Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Purification and characterization of murine long-term lympho-myeloid repopulating hemopoietic stem cells Szilvassy, Stephen Joseph 1990

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

Item Metadata

Download

Media
831-UBC_1990_A1 S94.pdf [ 9.66MB ]
Metadata
JSON: 831-1.0100545.json
JSON-LD: 831-1.0100545-ld.json
RDF/XML (Pretty): 831-1.0100545-rdf.xml
RDF/JSON: 831-1.0100545-rdf.json
Turtle: 831-1.0100545-turtle.txt
N-Triples: 831-1.0100545-rdf-ntriples.txt
Original Record: 831-1.0100545-source.json
Full Text
831-1.0100545-fulltext.txt
Citation
831-1.0100545.ris

Full Text

PURIFICATION A N D C H A R A C T E R I Z A T I O N O F M U R I N E L O N G - T E R M LYMPHO-MYELOED R E P O P U L A T M G HEMOPOIETIC S T E M C E L L S  by  STEPHEN JOSEPH SZILVASSY B.Sc, The University of British Columbia, 1985  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES (Department of Microbiology)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April 1990 © Stephen Joseph Szilvassy, 1990  In presenting  this thesis in partial fulfilment of the  requirements for an  advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference  and  study. I further agree that permission for extensive  copying of this thesis for scholarly purposes may department  or  by  his or  her  representatives.  be  It is understood  publication of this thesis for financial gain shall not be permission.  Department of  Microbiology  The University of British Columbia Vancouver, Canada Date  DE-6  (2/88)  M a y 3n i QQO  granted by  the head of  my  that  or  copying  allowed without my  written  -11 -  ABSTRACT The hemopoietic system Is organized as a hierarchy of hemopoietic cell populations distinguished by differences i n their proliferation and differentiation potential. Studies using short-term i n vitro and i n vivo assays based on colony formation i n semi-solid medium, or i n the spleens of lethally irradiated mice, respectively, have shown that these procedures detect primarily lineage-restricted progenitor types and have provided m u c h information about the characteristics and regulation of such cells. Assessment of lymphoid and myeloid tissue reconstitution after more prolonged periods following transplantation has established the existence of a more primitive stem cell type; however, the retrospective nature of these complex analyses has Impeded characterization and purification of these cells. My first objective was to develop a procedure for the selective isolation of stem cells with short-term i n vitro and i n vivo multilineage differentiation potential. For this I devised a single-step, four-parameter fluorescence activated cell sorting procedure In which cells were selected according to their forward and orthogonal light-scattering properties, and their surface expression of the Thy-1 and H-2K  antigens. Application of this procedure to marrow cells from mice treated 4 days  previously with 150 mg/kg of 5-fluorouracil showed that it could be used to sort a subpopulation that was enriched 100-fold i n CFU-GEMM and i n which 1 i n 4 cells was a day 12 CFU-S. To determine the extent to which stem cells with long-term lympho-myeloid repopulating potential had been copurified, I undertook to develop a quantitative procedure that might allow this primitive cell population to be measured and hence characterized on a routine basis. This required an assay that would detect donor-derived hemopoiesis exclusively, and that was sensitive enough for the detection of limiting numbers of cells with long-term lympho-myeloid repopulating potential. This was shown to be possible using a competitive repopulation assay i n which lethally irradiated female recipients were transplanted with male "test" cells together with a second suspension of female cells with adequate short-term repopulating activity but greatly diminished long-term repopulating potential. These sex differences were then used to  - Ill specifically identify the 5 week progeny of stem cells i n the test suspension. Assessment of the sorted day 4 5-FU marrow population revealed that it was capable of repopulatlng all hemopoietic organs after transplantation and that an enrichment of 30-fold over unseparated, 5-FU-treated marrow had been achieved. My second objective was to determine whether the competitive long-term lymphoid and myeloid repopulation obtained with these sorted cells was due to the activity of Individual stem cells with a dual potential for lymphopoiesis and myelopoiesis. For this I used retroviralinfection to uniquely m a r k sorted cells In vitro, and then transplanted them i n sufficiently low numbers to allow individual regenerated clones to be detected and analyzed. In some mice, distribution of cells with the same unique integration marker i n different lymphoid and myeloid cell populations established the presence of lympho-myeloid stem cells i n the original sorted population. In addition, clones with restricted tissue distributions were also documented. My final objective was to investigate whether the competitive repopulation assay was i n fact able to serve as a procedure for the exclusive quantitation of long-term lympho-myeloid repopulating stem cells. A limiting dilution approach was used to compare the frequency of hemopoietic stem cells (competitive repopulatlng units, CRU) i n marrow obtained from a variety of sources, using >20%  repopulation by male cells at 5 or 10 weeks post-transplantation as the  end point. The results obtained were largely independent of the time of analysis, and whether repopulation of recipient marrow or thymus was evaluated, suggesting that either can be used i n this assay to quantitate a hemopoietic stem cell with the potential to regenerate both lymphoid and myeloid systems.  These studies have provided procedures for the detection, quantitation and selective enrichment of the most primitive stem cells i n the murine hemopoietic system which have competitive long-term lympho-myeloid repopulatlng ability. The availability of these procedures should facilitate the development of additional purification steps leading to the isolation of these cells as homogeneous suspensions, and their further use as targets for  -ivretrovlrus-medlated gene transfer to determine the genetic basis of their activation, determination and neoplastic transformation.  - V -  TABU; OF CONTENTS Page ABSTRACT  ii  LIST OF TABLES  viii  LIST OF FIGURES  ix  ACKNOWLEDGEMENTS Chapter I  xii  INTRODUCTION 1)  The Structure of the Hemopoietic System A) An Overview B) Hemopoietic Cells Defined by Short-Term Assays a) b)  Cells with the Characteristics of Hemopoietic Progenitor Cells Cells with the Characteristics of both Hemopoietic Stem and Progenitor Cells  1 1 4 4 6  C)  Hemopoietic Cells Defined by Long-Term Assays  12  a) b) c)  Repopulation of Lethally Irradiated Mice Repopulation of W/W Mice Long-Term Bone Marrow Cultures as an Assay for Stem Cells Evidence for Human Repopulating Stem Cells  14 16  d)  v  17 20  2)  Regulation of the Hemopoietic System A) Regulation By Cellular Interactions B) Regulation By Soluble Growth Factors  21 22 24  3)  The Isolation of Hemopoietic Cells A) Stem Cell Purification B) Differentiation Antigens Expressed on Primitive Hemopoietic Cells  29 30  a) b) c) d) e) f) g) h) i)  Theta Antigen (Thy-1) Major Histocompatibility Complex Class I Antigens (H-2K) Major Histocompatibility Complex Class II Antigens (la) Qa Antigens Wheat Germ Agglutinin (WGA) Binding Sites Rhodamine-123 (Rh-123) Uptake Stem Cell Antigen-1 (Sea-1) Human Major Histocompatibility Complex Class II Antigens (HLA-DR) Human Cluster of Differentiation-34 Antigen (CD34, My-10)  36 36 37 38 39 40 40 41 42 43  - vi 4)  Thesis Objectives and General Strategy References  Chapter II  , 47  MATERIALS AND METHODS 1)  Cells A) Animals B) Preparation of Marrow Cell Suspensions  63 63 63  2)  Hemopoietic Cell Purification A) Monoclonal Antibodies B) Indirect Double-Immunostainlng of Bone Marrow Cells C) FACS Analysis and Sorting  64 64 65 65  3)  Assays A) Methylcellulose Assays B) CFU-S Assays C) Competitive Repopulation Assays  66 66 67 69  4)  Retroviral Marking of Purified Marrow Cells A) Recombinant Retrovirus B) Infection of Purified Marrow Cells C) Spleen Colony Analysis D) Separation of Marrow Macrophage and Splenic Lymphocyte Subpoplations  71 71 71 72  Molecular Analyses A) Southern Analysis with Y-Specific Probe B) Southern Analysis with Neo -Specific Probe C) Spot Blot Analysis  73 73 74 74  References  76  5)  r  Chapter III  44  72  ISOLATION IN A SINGLE STEP OF A HIGHLY ENRICHED MURINE HEMOPOIETIC STEM CELL POPULATION WITH COMPETITIVE LONG-TERM REPOPULATING ABILITY 1)  Introduction  2)  Results A) Light-Scatter Properties, Thy-1, and H-2K Antigen Expression of In Vitro Clonogenic Cells B) Characterization of Cells with Competitive Long-Term Repopulating Ability C) Enrichment of Primitive Hemopoietic Cells After Four Parameter Sorting  87  Discussion  94  References  97  3)  78  80 85  - vii Chapter IV  RETROVIRUS-MEDIATED GENE TRANSFER TO PURIFIED HEMOPOIETIC STEM CELLS WITH LONG-TERM LYMPHO-MYELOPOIETIC REPOPULATING ABILITY 1)  Introduction  2)  Results A) Transfer of the Neo Gene to Purified Stem Cells  99  r  B) 3)  Chapter V  r  Discussion  References VALIDATION OF A NEW ASSAY FOR THE LYMPHO-MYELOID HEMOPOIETIC STEM CELL USING A COMPETITIVE REPOPULATION STRATEGY  108 111  1)  Introduction  112  2)  Results A) Origin of Regenerated Cells in the Competitive Long-Term Repopulation Assay B) Selection of Endpoint Time C) Comparison of Thymus Versus Marrow Analyses  114 117  Discussion  123  References  129  3)  Chapter VI  Analysis of Individual Neo -Marked Clones  100 102  SUMMARY AND FUTURE DIRECTIONS References  119  131 139  -viiiLIST OF TABLES Page  TABLE I  Hemopoietic Growth Factors (CSFs and Interleukins).  