by incubation of near confluent viral producer cells for 20 hours in CL mediacontaining 10% FCS and 5% CS. The media was harvested, filtered (O.2211m filter, Millipore,Bedford, MA) and then supplemented with an additional 10% (v/v) FCS, 4p.g/ml of polybrene(Sigma), 10% (v/v) agar stimulated human leukocyte conditioned medium (14) and 5% (v/v)pokeweed mitogen stimulated spleen cell conditioned media (15). Day 4 5-PU bone marrowcells were then resuspended in the viral conditioned media using 3-5 x i06 bone marrow cellsper 100mm petri dish containing 10 mL of filtered viral conditioned media. After 6 to 8 hoursat 37°C in 5% C02 and a humidified atmosphere, the media was replaced with an equal volume82of freshly prepared virus-containing medium, and the cultures incubated for a further 12 to 14hours. Dishes were scraped with a rubber policeman to remove adherent cells which werethen combined with non-adherent cells and washed in 2% FCS in PBS and subsequently usedfor initiation of long-term bone marrow cultures or in vivo reconstitution as described below.Co-cultivation infection protocol: Infections of bone marrow cells by co-cultivationwere carried out by addition of 3-5 x 106 day-4 5FU bone marrow cells from either C3H orB6C3F1 mice onto sub-confluent irradiated viral producer cells (15 Gy X-ray) in a 100mmtissue culture dish in 10 mL of a media containing 5% (v/v) Cs, 10% (v/v) FCS, 4p.g/nil ofpolybrene (Sigma), 10% (vlv) agar stimulated human leukocyte conditioned medium and 5%(v/v) pokeweed mitogen stimulated spleen cell conditioned media. After 6 to 8 hours the mediawas changed by pipetting off the non adherent cells, pelleting and resuspending in mediaconditioned as described above. After a further 12 to 14 hours of co-cultivation non-adherentbone marrow cells were removed by gentle pipetting and washed in 2% FCS in PBS fortransplantation or initiation of long-term cultures as described. In all cases viral producer cellswere maintained in selective media which was replaced with DMEM plus appropriate serum upto 2 to 7 days prior to use in cocultivation. Producer cells were irradiated (15 Gy X-ray)immediately prior to use.C) Long-Term Bone Marrow Culture.Male bone marrow cells were retrovirally infected with the TKneol9virus generatedfrom a i-2 producer cell line by the supematant infection protocol described above. After theinfection period harvested cells were washed in 2% FCS in PBS, and 3 x 106 cells resupendedin 4 mL of LTC media (Alpha media supplemented with 10% horse serum, 10% fetal calfserum, 10 M hydrocortisone succinate and i04 M f3-mercaptoethanol) , and added to83previously established 3 week old irradiated (l5Gy X-ray) female long-term culture adherentlayers. These adherent layers were obtained from Dexter long-term marrow cultures (16) setup in 25cm flasks with an initial inoculum of 3 x i07 normal female marrow cells in LTCmedia. Cultures were incubated at 33°C, 5% C02 in a humidified atmosphere and given halfmedia changes weekly.D) Clonal Analysis of Repopulating Cells Recovered From Long-Term Cultures.Long-term marrow cultures were analysed 4 weeks after initiation with retrovirallyinfected Day-4 5-FU bone marrow. Adherent layers were removed with a rubber policeman,washed, and passaged through a 21g needle. Lethally irradiated recipients (8gy, 250 kVp Xray ) were given either 1-2 x 106 cells directly by i.v. injection or received the proportionsdescribed for the competitive repopulation assay (see below), plus 2 x i0 syngeneic“compromised” female marrow cells that had been previously subjected to two cycles of serialmarrow transplantation and regeneration (17,18).For cultures which were used to assess recovery of repopulating cells in thenonadherent fraction over time, all of the medium and nonadherent cells were removed weeklyand replaced either with fresh medium alone, or with LTC medium containing 25u/mL ofrecombinant mouse Interleukin-3 (IL-3) (Biogen). The nonadherent cells removed after 3, 5, 6and 7 weeks of culture were then injected into irradiated recipients. Recipient mice transplantedwith cells from long-term cultures were sacrificed 5 weeks to 7 months post transplant andbone marrow, spleen and thymus tissues were used for DNA extraction and Southern blotanalysis. For some animals a more detailed analysis of fractionated tissues and subpopulationswas performed. For those animals, thymus and lymph nodes were processed directly forDNA. Bone marrow was divided into three portions. One portion was used directly for DNA84extraction, the other two fractions were used to obtain pure myeloid cell populations asdescribed in detail previously (19). The second portion of marrow was cultured in Alphamedia, 15% FCS, 20% WEHI-3 conditioned media (CM) as a source of IL-3 for 3 weeks.This portion had >90% mast cell morphology when stained with May-Grunwald-Geimsa. Thethird bone marrow portion was cultured for 48 hours in Alpha medium, 15% FCS, 10%WEHT-3 CM, and 35% L-cell CM as a source of M-CSF (20). The non-adherent cells werethen removed and replated in 35% LCM for 7-10 days to give a highly purified source ofmacrophages as determined by morphology and indirect immunoperoxidase staining with anantibody specific to macrophages (Mac-i).To obtain pure B and T lymphocytes for DNA, spleen cells were fractionated by adding1/2 of the spleen in 1 mL DMEM, 10% FCS to a 3-mL nylon wool column, which wasincubated for 1 hr at 37°C prior to elution by extensive washing to obtain the non-adherentfraction. This nylon wool non-adherent fraction was determined to be > 90% T-cells asdescribed previously (21), when screened by FACS after staining with anti-Thy 1 antibody.Nylon-wool adherent cells were removed by gentle agitation for 2-3 mm in PBS containing10mM EDTA. B lymphocytes were isolated from this fraction by panning these cells for 1 hrat 37C in 100-mm-diameter plastic dishes (<108 cells per dish) precoated with unpurifiedrabbit anti-mouse immunoglobulin (22). After washing away non-specifically bound cells, theadherent B lymphocytes were removed and frozen for DNA extraction or used for FACSscreening. Staining with an antibody specific for B lymphocytes (B220 ) indicated thispopulation was >90% pure. The remaining 1/2 of the spleen was divided into 3 portions. Onewas used directly for DNA extraction, and the other two were used to provide pure mast celland macrophage populations as described for the bone marrow.85E) Limiting Dilution Analysis of Competitively Repopulating Cells.Total non-adherent and adherent cells were recovered from individual 4 week oldmarrow cultures established with Day-4 5-FU retrovirally infected male marrow cells. Groupsof lethally irradiated (800-850 cGy of total body X-irradiation; 124 cGy/min) female mice (4-8mice per group) were then injected i.v. with various proportions of a single culture cells (1/45,1/150, 1/450, 1/1500) together with 2 x compromised marrow cells from female mice.Compromised cells had been subjected to two previous rounds of transplantation andregeneration in female mice (23). This was performed by transplanting 106 female bonemarrow cells into a group of primary lethally irradiated (800 to 850 cGy) female recipients.This group of mice were used to donate 106 cells to lethally irradiated secondary femalerecipients 5-8 weeks later. Bone marrow from these secondary mice was used as thecompromised female marrow source within 3 months post transplant. Compromised cellsalone (10) were able to allow irradiated recipients to survive long-term, but doses of up to i06could not outcompete the repopulating ability of day-4 5-FU marrow (23). Recipients ofmixed donor long-term culture derived cells and female compromised cells were analysed at 5weeks and 7 months post transplant for male reconstitution (see below). DNA was preparedfrom bone marrow, spleen and thymus, and the presence of male contribution detected bySouthern blot analysis with a fragment from plasmid pY2 as described below in the Materialsand Methods. Male contribution to total DNA content was detectable as low as 1%, and wasreproducibly detectable at greater than 5%. Tissue was scored as positive in the assay if acontribution of male cells was found to be greater than 5%. Competitive repopulating units(CRU) frequencies were calculated from the proportion of negative recipients by limitingdilution analysis procedures (based on Poisson statistics) (24).86F) CFU-S Assays.Spleen colonies were generated immediately after retroviral infection of bone marrowcells, or after 1-2 weeks in LTC (collected during half weekly media change) as a method toassess gene transfer efficiencies to early hemopoietic precursors. For the generation ofdistinguishable spleen colonies, female mice received 800 to 850 cGy total body irradiation,followed by injection i.v. with 1-5 x i0 retrovirally infected ( supematant or co-cultivation,see above) or cultured bone marrow cells per mouse. This dose of irradiation was sufficient tocompletely eliminate endogenous spleen colony formation (to< 0.1/spleen). Recipient micewere sacrificed 12 days later and individual macroscopic spleen colonies dissected andhomogenised for DNA extaction and Southern blot analysis as described below.G) Methylcellulose Assays.Bone marrow cells (3 x 104)were plated in 35 mm petri dishes (Greiner, Germany) in1.1 mL culture mixtures consisting of 0.8% methylcellulose in alpha medium supplementedwith 30% FCS, 1% bovine serum albumin (BSA), i0 M j3-mercaptoethanol, 3 U/mLpartially purified human urinary erythropoietin (Terry Fox Laboratories), 2% pokeweedmitogen-stimulated mouse spleen cell conditioned medium (PWM-SCCM) and 10% agarstimulated human leukocyte conditioned medium (LCM). PWM-SCCM and LCM wereobtained from media preparation service of the Terry Fox Laboratory, Vancouver. Cultureswere incubated at 37°C in a humidified atmosphere of 5% C02, and colonies scored after 12-14 days. Colonies were scored based on morphological appearance as described previously(25).873) MOLECULAR ANALYSIS.A) Southern Blot Analysis.High molecular weight DNA was isolated from various tissues by sodium dodecylsulfate (SDS)/proteinase K digestion followed by phenol/chloroform extractions (26). Afterdialysis against 1X Tris-EDTA buffer (3mM Tris, 0.2 mM EDTA, pH 7.5, TE), 10 pgsamples of DNA were digested with Flindifi or EcoRI at 15U/p.g for 4-12 hours at 37°C in thebuffer recommended by the manufacturer (Bethesda Research Laboratories (BRL),Gaithersburg, MD). Samples of male and female DNA from normal B6C3F1 mice, and viralproducer cell lines were used as positive and negative controls. After ethanol precipitation,DNA was dissolved in 15 .tL of TE buffer, and electrophoresed through a 0.8% agarose gel.Gels were then treated with 01 N HC1 for 15 minutes followed by two 35 minute treatments ofSolution 1 (0.5M NaOH, 1.5M NaC1) and Solution 2 (1M Tris pH 7.0, 2M NaCl)respectively. DNA was then transferred to nylon membranes (Zeta-Probe, Bio-RadLaboratories, Richmond, CA) in 20 X SSC by standard blotting methods. Membranes wereprehybridized for 4 hours at 60°C in 40 mL of a solution containing 0.9M NaC1, 10%formamide, 1% SDS, 2mM EDTA, 1% nonfat dried milk (Carnation Milk) and 0.5 mg/mLdenatured salmon sperm DNA. Hybridization conditions were the same except for the additionof 10% dextran sulfate (Sigma Chemical Co., St. Louis). Membranes were hybridized with adenatured EcoRT/BaniHI fragment from pMClneo containing only neo specific sequences thatwas 32P labelled oligonucleotide labelled using the multiprime labelling method (27). Usingthis procedure, the specific activity of labelled products was routinely between 108 and i09dpni/jig of DNA. After hybridization for 18 to 20 hours, membranes were washed at a finalstringency of 0.1% SDS, O.3XSSC (20 X SSC is NaC1 3M, Na Citrate 0.3M, pH 7), and880.1% sodium pyrophosphate at 60°C twice for 30 minutes. Autoradiography was performedwith Kodak XAR-5 film and an intensifying screen at -70°C for either 1,3, or 14 days.Membranes were stripped for re-probing by boiling in 1% SDS, and washing for 40 minutes,then reprobed with either a 833 base pair SacJJSspI GM-CSF cDNA fragment from plasmidpGM3.2 for an internal control for DNA loading (28), T-cell receptor 1 constant region probefrom plasmid 86T5 (29) ) in order to determine T-cell receptor rearrangement in some tissues,or a 720 base pair MboI Y-specific probe from plasmid pY2 (30).B) Northern Blot Analysis.Total cellular RNA was isolated by lysing tissues in a guanidine isothiocyanate (GIT)solution (4 M GIT, 25mM sodium acetate, pH 6), and cenirifugation through a CsCl gradient.5.7 M CsC1, 25 mM sodium acetate, pH 6) (7). RNA pellets were resuspended in 0.3 M Naacetate pH 6, ethanol precipitated, dried and resuspended in 1 X TE (10mM Tris, 1 mMEDTA, pH 7.0), and stored at -70°C. RNA (5-10 p.g) was then separated by electrophoresisthrough a 1.2% agarose gel containing 5% formaldehyde and 1X MOPS buffer ( 0.36 MNaMOPS, 10 mM EDTA, pH 7.0), and transferred directly to Zetaprobe in lox SSC.Membranes were hybridized to a 524 base pair BamHJ/XmnI JL-7 cDNA (8) fragment as aprobe in a hybridization solution containing 40% foñnarnide, 1% SDS, 0.1M sodiumphosphate, 0.2 M EDTA pH 7.2 and 1 mg/mL BSA at 42°C for 16 hours. Blots were washedin a final stringency of 0.3 x SSPE (20X SSPE is NaC1 3M, Na Phosphate 0.2M, EDTA20mM, pH 7.4), 1% SDS at 55°C for 40 minutes.894) IMMUNOLOGICAL ANALYSES.A) Antibodies.Lyt-2, the antibody against CD8 (31) was purchased from Becton Dickinson(Moutainview, Ca), as a fluorescein (FITC) conjugate. L3T4, the antibody against CD4 (32)was purchased from Becton Dickinson as a phycoerythrin (PE) conjugate.A hybridoma producing MAb to heat stable antigen M1/69 (33) , and hybridoma clone145-2C1 1 that produces MAb to CD3(34) from J. Levy (University of British Columbia,Vancouver) were grown and antibody rich supernatant collected, purified on a protein Acolumn and F1TC (Sigma) conjugated as described elsewhere (35). HO-13-14 hybridoma cellline (Ledbetter 1979), producing the antibody to Thyl.2 was purchased from ATCC, and theantibody purified and biotynlylated from supematants (35).R3A-3A1/6C (36) producing MAb to B220 was obtained from the American TypeCulture Collection (Rockville, MD) and coupled (as described below) to RG7/9. 