25  TABLE II  Reduction in CFU-S Detectable After Double Staining of Marrow Cells.  68  TABLE III  Frequencies of Clonogenic Progenitors in Unstained Suspensions of Normal and Compromised Marrow Cells.  70  TABLE IV  In Vitro and In Vivo Assayable Clonogenic Cell Content of Marrow Cells Isolated by Four Parameter Sorting.  90  TABLE V  Summary of 12 Mice Transplanted With Retrovirally Marked Purified Male Repopulating Stem Cells.  103  TABLE VI  Comparison of the Frequency of CRU in Different Marrow Cell Populations Assessed Using Different Endpoints.  121  TABLE VII  Test of Independence of Marrow and Thymus Repopulation by Male Test Cells When Co-Injected With 2 x 10 Compromised Female Marrow Cells.  124  5  - Ix LIST OF FIGURES Page  FIGURE 1  FIGURE 2  FIGURE 3  Schematic representation of the organization of the hemopoietic system showing both functional and developmental compartmentalization.  3  Schematic representation of hemopoietic cell maturity vs. self-renewal capacity, proliferative potential and cycling status.  13  Schematic representation of positive and negative selection approaches to stem (S) cell purification.  32  FIGURE 4A The major components of the Fluorescence Activated Cell Sorter (FACS).  34  FIGURE 4B General principles of operation of the FACS.  35  FIGURE 5  FIGURE 6  FIGURE 7  FIGURE 8  Representative FLS profiles of normal and day 4 5-FU marrow cells (panels A and B, respectively). Distributions of CFU-GEMM, CFU-GM and BFU-E in the FLS profiles of normal and day 4 5-FU marrow (panels C and D, respectively). Normal frequencies of CFU-GEMM, CFU-GM and BFU-E in unfractionated normal and day 4 5-FU marrow (panel E).  81  Representative OLS profiles of normal and day 4 5-FU marrow cells (panels A and B, respectively). Distributions of CFU-GEMM, CFU-GM and BFU-E in the OLS profiles of normal and day 4 5-FU marrow (panels C and D, respectively).  83  Representative Thy-1.2 fluorescence profiles of normal and day 4 5-FU marrow cells (panels A and B, respectively). Distributions of CFU-GEMM, CFU-GM and BFU-E in the Thy-1.2 profiles of normal and day 4 5-FU marrow (panels C and D, respectively).  84  Representative H-2K° fluorescence profiles of normal or day 4 5-FU marrow cells (panel A). Distributions of CFU-GEMM, CFU-GM and BFU-E in the H-2K profiles of normal and day 4 5-FU marrow (panels B and C, respectively).  86  Spot blot analysis of DNA from tissues of two female mice 5 weeks after transplantation of 10 male day 4 5-FU marrow cells sorted from the same regions of the OLS profile shown in Figure 2, together with 2 x 10^ compromised female marrow cells.  88  Contour plot of Thy-1.2 (green fluorescence) versus H-2K (red fluorescence) of double stained day 4 5-FU marrow cells after gating for FLS and OLS as described in the text.  89  b  FIGURE 9  4  FIGURE 10  D  - X -  FIGURE 11  FIGURE 12  FIGURE 13  FIGURE 14  FIGURE 15  FIGURE 16  FIGURE 17  Demonstration by Southern analysis of male DNA in bone marrow (b), spleen (s), and thymus (t) of female mice transplanted with 500 double stained male day 4 5-FU marrow cells isolated by four-parameter sorting (as described in the text) together with 10^ compromised female marrow cells.  92  Percent of recipients (8 animals per group) showing >20% repopulation of the marrow with male cells 35 days after transplantation of varying numbers of male day 4 5-FU marrow cells before (O ) or after (• ) staining and four-parameter sorting, together with 2 x 10^ compromised female marrow cells.  93  Southern analysis of DNA from peripheral blood leukocytes isolated from 19 female mice 35 days after transplantation with 300 (mice 2.1-2.15) or 500 (mice 2.16-2.19) TKneol9-infected, purified, male day 4 5-FU marrow cells together with 2 x 1 0 ^ compromised female marrow cells.  101  Southern analysis of DNA from individual day 12 spleen colonies generated from the marrows of mice 2.4 and 2.10 sacrificed 49 days after transplantation.  104  Southern analysis of DNA from the bone marrow (b), spleen (s), and thymus (t) of five female mice sacrificed 35-196 days (35d-196d) after transplantation (post Tx) with 1,000 (mouse 1.3), 250 (mouse 1.4), or 300 (mice 2.4, 2.5, and 2.15) TKneol9-infected, purified, male day 4 5-FU marrow cells together with 2 x 10* compromised female marrow cells.  106  Presence of a common retrovirally marked clone In bone marrow (bm), spleen (spl), thymus (thy), lymph node (LN), and separately Isolated marrow macrophage (bm mac), splenic B (spl B) and T (spl T) lymphocytes from mouse 1.7 sacrificed 121 days (12Id) after transplantation (post Tx).  107  Schematic representation of the competitive long-term repopulation assay.  115  FIGURE 18A Southern analyses of bone marrow and thymus cells from lethally irradiated mice transplanted 10 weeks previously with 2 x 10^ compromised marrow cells.  116  FIGURE 18B Southern analyses of bone marrow and thymus cells from lethally irradiated mice transplanted 10 weeks previously with 10 day 4 5-FU cells and 2 x 10 compromised marrow cells.  118  4  FIGURE 19  5  Percent of recipients (8 animals per group) showing >5% repopulation of the marrow with male test cells 10 weeks after transplantation of varying numbers of male day 4 5-FU marrow cells together with 2 x 10^ female compromised marrow cells.  120  -xi FIGURE 20  Percent of recipients (8 animals per group) showing >5% repopulation of the thymus with male test cells 10 weeks after transplantation of varying numbers of male day 4 5-FU marrow cells together with 2 x 10^ female compromised marrow cells.  122  -xii-  ACKNOWLEDGEMENTS  I w i s h to express m y sincere gratitude: to m y s u p e r v i s o r , Dr. A l l e n C. E a v e s f o r t h e o p p o r t u n i t y t o u n d e r t a k e m y g r a d u a t e t r a i n i n g i n t h e T e r r y F o x L a b o r a t o r y f o r H e m a t o l o g y / Oncology. to Dr. C o n n i e J . E a v e s f o r h e r e n t h u s i a s t i c s u p p o r t a n d g u i d a n c e t h r o u g h o u t m y project, a n d for critically reviewing t h i s thesis. to Dr. P e t e r M. L a n s d o r p , Dr. R K e i t h H u m p h r i e s , a n d fellow g r a d u a t e s t u d e n t C h r i s F r a s e r for h e l p f u l d i s c u s s i o n s a n d active c o l l a b o r a t i o n t h r o u g h o u t t h i s project. to Dr. George B. S p i e g e l m a n , Dr. H u n g - S i a Teh, a n d Dr. F r a n k T u f a r o f o r s e r v i n g o n m y c o m p r e h e n s i v e a n d g r a d u a t e c o m m i t t e e s i n t h e D e p a r t m e n t of M i c r o b i o l o g y ; a n d Dr. J o h n W. S c h r a d e r ( B i o m e d i c a l R e s e a r c h C e n t r e , U.B.C.), Dr. D o n B r u n e t t e ( D e p a r t m e n t of O r a l Biology, U.B.C.), a n d Dr. R o b e r t A. P h i l l i p s ( D i v i s i o n o f I m m u n o l o g y a n d C a n c e r , H o s p i t a l f o r S i c k C h i l d r e n , Toronto) f o r s e r v i n g o n m y t h e s i s e x a m i n i n g committee. to G r a c e L i m a a n d C a m S m i t h for expert t e c h n i c a l a s s i s t a n c e , S a m u e l A b r a h a m f o r t e a c h i n g m e m o l e c u l a r biology, F r e d J e n s e n for t a k i n g c a r e of m y four-legged f r i e n d s , a n d e s p e c i a l l y W i e s l a w a D r a g o w s k a f o r t i r e l e s s F A C S a n a l y s i s a n d l o n g sorts. to D o n H e n k e l m a n f o r s t a t i s t i c a l a n a l y s i s , a n d , together w i t h I s a b e l H a r r i s o n a n d R o b e r t M o o n l e , f o r h e l p w i t h t h e p r e p a r a t i o n of t h i s t h e s i s , P a t t y R o s t e n f o r p h o t o g r a p h y , a n d M i c h e l l e C o u l o m b e , S t e p h a n i e H u d s o n , C h r i s Freer, a n d J e n n y F o r s t v e d f o r t h e t y p i n g of m a n u s c r i p t s . to T h e N a t i o n a l C a n c e r I n s t i t u t e of C a n a d a f o r f i n a n c i a l s u p p o r t . a n d above a l l , t o m y wife W e n d y a n d m y f a m i l y f o r p a t i e n c e a n d u n d e r s t a n d i n g .  -1 -  C H A P T E R  I  INTRODUCTION  1)  T H E S T R U C T U R E O F T H E HEMOPOIETIC SYSTEM.  A)  AN  OVERVIEW.  Hemopoiesis is the formation of blood; a system composed of a large number of mature cell types, each exhibiting highly specialized functions such as oxygen and carbon dioxide transport (erythrocytes), response to foreign organisms and their products (granulocytes and macrophages), blood clotting (thrombocytes), antibody production (B-lymphocytes and plasma cells) and cell-mediated immunity (T-lymphocytes). In the normal adult, the bone marrow is the major site of hemopoiesis, having recruited ancestral hemopoietic stem cells from the embryonic yolk-sac and fetal liver (1), and contains cells from many hemopoietic cell lineages at all stages of differentiation. Due to the early developmental separation of the different lineages of blood cells, the hemopoietic system can be functionally divided into two distinct arms, termed myeloid and lymphoid. The cells of the myeloid arm are produced in the bone marrow (2) and can be further divided into the erythroid, granulocytic and megakaryocytic lineages which tenninate i n the production of erythrocytes, granulocytes/monocytes/macrophages and platelets, respectively. The cells of the lymphoid arm are produced to varying degrees in the bone marrow, spleen, thymus and lymph nodes and can be divided into the B and T-lymphoid lineages. Of particular relevence to the present thesis is the fact that all of the mature myeloid blood cell types represent "end" cells. Once fully differentiated and released from the marrow, these cells remain functionally active for only a few days or weeks before being destroyed and  - 2broken down (3,4). Maintenance of stable numbers of these cells i n the blood and tissues throughout adult life therefore depends upon the continued production of new cells from a pool of undifferentiated, self-maintaining, proliferating precursors. Several lines of evidence i n both mouse and m a n indicate that the myeloid arm of the hemopoietic system can be subdivided developmentally into four major cell compartments which define a hierarchy of cells with decreasing differentlative and/or proliferative potential (Figure 1, and reviewed i n (5)). The most primitive of these is comprised of totipotent hemopoietic stem cells, defined by their capacity to regenerate and maintain both the lymphoid and myeloid arms of the hemopoietic system following transplantation, as well as having an extensive capacity to regenerate themselves (a process referred to as self-renewal). Stem cells give rise to cohorts of progenitors whose differentiation potentialities are restricted to a more limited range of lymphoid or myeloid lineages. In addition to limits i n their differentlative potential, progenitors have a lower capacity for proliferation and self-renewal. Both stem cells and progenitors lack any unique morphological features that would enable their separate identification. They must therefore be detected retrospectively on the basis of their developmental properties, expressed either through colony formation i n vitro or tissue repopulation i n vivo. As progenitors divide and differentiate they eventually begin to acquire distinct morphologic characteristics that enable their recognition as the immediate precursors of specific mature blood cell types. Their terminal differentiation is defined by the repopulation of the final compartment of end cells. Hemopoietic cell development is likely to be a process of continuous change. Therefore, although this model of the hemopoietic system focusses on the differences between cells i n each of the compartments visualized above, the borders between each compartment are probably not distinct and the cells i n this spectrum likely represent a continuum, with the most primitive cells having the highest self-renewal and differentiation potential, and later stages showing a progressive loss of both these properties. In vivo studies with cycle-active cytotoxic drugs have suggested that In the normal adult most of the more primitive cells are i n  -3 -  LYMPHOMYELOID STEM  GO  CELL  UJ O  STEM  ' oo  MYELOID  LYMPHOID  STEM  CELL  CELL  oo  _i _i  B-LYMPHCID  PROGENITOR  LU o  GRANULOCYTE/ MACROPHAGE PROGENITOR  T.LYMPHOID PROGENITOR  ERYTHROID PROGENITOR  MEGAKARYOCYTE PROGENITOR  GC  o o  o  a: Proerythroblast  Myeloblast  Megakaryoblast  oo  _J —i LU  —  o  Early normoblast  Promyelocyte  Promonoblast  CE  Myelocytes -  0 00  Megakaryocyte  Basophil  Neutrophil  Eosinophil  Late  3 O LU  Intermediate normoblast  CC  I Metamyelocyte!  CC  a.  6)  normoblast  Juvenile  Reticulocyte  Monocyte  Segmented  6>  (9  B-lymphocyt>  T-lymphocyte  a  o  00  _l _l  LU  U  a  Erythrocyte Monocyte Tissue macrophage  Figure 1.  o  z  Polymorphonuclears (granulocytes)  Platelets  Schematic representation of the organization of the hemopoietic system showing both functional and developmental compartmentalization.  LU  -4-  a non-proliferating or quiescent state (G -phase of the cell cycle) but can be triggered into 0  S-phase by hematologic or immunologic stress (6). In contrast, most mature progenitors are actively cycling even i n the absence of apparent stimulation (7,8). Thus, the key function of the stem cell compartment is to maintain a n appropriate production of mature cells by responding to extrinsic factors that shift the balance between quiescence and proliferation. In this manner they modulate the number of cells feeding into the system for subsequent amplification a n d lineage-specific regulation.  B)  HEMOPOIETIC C E L L S D E F I N E D B Y SHORT-TERM ASSAYS.  a)  Cells with the Characteristics of Hemopoietic Progenitor Cells.  Information about the more mature compartments of the hemopoietic system has been derived primarily from the use of i n vitro culture systems that support the proliferation and differentiation of hemopoietic cells i n isolated colonies (Reviewed i n (9)). These involve suspending the cells to be evaluated i n a semi-solid culture medium (eg. soft-agar, methylcellulose or plasma clot) supplemented with appropriate nutrients, serum (or the essential components thereof) and specific hemopoietic growth factors (although originally provided as poorly defined 'conditioned* media, many of the active molecules have now been purified, cloned a n d are available i n recombinant form). Since the colonies which grow i n these culture systems are derived from isolated single cells (5), the morphological properties of the colony (or clone) itself can be used to characterize the biological properties of its progenitor. Lineage-restricted colony-forming cells are distinguished by the number (colony size; indicative of proliferative potential) and ultimate type (lineage; indicative of differentiation potential) of progeny they c a n produce. The time at which mature progeny first appear allows further assignment of the position of the progenitor cell within a specific lineage; the longer the time after plating before mature cells appear, the more primitive the progenitor (10,11). The  -5 -  addition of different hemopoietic growth factors enables the simultaneous growth and differentiation of cells from all three myeloid lineages, and hence colonies containing cells from one or more of these can be obtained (12-14). Culture conditions that allow the growth of many of the various classes of lineage-restricted progenitors have been described for several mammalian species including rat, cat, dog, sheep, and cow as well as mouse and h u m a n (reviewed In (15)). Uni- or bipotent progenitors are the most frequent type of clonogenic cell encountered In hemopoietic tissue. These progenitors appear to have lost the capacity for self-renewal as they are unable to generate secondary colonies in replate assays (16). In the mouse, virtually all lineage-restricted clonogenic cells appear to be actively cycling under steady-state conditions since most are killed by In vivo exposure to drugs such as 5-fluorouracil (5-FU) that are selectively cytotoxic for cycling cells (17,18). T h u s clonogenic progenitors are considered to serve primarily as an amplification compartment to allow the production of a large number of end cells from a small number of active stem cells. Nevertheless, they still lack any unique morphological features and therefore are routinely detected by functional assays as described above. Progenitor cells giving rise to colonies consisting exclusively of granulocytes  and  monocytes/macrophages are called granulocyte-macrophage colony-forming units (CFU-GM). CFU-GM become further restricted to differentiate into either C F U - G or C F U - M , which give rise to colonies consisting solely of granulocytes or monocytes/macrophages respectively (19-21). Three distinct classes of erythroid progenitor cells have been identified (12). The primitive burst-forming unit-erythroid (primitive BFU-E) forms multi-focal colonies (bursts) containing many clusters of hemoglobinized red cells. Mature B F U - E also generate more than two (but usually less than ten) clusters of erythroblasts and these are detected several days earlier In culture. In the mouse, most primitive BFU-E and even some mature BFU-E are not erythroidrestricted as the colonies they produce often contain megakaryocytes, macrophages and mast cells (22). The third class of erythroid-restricted progenitors with the least proliferative  -6 potential is the colony-forming unit-erythroid (CFU-E) which forms small, single (or double) clusters of 8-64 hemoglobinized red cells after a relatively short period i n culture (23,24). CFU-E are the direct progeny of mature BFU-E which i n t u r n are derived from primitive BFU-E. Megakaryocyte progenitors have been classified into megakaryocyte burst (BFU-Mk) and megakaryocyte colony (CFU-Mk)-forming units i n a manner similar to the progenitors of the erythroid series (25,26). Progenitors of lymphoid cells have been more difficult to grow i n vitro and colony assays for pre-B or pre-T cells are not available. Modifications i n the culture conditions used for myeloid colony assays have, however, allowed the identification of both murine C F U - B (27) and C F U - T (28) which simply represent the activation and proliferation of mature B-cells and T-cells i n semi-solid media.  