1 producingMAb to rat Ig-kappa chain (37), which had been purified and FITC conjugated. Polyclonalantisera against mouse immunoglobulin (Ig), was made by immunizing rabbits with mouse 1g.Blood was collected and serum passed over a protein A column to purify rabbit anti-mouse Ig,which was then FITC conjugated (35).B) FACS Analysis and Cell Sorting.The binding of antibodies to the surface of various cell populations was determined byindirect fluorescence staining and analysis on a FACScan (Becton-Dickinson, Oxnard, CA).Tissues (spleen, thymus or lymph node) were teased apart and cells passaged through an 18G90needle in order to obtain a single cell suspension (5 X cells). Aliquots were resuspendedin 0.2 mL of supematants containing specific antibodies for 30 mm. at 0°C. If required,second antibodies were conjugated under the same conditions after repeated washes in Hanksbalanced salt solution, plus 5% FCS. Cells were washed and resuspended in Hanks with 2%FCS and 1 .tg/mL propidium iodide. Isolation of populations was performed by indirectfluorescence staining by binding of monoclonal antibodies to the cell surface then sorting on aFACStar equipped with dual laser (Becton-Dickinson, Oxnard, CA). Propidium iodide wasused in all cases to stain dead cells, which were gated out during FACS analysis and sorting.C) IL-7 Activity AssayIL-7 production by viral producer cell line JZenmlL7tkneo was determined by testingthe ability of filtered supernatant conditioned for 24 hours to stimulate proliferation of CD4/CD8 (DN) thymocytes to proliferate in culture. DN thymocytes were isolated by mixingthymocytes, after lysing red blood cells (0.83% NH4C1 and 0.1% NaFICO3, pH7; 3 mm.4°C), with a 1:1 mixture of anti-CD4 and anti-CD8 MAb for 45 mm on ice, followed bywashes (3X) in PBS, 5% FCS. Cells were then incubated at 4°C for 40 mm on dishes precoated with RG7/9. 1 MAb. After recovering non-adherent cells the procedure was repeated.Some cells were tested for purity by incubating again with anti CD4 and CD8, followed byFITC conjugated goat (Fab)2 anti-rat Ig (heavy and light chains), and analysed on a FACScanafter staining with propidium iodide (10ig/mL). The remainder of DN thymocytes were testedfor response to viral supematants from JZenmlL7tkneo, JZen-neo or media alone as negativecontrols or various concentrations of recombinant murine IL-7 for positive controls in thethymidline incorporation assay.91Cell suspensions (5 x i0/ well) were plated in triplicate cultures in round bottommicrotitre wells (Flow Laboratories, McLean, VA) in 0.1 mL of RPMI 1640 medium, 5%FCS, 50.tM 2-mercaptoethanol and antibiotics. Cultures were treated either by using microtitrewells pre-incubated with anti-CD3 antibody, including IL-7 (10 .tg/mL) in the media, or withmedia alone as a control. All cultures were incubated for 68 hours at 37°C, then pulsed withlp.Ci3H-methyl-TdR(3H-TdR) for an additional 6 hours. Radioactive counts weredetermined using a 1205 BetaPlate liquid scintillation c9unter (LKB, Wallac Finland).D) White Blood Cell CountsPeripheral blood was sampled from mice 4-16 weeks post bone marrow transplant bycollecting tail blood into heparinized capillary tubes. Blood was used either to generate smearson glass slides for morphological analysis or used to determine peripheral white blood cellcounts. Peripheral blood smears were air dried then stained in Geimsa stain for 15-20seconds, followed by washes in distilled water for 0.5- 1.0 mm. White blood cell counts weredetermined on a hemocytometer after lysing peripheral blood by mixing 1 volume of bloodwith 9 volumes of 3% acetic acid.92REFERENCES1. Vennstrom B, Kahn P, Adkins B, Enrietto P. Hayman MJ, Graf T, Luciw P.Transformation of mammalian fibroblasts and macrophages in vitro by a murineretrovirus encoding an avian v-myc oncogene. EMBO J 3:3223 (1984)2. Wagner EF, Vanek M, Vennstrom B. Transfer of genes into embryonal carcinoma cellsby retrovirus infection: efficient expression from an internal promoter. EMBO J 4:663(1985).3. Hughes PFD, Eaves CJ, Hogge DE, Humphries RK. High efficiency gene transfer tohuman hemopoietic cells maintained in long term marrow culture. Blood 74:1915(1989).4. Johnson GR, Gonda TJ, Metcalf D, Hariharan 1K, Cory S. A lethal myeloproliferativesyndrome in mice transplanted with bone marrow cells infected with a retrovirusexpressing granulocyte-macrophage colony stimulating factor. EMBO J 8:441 (1989).5. Cepko CL, Roberts BE, Mulligan RC.Construction and applicatons of. a highlytransmissible murine retrovirus shuttle vector. Cell 37:1053 (1984).6. Thomas KR, Capecchi MR. Site-directed mutagenesis by gene targeting in mouseembryo-derived stem cells. Cell 51:503 (1987)7. Davis LG, Dibner MD, Battey iF. Basic Methods in Molecular Biology. New York:Elsevier (1986).8. Namen AE, Lupton S, Hjerrild K, Wignall J, Mochizuki DY, Schmierer A, Mosley B,March CJ, Urdal D, Gillis 5, Cosman D, Goodwin RG. Stimulation of B-cellprogenitors by cloned murine interleukin-7. Nature 333:57 1 (1988).9. Miller AD, Trauber DR, Buttimore C. Factors involved in the production of helpervirus-free retrovirus vectors. Somatic Cell Mol Genet 12:175 (1986).10. Miller AD, Buttimore C. Redesign of retrovirus packaging cell lines to avoidrecombination leading to helper virus production. Mol Cell Biol 6:2895 (1986).11. Mann R, Mulligan RC, Baltimore D. Construction of a retrovirus packaging mutant andits use to produce helper-free defective retrovirus. Cell 33:153 (1983).12. Markowitz D, Goff 5, Bank A. A safe packagin line for gene transfer: separating viralgenes on two different plasmids. J Virol 62:1120 (1988).13. Gross L. Oncogenic viruses. 2nd ëd. New York: Permagon Press (1970).14. Gregory CJ, Eaves AC. Human marrow cells capable of erythropoietic differentiationin vitro. Definition of three erythroid colony responses. Blood 49:855 (1977).9315. Murthy SC, Eaves CJ, Krystal G. A simple 3-step purification procedure forinterleukin-3 involving absorption to fixed cells. Exp. Hematol. 17:997 (1989).16. Dexter TM, Allen TD, Lajtha LG. Conditions controlling the proliferation ofhemopoietic stem cells in vitro. J Cell Physiol 9 1:335 (1977).17. Szilvassy SJ, Lansdorp PM, Humphries RK, Eaves AC, Eaves CJ. Isolation in asingle step of a highly enriched murine hematopoietic stem cell population withcompetitive long-term repopulating ability. Blood 74:930 (1989).18. Harrison DE, Astle CM, Delaittre JA. Loss of proliferative capacity inimmunohemopoietic stem cells caused by serial transplantation rather than aging. 3 ExpMed 147:1526 (1978).19. Keller G, Snodgrass R. Life span of multipotential hematopoietic stem cells in vivo. JExp Med 171:1407 (1990).20. Stanley ER, Heard PM. Factors regulating macrophage production and growth.Purification of some properties of the colony stimulating factor from mediumconditioned by mouse L cells. J Biol Chem 252:4305 (1977).21. Dougherty GJ, Allen CA, Hogg NM. Applications of immunological techniques to thestudy of the tumor-host relationship. In: Weir DM. Handbook of ExperimentalImmunology. Applications of Immunulogical Methods in Biomedical Sciences. Oxford:Blackwell (1986).22. Chan P-Y, Takei F. Molecular cloning and characterization of a novel murine T-cellsurface antigen, YE1/48. 3 Immunol 1727:142(1989).23. Szilvassy SJ, Humphries RK, Lansdorp PM, Eaves AC, Eaves CJ. Quantitative assayfor totipotent reconstituting hematopoietic stem cells by a competitive repopulationstrategy. Proc Natl Acad Sci U S A 87:8736 (1990).24. Taswell C. Limiting dilution assays for the determination of immunoincompetent cellfrequencies. I. Data analysis. J Immunol 126:1614 (1981).25. Humphries RK, Eaves AC, Eaves CJ. Self-renewal of hemopoietic stem cells duringmixed colony formation in vitro. Proc Nat! Acad Sci USA 78:3629 (1981).26. Maniatis T, Fritsch P, Sambrook 3. Molecular Cloning: A Laboratory Manual. NewYork: Cold Spring Harbor (1982).27. Feinburg AP, Vogelstein B. A technique for radiolabelling DNA restrictionendonuclease fragments to high specific activity. Anal Biochem 132: 6-13, (1983)28. Gough NM, GoughJ, Metcalf J, Kelso D, Grail D, Nicola NA, Burgess AW, DunnAR. Molecular cloning of a cDNA encoding a murine hemopoietic growth regulator,granulocyte-macrophage colony stimulating factor. Nature 309:763 (1984).9429. Hedrick SM, Nielsen EA, Kavaler J, Cohen DI, Davis MM. Sequence relationshipsbetween putative T-cell receptor polypeptides and immunoglobulins. Nature 308:153(1984).30. Lamar EE, Palmer E. Evidence that the Y chromosome exists in two polymorphicforms in inbred strains. Cell 37:17 1 (1984).31. Ledbetter JA, Herzenberg LA. Xenogeneic monoclonal anytibodis to mouse lymphoiddifferentiation antigens. Immunol. Rev. 47:63. (1979).32. Dialynas DP, Wilde DB, Marrack P, Pierres A, Wall P, Havran G, Otten G, Pierres M,Fitch FW. Characterization of the AlPine antigenic determinant, designated L3Ta,recognized by monoclonal antibody GK 1.5. Immunol. Rev. 74:29 (1984).33. Springer T, Galfre G, Secher D, Milstein C. Monoclonal xenogeneic antibodies tomouse Ig allotypes H-2 and Ia antigens. Eur. 3. Immunol. 8:539 (1978).34. Leo 0, Foo M, Sachs DH, Samelson LE, Bluestone JA. Identification of a monoclonalantibody specific for a murine T3 polypeptide. Proc. Natl. Acad. Sci. U.S.A. 84:1374(1987).35. Hardy R. Purification and coupling of fluorescent proteins for use in flow cytometry.In: Weinn DM, Herzenberg LA, Blackwell C. Handbook of ExperimentalImmunology. Oxford: Blackwell Scientific Publications (1984).36. Coffman RL, Weissman IL. B220, a B cell specific member of the T-200 glycoproteinfamily. nature 289:681 (1981).37. Springer TG, Bhattacharva A, Cardoza if, Sanchez-Madrid F. Monoclonal antibodiesspecific for rat IgG1, IG2a and IG2b subclasses, and kappa chain monotypic andallotypic determinants. Hybridoma 1:257 (1982).95CHAPTER IIIEXPANSION IN VITRO OF RETROVIRALLY MARKED TOTIPOTENTHEMOPOIETIC STEM CELLS1) INTRODUCTION.Mature blood cell production depends on the continual activation of hemopoietic cellswith extensive self-renewal, proliferation and differentiation potential. The most primitive ofthese appear to be individually capable of maintaining normal numbers of a variety of lymphoidand myeloid cell types for many months as shown by clonal analysis techniques including Xlinked isoenzyme measurements (1-2), or genetic marking (3-11) in combination withembryonic or adult reconstitution strategies (12). The ability to expand primitive totipotenthemopoietic cells in vitro would provide a significant step towards the further analysis andmanipulation of stem cell behaviour and would also have important implications for genetherapy.In the presence of appropriate combinations of sera, media and other supplements,bone marrow cultures can be established that maintain hemopoiesis in vitro for many months(13). This is evidenced by the sustained production ofature myeloid elements and varioushemopoietic progenitors. In addition, such cultures have been shown to contain cells capableof reconstituting the lymphoid and myeloid elements of supralethally irradiated mice (14-16).The extent to which such repopulation arises from persistent totipotent hemopoietic stem cellsas opposed to cells with more restricted developmental potentialities is not known. In thepresent study retroviral marking was used to track the fate of individual repopulating96hemopoietic stem cells during and after 4 weeks maintenance in long-term marrow cultures.The results provided here are the first evidence that lympho-myeloid stem cells with long-termrepopulating potential can persist in vitro under these conditions. Moreover, in this report, it isnow shown that totipotent stem cells are stimulated to undergo self-renewal divisions in thesemarrow cultures thereby giving rise to daughter stem cells that can reconstitute both lymphoidand myeloid systems of multiple recipient animals.2) EXPERIMENTAL STRATEGY.A schematic representation for the protocol to detect maintenance of lympho-myeloidstem cells is outlined in Figure 7A and is described in detail in Materials and Methods. Bonemarrow cells were isolated from male B6C3F1 mice that had been injected with 5-FU 4 dayspreviously and the cells then exposed in vitro to supernatant from a cell line producing helper-free recombinant reirovirus carrying the neomycin resistance (ne&) gene under conditionspreviously shown to achieve efficient levels of gene transfer to lympho-myeloid repopulatingcells (17). Cells were washed and then some aliquots injected directly into irradiated syngeneicfemale recipients to generate spleen colonies. The remaining cells were cultured on irradiatedadherent cell layers of pre-established long-term marrow cultures of female origin. After twoor four weeks, cultured cells were harvested and then transplanted into lethally irradiated,syngeneic, female recipients. At 7 weeks post transplant bone marrow, spleen and thymuswere harvested for DNA and subsequent clonal analysis by Southern blotting. To determinewhether cells with in vivo lympho-myeloid repopulathig potential can also proliferate in long-term marrow cultures, cells from individual culture flasks were transplanted into several femalerecipients as outlined in Figure 7 B, and clonal analysis performed on the tissues of theserecipients 7 weeks post transplant.bone marrow cellsfrom 5-PU treated• miceInfect with retrovirus for 24 hrsCulture onpre-establjshedfeeders97I 2 weeks 4 weeksday 12 CPU-SDNAat 7 weeksharvest bone-marrow spleenthymus for DNAB. bone marrow cellsfrom 5-PU treatedmiceculture onpre-establishedfeedersat 7 weeksharvest bone-marrow spleenthymus for DNAFigure 7. Schematic representation of the protocols used to demonstrate maintenanceof lymphoid-myeloid stem cells (A.), and to demonstrate proliferation oflymphoid-myeloid stem cells (B.) in