b)  Cells with the Characteristics of both Hemopoietic Stem and Progenitor Cells.  In vitro clonogenic assays also allow the identification of pluripotent hemopoietic cells. Various terms for such cells exist and reflect the historically coincident recognition of suitable culture conditions i n different centres (14,29,30). C F U - G E M M (colony-forming unit granulocyte/erythrocyte/macrophage/megakaryocyte) is a term first used to describe h u m a n pluripotent clonogenic cells before it was widely appreciated that many of the larger erythroid "bursts" produced by primitive murine BFU-E also contained cells of other lineages which could be elicited with appropriate stimulation. This i n t u r n led to the definition of the macroscopic burst-forming cell (B-macro or CFU-GEMM-macro) defined by its ability to generate very large erythroid colonies (of > 1 0  4  cells) which were then shown to consistently  contain differentiated progeny of at least two and often three different myeloid lineages 12-14 days after plating (22). Pluripotent hemopoietic cells thus demonstrate a high (but variable) proliferative and differentiative capacity i n vitro. In addition, i n the mouse, some are capable of extensive, albeit limited, self-renewal as demonstrated by the ability of some of their clonal progeny to form secondary and even tertiary (but not quaternary) multi-lineage colonies of  -7 equivalent size i n replate assays (22). This suggests that at least some multipotent hemopoietic cells do not display (at least under conditions of colony formation) the high self-renewal potential characteristic of very primitive hemopoietic cells i n vivo. Whether these latter differences are due to differences i n the conditions to which the cells are exposed, or represent intrinsic biologic changes related to their differentiated state is not yet known. The lymphoid differentiation potential of CFU-GEMM also remains controversial. Although some researchers have reported the presence of T-lymphocytes i n large mixed colonies of h u m a n origin (31), these results have not been confirmed and the possibility of contamination not ruled out. Colony assays have been used to identify a variety of clonogenic cell "types" with extensive proliferative capacity. Bradley and Hodgson (32) have described a murine high proliferative potential colony-forming cell (HPP-CFC) which is able to generate large colonies (containing 10^ cells) of macrophages i n cultures stimulated by a source of macrophage colony-stimulating factor (M-CSF or CSF-1) and synergistic factor(s), of which one may be Interleukin-3 (IL-3) (33). Like CFU-GEMM, a significant proportion of these cells are relatively resistant to the effects of 5-FU i n vivo (32,34). Although the extremely large colonies generated by HPP-CFC demonstrate their extensive proliferative capacity, their limited ability to form secondary colonies In replate assays and their apparent restriction to the monocyte/macrophage lineage appears to separate HPP-CFC from the most primitive stem cells. Recently a n assay for h u m a n HPP-CFC has also been reported (35). It must be noted, however, that the assay conditions used to identify HPP-CFC are not conducive to the terminal differentiation of the progenitors of either the erythroid or megakaryocytic lineages and may thus prevent determination of their true potential and/or their degree of overlap with the high proliferative potential progenitors defined by the colonies they generate under different culture conditions. Following cytotoxic ablation of a n animal's hemopoietic system by radiation or chemicals, normal hemopoiesis can be re-established if the animal is infused with normal hemopoietic cells from bone marrow, for example (36). Thus, transplantable hemopoietic stem cells  -8 -  obviously exist: the problem is how this procedure can be used to recognize, quantltate and characterize them. The spleen colony assay first described by Till and McCulloch l n 1961 (37) was the first i n vivo assay developed for the quantitation of a primitive hemopoietic cell. This assay Involves the injection of limiting numbers of hemopoietic cells into lethally irradiated or genetically compromised histocompatlble recipient mice. Some of the injected cells "seed" to the spleen where they give rise to isolated, macroscopic colonies visible between 8 and 14 days post-transplantation. These cells were thus termed colony-forming units-spleen or CFU-S. The number of spleen colonies produced is linearly related to the number of bone marrow cells transplanted (37) and cytological analysis of individual spleen colonies generated from marrow cells chromosomally marked by irradiation has directly demonstrated that they originate from a single cell (ie. are clonal) (38,39). These observations have recently been confirmed and extended through the use of other marking systems such as retroviral integration (40,41). The majority of spleen colonies isolated 12 to 14 days post-transplantation are large (containing 10^ to 1 0 cells) and include erythroid cells, granulocytes, and megakaryocytes (16,42,43) 7  mdicating the extensive proliferative and differentiative capacity of the cell of origin. Although lymphoid cells are not found i n such-spleen colonies (4), it is not known whether this Is the result of a n inherent restriction of spleen colony-forming cells to myelopoiesis, or the lack of a n appropriate microenvironment for lymphopoiesis provided by the region of the spleen which supports myeloid cell differentiation. The self-renewal potential of some spleen colony-forming cells has been demonstrated by the ability of individually excised primary colonies to produce new spleen colonies containing multiple myeloid cell types upon injection Into irradiated secondary recipients (44). The spleen colony assay thus includes the detection of some cells which meet the operational criteria of extensive proliferative capacity, multilineage myeloid differentiation potential and some selfrenewal potential. Recent studies have shown that spleen colonies are derived from different subpopulations of CFU-S present in normal adult bone marrow. Careful comparisons of spleen  -9 -  colonies enummerated early (day 8-9) or late (day 12-14) following transplantation have revealed differences in both their differentiative and proliferative capacities (42), in their selfrenewal potential (44) and in their immunological characteristics (discussed in more detail in a later section). Spleen colonies apparent after 8-9 days are thus now specified as day 8-9 CFU-S and considered to be restricted in their differentiative and proliferative potential because they generate smaller, transient colonies that usually contain only erythroid cells (45). The majority of day 8-9 CFU-S show no evidence of self-renewal when their progeny are transplanted into secondary recipients (44). In normal adult marrow, day 8-9 CFU-S show considerable turnover as evidenced by the inactivation of the majority of these cells following exposure to 5-FU (17,46). The mixed colonies derived from some CFU-GEMM (macro) contain day 8-9 CFU-S (47,48) suggesting overlap between the former and the precursors of day 8-9 CFU-S (see below). Spleen colonies apparent 12-14 days after transplantation are derived from cells referred to as day 12-14 CFU-S and, in normal adult marrow, these show greater proliferative and differentiative potential than do day 8-9 CFU-S. Because of these properties, the diminished sensitivity of day 12-14 CFU-S to ablation by cycle active drugs relative to day 8-9 CFU-S (17), and the ability of day 12-14 CFU-S to generate day 8-9 CFU-S (10). the former are viewed to be the more primitive precursor of the latter. Although it has been reported that pure populations of day 12-14 spleen colony-forming cells can rescue mice from lethal doses of irradiation (49), whether the spleen-colony assay provides a direct measure of the most primitive stem cells with long-term myeloid and lymphoid repopulatlng potential is not known. Indeed, recent studies indicate that these two functions are associated with largely separable cell types (50,51). Thus, while day 12-14 CFU-S are probably developmentally more closely related to the most primitive transplantable stem cells than are day 8-9 CFU-S, neither are now presumed to overlap completely with such cells (52,53). A more detailed discussion of this issue Is given in a later section. Nevertheless, CFU-S assays have been very useful for obtaining information about primitive hemopoietic cells and for many years this was the only method for quantitating  - 10 such cells i n heterogeneous suspensions. It is, however, pertinent to note that the CFU-S assay is limited by the fact that only a small proportion (the seeding fraction, f (44)) of the injected CFU-S reach the spleen and ultimately proliferate there to generate a macroscopically visible colony. The majority of CFU-S become entrapped i n other organs such as the lung and liver where they are thought to die or at least be lost from contributing to short-term hemopoiesis. The seeriing fraction has been determined by secondary transplantation of the spleen 2-48 hours after injection of the primary cell suspension (44) but estimates of its value vary considerably (from 3-25%) depending on the time interval between transplants, and the source of stem cells (eg. lower f factor for marrow cells from mice treated with cycle active drugs (reviewed i n 16,54, and 55)). Extrapolation from spleen colony numbers to total cells with spleen colony-forming potential is therefore always accompanied by some uncertainty. Nakahata a n d Ogawa (56,57) and others (58,59) have described a clonal i n vitro assay for the growth of a cell with stem cell-like properties similar to CFU-S. This so-called "stem" (S-cell) or blast colony-forming cell is identified on the basis of its ability to generate small colonies containing 20 to a few hundred undifferentiated blast cells detected up to 21 days after plating i n semi-solid medium. Blast cell colonies remain small and undifferentiated (the cells within them being characterized by pseudopodia a n d a motile phenotype) by the time most other colonies i n the culture have reached maturity and are beginning to lyse. Studies i n which cells are exposed to 5-FU i n vivo or ^H-thymidine In vitro suggest that this is due to the delayed transition of blast colony progenitors from G  0  into S-phase (60). This is supported by  antibody neutralization experiments which indicate that survival i n G  Q  does not require early  acting hemopoietic regulators (61). Once blast colony-forming cells enter S-phase they rapidly proliferate i n the presence of certain hemopoietic growth factors (62). Replating studies have demonstrated that both whole blast colonies and individual cells micromanipulated from blast colonies are capable of generating occasional secondary colonies of blast cells when transferred to fresh culture medium (56,57), suggesting some S-cell self-renewal within the primary colony. However, even at early stages, both primary and secondary colonies consist primarily of  -11 clonogenic progenitors of limited, albeit diverse, differentiation potentialities (63). Blast cell colonies also contain day 8-9 CFU-S but not day 12-14 CFU-S (57,64) suggesting that they may represent a subset of CFU-GEMM (macro). The extent to which blast colony-forming cells can differentiate down the lymphoid pathway remains to be determined, hence the relationship between cells i n this compartment and the stem cells capable of both lymphoid a n d myeloid repopulation i n vivo remains unresolved. Recently, the growth of h u m a n marrow-derived blast cell progenitors with self-renewal and differentiation potentialities similar to their murine counterparts has been reported (58). These thus appear to represent the earliest h u m a n pluripotent hemopoietic cell yet grown i n semi-solid culture. The use of In vitro clonogenic assays has been invaluable i n elucidating the changes that accompany hemopoietic cell differentiation, and i n the characterization of the hemopoietic growth factors whose presence is required for this to occur. Validation of the hierarchical model of the organization of the hemopoietic system has been approached by attempts to characterize the different progenitor cell types with respect to multiple parameters, and most definitively by cell separation experiments, to allow assignment of different properties to different cell populations regardless of assay conditions, and to allow direct precursor-progeny relationships to be demonstrated. Measurements of progenitor cell size, buoyant density, i n vitro nutrient requirements, reactivity with various antibodies and lectins, sensitivity to specific growth factors, normal cycling status and replating experiments have all provided evidence of differences between populations of cells that generate different types of colonies i n vitro (23). This is particularly well illustrated i n the erythroid lineage where progressive increases i n sedimentation velocity, cycling status, and sensitivity to erythropoietin (Ep) can all be seen as primitive BFU-E differentiate into mature BFU-E and ultimately CFU-E (18,23). In summary, although i n vitro clonogenic assays have been invaluable i n elucidating the relationships between the progenitors of the various myeloid lineages, the relationships of these cells to those with stem cell functions i n vivo remain less clear. This arises no doubt from the problems inherent i n the dependence on rather crude operational criteria to compartmentalize  - 12 -  cells that i n fact represent regions of a continuum of overlapping changes in self-renewal, proliferative and differentiative potentialities (Figure 2), and the focus of colony assays on the rapid generation of recognlzeable (mature) cells.  C)  HEMOPOIETIC C E L L S D E F I N E D BY LONG-TERM ASSAYS.  Evidence for a population of cells in adult marrow with lymphopoietic as well as myelopoietic repopulating potential was provided as early as 1968 using radiation-Induced chromosomal changes (65), and more recently retroviral Integration (40.41,66) to uniquely mark individual primitive cells and allow tracking of their clonal progeny. Such studies showed that cells repopulating lymphoid organs such as the thymus, lymph nodes and spleen of lethally irradiated mice carried the same unique marker as cells present i n Individual spleen colonies indicating that all were commonly derived from a more primitive stem cell. In spite of their multipotentiality and demonstrable self-renewal potential most day 12-14 CFU-S, like most pluripotent in vitro clonogenic cells, are no longer thought to have long-term repopulating ability. Reports of the purification of day 12-14 CFU-S to near homogeneity using a variety of physical and immunological procedures (discussed in a later section) has enabled an analysis of the extent of likely overlap between these compartments. For example, Visser et al. have described the isolation of a population of cells enriched > 100-fold in day 12 CFU-S which are also able to protect lethally irradiated mice for at least 30 days (49). Similar experiments by Spangrude et al. also demonstrate an in vivo repopulating ability of a population of alternatively purified cells determined to form late spleen colonies with unit efficiency (67). These results suggest a repopulating potential of one month for at least some day 12 CFU-S, although the presence of another minor hemopoietic cell type responsible for this would not have been excluded by the way these experiments were performed and the inherent difficulties in quantitating cells detected by different in vivo endpoints.  PROGENITOR CELLS  STEM CELLS  ^  CFU-G  ^  CFU-M  CFU-GM  MATURE END CELLS -•  GRANULOCYTE  HPP-CFC S-CELL  MONOCYTE/MACROPHAGE  CFU-GEMM—) TOTIPOTENT HEMOPOIETIC STEM CELL Day 12-14 CFU-S  CFU-E  BFU-E BFU-Mk Day 8-9 CFU-S  •  CFU-Mk  -•  MEGAKARYOCYTE/PLATELET  CFU-T •  -•  T-LYMPHOCYTE  CFU-B  Figure 2.  ERYTHROCYTE  B-LYMPHOCYTE  Schematic representation of hemopoietic cell maturity vs. self-renewal capacity, proliferative potential and cycling status. (Modified from Reference (52)).  