long-term marrow cultures.infect with retrovirus for 24 hrs4wee983) RESULTS.A) Maintenance of Lvmpho-Mveloid Stem Cells.After 2 or 4 weeks cultured cells were harvested and transplanted into lethally irradiatedrecipients to detect maintenance of retrovirally marked CFU-S. 82% of initial marrow CFU-Sin these experiments showed integration of the neor gene (18 of 22 12-day-old spleen colonieswere neor positive). A high frequency of neo’ - positive CFU-S (44%, 14 of 32) was alsodetected in the cultures after two to four weeks. Analysis of total DNA from the cells in 4week old cultures using a restriction enzyme that cuts once in the proviral genome (to allowdetection in Southern blots of unique integration fragments) revealed a complex pattern ofintegration events in individual cultures consistent with the presence of multiple active clones ineach flask (Figure 8). Southern analysis of DNA from the regenerated marrow, spleen andthymus of recipients of these same cultured cells (one recipient per culture) assessed 6 to 7weeks after injection of the cells typically showed more than 50% transplant-derived (male)cells in all three tissues. The presence of a single, uniquely marked clone in all three tissueswas also documented (14 of 23 recipients showed 5% and up to 40% of marked tissue DNAassuming the presence of one copy of the neor gene, established by comparison to the intensityof the signal obtained when the same blots were rehybridized with a murine GM-CSF probe(18)). Figure 8 shows the results of a representative experiment in which three of five miceshowed this pattern. Marrow from one of these mice (mouse 3) was further transplanted intosecondary irradiated female recipients and day 12 spleen colonies generated. All secondaryspleen colonies were found to be both male and marked by the same integration fragment seenin the marrow of the primary recipient (Figure 8), thus confirming that the marked cells had99Ml M2 M3Clonal analysis by Southern blot of hematopoietic tissues from three mice (Ml,M2, M3) 45 days after reconstitution with cells from the adherent or non-adherent fractions of separate 4 week old long-term marrow cultures. Lanes Naand Ad, DNA from non-adherent or adherent fractions of the long-term cultureused for donor cells to Ml; lanes b, s and t, DNA from bone marrow, spleen orthymus of mice Ml, M2 or M3; secondary spleen colonies 1-4 derived fromCFU-S assay of marrow of mouse 3; male and female control lanes fromnormal mouse spleens. Top panel, Hindill digested DNA hybridized to a neospecific probe; bottom panel, the same blot rehybridized to a Y-specific probeshowing a 3.1 kb male specific band.________2° spl.col.NaAdb s tc( ‘b s I b s t 1 2 3 423 —9.4 —6.6 —kb. neoprobe. • probeFigure 8.100been derived from a single male cell in the original population used to initiate the cultures. Theremaining two mice in this experiment did not contain detectable levels of marked cells amongstthe male population in either the marrow or thymus, although in one, marked spleen cells werefound.B) Proliferation of totipotent stem cells.In order to determine if repopulating stem cells had undergone proliferation in long termcultures, individual cultures were transplanted into multiple lethally irradiated recipients.Multiple recipients of cells from a single flask showed repopulation of both the marrow andthymus by the same retrovirally marked clone in three of four such experiments (Figure 9).The results illustrated by Flask A are particularly striking. In all five primaryrecipients, most of the regenerated hemopoietic cells were male and all showed a prominentretrovirally marked clone in both lymphoid and myeloid lineages. In three of these mice(mouse 1, mouse 4 and mouse 5) the same unique 6.4 kb Hind III proviral fragment wasdetected. Reanalysis of the same DNA after digestion with a different restriction enzyme(EcoR 1) that, like Hind ifi, cuts once in the retroviral genome but at a different site, yielded aunique 15.5 kb integration fragment, again common to all 9 tissue samples (data not shown).Reiroviral proviruses were found to be intact in these cells and in all subsequent recipients.This was determined by Southern blot analysis using restriction enzymes that release the entireprovirus (Kpn I) or the neoR insert (Bam Hi) (data not shown). A second fragment of 8.7 kbwas seen in both mouse 1 (in marrow) and mouse 4 (in spleen). This second shared fragmentmust also have been due to the proliferation first in vitro and then in vivo of another markedstem cell as shown by analysis of three secondary spleen colonies generated from the marrowof mouse 1. Two of these spleen colonies (“a” and “c”, Figure 9) were found to be retrovirallyMl M2 M3 M4 MSbsthIbsI1bSlhIb5tIIk 1bstId”.V2° spleen coloniesMl M2 M31 b c’I”l’a b c’Ml M2 M3 M4____ ____ ________mkQ ClflIb S fl $ tub s tiFigure 9.. Clonal analysis of multiple mice transplanted 45 days previously with either theadherent layer or non-adherent fraction of single long-term cultures A,B and C.Recipients received either 2 x 106 cells from the adherent fraction or 5 x 106cells from the non-adherent fraction. Bone marrow (b), spleen (s), and thymus(t) DNA were assessed for proviral integration sites by digestion with Hindifi(or EcoR 1 where indicated) and hybridization to a neo specific probe; Blotswere re-probed with a Y-specific fragment for assessment of donor origin.Individual secondary spleen colonies (a-c) derived from CFU-S assays ofmarrow from mice Ml, M2, or M3 reconstituted by Flask A were similarlyanalyzed. Mk is a control lane derived from Hindifi digested DNA of the viralproducer cell line tkneol9psi-2 showing multiple proviral integration sites.101Flask A23 —9.4—6.6—4.4—kb— neoI23—6.6—4.4—kbneo— VFlask B Flask CMl M2—fl -flMl M2-i1 i1-.—— Vneo23-2.9.4—EcoRlHind Ill102marked although the fragments characteristic of each were different. Spleen colony “a”contained the 8.7 kb fragment, whereas spleen colony “c” contained the 6.4 kb fragment. Allthree spleen colonies were male. Similarly, for the other two primary recipients of Flask Acells (i.e. mouse 2 and mouse 3), the clones marked by 5.4 kb and 5.8 kb fragments,respectively, were able to generate secondary spleen colonies (Figure 9, top right panel).Three of five long- term recipients of marrow cells frorn mice 1 through 5, (Flask A), werekept for 50 days after transplantation prior to sacrifice for tissue analysis (Figure 10, mouse2*, 3* and 4*). Two of these showed regeneration of the original donor-derived clones in thetissues of secondary recipients. These results demonstrate the ability of retrovirally markedlympho-myeloid stem cells to maintain their very extensive proliferative and differentiationpotentialities even after maintenance in culture for up to four weeks. Furthermore, in the caseof Flask A, the presence of a common clone in multiple lineages of several recipients indicatesthat lympho-myeloid stem cells can undergo self-renewal in culture prior to transplantation.Similar findings were obtained from analyses of recipients reconstituted with cells fromFlask B. In this case three of four recipients (mouse 2, mouse 3 and mouse 4) showed acommon 6.2 kb Hind III fragment in one or more tissues (Figure 9, Flask B). Digestion withEcoR 1 confirmed the presence of a common 6.0 kb band (data not shown). Interestingly, inthese recipients substantial variation in the distribution of the neor positive clones in differenttissues was seen. In mouse 2, bone marrow and spleen were marginally repopulated with thisclone, whereas the thymus was strongly marked. In mouse 4, the neor signal in the spleenwas intense, but in the marrow and thymus it was very weak. In mouse 3 a faint neor signalwas observed only in the spleen. In contrast, the proportion of male cells in all tissuesanalyzed from each of these recipients was similar, indicating a significant contribution ofother, unmarked clones to at least some lineages in many instances (e.g. in the marrow andthymus of mouse 4).103A2* A3* A4*Clonal analysis of long-term secondary reconstituted mice. Irradiatedsecondary mice received 5 x i05 to 1 x 106 bone marrow cells from primaryreconstituted recipients (at day 45), and were sacrificed 50 days post-transplantfor tissue analysis. A2*, A3* and A4* are secondary recipients of bonemarrow from mouse 2, 3, and 4 respectively from long-term culture A (Fig 2).Southern analysis was carried out as in Fig 8 and Fig 9 legends. Thymic cellsfrom all secondary recipients were determined to be >90% T-cells whenscreened by FACS after staining with an anti-CD3 antibody, whereas an antibody specific to macrophages (Mac-i) failed to detect positive cells.Abbreviations: b, bone marrow; s, spleen; t, thymus; In, lymphnode; C,Hindlil digested DNA from tkneo 1 9psi-2 producer cell line.23s t1 ‘b S t In’’b s C9.46.64.4Figure 10.104Recipients of cells from a third culture (Flask C, Figure 9) revealed a more complexpattern of hemopoietic reconstitution by retrovirally marked cells. In two recipients, the sametwo clones appeared to be present. Consistent intensities and co-segregation of restrictionfragments suggested that this was due to the presence of two clones marked by multipleintegration events rather than several independent clones (Figure 9). One of these clonesmarked by multiple Hind Ill fragments and, likewise, multiple EcoR I fragments, (Flask C,bands indicated by asterisks) appeared to have been derived originally from a totipotential cellsince it gave rise to both marrow and thymus cells in mouse 1, although it appeared to havecontributed only to the marrow of mouse 2 at the time of sacrifice. The other clone originatingfrom a cell in Flask C that was also multiply marked (Flask C, bands indicated by dots), wasfound in both the marrow and the thymus of mouse 1 but appeared restricted to the thymus ofmouse 2. These findings illustrate the lineage or tissue restriction of clones that is frequentlyobserved when recipients are analyzed at a single time point (7-1 1). This apparent restrictionmay simply reflect the detection limit of small subpopulations using Southern analysis or thedifferent turnover kinetics of mature lymphoid and myeloid cell types in vivo. Alternatively, itmay reflect the generation in culture of stem cells that have retained extensive repopulatingability but that have become developmentally restricted (19).4) DISCUSSION.This study demonstrates that conventional retroviral marking techniques can readilydetect the persistence of totipotent lympho-myeloid stem cells in 4 week old long-term marrowcultures. Furthermore, the strategy of transplanting multiple recipients with the contents of asingle flask has made it possible to obtain evidence of lympho-myeloid stem cell proliferationin this culture system. In most instances, clonal regeneration of both lymphoid and myeloid105tissues was seen in mice sacrificed 6-7 weeks after transplantation of cultured cells. Moreoverin several instances, both short-term (spleen colony formation) and long-term (up to 7 weeks)regeneration of hemopoiesis by retrovirally marked cells in secondary recipients wasdemonstrable. At least some lympho-myeloid cells harvested from long-term marrow culturesmust, therefore, have retained a very extensive potential for self-maintenance, sufficient toallow them to sustain hemopoiesis at a significant level for more than three months followingtransplantation in vivo. The only alternative interpretation, i.e. of infection of initiallytotipotent stem cells with subsequent in vitro expansion of both myeloid and restricted stem cellprogeny seems unlikely given the high frequency with which individual animals containedclonal populations in both lymphoid and myeloid tissues. Additional experiments includinguse of competitive repopulation assays (17) are addressed in Chaper IV, demonstrating that theamplified totipotent cells indentified here represent the most primitive of lympho-myeloidreconstituting stem cells.The ability to detect lympho-myeloid stem cell proliferation in culture makes possiblethe further investigation of the factors to which these primitive cells respond, as well as thenature of this response. Experiments designed to assess the potential involvement of one suchgrowth factor (Interleukin-7) in the early stages of hemopoietic commitment are outlined inChapter V. Previous analyses of pluripotent hemopoietic cell commitment during colonyformation either in vitro or in vivo have provided data consistent with a stochastic model ofstem cell renewal and differentiation (20,21). Analogous studies of the progeny of individualpluripotent cells generated under conditions of long-term marrow culture have not beendescribed, although it has been suggested that such conditions may favour the accumulation oflymphoid- restricted stem cells (22). The present studies thus serve as a starting point fordelineating the earliest stages of hemopoietic cell development. They also provide impetus forthe utilization of long-term marrow cultures for expansion of transplantable human hemopoietic106stem cells in vitro, in particular for therapeutic applications requiring the biologic or geneticmanipulation of hemopoietic stem cells in vitro.107REFERENCES1. Fialkow PJ. Cell lineages in hematopoietic neoplasia studied with glucose-6-phosphatedehydrogenase cell markers. J Cell Physiol Suppl 1:37 (1982).2. Nakano T, Waki N, Asai H, Kitamura Y. Lymphoid differentiation of thehematopoietic stem cell that reconstitutes total erythropoiesis of a genetically anemicW/Wv mouse. Blood 73:1175 (1989).3. Wu AM, Till JE, Siminovitch L, McCulloch EA. Cytological evidence for arelationship between normal hematopoietic colony-forming cells and cells of thelymphoid system. J Exp Med 127:455 (1968).4. Edwards GE, Miller RG, Phillips RA. Differentiation of rosette-forming cells frommyeloid stem cells. J Immunol 105:7 19 (1970).5. Abramson S, Miller RG, Phillips RA. The identification in adult bone marrow ofpluripotent and restricted stem cells of the myeloid and lymphoid systems. 