CO  - 14 -  Evidence against such a relationship was already suggested by experiments performed in the 1960's and 1970's. Siminovitch et al. (68) and others (69,70) had shown that compared to normal marrow, a single transplantation causes a 3-7 fold decline i n the (initial) repopulatlng ability of a marrow graft and after four to six successive serial transplantations the ability of marrow cells to repopulate another recipient is reduced to essentially undetectable levels. Similar losses of repopulatlng ability c a n be seen i n mice subjected to several cycles of sublethal Irradiation (71), or In genetically compromised W/W^  mice that have been cured of  their anemia by transplantation of +/+ marrow (71). Despite the decline in long-term repopulatlng ability, an essentially normal short-term repopulatlng cell population may be present as measured by maintenance of normal numbers of day 9 and day 12 CFU-S In these mice (69,72). Conversely, W/W  v  mice clearly have stem cells but no assayable day 12 CFU-S  (73). Such examples illustrate how the definition of cells by performance in an assay may be too narrow or even misleading. Recent reports of the differential purification of day 12 CFU-S and more primitive cells (termed pre-CFU-S) whose frequency more accurately predicts marrow repopulatlng ability (50,51) further support the idea that most day 12 CFU-S should not be placed i n the most primitive stem cell compartment.  a)  Repopulation of Lethally Irradiated Mice.  Regeneration and maintenance of hemopoiesis following transplantation of cells Into mice Is the definitive assay for repopulatlng potential. Exposure to high doses of whole-body Ionizing radiation results In a time-dependant, logarithmic reduction in the number of surviving stem cells i n all hemopoietic organs (74). Cell death occurs as a result of unacceptable levels of unrepaired DNA breaks in some cases after one or two initial, faulty cell divisions. The proportion of stem cells killed in this way is further influenced by whether radiation Is administered i n a single or fractionated dose (75); the latter schedule enabling elimination of more stem cells while minimizing damage to other proliferating compartments  - 15 -  such as the gut due to corresponding differences In the abilities of these cells to repair sublethal damage (76). Irradiated recipients die rapidly (within 2 weeks) due to infection and hemorrhaging in the absence of a protective graft containing histocompatible. repopulatlng stem cells. By transplanting graded numbers of marrow cells, the frequency of stem cells able to rescue a proportion of the irradiated recipients can be determined. End-points such as the number of marrow cells required to promote the 30 day survival of 5 0 % of transplanted recipients are commonly used (74), although more sophisticated measures of repopulatlng ability such as the number of reconstituted marrow CFU-GM or CFU-S (13 days post-transplant) per cell injected have also been reported (77). Historically, these studies have generally yielded estimates of the frequency of repopulatlng stem cells In normal adult marrow of between 1 In 1 0  4  and 10^ total nucleated cells.  A significant difficulty in Interpreting most reported 30 day survival data is that the contribution of the transplanted cells has not usually been unambiguously and concurrently established. For example, even after "lethal" doses of irradiation regenerated hemopoiesis may be of host as well as donor origin. Hence survival is not necessarily indicative of long-term engraftment by donor cells. The presence of donor cells is best determined using appropriate genetic markers. In mice, such markers have included random radiation-induced chromosomal changes (that do not affect viability or differentiation) detected by cytogenetic analysis of metaphases generated in vitro from repopulated tissues (78), sex differences detected by cytogenetics or Southern analysis using a Y-chromosome specific probe (79), congenlc allotype markers (80) or naturally occurring isoenzyme or hemoglobin variants (81,82) distinguishable by biochemical analysis of repopulated tissue extracts. The restricted expression of some markers, such as hemoglobin, to specific lineages or stages of differentiation, and the limited numerical spectrum of other markers has (at least until recently) restricted the utility of most of these techniques. In general they have thus been used only to examine the dynamics of clonal expansion by transplanted populations which contain many cells with the potential to differentiate into the cell type that can be  may  - 16 -  distinguished. The recent development of methods for irifecting pluripotent hemopoietic stem cells i n vitro with retroviruses overcomes many of the above problems (83). By analysis of DNA  molecular  from the progeny of transplanted, infected cells, the capacity of single cells to  repopulate multiple lymphoid and myeloid lineages in vivo has been established (40). The random, multiple integration pattern of the provirus provides the added advantage of allowing the proliferation of many different, uniquely marked stem cells to be followed simultaneously. The main limitation is the inability of single cells to be tracked by this type of marking, at least with currently available vectors. Thus lineage analysis requires the use of additional methods to obtain purified populations of a given hemopoietic cell type. Studies i n which fixed numbers of bone marrow cells are serially transplanted at 8-10 week intervals into lethally irradiated recipients have suggested that even one transplant  may  be a m u c h more severe stress on the hemopoietic stem cell population than that which occurs normally during the lifetime (84). Much larger reductions in repopulating ability are seen with increasing numbers of serial transplantations than with increasing age (69). Since studies using serial transplantation of spleen or marrow cells into irradiated hosts also result i n a dramatic loss of CFU-S (44), others have looked for assay systems that do not require heavy irradiation. One such system is the W/W  b)  Repopulation of W/W  v  v  mouse mutant.  Mice.  Mice carrying mutations at the White Spotting (W) locus (located on chromosome 5) are characterized by severe macrocytic anemia, lack of hair pigmentation and sterility (85). Effects of mutations at this locus are thus pleiotropic, and i n part may be attributed to interruptions in the proliferation and/or migration of cells early i n embryogenesis (86). The effect in the hemopoietic system is not only an anemia, but also an intrinsic defect i n the earliest stem cell types such that W/W  v  mice do not contain cells that will give rise to macroscopic spleen  colonies in a CFU-S assay, but will support the growth of wild-type spleen colonies even in the  - 17 -  absence of any treatment, and will eventually have all of their blood cells replaced by wild-type cells following Injection of the latter (87,88). The anemia of W/W  v  mice can thus be  permanently cured by grafting histocompatible normal (+/+) marrow stem cells. The W locus Is now thought to be the same as that encoding the c-kit proto-oncogene which encodes a receptor-like surface component (89,90), although the significance of this recent finding is not yet clear. A significant advantage of the W/W  v  mouse is that its use as a recipient avoids the  unpredictable effects of irradiation on survival and host stem cell recovery. A s mentioned above, studies In W/W  v  mice also support the Idea that day 12-14 CFU-S are not true stem  cells. Boggs et al. (91) used a limiting dilution method to show that the number of stem cells with the proliferative capacity required to cure W/W  v  mice was fewer than that which would be  predicted from CFU-S numbers. Differences In the frequency of the cells i n these compartments was also found to be more pronounced when 5-FU treated marrow was used as the source of +/+ stem cells (92).  c)  Long-Term Bone Marrow Cultures as an Assay for Stem Cells.  The availability of in vivo assays for the study of murine hemopoiesis has added physiological relevance to the concepts described but these assays remain poorly suited to the sort of manipulation required for controlled studies of hemopoietic stem cell regulation. Analysis of the specific cellular and molecular events underlying the control of hemopoiesis may be better served by the use of well-defined in vitro systems. In 1977, Dexter et al. (93) described for the first time the development of a liquid culture system which if maintained at 33°C allowed murine pluripotent stem cells to be maintained for several months In the absence of any exogenously supplied hemopoietic growth factors. By seeding bone marrow cells at high density ( 1 0 cells/mL) In a medium containing horse serum (and now also hydrocortisone) i n 6  addition to the fetal calf serum normally present i n semi-solid cultures, a confluent adherent  - 18 -  layer of flbroblastoid cells formed within several weeks. Cultures were then "recharged" by the addition of fresh marrow and subsequently "fed" by weekly replacement of half of the medium with fresh, similarly supplemented medium. The concomitant removal of half of the nonadherent cells provided an opportunity for their ennumeration, histochemical analysis or assessment i n in vitro or in vivo clonogenic assays. Since its Initial description, this long-term bone marrow culture (LTBMC) system has been studied extensively. The adherent "stromal" layer is now known to contain a variety of mesenchymal elements including fibroblasts (94), endothelial cells (95), adipocytes (96) and their extracellular components (collagens I, III and IV (94,97). laminin (98) and fibronectin (99)). Together these provide a complex matrix in which more primitive hemopoietic cells as well as macrophages are enmeshed (100) i n a fashion thought to simulate the interactions of these components with the marrow stroma in vivo. Such interactions appear to be important i n this system both for the maintenance of the primitive cells and their sustained output of mature myeloid cells (predominantly granulocytes and macrophages) (101). By comparing the initial number of hemopoietic cells seeded into the cultures with the number present i n both adherent and non-adherent fractions at intervals thereafter, both differentiation and amplification of various progenitors has been demonstrated (100). In addition to supporting the proliferation of CFU-GM. these cultures have also been demonstrated to amplify or maintain more primitive cells such as CFU-GEMM (macro) (102) and day 12 CFU-S (103). the cells responsible for competitive erythropoietic repopulation (104), and the cells capable of repopulating both B and T-lymphoid systems i n vivo (105). Although differentiation of mature granulocytes and macrophages occurs continuously, differentiation along all other lineages, including the erythroid, megakaryocyte, and either of the lymphoid pathways is not normally supported i n the L T B M C system as originally described. In 1980, similar conditions were found to allow the long-term maintenance of clonogenic progenitors and granulopoiesis i n cultures established from h u m a n bone marrow samples which behave like and appear to resemble murine L T B M C as far as can be determined (106-108).  - 19 -  Hemopoiesis In L T B M C is associated with the mesenchymal cells of the adherent layer i n which "foci" of differentiating hemopoietic cells (referred to as "cobblestone areas" (106)) may be seen. Quantitation of the most primitive clonogenic cells requires the physical or enzymatic dissociation of this network to obtain a single cell suspension suitable for assay. Studies of the turnover of hemopoietic cells i n the adherent layer of both murine and h u m a n L T B M C (108) have revealed a behaviour and regulation that may reflect the control of hemopoiesis i n vivo. Assessment of the cell cycle status of progenitors In the non-adherent and adherent fractions using the ^H-thymidine suicide assay to measure the proportion of cells in S-phase (109) has revealed that day 8-10 CFU-S oscillate between a cycling and non-cycling state dictated by the perturbation of the cultures associated with each weekly medium change (110). All CFU-S were found to enter S-phase within 1-2 days of feeding and. if the cultures remained undisturbed, returned to a non-cycling state by the time of the next feeding 5-6 days later. More mature progenitors such as CFU-GM remained continuously i n cycle in these cultures (110). Studies using h u m a n L T B M C s have revealed a similar pattern of alternating proliferation and quiescence. This was exclusive to the most primitive progenitor classes (CFU-GM-derived colonies of >500 cells, and primitive BFU-E-derived colonies of >8 erythroblast clusters), although In this case the pattern was confined to these progenitors only in the adherent layer (111). Moreover a change In the cycling status of these cells could not be obtained by agitating the culture or simply removing and replacing the pre-existing medium (so-called "mock feeding"). Rather, it was found that in human L T B M C activation of the primitive progenitors in the adherent layer requires the addition of a factor present In fresh horse serum (112). The addition of horse serum can, however, be duplicated by the addition of other factors such as Interleukin-1 (IL-1) and platelet-derived growth factor (PDGF) to mock-fed cultures (112). This suggests their Indirect role as mesenchymal cell activators which could reflect how perturbation of the marrow microenvironment might regulate hemopoiesis i n vivo. Since the maintenance and regulation of the various populations of progenitors detected in h u m a n L T B M C s closely mimics that seen i n murine L T B M C where stem cell maintenance  - 20 can be demonstrated to occur for several weeks (113), the use of L T B M C may offer a strategy for quantitatlng hemopoietic stem cells by virtue of their expressed proliferative activity i n vitro (114,115). The recent adaptation of h u m a n L T B M C for the purging of Philadelphia (Ph *) chromosome-positive leukemic cells from the marrow of patients with chronic myelogenous leukemia (CML) prior to the clinical use of such marrow for autologous bone marrow transplantation has suggested that h u m a n stem cells with at least short-term regenerative potential c a n be maintained i n vitro for at least 10 days (116). It seems likely that the hemopoiesis obtained i n h u m a n L T B M C reflects the maintenance a n d differentiative activity of a very primitive hemopoietic cell. Recent purification of h u m a n hemopoietic cells capable of generating clonogenic cells measured after 5 weeks i n L T B M C (the so-called "long-term cultureinitiating cells") (117) should enable more direct assessment of the various types of very primitive h u m a n hemopoietic cells.  d)  Evidence for H u m a n Repopulating Stem Cells.  The lack of comparable i n vivo assays for the long-term repopulating stem cell i n humans has made the study of hemopoiesis i n m a n difficult. Nevertheless, several lines of evidence support the existance of lympho-myeloid repopulating stem cell populations analogous to those identified i n the murine system. Studies of a number of females heterozygous for the Xchromosome linked gene encoding the enzyme glucose-6-phosphate dehydrogenase (G6PD), who also h a d a hematologic malignancy such as sideroblastic anemia (118) or C M L (119) have demonstrated the pluripotent stem cell origin of the neoplastic clone, including the involvement of lymphoid as well as myeloid cells. More recently, analysis of differentially methylated restriction fragment length polymorphisms i n X-linked genes such as those encoding the enzymes hypoxanthine phosphoribosyltransferase (HPRT) and phosphoglycerate kinase (PGK) have indicated a monoclonal origin of the normal (donor-derived) hemopoiesis (including lymphoid as well as myeloid cells) obtained i n two bone marrow transplant patients who h a d  been Injected with allogeneic marrow (120). It therefore appears that hemopoietic stem cells with repopulatlng ability comparible to those identified in the murine system are present i n adult h u m a n marrow as well.  2)  REGULATION O F T H E HEMOPOIETIC SYSTEM.  Clearly a system which possesses such a large capacity for proliferation and differentiation must be subject to a multiplicity of regulatory signals which mediate a n appropriate numerical output of each of the mature cell types. This regulation must enable the system not only to respond to hematologic and Immunologic stresses by the activation of primitive stem cells, but also by balancing the production of mature cells In a lineage-specific manner. Four general classes of potentially regulatable stem cell transitions are recognized: a) quiescence versus proliferation, b) self-renewal versus differentiation, c) restriction of differentiation potential, and d) viability versus death. For each of these transitions, the possibility of control mediated by mechanisms intrinsic to the stem cell (eg. genetic), and/or extrinsic factors present i n it's microenvironment, exist. With respect to the latter, the mechanism of control may be directive, itself specifying the decision made by the stem cell, or permissive, merely allowing the expression of a decision determined by some alternate mechanism. At later stages of differentiation, control of proliferative state, choice of differentiation potential and self-renewal may no longer be options for the cells and regulation would then depend exclusively on the control of cell viability. A variety of hemopoietic growth factors are known to exist and interact, at times i n a synergistic fashion, to modulate the output of differentiated cells from cells detected i n clonogenic assays (reviewed i n (4)). The growth regulatory properties of such factors requires that they, i n turn, be regulated so as to prevent situations of unrestricted growth. Indeed, examples of imbalances i n regulatory processes have been associated with leukemic transformation, as well as a number of other hematologic and immunodeficiency disorders. In  - 22 -  addition to regulation by soluble growth factors, hemopoietic cells may be regulated through direct interactions with bone marrow cells. This is indicated i n vitro through the use of longterm cultures. In the next section some of the evidence for these mechanisms will be reviewed.  A)  R E G U L A T I O N BY C E L L U L A R INTERACTIONS.  