3 Exp Med145:1567 (1977).6. Dick JE, Magli MC, Huszar D, Phillips RA, Bernstein A. Introduction of a selectablegene into primitive stem cells capable of long-term reconstitution of the hematopoieticsystem of W/Wv mice. Cell 42:71 (1985).7. Keller G, Paige C, Gilboa E, Wagner EF. Expression of a foreign gene in myeloid andlymphoid cells derived from multipotent haematopoietic precursors. Nature 318:149(1985).8. Lemischka IR, Raulet DII, Mulligan RC. Developmental potential and dynamicbehaviour of hematopoietic stem cells. Cell 45:9 17 (1986).9. Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of micereconstituted with retrovirus infected stem cells. EMBO 6:3955 (1987).10. Capel B, Hawley R, Covarrubias L, Hawley T, Mintz B. Clonal contributions of smallnumbers of retrovirally marked hematopoietic stem cells engrafted in unirradiatedneonatal W/Wv mice. Proc Natl Acad Sci USA 86:4564 (1989).11. Jordan CT, Lemischka IR. Clonal and systemic analysis of long-term hematopoiesis inthe mouse. Genes & Dev 4:220 (1990).12. Mintz B, Anthony K, Litwin S. Monoclonal derivation of mouse myeloid andlymphoid lineages from totipotent hematopoietic stem cells experimentally engrafted infetal hosts. Proc Natl Acad Sci USA 81.7835.(1984).13. Eaves C, Coulombel L, Eaves A. Analysis of hemopoiesis in long-term human marrowcultures, in Killman SV-AA, Cronkite EP, Muller-Berat CN (eds): Haemopoietic StemCells. Characterization, proliferation, regulation, Copenhagen, Munskgaard, p 287(1983).10814. Phillips RA, Jones EV, Miller RG. Differentiative potential of hematopoietic stem cells,in Golde GW, Cline MJ, Metcalf D, Fox CF (eds): Hematopoietic Cell Differentiation,New York, Academic Press, p 63 (1978).15. Schrader JW, Schrader S. In vitro studies on lymphocyte differentiation. I. Long-termin vitro culture of cells giving rise to functional lymphocytes in irradiated mice. J ExpMed 148:823 (1978).16. Harrison DE, Lemer CP, Spooncer E: Erythropoietic repopulating ability of stem cellsfrom long-term marrow culture. Blood 69: 1021 (1987).17. Szilvassy SJ, Fraser CC, Eaves CJ, Lansdorp PM, Eaves AC, Humphries RK.Retrovirus-mediated gene transfer to purified hemopoietic stem cells with long-termlympho-myelopoietic repopulating ability. Proc Nail Acad Sci USA 86:8798 (1989).18. Gough NM, Gough J, Metcalf D, Kelso A, Grail D, Nicola NA, Burgess AW, DunnAR. Molecular cloning of cDNA encoding a murine haematopoietic growth regulator,granulocyte-macrophage colony stimulating factor. Nature 309:763 (1984).19. Dorshldnd K. In vitro differentiation of B lymphocytes from primitive hematopoieticprecursors present in long-term bone marrow cultures. J Immunol 136:422 (1987).20. Till JE, McCulloch EA, Siminovitch L: A stochastic model of stem cell proliferationbased on the growth of spleen colony-forming cells. Proc Nail Acad Sci USA 51:29(1964).21. Humphries RK, Eaves AC, Eaves CJ: Self-renewal of hematopoietic stem cells duringmixed colony formation in vitro. Proc Nail Acad Sci USA 78:3629 (1981).22. Fulop, GM, Phillips RA: Use of scid mice to identify and quantitate lymphoidrestricted stem cells in long-term bone marrow cultures. Blood 74:1537 (1989).109CHAPTER IVPROLIFERATION OF TOTIPOTENT HEMOPOIETIC STEM CELLS INVITRO WITH RETENTION OF LONG-TERM COMPETITIVE IN VIVORECONSTITUTING ABILITY.1) INTRODUCTION.The regenerative capacity and life-long maintenance of the hematopoietic system isdependent on a primitive subpopulation of stem cells with extensive self-renewal, proliferativeand differentiation potential. Totipotent hematopoietic stem cells with the capacity to clonallyrepopulate both lymphoid and myeloid tissues in myeloablated recipients have beendocumented in both mouse and man by a number of experimental strategies.. These include theuse of radiation-induced chromosomal markers (1), naturally occurring electrophoretic variants(2) and markers introduced by retroviral insertion (3-9). While such cells may represent themost primitive type in adult marrow, considerable evidence points to a hierarchy oftransplantable hematopoietic cells with differing potential for subsequently sustaininghematopoiesis in vivo (10-15). Cells with long-term hematopoietic reconstituting ability can bedistinguished from cells that generate mature progeny in short-term in vivo or in vitroclonogenic assays by a number of physical and biological properties (16-21). Time coursestudies of the pattern of hemopoietic reconstitution in recipients of retrovirally- marked cellshave revealed examples of long-term reconstitution by small numbers of totipotent stem cells(22,23) under conditions where multiple, fluctuating short-lived clones are seen for up to 4months after transplantation. These results are consistent with a hierarchical structure in thestem cell populations transplanted but also do not exclude the possibility that environmental or110stochastic mechanisms may account in part or even wholly for the observed heterogeneity inbehaviour of potentially distinct hematopoietic cell types.Methods for maintaining, expanding and following the fate of primitive hemopoieticcells in vitro represent key requirements for the investigation of these questions. In ChapterIII, it was demonstrated that the system for the long-term culture (LTC) of mouse bone marrowintroduced by Dexter et al (24) maintains cells capable of lymphoid and myeloid reconstitutionin irradiated recipients (see Chapter 3). In contrast to this, a diminution in CFU-S self-renewal capacity (25) and competitive erythroid repopulating ability (26) of LTC-derived cellscompared to fresh marrow cells has also been reported. The present study combines the use ofa rigorous limiting dilution assay for competitive repopulating units (CRU) with retroviralmarking of the initial cell population and a long term post-transplant assessment to determine iftrue long-term repopulating cells survive and proliferate for extended times in vitro.2) EXPERIMENTAL STRATEGY.Initial experiments were aimed at analysing developmental and long-term repopulatingpotential of retrovirally marked stem cells maintained in long-term cultures. To do this lethallyirradiated mice were reconstituted with cells from 4 week old long-term cultures that had beenestablished with retrovirally infected bone marrow as described in detail in the Materials andMethods. Multiple tissues and enriched populations ofeells from different hemopoieticlineages were then obtained 5 months post transplant for DNA and subsequent clonal analysisby Southern blotting.Next, quantitation of competitive repopulating units (CRU) as well as thedevelopmental potential of retrovirally marked CRU in individual cultures were determined bytransplanting retrovirally marked male long-term culturederived cells at limiting dilution into111inadiated female recipients together with 2 x i05 female cells that had been compromised bytwo previous cycles of transplantation as described in detail in Materials and Methods. Thefrequency of CRU was then determined using Poisson statistics by scoring assay recipientsnegative for reconstitution with long-term culture derived male cells. Tissues from assayrecipients were also analysed for clonal contributions b’ retroviral integration markersobserved by Southern blotting.Finally, serial analysis was performed on long-term cultures for maintenance ofrepopulating stem cell clones by injecting portions of the whole non-adherent fraction intomultiple recipients from weeks 3-7 after culture initiation. DNA from tissues of recipients werethen analysed by Southern blot analysis 5 months post transplant for contributions of stem cellclones to hemopoiesis.3) RESULTS.A) In Vitro Recovery of Long Term Repopulating Cells.To determine whether totipotent cells capable of long-term repopulation could bedemonstrated in 4 week-old long-term cultures initiated with retroviraily infected 5-FIJ cells,aliquots were injected into irradiated recipients. These were then sacrificed 5 months later andassessed for the presence of common clones of culture-derived cells in various myeloid andlymphoid cell populations. Results for two representative recipients are shown in Figures 1 1Aand B. Figure 1 1A shows the presence of a dominant clone in both myeloid (unseparatedmarrow, marrow-derived mast cells and marrow-or spleen-derived macrophages), and1120—— —DI. = dD JDFigure 11. Presence of unique reiroviral insertion fragments in bone marrow (BM), spleen(Spi), separately isolated macrophage () and mast cell populations (mast),separated splenic T and B lymphocyte populations (Spi B, Spi T), thymus(Thy), and lymph node (LN) tissues of 2 mice (A and B) detected by Southernblot analysis of HindIll digested DNA. Membranes were hybridized to a neorspecific probe. Lethally irradiated mice received 1-2 x 106 cells fromretrovirally infected 4 week-old long-term cultures (i.e. approximately 1/10 of aculture initiated from 3 x 106 Day 45-PU marrow cells), and were analysed 5months post-transplant. Lane M is DNA from the viral producer cell lineTkneol9NJ-2. All lanes contain 5-10 .tg DNA.A.——113lymphoid cell populations (isolated splenic T and B cells, as well as thymus and lymph nodecells). Comparison of the intensity of the provirus band with that seen in a clonally derived3T3cell line generated after infection with the same virus indicates that the retrovirally marked stemcell clone in this mouse was contributing to at least 80% of total hematopoiesis at the time ofsacrifice. In the second mouse illustrated (Figure 1 1B) at least two marked clones were foundto be present in all hematopoietic cell populations exaitñned. Again, the intensities of thebands of the major clone (asterisks) indicates a dominaht contribution to reconstitution in each.B) Ouantitation of Competitive Repopulating Units (CRU’ in LTC.To quantitate changes in the total number of repopulating cells resulting frommaintenance of the initial cells under LTC conditions a recently described limiting dilutionassay for competitive repopulating cells (27) was employed as detailed in the Methods anddiagramed in Figure 12. Data for three experiments are plotted in Figure 13 and derived CRUvalues are summarized in Table II where they have also been compared with published valuesfor CRU frequencies in freshly isolated Day 4 5-FLJ cells also derived from analysis ofrecipients sacrificed 5 weeks post transplantation. It can be seen that the frequency of CRU in4 week-old LTC is the same whether the calculations are based on marrow or thymusrepopulation as shown previously for fresh or partially purified marrow cells (27). Whencombitied, an overall decrease in CRU frequency after 4 weeks in LTC of approximately 7-foldis apparent. In two of these experiments, CRU frequencies in the 4 week- old LTC were alsodetermined in additional recipients given aliquots of the same cells but sacrificed 7 months posttransplant. Because fewer recipients were available forevaluation at this later time, the derivedCRU frequencies are less precise. Nevertheless, the CRU values derived from 7 month tissuerepopulation data were again not significantly different whether marrow or thymus was114bone moiiow cells1/45 lb of flaskMIll. •I Quantilale CPU byLimiting Dilution Analysis4 weeks 5 weeks•l•or 7 monthsIndividual tiC1/150th of flask+ Assess CPU poliferafloncompromised cells by clond analysis 04 vIralIntegration1/450th of flask1/1500 lb of flaskFigure 12. Outline for the quantitation of CRU in single long-term marrow cultures.Retrovirally marked male LTC cells were injected at limiting dilutions intoirradiated recipients together with female compromised cells. The proportion ofanimals positive for male repopulation was determined 5 weeks and 7 monthspost transplant, and the clonal contributions from the LTC derived stem cellsdetermined by retroviral integration events.0a.0w0zC)0C0t0a.01115Bone Marrow Thymus10.37-I‘.‘j.\\.AT— II A— I— I0.10 10000 20000 30000Figure 13.I I I I0 10000 20000 30000Initial Cells per LTC Initial Cells per LTCProportion of mice (4-8 animals per group) negative for reconstitution (<5%) ofmarrow or thymus with male cells 5 weeks after transplantation with varyingproportions of a 4-week old long-term culture initiated with Day 4 5-FU cells,together with 2X105 compromised female marrow cells. The proportion ofreconstitution by male cells was detennined by Southern blot analysis ofmarrow and thymus cell DNA from each recipient using the Y-specific probe,pY2. Circles, squares and triangles represent results from analyses of threeseparate long term cultures. A straight line fit to the combined data based onmaximum likelihood analysis is shown by the solid line. The broken linerepresents previously published data for similar analyses of fresh Day 4 5-FUmarrow cells.116Table II. The Frequency of CRU in 4 Week Old Long-Term Bone Marrow Cultures.Hemopoletic Tissue Analyzedfor LTC Derived HemopoiesisTime of Analysis Bone Marrow Thymus Data Pooled for BoneMarrow and Thymus5 weeks 12,000 14.500 13,000post-transplantation (8,800-18.000) (8,900-23,600) (9,600- 17,800)7 months 25.000 2 1.000 23,400post-transplantation (14.000-45,400) (10,200-43.100) (14,800-37.000)Values shown are the reciprocal of the CRU frequency expressed relative to the numberof cells used to initiate the long-term culture 4 weeks previously. 95% confidence limitsdefined by 2SE are shown in parentheses. Cells from 3 long term cultures were analysed inrecipients sacrificed 5 weeks post transplant (20 to 28 mice per experiment); and for two ofthese cultures CRU measurements were also derived from assessment of recipients sacrificed 7months post transplant (16 to 22 mice per experiment).Reciprocal CRU frequencies in fresh Day 4 5-FU marrow have been previouslydetermined to be 2700 (1300-5700) and 1300 (540-3100) from bone marrow or thymusrepopulation analyzed separately, or 2,000 (1200-3500) when both read-outs are combined.117analyzed and were also not noticeably different from the values obtained from mice analyzedonly 5 weeks post transplant (Table II).C) Evidence of Totipotent Stem Cell Amplification In Vitro.Because all LTC in these studies were established with retrovirally infected Day-4 5-PU marrow cells, and the contents of individual cultures were then assayed in multiplerecipients, it was possible to analyze their tissues for the number and distribution of markedclones in different animals and in some cases, to relate the patterns observed to the number ofCRU injected. Table III summarizes the characteristics of all marked clones regenerated inrecipients of cells from 8 individual 4-week LTC that were individually assayed in multiplerecipients. Of a total of 46 uniquely identifiable clones, 9 were observed to contribute torepopulation of 2 or more mice and 8 of these included clones that repopulated both myeloidand lymphoid tissues (amplified clones I-DC, Table ifi). A further 37 clones were identified inonly one recipient but 14 of these showed contributions to both myeloid and lymphoid tissues(single clones, Table ill).Figure 14 shows four examples of a common clone contributing to the hematopoieticreconstitution of multiple recipients of cells from a single flask, indicating amplification oftotipotent stem cells in the original retrovirus infected input population. For simplicity it issuggested that such clones be termed as sibling stem cell clones because they are derived froma single parental precursor.Figure 14A shows an experiment in which reconstitution of lymphoid and/or myeloidtissues by sibling clones (clone VII in Table ifi) was detected in 2 CRU assay mice sacrificed 5weeks post transplant (mouse 1/150 A and mouse 1/150 B) as well as in 2 of 6 recipients ofthe same LTC (i.e. 1/20 of the culture) but without additional compromised cells and sacrificed118Table III. Tissue Distribution of Marked Clones in Recipients of LTC Cells.marrow spleen spleenb marrowCspleen thymus thymussingle 11 8 4 14clonesamplifiedclonesA 1 1B 1 1C 2D 1 1E 1 3F 1 1G 2 3H 3I 1Totals 17 13 6 27a. Apparently myelold-restricted repopulatlon.b. Apparently lymphoid-restricted repopulation.c. Myelold and lymphoid repopulation.d. Of 63 recipients identified with clonal markers, 20 were analysed 5 weeks posttransplant, and 43 were analysed 5-7 months post transplant.119S week 7 month1/150 I/ISO‘B S T’ ‘II S I’ NIFigure 14. Demonstration of unique retroviral insertion fragments in various hematopoietictissues of multiple mice sacrificed either 5 weeks or 7 months aftertransplantation of cells from single 4-week old long-term cultures initiated withretrovirally infected Day 4 5-FIJ marrow cells. Panels A,B,C and D showselected results from analysis of recipients of cells from 4 different long-termcultures. Irradiated recipients received cells either under standardtransplantation conditions (A, 7 months and D), or under CRU assay conditions(A, B and C). In the latter case 1/45, 1/150, or 1/450 of the original long-termculture was injected and this corresponded to the transplantation ofapproximately 6, 2 or 0.7 CRU per recipient. Bone marrow (B), spleen (S),and thymus (T) DNA were assessed for proviral integration sites by digestionwith Hindu and hybridization to a neo-specific probe. Blots were re-probedwith the Y-specific pY2 probe for assessment of donor origin. M is a controllane as described in Fig. 11.5 week 7 month1/150 A 1/150 B 1/20 A 1/20B‘B S T’ ‘B S T’ ‘B S T’ ‘B S T’A.—neoB.INLTV, V V--‘C. D.ST’B‘B S T’1/45 A 1/450 B‘B S T’ ‘B S T’. wneod I-1207 months post transplant (mouse 1/20 A and mouse 1/20 B). Mouse 1/150B which wascompetitively repopulated following injection of limiting numbers of CRU (2 per recipient)demonstrates repopulation of marrow, spleen and thymus by a common stem cell. In mouse1/150A and in mouse 1/20A cells with the same clonal marker were detectable only in themarrow and spleen. In mouse 1/20B, the sibling clone detected appeared restricted to thespleen and thymus.Further evidence of expansion of totipotent stem cells in LTC is shown by therecipients analyzed in Figure 14B. Here, analysis of tissues of 2 mice which had been injectedunder competitive conditions with approximately 2 CRU from the same flask revealed thepresence of sibling clones (clone V in Table Ill) in the marrow, spleen and thymus tissues atboth 5 weeks and 7 months post transplant. These results also extend the CRU quantitationstudies that suggest detection of the same type of totipotent cell at either 5 weeks or 7 monthspost transplant when CRU assay conditions are used.In recipients of cells from some LTC, analysis of retrovirãl insertion fragmentsrevealed marked sibling clones that showed a predominant or restricted distribution to eitherlymphoid or myeloid tissues. An example of 2 apparently lymphoid-predominating siblingclones is presented in Figure 14C (clone IV in Table ifi). Even though mouse A received 10times as much LTC cells as mouse B, in both mice the presence of marked and male cells isreadily apparent only in the spleen and the thymus. In a longer x-ray exposure this clone wasdetected at low levels in the marrow of mouse 1/45A. This suggests a greatly diminishedmyeloid repopulating potential in the cells harvested from this particular LTC, reconstitution ofthe recipients’ marrow thus being dependent on the activity of residual host cells. Figure 14Dsuggests the presence of apparently myeloid-reslricted sibling clones (clone Vifi in Table III) in2 mice (A and B) injected with LTC cells (under noncompetitive conditions) and sacrificed 7months post transplant. Both of these mice show a significant reconstitution of the marrow121and spleen with sibling stem cells characterized by multiple integration fragments. Mouse Bwas completely devoid of these markers in the thymus in spite of evidence of equalcontributions of male cells to all tissues indicating repopulation by other unmarked LTCderived cells. Similar analyses of the thymus of mouse A was not possible due to insufficientDNA.D) Serial Studies of Repopulating Stem Cell Clones During LTC.To determine whether the proliferation of repopulating stem cells occurs in the LTCsystem over extended periods of time, the nonadherent cells from some LTC were sampledserially and each time injected into multiple recipients. (In some cases, the medium in thesecultures was supplemented with 25 u/mi of recombinant IL-3, although this proved to have nosignificant effect on either total cell or progenitor numbers by comparison to parallel cultures towhich no IL-3 was added). Figure 15 shows the results from 2 such experiments in whicheach lethally inadiated female mouse received one tenth of the nonadherent cells from a singleLTC removed between 3 and 5 or 7 weeks after initiatii. All recipients were sacrificed 5months post transplant for tissue analysis. The same clonal pattern in at least one of thehematopoietic tissues examined was observed in 7 of the 11 mice reconstituted by cells fromone LTC (shown in Figure iSA), and by 3 of the 9 mice reconstituted by cells from the otherLTC (shown in Figure 15B). At early time points (weeks 3 and 5, Fig. 15A, and week 3,Fig. 15B), the sibling stem cells detected exhibited significant totipotent and long-termreconstituting capacity. At later time points, additionaPsibling stem cells defined by their long-term reconstituting capacity could still be detected but the tissue distribution of the progenydetected after 5 months was limited predominantly to the spleen, as indicated by assessment ofthe proportion of either male or retrovirally marked cells in marrow, spleen, and thymus.wk 7m 7.1 m 7.2 m 7.3 m 7.4bst bst bst bst114-. NIW NilFigure 15. Serial analysis of repopulating cells in the nonadherent fl-action of single long-term cultures (A and B) assessed after 3,5,6 and 7 weeks by injection intomultiple recipients. Each recipient was injected with 1 x 106 non-adherent cells(i.e. approximately 1/10 of a culture initiated with 3 x 106 Day 4 5-FU cells)and was sacrificed 5 months post transplant for analysis of unique retrovIra!insertion fragments and male DNA as described in Fig. 14.122wk 3m 3.1 m 3.2bst bstwk5 wk6m 5.1 m 6.1bst bstABSIwk 3m 3.1bstwk 5m 5.1 m 5.2bst bst“(-II1234) DISCUSSION.In this study both retroviral marking and quantitative assays for competitiverepopulating cells (CRU) were used to delineate the behaviour of totipotent hemopoietic stemcells under conditions of LTC. Analysis of mice injected with cells from 4-week old culturesrevealed the presence of marked progeny in reconstituted tissues for periods of at least 7months. A high proportion of all clones detected (50%, Table II), were represented in bothlymphoid and myeloid populations at the time of analysis. In some instances this occurredeven when limiting numbers of CRU were injected (<2 per recipient), thus providing directevidence of totipotent stem cell maintenance in LTC.The repeated demonstration of the same retroviral insertion fragments in hemopoietictissues of different mice injected with cells from the same LTC further showed that at leastsome of the totipotent cells in Day 4 5-FU marrow undtrgo clonal expansion in vitro withpreservation of both their long-term and competitive lymphoid and myeloid repopulatingability. The fact that such marked sibling clones could be detected in CRU assay recipients of<2 CRU assessed either 5 weeks or 7 months post transplant is consistent with previousevidence that CRU in Day 4 5-FU marrow can be readily infected with the supematantprocedure used and that most if not all CRU are totipotent cells capable of sustaininghemopoiesis for 7 months or more. This is further supported by the finding of the same CRUfrequency regardless of the time of recipient assessment from 5 weeks to 7 months posttransplant and regardless of whether the marrow or thymus was used to assess repopulatingpotential. These results are also a strong indication of the capacity of the 5 week CRU assay todetect stem cells with long-term repopulating potential. Unlike previous studies quantitatingstem cells in irradiated scid mice (28), enrichment in LTC of repopulating stem cells withlymphoid restricted differentiation potential was not found. Neither, however, was a124consistent decline in the number of repopulating stem cells active at 5 weeks and 7 months posttransplant observed as described previously (23). One explanation for both of thesedifferences is that the number and genotype of cells present during the initial period ofhematologic recovery may influence both the type of progeny produced and the rapidity withwhich they appear. Evidence in support of this possibility comes from the different rates ofappearance of +1+ cells in deficient lineages following their transplantation into W/W” mice(29). In addition, allophenic mice created from 2 differing genotypes can show a consistentlyunequal contribution of each genotype to hemopoiesis (‘9).Quantitation of CRU numbers after 4 weeks in LTC revealed a slow decline toapproximately 15% of the input value. Since a high proportion of those present after 4 weekscould be shown to represent clonal derivatives of initially marked CRU, the behaviour ofindividual CRU in LTC may be very heterogeneous, with some achieving extensiveamplification even in the face of concurrent mechanisms leading to a net loss of CRU,presumably due to their differentiation and/or death.Finally, evidence for the continuous turnover and self-renewal of totipotent stem cellsin LTC over a period of several weeks was demonstrated by analysis of mice injected withserially sampled nonadherent cells from two separate LTC. In both, daughter stem cells thatstill possessed totipotent long-term repopulating potential continued to be produced for up to 4weeks although thereafter, repopulation was only seen in the spleen.Together these results provide an important starting point for further delineation of themolecular mechanisms required to support the proliferation of totipotent hemopoietic stem cellsin vitro and of the effects of clonal expansion on the proliferative and developmental capacity ofindividual stem cells. They also provide the opportunity to explore potential growth factorcandidates that may influence developmental decisions in repopulating cells. The results inChapter V suggest that Interleukin-7 is one such candidate. These experiments will also be125aided by the recent development of methods for attaining even higher levels of gene transferefficiency to repopulating cells (30).The present studies also have implications for clinical bone marrow transplantation.Persistence of very primitive cells with long-term in vitro repopulating ability in LTC ofhuman marrow has been shown to be analogous to the kinetics of CRU maintenance describedhere (31). It is therefore not unreasonable to assume the operation in human LTC of simiiarlycompeting mechanisms of stem cell proliferation and decline. It should therefore also bepossible by appropriate manipulation to optimize conditions that favour expansion oftransplantable human hemopoietic stem cells in vitro, particularly where selection of clonesover prolonged periods may be therapeutically advantageous; for example, for certain genetherapy protocols.126REFERENCES1. Abramson S, Miller RG, Phillips RA. The identification in adult bone marrow ofpluripotent and restricted stem cells of the myeloid and lymphoid systems. 3 Exp Med145:1567 (1977).2. Mintz B, Anthony K, Litwin S. Monoclonal derivation of mouse myeloid andlymphoid lineages from totipotent hematopoieti stem cells experimentally engrafted infetal hosts. Proc Natl Acad Sci U S A 8 1:7835 (1984).3. Keller G, Paige C, Gilboa E, Wagner EF. Expression of a foreign gene in myeloid andlymphoid cells derived from multipotent haematopoietic precursors. Nature 318:149(1985).4. Dick JE, Magli MC, Huszar D, Phillips RA, Bernstein A. Introduction of a selectablegene into primitive stem cells capable of long-term reconstitution of the hemopoieticsystem of W/W” mice. Cell 42:7 1 (1985).5. Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamicbehavior of hematopoietic stem cells. Cell 45:917 (1986).6. Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of micereconstituted with retrovirus-infected stem cells. EMBO J 6:3955 (1987).7. Cape! B, Hawley R, Covarrubias L, Hawley T, Mintz B. Clonal contributions of smallnumbers of retrovirally marked hematopoietic stem cells engrafted in unirradiatedneonatal W/W” mice. Proc Nat! Acad Sci U S A 86:4564 (1989).8. Capel B, Hawley RG, Mintz B. Long-and short-lived clones individually identifiedwith reiroviral integration markers. Blood 75:2267 (1990).9. Van Zant G, Chen J-J, Scott-Micus K. Developmental potential of hematopoietic stemcells determined using retrovirally marked allophenic marrow. Blood 77 :756 (1991).10. Hellman S, Botnick LE, Hannon EC, Vigneulle RM. Proliferative capacity of murinehemopoietic stem cells. Proc Natl Acad Sci U S A 75 Suppl 1:490 (1978).11. Cudkowicz G, Upton AC, Shearer GM. Lymphocyte content and proliferative capacityof serially transplanted mouse bone marrow. Nature 201:165 (1982).12. Ross EAM, Anderson N, Micklem HS. Serial depletion and regeneration of the murinehematopoietic system. Implications for hematopoietic organization and the study ofcellular aging. J Exp Med 147:432 (1982). .13. Jones RJ, Celano P, Sharkis SJ, Sensenbrenner LL. Two phases of engraftmentestablished by serial bone marrow transplantation in mice. Blood 73:397 (1989).12714. Mauch P, Heliman S. Loss of hematopoietic stem cell self-renewal after bone marrowtransplantation. Blood 74:872 (1989).15. Harrison DE, Stone M, Astle CM. Effects of transplantation on the primitiveimmunohematopoietic stem cell. 3 Exp Med 172:43 1 (1990).16. Hodgson GS, Bradley TR. Properties of hematopoietic stem cells surviving 5-fluorouracil treatment: Evidence for a pre-CFU-S cell? Nature 28 1:381 (1979).17. Visser JWM, Bauman JGJ, Mulder AFT, Eliason IF, de Leeuw AM. Isolation ofmurine pluripotent hemopoietic stem cells. J Exp Med 59:1576 (1984).18. Spangrude GJ, Heimfeld 5, Weissman IL. Purification and characterization of mousehematopoietic stem cells. Science 24 1:58 (1988).19. Szilvassy SJ, Lansdorp PM, Humphries RK, Eaves AC, Eaves CJ. Isolation in asingle step of a highly enriched murine hematopoietic stem cell population withcompetitive long-term repopulating ability. Blood 74:930 (1989).20. Ploemacher RE, Brons RHC. Separation of CFIJ-S from primitive cells responsible forreconstitution of the bone marrow hemopoietic stem cell compartment followingirradiation: Evidence for a pre-CFU-S cell. Exp Hematol 17:263 (1989).21. Jones RJ, Wagner JE, Celano P, Zicha MS, Sharkis SJ. Separation of pluripotenthaematopoietic stem cells from spleen colony-forming cells [letter]. Nature 347:188(1990).22. Keller G, Snodgrass R. Life span of multipoterfial hematopoietic stem cells in vivo. JExp Med 171:1407 (1990).23. Jordan CT, Lemischka JR. Clonal and systemic analysis of long-term hematopoiesis inthe mouse. Genes Dev 4:220 (1990).24. Dexter TM, Allen TD, Lajtha LG. Conditions controlling the proliferation ofhaemopoietic stem cells in vitro. J Cell Physiol 9 1:335 (1977).25. Mauch P, Greenberger JS, Botnick L, Hannon , Heliman S. Evidence for structuredvariation in self-renewal capacity within long-term bone marrow cultures. Proc NatlAcad Sci USA 77:2927 (1980).26. Harrison DE, Lemer CP, Spooncer E: Erythropoietic repopulating ability of stem cellsfrom long-term marrow culture. Blood 69:1021(1987).27. Szilvassy SJ, Humphries RK, Lansdorp PM, Eaves AC, Eaves CJ. Quantitative assayfor totipotent reconstituting hematopoietic stem cells by a competitive repopulationstrategy. Proc Natl Acad Sci U S A 87:8736 (1990).28. Fulop GM, Phillips RA. Use of scid mice to identify and quantitate lymphoid restrictedstem cells in long-term marrow cultures. Blood 74:1537 (1989).12829. Barker JE, Braun J, McFarland-Starr EC. Erythtocyte replacement precedes leukocytereplacement during repopulation of W/Wv mice with limiting dilutions of +1+ donormarrow cells. Proc Nati Acad Sci U S A 85:7332 (1988).30. Bodine DM, Karisson 5, Nienhuis AW. Combination of interleukins 3 and 6 preservesstem cell function in culture and enhances retrovirus-mediated gene transfer intohematopoietic stem cells. Proc Natl Acad Sci U S A 86:8897 (1989).31. Sutherland HJ, Eaves CJ, Lansdorp PM, Thacker 3D, Hogge DE. Differentialregulation of primitive human hematopoietic cells in long-term cultures maintained ongenetically engineered murine siromal cells. Blood (in press).129CHAPTER VALTERATIONS IN LYMPHOPOIESIS FOLLOWING HEMOPOTETICRECONSTITUTION WITH INTERLEUKIN-7 VIRUS INFECTED BONEMARROW1) INTRODUCTION.The common origin of lymphoid and myeloid cells from totipotent stem cells has beenwell documented through the use of genetic markers. Totipotent cells give rise to cells that areconmiitted to either myelopoiesis or lymphopoiesis. Whether stem cells with long term self-maintaining capacity but restricted differentiation capacity exist remains controversial. Assummarized in Chapter I some evidence points to the existence of such cells. Whether theseobservations reflect developmental resthction, or differential recruitment and amplification ofcommitted cells remains unclear. In addition, it is not known if factors exist that may influencethe decision of totipotent stem cells to commit to a particular lineage. The present studies wereinitiated to explore these questions. Work presented in previous chapters and by othersdemonstrated the feasibility of using recombinant retroviruses to introduce genes into totipotenthemopoietic cells with retention of their reconstituting potential. As a starting point for testingthe possibility of influencing lymphoid versus myeloid commitment decisions by totipotentstem cells a retroviral vector was constructed carrying the interleukin-7 (IL-7) gene.IL-7 can modulate early stages of both B-lymphopoiesis and T-lymphopoiesis(reviewed in (1)). IL-7 was originally purified (2) and cloned (3) from an SV4O transformedmurine bone marrow stromal cell line as a potent inducer of proliferation but not differentiationof bone-marrow derived pre-B cells in Whitlock-Witte cultures (4). In addition to its effects onB-cell proliferation, increasing evidence favors the notion that IL-7 may play an important role130in pre-T cell development. IL-7 has been shown to be co-mitogenic for murine T cells in thepresence of PMA or Concanavalin A (5,6), moreover effects on earlier stages of T-cells havebecome evident with studies showing direct stimulation of the proliferation of CD4CD8 (DN)thymocytes (7), and maintenance of triple negative (CD3CD4CD8jthymic progenitors inculture (8). The potential of IL-7 to influence either the proliferation or differentiation oftotipotent stem cells is not known.Lethally irradiated mice were therefore injected with bone marrow infected with arecombinant retrovirus capable of expressing high levels of murine IL-7 message. The originalgoal was to then monitor CRU and their differentiation capacity upon transplantation tosecondary recipients, i.e. to determine whether autocrine expression of IL-7 would allow theselective amplification of lymphoid restricted stem cells during the regeneration ofhemopoiesis. This overall goal however, was precluded by the observation of grossperturbations and morbidity due to appearance of a lymphoproliferative disorder in a largeproportion of the mice. This chapter describes these experiments and an analysis of the resultsobtained.2) EXPERIMENTAL STRATEGY.Construction and production of recombinant retroviruses JZen-neo and JZenmlL7tkneoare described in detail in Materials and Methods as are protocols for infection of bone marrowwith recombinant viruses, reconstitution of lethally irradiated recipients and subsequentmolecular and immunological analysis.1313) RESULTS.A) Phenotvpic Heterogeneity in Disease Presentation of Mice Reconstituted withmlL-7 Virus Infected Bone Marrow.Three experiments were performed in which mice were reconstituted with day-4 5-FUbone marrow following infection with JZen-neo or JZen-mlL7tkneo viruses (Fig.16) by cocultivation. Gene transfer efficiencies were initially assessed by Southern analysis of day 12spleen colonies in recipients of limited numbers of mIL-7 infected bone marrow cells. Allspleen colonies analysed (12 of 12) were positive for the transduced mJL-7 cDNA (data notshown). Of 31 mice reconstituted with IL-7 virus infected bone marrow in three separateexperiments, 7 (23%) developed various noticeable morbidity within 4-16 weeks posttransplant, 6 of these were sacrificed for molecular and immunological analysis of tissues.None of 18 control mice in these experiments transplanted with JZenneo-infected bone marrowdeveloped phenotypic abnormalities in this time frame. The four major abnormalities observedin the IL-7 mice included skin lesions characterized by loss of hair, open sores and scar tissue,massive lymph nodes, peritoneal ascites, or hind limb paralysis with mice eventually becomingmoribund. Peripheral blood analysis and autopsy revea’ed that these mice had elevated whiteblood cell counts ranging from 2-8 x celIJmL, and enlarged spleens. Findings in 5 of thesemice plus one mouse that appeared initially to have no apparent disease (mouse 4) aresummarized in Table IV. Mouse 2 appeared moribund at 6 weeks post transplant, and prior tosacrifice had elevated peripheral blood white cell count of 6 x 107/mL. Peripheral bloodsmears indicated that the circulating cells were large lymphocytes with a blast morphology (datanot shown). Mouse 1 presented at 4 weeks with hind limb paralysis and severe skin lesions,132JZen-neoSD SAEJLLToH In dill3’ LTRFigure 16. Schematic representation of JZen-neo and JZenmlL7tkneo proviruses. SD andSA denote splice donor and splice acceptor sites repectively. NEO representsneomycin phosphotransferase coding sequences. The coding sequences formurine lnterleukin-7 (mIL-7) are followed by a thymidine kinase promotedNEO coding sequence (tkneo). Construction of these viruses is described indetail in Chapter II.JZenmlL7tkneo133TABLE IV. White Cell Count and Phenotype Observed in Mice Reconstituted with JZenmlL7tkneo Infected Marrow.wks post txn p.b. white cell cnt. findingsmouseX 106 permL1 6 20 lrg. spi., paral., SL2 6 60 kg. spl., LN morib.3 12 20 kg. LN, infilt.,paral4 12 ND lrg.LN.5 8 24 kg. saliv. LN.6 16 ND IrgLN,edemacont.mean 8 7±5Designation of mouse 1 through 6 are maintained as JZen mlL7tlcneo infected marrowrecipients throughout the text. Control mean (cont. mean) refers to 3 mice reconstituted withJZen-neo infected bone marrow, and peripheral blood samples analysed for white cell count 8weeks post transplant. Phenotypes were those observed at time of sacrifice(wks post txn) andare abbreviated as follows: kg. spi. and kg. LN refers to enlarged spleen and lymph noderespectively. Enlarged lymph node sizes ranged from 5-10mm in diameter. In the case ofmouse 3 the lymph node appeared to be infiltrating the surrounding muscle tissue (infilt.).paral. refers to near total hind limb paralysis, SL refers to skin lesions with accompanying lossof hair located primarily on the back in mouse 1 plu other mice not shown on this Table.Morib.refers to a moribund state. wks post txn and p.b. white cell cnt refer to the number ofweeks post transplantation mice were sacrificed, and peripheral blood white cell countsrespectively.134although the peripheral blood counts were only slightly elevated (2 x 107/mL), bone marrowsmears indicated a massive infiltration of lymphocytes. Both mouse 1 and mouse 2 had a 5-10fold increase in CFU-GEMM circulating in the peripheral blood compared to control mice(300/mL and 288/mL respectively, compared to 60/rnL and 33ImL for 2 control mice).Although in mouse 2 in vitro clonogenic progenitor numbers were normal in the marrow, thebone marrow of mouse 1 was essentially devoid of CFU-C.B) Clonal Analysis and IL-7 Expression.In order to demonstrate the presence of integrated recombinant virus and determine thenature of clonal reconstitution in IL-7 mice, Southern analysis of DNA from the bone marrow,spleen, thymus and lymph nodes of 5 mice with gross phenotypic abnormalities followingreconstitution with mJL-7 infected bone marrow was performed using the restriction enzymeHindu that cuts once in the proviral genome and the blot probed with a neo1 specific DNAfragment to allow detection of unique integration fragments. In all cases examined multipleintegrations were observed in regenerated tissues due either to multiple cellular clones, ormultiple integrations within a clone, with at least one major clone predominating (Fig. 17;mouse 1-5). Intensities of the dominant clones in lymphoid tissues indicates reconstitution wasnearly 100% when compared to the DNA of the viral producer cell lines (Fig 17; N,7) In thecases where large to massive lymph nodes were observed (Fig.17; mouse 2,3, 5; L), the clonalpatterns were the same as other tissues, even when multiple sites were sampled (Fig. 17;mouse 3; L1-3). Mouse 5, which had a massive salivary lymph node and spleen was unusualin that the single dominant clone found in the lymph nodes was also found in the spleen andbone marrow, but was only faintly conthbuting to thymic repopulation. This may representpreferential expansion of a clone due to IL-7 expression, or tissue restriction from a totipotent135mouse 1 mouse 2 mouse 3‘B S T”B S T LN”B S T Li L2 L3’ ‘B S T 1.1 L2’ N; ‘-Figure 17. Presence of unique retroviral insertion fragments in bone marrow (B), spleen(S), thymus (T), and pooled or individual lymph node (LN and L1-L3respectively) tissues of mouse 1,2 3 and 5 detected by Southern blot analysis ofHindifi digested DNA. Lethally irradiated mice received 1 x 106JZenmlL7tkneo infected bone marrow cells as described in Materials andMethods. Membranes were hybridized with a neor specific probe (NEO).Blots were re-probed with a T-cell receptor 131 constant region probe (TCR) forassessment of TCR rearrangement in those tissues. N and 7 are control DNA’sfrom a JZen-neo viral producer clone and a JZenmlL7tkneo viral producer clonerespectively.mouse 57,I.NEOTCR--ia- ——.136or lineage-restricted stem cell as observed in previous retroviral marking studies (Lemischka1986) when recipients were analyzed at early time points.mIL-7 mRNA was not detected by Northern blot analysis (Fig. 18 ) in the marrow,thymus, or spleen of a JZen-neo reconstituted mouse ( neo mouse; B,T,S), in the lymph nodeof a normal mouse, or in the JZen-neo viral producer cell line (Fig 18;NL and N respectively).High levels of mIL-7 RNA were however, readily detected in the three lymph node samplesanalysed from mouse 3, and in the marrow, spleen, thymus, and lymph node of mouse 5 (Fig.18).C) Perturbations in Early T-Lvmphopoiesis in IL-7 mice.In an effort to better define the nature of cells expanded in affected mice, FACSanalysis was performed on thymus, spleen, and lymph node of mice 1-5 with a panel ofantibodies specific to T-cells (CD4, CD8) and B cells (B220, surface immunoglobulin).Representative profiles for these and controls are presented in Figure 19. Table V summarizesFACS data for all 5 mice. In all cases examined there was a significant difference in Tlymphocyte distribution in at least one tissue in IL-7 mice compared to control neo mice. Largevariability in the relative frequency of T-cells expressin CD4 or CD8 was observed in IL-7mice, however a number of trends are apparent. In no case was an enlarged thymus observedin mice whose peripheral lymphocyte populations were altered. Thymuses exhibited a generaldecrease in the proportion of double positive, CD4+CD8+ (DP) cells and an increase inproportion of double negative, CD4CD8 (DN) and single CD4 or CD8 cells (Fig 19, m2and m3; Table V Thymus, 1-3) when compared to control JZen-neo mice (Fig. 19 ; Table VThymus, A-B). The most marked example of this was observed in mouse 3 (Fig. 19;Table V,m3), with a large decrease in DP (37% vs 84% in control mouse A) and an accompanying9.5—7.5—4.4—2.4—1.4—mouse 3‘Li L2 L3’ rB TFigure 18. Northern blot analysis of IL-7 mRNA expression in tissues of recipients ofJZenmlL7tkneo infected bone marrow (mouse 3 and mouse 5), and a recipientof JZen-neo infected bone marrow (neo mouse). Tissue legends are asdescribed in Figure 16. NL is RNA from pooled lymph nodes from normalunirradiated mice. N and 7 are RNA samples from a JZen-neo viral producerclone and a JZenmlL7tkneo viral producer clone respectively. 5-10 jig of RNAwas separated by formaldehyde gel, blotted and probed with an IL-7 cDNAspecific probe. D is 200pg of a denatured mIL-7 cDNA fragment.137mouse 5 neo mouseS Li L2’ ‘B T 5’ NL D NII101004_10:id3:— 210210:011(111 11111121113h410 10 10 10410id3.210Figure 19. Sample FACS analysis of CD4 (vertical axis) and CD8 (horizontal axis)expression in thymus (THY) spleen (SPL) and lymph node (LN) of a lethallyirradiated mouse repopulated with JZen-neo infected bone marrow (NEO) ormice reconstituted with JZenmlL7tkneo infected bone marrow (IL-7). Sampleprofiles from mouse 2, 3 and 5 (m2, m3 and m5) are indicated. Cells werestained with F1TC conjugated anti-CD8 and PE- conjugated anti-CD4. Cellswere then analysed in a FACScan (Becton Dickinson).138410NEO IL-7 IL-7—Im2 I• V102 4101- IV4I III 1’ I I I 11121 I I I 1131 I I I10 10 10 10 11m3-LI-’I 2I 1111111 I 1111121 I 1111131 I 1111 410 10 10 10I I I I10010-ATHYSPLLNI 1111111 11111121 11111311 [TI 410 10 10 10410:1 d3..210’10:0•m2 10: m5..Id3• V —2 -—••V••;:V.e 1-,0 2 3 0 210 10 10 10 10 10 10 10A10:1 d3.210:10:10-1 o0!L2Vm3V• .— ——_. VV•*• V--:.VV- I10.10Em5-tV:•VV.AV.Id3 I___21010 I- V.IVV-’:VcjV- VVVVVVVVVII:V V0’’’’ 1’’’••’• 2’’’’’°3’’’’ 410 10 10 10 1001111 1, 2”’”3’’’10 10 10 10 10139TABLE V. Summary of Distribution of Cell Phenotypes in Tissues of Mice ReconstitutedWith JZenmlL7tkneo or JZen-neo Infected Bone Marrow.CD4CD8+1+-7-Thymus1112131415118 14 39 8 116157 1 361 56 37 85 84151617 6 2Spleen111213141513 10 32 14 225 33 27 5 198488 11132 0263 2 8145408153IAIBI20 168 60.4 0.571 78B220 NDNDI6 2 4 4 3 NDND23 ND 37 41 41 NDND13 5 7 22 9Ig ND ND ND ND ND ND ND 43 16 56 30 40 6467 NDND 721 7 33 31Thymus, spleen and lymph nodes from mice 1-5 reconstituted with JZenmlL7tkneoinfected bone marrow and mice A and B with JZen-neo infected bone marrow.were analysedfor cell surface antigen expression by anti-body binding and FACScan as described inMaterials and Methods. Mice A and B are age matched controls analysed in parallel with mice3 and 5 respectively. Values are given for the percentage of cells positive for each antigentested. +1+ and -I- refer to CD4+/CD8+ and CD4-ICD8- populations respectively.9942Lymph node11121314151 IAIBINDND2O 4813 49 54NDND57 13 37 17 17ND ND 12 3 47 0.7 0.6NDNDII 36 3 34 28140increase in DN (17% vs 3%) and CD4 cells (39% vs 9%). Mouse 3 thymus cells expressingB220 were increased 4-fold when compared to controls, and a large portion of thymocyteslacked T-cell receptor gene (TCR) rearrangement when analysed by Southern blot (Fig 19,mouse 3, T). These findings are consistent with an expanded population of B-lineage cells inthe thymus.The most dramatic effects were observed in T cell subsets of spleen and lymph nodesof IL-7 mice, where large increases in total cell number and tissue size occurred. In bothspleen and lymph nodes the predominant differences compared to control animals wereincreases in DP and CD8+ lymphocytes, with an associated decrease in the proportion of DNcells in these tissues (Fig.19, Spl, ml, m2, m3 and m5; LN, m3 and m5; Table V, 1-3,5,spleen and lymph node). A significant proportion of lymph-node cells retained T-cell (3-chainreceptors in germ line configuration when analysed by Southern blot analysis (Fig. 17, mouse3 and 5, Ll-L3). This may be due at least in part to the presence of B220 cells (B cells)which represent from 7-13% of the lymph node (Table V). No predominant TCR rearrangedclones were detected in IL-7 mice suggesting TCR rearranged T cells had undergoneoligoclonal or polyclonal expansion. An exception to this was mouse 3 (Fig. 17) whose spleencells showed 3 unique TCR (3 chain bands other than the germ-line fragment.Although some general trends in distributions of amplified T-cell populations wereobserved, there were variations with regards to disease presentation,or predominant cellularphenotype. Mouse 4 did not originally present with an obvious phenotypic abnormality, butupon sacrifice, enlarged lymph-nodes and a massive mesenteric lymph node were observed.FACS analysis showed that all CD4-CD8 populations increased within the enlarged lymphnodes of this mouse when compared to JZen-neo infected controls.(Table V).Mouse 3 and 5 developed massive lymph-nodes, noticeable within 12 weeks posttransplant. FACS analysis in both mice indicates the enlargement was predominantly due to141double positive (CD4CD8j lymphocytes (Fig. 19, LN, m3 and m5). Further analysis oflymph node cells from mouse 5 revealed that a significant proportion were positive for bothheat stable antigen (Mi-69) and Thy-i (Fig 20A), and were CD3+ (Fig.20B) suggesting thatthe predominant cell type in this lymph node was similar to immature CD4hiCD8hicD3lOThyi+HSA thymocytes, and very different than a matched control mouse reconstituted withJZen-neo virus in which lymph node cells are predominantly double negative or single CD4-CD8tCD3hiThy 1 +HsAlo.Proliferation assays were performed in order to determine if the observed expansionwithin the lymph node was localized or due to an external input of cells. The proliferativeresponse of mouse 5 lymph node cells to anti-CD3 was also determined (Fig. 20C), and foundto be low compared to the control mouse which had a significant proliferative response. Mediaalone was unable to elicit proliferation from control lymph node, mouse 5 lymph node cellshowever, demonstrated significant proliferation in absence of exogenous stimulation.A sixth mouse presented with marked abdominal swelling at 16 weeks post transplant.Analysis upon sacrifice revealed peritoneal and pleural ascites (5 x iO cells recovered from theperitoneal lavage), and an enlarged mesenchymal lymph node. FACS analysis of ascites cells(Fig. 21A) revealed predominant CD4CD8 or CD4 or CD8 single positive cells. Theseascites cells also showed proliferation in vitro in the presence of media alone (Fig. 2iB).Interestingly, FACS profiles for CD4 and CD8 in this mouse did not have the shift towardCD4CD8 cells (see Fig. 19). Proliferation assays for sorted single and double positive Tcells from this lymph node showed that while CD4 and CD4/CD8 cells could proliferate inabsence of added stimulation, single CD8+ cells required anti-CD3 to induce a response (Fig.21B).142Figure 20.120-100-C.,80.-- 60-z0 40-C.)20.Cell surface phenotypic analysis and proliferative responses of lymph node cellsfrom a mouse repopulated with JZen-neo infected bone marrow (NEO) ormouse 5 reconstituted with JZenmiL7tkneo infected bone marrow (IL-7).Other phenotypic and cellular characteristics of mouse 5 are described in TableIll and IV, as well as in Figures 15, 16 and 17. Lymph node cells were stainedwith FITC conjugated M169 and biotin avidin-FITC anti-Thyl (A), or withFITC-anti-CD3 (B). For proliferation assays (C), 5 x lO cells were culturedfor 3 days in triplicate cultures in wells pre-treated with anti-CD3 antibody, IL-7(lOp.gImL), or media alone. Cultures were pulsed with 1 .LCi 3H-TdR andharvested 6 hours later. The mean of triplicate cultures with their standarderrors are indicated.ANEOI IIL-710:1 d.M169 210’ic0•10-A.2 :310 10 10 10THY-iB.CD3C’JantiC0311-7 media - antiC 03143CD41LN AscitesA.Figure 21. FACS analysis of CD4 and CD8 expression and proliferative responses inlymph node (LN) and ascites cells of mouse 6, reconstituted withJZenmlL7tkneo infected bone marrow, and sacrificed 16 weeks post transplant.(A) Cells were stained with FITC conjugated anti-CD8 and PE- conjugated antiCD4. Ascites cells were then analysed on a FACScan, and lymph-node cellssorted for analysis and collection on a FACStar (Becton Dickinson). (B)Proliferation assays for ascites cells and lymph-node sub-populations wereperformed as described in Figure 20.CD8counts xanti-CD3 IL-7 mediaB.ascites 37.0 j 3.5 28.0 3.7 32.4 ±. 5.3LN CD4+ 51.7 ±. 9.1 21.4 j. 1.9 26.7 j 1.9LN CD8+ 10.5 j. 1.3 0.5 j 0.3 0LN +1+ ND ND 48.1 18.01444) DISCUSSION.Interleukin-7, was originally cloned as a potent inducer of proliferation, but notdifferentiation of lymphoid cells at the pre-B cell stage of development (9,3). Recent attemptshave been made to determine if IL-7 may play a causative role in the generation of lymphoidtumors (14,15). Pre- B cell lines that were tested lacked endogenous IL-7 expression andattempts to induce an autocrine transformation by infection of cell lines with recombinant IL-7retroviruses were either unsuccessful (10), or required secondary genetic events (11). IL-7 hasbeen shown to accelerate T and B cell repopulation in mice with lymphopenia (12). Similarly,IL-7 transgenic mice have been reported to have higher numbers of pre-B cells, mature B cellsas well as all subsets of T-cells (13). We found no evidence for gross disproportionate pre-Bcell amplification in the hemopoietic tissues of the IL-7 mice we analysed, and generally thenumber of B220-i- cells did not exceed those expressing 1g. Given the extent of increasecellularity in the spleen and lymph-nodes of some animals, B220 and Ig expression suggeststhe overall numbers of B cells increased proportionately.While the bone marrow is the primary site of B-lymphocyte differentiation in adultmammals (14), T-lymphocyte development begins with migration of early hemopoietic stemcells to the thymus. These cells do not express the T-cell receptor (TCR), CD4 or CD8, but areThy1b0 (15) and rapidly acquire high levels of heat stable antigen (HSAhigh), which ismaintained until maturation is complete (16,17). Thymocytes proceed through an intermediateCD3medCD4+CD8+ stage prior to becoming 3h1gh and single CD4 or CD8 positive(reviewed in (18)). Although these stages of early lymphocyte development are becomingbetter defined, many of the regulatory signals involved are unknown. Nevertheless there is asubstantial amount of evidence that IL-7 can influence the early stages of T-lymphocyte145development. For example IL-7 is known to stimulate he proliferation of CD4-CD8- (DN)thymocytes (7) and day 15 fetal thymocytes (19,20) in vitro.Our results suggest that a major target in vivo of dysregulated IL-7 expression can beearly T-lymphocytes. We found that the predominant feature in mice undergoing hemopoieticreconstitution with IL-7 infected marrow was the accumulation of CD4CD8 (DP) and CD4CD8 T- cells in lymph nodes and spleen. Further analysis of lymph node cells demonstratedthat a high proportion of cells also expressed low levels of CD3 and high levels of HSA(M1/69), suggestive of an immature T-lymphocyte phenotype. The origin of these cells is notknown. Thymocyte precursors to DP CD310cells may be eitherCD31OCD40 8 orCD310CD48hi (reviewed in (21)). It is possible that these cells or their precursorsunderwent extensive proliferation in the thymus of IL-7 mice and were then exported to theperiphery. Fractionation of adult DN thymocytes into various sub-populations has shown thatIL-7 is sufficient to induce proliferation of CD3+ cells without subsequent differentiation(22,23). Triple negative cells (CD3CD4CD81L-2Rj, do not proliferate, but can bemaintained in vitro in the presence of IL-7 (8). A large proportion of TN IL-2R cellshowever, differentiate into CD4+CD8+ cells within 24 hours in vitro with or without IL-7(15,24). In most cases we observed only minor alterations in the proportion of thymocytesubsets, usually an increase in DN thymocytes, but in no case was an enlarged thymusobserved. It is also possible that T-cell precursors in the periphery are undergoing extensiveproliferation. IL-7 does not normally maintain the viability of DP thymocytes in vitro (19);wefound however that a high proportion of lymph node cells in IL-7 mice were proliferatingwithout added stimulation. Sorting of these cells indicated that in at least one recipient DPcells, CD4+ CD8- , but not CD4 CD8+ cells were proliferating autonomously.Other studies using retrovirally mediated gene transfer to over express normal growthfactors have similarly demonstrated expansion of particular cellular lineages. Transplantation146of bone marrow cells infected with either IL-3 or GM-CSF recombinant retroviruses resulted innon-neoplastic lethal myeloproliferative syndromes (25,26). IL-6 under similar conditions,resulted in a disease state similar to that of multicentric Castlemans disease with massivesplenomegaly and peripheral lymphadenopathy primarily due to plasma cell infiltration (27).The recipients of IL-7 virus infected bone marrow described here developed a severelymphoproliferative disorder with hyperplastic lymph nodes and splenomegaly, commonly dueto expansion of T-cells with a primitive thymic cell phenotype.Further studies analyzing and isolating cells from tissues of IL-7 mice using antibodiesthat recognize early lymphoid populations and repopulating cells, together with functionalstudies of these cells repopulating capacities in secondary recipients may give some insight intoa potential role of IL-7 at these stages. Further resolution of these questions may require invitro studies in which bone marrow stem cells are exposed to IL-7 and then assessed in vivofor repopulating capacity. In addition, it may be important to begin with a more defined targetpopulation for retroviral infections, particularly purified stem cells, in order to better defme thestage at which perturbations in development are important. Finally it will be important todetermine if the observed expansion of early T-lymphopoiesis is occurring in the thymus orextrathymically. This may be accomplished by reconstituting genetically deficient athymic micewith IL-7 infected bone marrow. These studies will be necessary to determine if the IL-7mouse will serve as a model to study the very earliest stages of hemopoiesis, and define afunctional role for IL-7 in these events.147REFERENCES1. Henney CS. Interleukin 7: effects on early events in lymphopoiesis. Immunol Today10: 170 (1989).2. Namen AE, Schmierer AE, March CJ, Overell RW, Park LS, Urdal DL, MochizukiDY. B cell precursor growth promoting activity. Purification and characterization of agrowth factor active on lymphocyte precursors. 3 Exp Med 167:988 (1988).3. Namen AE, Lupton 5, Hjerrild K, Wignall 3, Mochizuki DY, Schmierer A, Mosley B,March CJ, Urdal D, Gillis S, Cosman D, Goodwin RG. Stimulation of B-cellprogenitors by cloned murine interleukin-7. Nature (London) 333:57 1 (1988)4. Whitlock CA, Witte ON. Long-term culture of B-lymphocytes and their precursorsfrom murine bone marrow. Proc.NathAcad.Sci.USA 79:3608 (1982).5. Morrissey PJ, Goodwin RG, Nordan RP, Anderson D, Grabstein KH, Cosman D,Sims J, Lupton S, Acres B, Reed SG, Mochizuki D, Eisenman J, Conlon PJ, NamenAE.Recombinant interleukin-7, pre-B cell growth factor, has co-stimulatory activity onpurified mature T cells. J Exp Med 169:707 (1989).6. Chazen GD, Pereira GMB, LeGros G, Gillis S,, Shevack EM. Interleukin-7 is a T-cellgrowth factor. Proc Natl Acad Sci USA 86:5923 (1989).7. Okasaki H, Ito M, Sudo T, Hattori M, Kamo S Katsura Y, Minato N. IL-7 promotesthymocyte proliferation and maintains immunocompetent thymocytes bearing (43 orT-cell receptor in vitro: synergism with IL-2. J Immunol 143:29 17 (1989).8. Suda T, Ziotnick A. IL-7 maintains the T-cell precursor potential of CD3-CD4-CD8-thymocytes. J Immunol 146:3068 (1991).9. Lee G, Namen AE, Gillis S, Ellingsworth LR, Kincade PW. Normal B cell precursorsresponsive to recombinant murine IL-7 and inhibition of IL-7 activity by transforminggrowth factor-13. 3 Immunol 142:3875 (1989).10. Young JC, Gishizky ML, Witte ON. Hyperexpression of interleukin-7 is not necessaryor sufficient for transformation of a pre-B lymphoid cell line. Mol Cell Biol 11:854(1991).11. Overell RW, Clark L, Lynch D, Jerzy R, Schmierer A, Weisser KE, Namen AE,Goodwin RG. Interleukin-7 reiroviruses transform pre-B cells by an autocrinemechanism not evident in Abelson Murine Leukemia Virus transformants. Mol CellBiol 11:1590 (1991).12. Morrissey P3, Conlon P, Braddy 5, Williams DE, Namen A, Mochizuki DY.Administration of IL-7 to mice with cyclophosphomide-induced lymphopeniaasselerates lymphocyte repopulation. 3 Immunol 146:1547 (1991).13. Samardis I, Casorati G, Traunecker A, Iglesias A, Gutierrez JC, Muller V, Palacios R.Development of lymphocytes in IL-7 iransgenic mice. Eur J Immunol 21:453 (1991).14814. Osmand DG: Production and differentiation of B lymphocytes in the bone marrow. InBattisto JK,Knight KL (eds): Immunoglobulin genes and B-cell differentaition. NewYork: Elsevier-North Holland Inc. (1980).15. Nakano N, Hardy RP, Kishimoto T. Identification of intrathymic T progenitor cells byexpression of Thy-i, 11-2 receptor and CD3. Eur J Immunol 17:1567 (1987).16. Takei F, Secher DS, Milstein C, Springer T. Use of a monoclonal antibody specificallynon-reactive with T cells to delineate lymphocyte subpopulations. Immunology 42:371(1981).17. Crispe IN, Bevan MI. Expression and functional significance of the Ii id marker onmouse thymocytes. J Immuni. 138:2013. (1987).18. Boyd RL, Hugo P. Towards an integrated view of thymopoiesis. Immunol. Today12:71 (1991).19. Murray R, Wrighton N, Lee F, Zlotnick A. Interleukin 7 is a thymocyte growth andmaintenance factor for mature and immature thymocytes. Int Immunol 1:526 (1989).20. Suda T, Murray R, Fischer M, Yokota T, Zlotmck A. Tumor necrosis factor-tx andP40 induce day 15 murine fetal thymocytes proliferation in combination with IL-2. JImmunol 144:1783 (1990).21. Nikolic-Zugic J. Phenotypic and functional stages in the inirathymic development ofcxf T cell. Immunol Today 12:65 (1991).22. Suda T, Murray R, Guidos C, Ziotnick A. Growth promoting activity of IL-ia, IL-6,and tumor necrosis factor-a in combination with IL-2, IL-4, or IL-7 on murinethymocytes. J Immunol 144:3039 (1990 ).23. Vissinga CS, Fatur-Saunders DJ, Takei F. Dual role of IL-7 in the growth anddifferentiation of immature thymocytes. in press (1991)24. Wilson A, Petrie HT, Scollay R, Sortman K. The acquisition of CD4 and CD8 duringthe differentaition of early thymocytes in short-term culture. mt Immunol 1:605 (1989).25. Chang JM, Metcalf D, Lang RA, Gonda TI, Johnson GR. Nonneoplasticmyeloproliferative syndrome induces by dysregulated Multi-CSF (IL-3) expression.Blood 73:1487 (1989).26. Johnson, G.R., Gonda, T.J., Metcalf, D., Hariharan, I.K., and S. Cory. A lethalmyeloproliferative syndrome in mice transplanted with bone marrow cells infected witha retrovirus expressing granulocyte-macrophage colony stimulating factor. EMBO J8:441 (1989).V14927. Brandt SJ, Bodine DM, Dunbar CE, Neinhuis AW. Dysregulated interleukin 6expression produces a syndrome resembling Castleman’s disease in mice. J Clin Invest86:592 (1990).150CHAPTER VISUMMARY AND FUTURE DIRECTIONSHemopoiesis is a complex process involving multiple regulators acting singly or incombination to control each step (reviewed in (1)). The biological properties of totipotenthemopoietic stem cells, the most primitive cells which are capable of long-term in vivo bloodcell production, are slowly becoming better defined but a large number of questions have yet tobe answered. These questions primarily concern the regulatory mechanisms of stem cells thatgovern their maintenance and use. The initial stages of hemopoietic development when stemcells begin to proliferate and either remain totipotent or commit to myeloid or lymphoid lineagesare poorly understood. Clearly, assays are required that will help define these stages andidentify factors that influence these events.A key development of the work presented in this thesis is a model system whichcombines retroviral marking of hemopoietic stem cells with a prolonged in vitro culture period,and a rigorous assay for the in vivo repopulating ability of the culture derived cells. Using thisapproach the results presented in Chapters ifi and IV have demonstrated for the first time the invitro maintenance of stem cells with both lymphoid and myeloid repopulating ability. Moreimportantly however, is the demonstration that despite an overall net decline in stem cellnumbers in long-term culture, some individual totipotent stem cells undergo a significant clonalamplification while retaining competitive long-term repopulating ability.The ability to track in vitro expansion of stem cells in this model system has importantimplications to the current theories of early hernopoiesis. A number of questions can beapproached using this model. For example, a question remains as to whether or not a truelineage restricted self-renewing population of stem cells exists. Although a selectively151expanded population of lymphoid restricted stem cells was not observed as previouslysuggested by others (2), a lymphoid-lineage restricted developmental pattern was observed in anumber of instances (see Table ifi). Equally interesting will be the future detailed definition ofthe kinetics of CRU in vitro. Also yet to be resolved is the question of whether the observed invitro stem cell decline with a concurrent clonal expansion of some repopulating cells is areflection of heterogeneity in the starting CRU population or a result of induced or intrinsicvariations in their behavior in the long-term culture system.These fundamental questions of stem cell biology establish the basis for futureexperiments using this in vitro model system that may test the potential biological response oftotipotent long-term reconstituting cells to defined growth factors. The two responses that maybe detected in such experiments are first proliferative, and second developmental. A proposedschematic approach to these experiments is outlined in Figure 22. Retrovirally marked bonemarrow could be inoculated into mini LTC at limiting dilution in order to seed a single CRU(stem cell) into each well. One might then attempt to detect progeny CRU being releasedcontinuously into the non-adherent fraction by injecting these into multiple recipients. Duringculture, the cells could then be manipulated by growth factors either by adding them directly tothe media, or by using feeder layers genetically engineered to produce a specific growth factor.Such experiments would allow information about CRU maintenance and amplification to beobtained from estimating CRU numbers before and after culture. In addition, changes indevelopmental potential could then be explored by analysis of retrovirally marked clonalcontributions to the lymphoid and myeloid tissues of the recipients.Such a protocol has exciting potential not only for determining effects of growthfactors, but also for experiments where a clonally derived population of proliferating stem cellsmay be useful, such as certain gene replacement protocols. This potential may also not berestricted to murine studies, in which case it could have implications for clinical bone marrow0152SouthernblotFigure 22. Schematic representation for in vitro clonal expansion and manipulation of anindividual stem cell, followed by in vivo assessment of progeny stem cellsrepopulating capacity.stem celltransplantB ST0— — —— — —— — —0markedrepopulatingstem cell0—— in vitro— amplification— —— —— —— — —— — —— — —— —— —— —153transplantation. Results from human long term cultures suggest that it is not unreasonable toassume that human LTC have similar mechanisms of stem cell maintenance and decline (3). Ifthis is the case then it may be possible to manipulate human hemopoietic stem cells in vitroparticularly where selection of clones may be therapeutically advantageous.In a second study described in Chapter V. experiments were undertaken as an initialstep towards the proposed experiments described above. In this case the IL-7 molecule waschosen as a candidate regulator of stem cell development. In these initial experiments micewere reconstituted with bone marrow that had been infected with a recombinant retroviruscarrying the IL-7 gene. The key finding in these experiments was a marked increase of a celltype in extrathymic tissue normally primarily restricted to the thymus. Such an affect mayhave been due to either extrathymic expansion of these cells, or due to proliferation within thethymus and subsequent export to the periphery. It will be of great interest to determine if the invivo effects observed are due at least in part to stimulatjon of earlier cells in the lymphoidpathway, and perhaps repopulating cells. Ultimately however, the goal of these studies was toextend these findings to the in vitro model, where the long-term culture stromal layer could begenetically engineered using the recombinant 1L-7 virus, and the subsequent affect on CRUexpansion and repopulating potential determined.In summary these studies have demonstrated for the first time the in vitro maintenanceand proliferation of totipotent hemopoietic stem cells with long term in vivo repopulatingpotential. These studies set the stage for future experiments designed to help furtherunderstand the intrinsic properties of these cells as well as the early events that regulate theirdevelopment. The procedures described here should facilitate further in vitro and in vivostudies in the early events in hemopoiesis.154REFERENCES1. Metcalf D. The molecular control of cell division, differentiation commitment andmaturation in haemopoietic cells. Nature 339:27 (1989).2. Fulop GM, Phillips RA. Use of scid mice to identify and quantitate lymphoid restrictedstem cells in long-term marrow cultures. Blood74: 1537 (1989).3. Sutherland HJ, Eaves CJ, Lansdorp PM, Thacker ID, Hogge DE. Differentialregulation of primitive human hematopoietic cells in long-term cultures maintained ongenetically engineered murine stromal cells. Blood (in press)."@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "1992-05"@en ; edm:isShownAt "10.14288/1.0098877"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Genetics"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "In vitro growth and genetic manipulation of hemopoietic repopulating cells"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/3279"@en .