Since early observations that the cellular composition of individual spleen colonies varies with their specific location i n the spleen, and the subsequent proposal of the HemopoieticInductive Microenvironment (HIM) model by Curry and Trentin i n 1967 (42) the stroma has been implicated i n the regulation of pluripotent stem cells. Hemopoietic cells have been observed to be physically associated with stromal cells i n several i n vivo and i n vitro settings. Ultrastructural analysis of the organization of bone marrow has revealed a number of consistent and specific localizations for both mature cells such as megakaryocytes (typically associated with parasinal endothelial cells) (121), and primitive progenitors such as  CFU-  G E M M and CFU-S (the most primitive of which are found associated with mesenchymal cells lining the endosteum) (122,123). A preferential binding and subsequent differentiation of a proportion of primitive BFU-E, but not CFU-GM, on cloned fibroblastoid stromal cell lines has been demonstrated and this interaction appears to be mediated through fibronectin (FN) and i s inhibitable by anti-FN antibodies (124). The IL-3 dependent multipotential stem cell line FDCP-mlx has been shown to proliferate and differentiate along several myeloid lineages i n the absence of IL-3 when allowed to infiltrate irradiated long-term culture adherent layers (125). This suggests a supportive role of the stroma i n this system perhaps i n part by directing a localized supply of extracellular matrix (ECM) or cell surface bound growth factors. The finding that some growth factors can be membrane components (eg. M-CSF) (126), or can be bound to ECM  (eg. GM-CSF) (127) provides further support for this idea. In mice, the transplantation of  small fragments of bone marrow to heterotopic sites such as the kidney or s k i n of histocompatible recipients results i n the subsequent colonization of the implant with  - 23 hemopoietic cells of recipient origin (128). Thus, while the fact that multipotential hemopoietic progenitors (CFU-GEMM) differentiate i n semi-solid cultures i n the absence of a structurally intact rnicroenviroriment suggests that direct interactions with stromal cells are not obligatory for either stem cell self-renewal or commitment, such interactions do appear to play a major role i n the regulation of early hemopoietic cells i n vivo. The long-term culture system is particularly amenable to the study of the role of the microenvironment i n hemopoiesis. Studies which demonstrate both a differential localization (108) and activation (110,111) of primitive progenitors i n the adherent layer of both murine and h u m a n L T B M C s suggest a potentially important role for the stroma. Similar associations are found to take place i n long-term cultures which have been modified to support the growth of B-lymphoid progenitors (129). However, the ability to 'switch' the culture conditions of established myeloid (so-called "Dexter") long-term cultures so that they support B-lymphopoiesis (so-called "Whitlock-Witte" cultures) and not myelopoiesis (130,131) indicates that the cells i n the bone marrow that are responsible for maintaining these various lineages are not themselves restricted for this role developmentally, but rather may respond to external conditions that alter their phenotype and function leading to an apparent change i n the microenvironment. Studies of the supportive capacity of the marrow stroma have been well served by the availability of mouse mutants with a hemopoietic microenvironmental defect. Mice carrying certain mutations at the Steel (SI) locus are characterized by severe macrocytic anemia, lack of hair pigmentation, increased sensitivity to radiation and sterility (16). Steel-Dickie (Sl/Sl^) mice contain hemopoietic stem cells which are essentially indistinguishable from normal  +/+  cells with respect to their radiosensitivity, spleen colony-forming capacity and their ability to restore normal hemopoietic function i n lethally irradiated normal or unirradiated W/W  v  mice.  However, when S l / S l ^ mice are used as recipients of +/+ stem cells, the transplanted cells do not form spleen colonies and normal hemopoiesis is not restored (132). This can only be accomplished by the subcutaneous transplantation of whole +/+  (or even W/W ) v  bone marrow  - 24 -  or splenic tissue (133). Such findings indicate the defect In the S l / S l mouse to be associated a  with the hemopoietic microenvironment. Transplanted S l / S l tissue Is maintained In +/+ d  recipients but the graft fails to support erythropolesis (134). These findings indicate a defective Interaction between S1/S1°- matrix and normal erythroid precursors.  B)  REGULATION BY SOLUBLE GROWTH FACTORS.  The differentiation of stem cells into committed progenitors that subsequently proliferate and mature into the blood cells that are released into the circulation are all processes that may be regulated by a group of growth factors known collectively as colony-stimulating factors (CSFs and/or Interleuklns). Hemopoietic growth factors (except for erythropoietin) were initially described in the early 1960's with the first successful reports of techniques for the clonal growth of murine bone marrow cells in vitro. Since that time, our limited understanding of "biological activities' (or so-called colony-stimulating activities (CSA)) present in crude conditioned media has given way to the discovery of a large spectrum of factors. Many of these have now been purified to homogeneity, characterized, cloned and expressed as recombinant proteins. A number of general observations can be made about the nature of the hemopoietic growth factors (HGFs) and their biological specificities (Reviewed in (4), and summarized in Table 1). All the HGFs are relatively small (subunit Mr 14-40 kD) monomeric acidic glycoproteins (except macrophage-CSF (M-CSF) which Is a homodimer). Although carbohydrate is not necessary for the biological activity of most of the HGFs, it may enhance their solubility and stability. Most HGFs contain one or more disulfide bonds which are essential for bioactivity (135). The HGFs can have multiple levels of biological activities, being necessary for cell survival, proliferation, differentiation and maturation. Some of their activities appear to reflect the hierarchical organization of the target cells they stimulate, and all show a surprising degree of overlap In the range of cells they can act upon. Some are essentially lineage-restricted In  Table I. Hemopoietic Growth Factors (CSFs and Interleuklns)  Name  Abbreviations  Other Common Names  Major Hemopoietic Lineage Stimulated  Erythropoietin  Epo  Erythroid  Macrophage colony-stimulating factor  M-CSF, CSF-1  Monocyte/macrophage  Granulocyte colony-stimulating factor  G-CSF, CSF-a  Neutrophil  Granulocyte-macrophage colonystimulating factor  GM-CSF, CSF-P  Pluripoletin  Neutrophil/macrophage  Interleukin-1 alpha  IL-1 a  Hemopoietln-1 (H-l)  Co-stimulator of early cells, T cells  Interleukin-1 beta  IL-1 P  Lymphocyte activating factor  Co-stimulator of early cells, T cells  Interleukin-2  IL-2  T cell growth factor  T cells  Interleukin-3  IL-3  Multi-lineage colony stimulating factor (multl-CSF), persisting cell factor, hemopoietln-2 (H-2), hemopoietic cell growth factor (HCGF), mast cell growth factor (MCGF)  Most myeloid lineages  Interleukin-4  IL-4  B cell stimulating factor-1 (BCSF-1), B cell differentiation factor  B cells, T cells  Interleukin-5  IL-5  T cell replacing factor, B cell growth factor-2 (BCGF-2), B cell differentiation factor  eosinophil differentiation, B cells  Interleukln-6  IL-6  B cell stimulating factor-2 (BCSF-2), hybridoma growth factor, plasmacytoma growth factor  B cells, T cells, co-stimulator of early cells  Interleukln-7  IL-7  Interleukin-8  1L-8  Interleukin-9  IL-9  Pre-B cells, T cells Monocyte-derived neutrophil chemotactic factor (MDNCF), monocyte-derived neutrophil-activating peptide (MONAP), NAP-1  Neutrophils, T cells  Early erythroid  to Ol  - 26 -  their actions as single factors and support the terminal differentiation of particular unl-potent progenitor cells (eg. M-CSF (136), granulocyte-CSF (G-CSF) (137) and erythropoietin (Epo) (138)), whereas many can also stimulate multipotential hemopoietic cells and mediate amplification of more primitive compartments either alone or in combination with other factors (eg. G-CSF (139). granulocyte-macrophage-CSF (GM-CSF) (140) and Multi-CSF (or IL-3) (141)). The types of progenitor cells that are stimulated to proliferate and differentiate by each H G F are also dependent on its concentration (135). Many of the various actions of the crude preparations of H G F s have been confirmed for the purified native or recombinant, and i n some cases nongrycosylated, forms of the molecules allowing definitive assignment of specific actions to particular molecules. For example, analysis of single developing cells micromanipulated from highly self-renewing murine blast cell colonies (57), and highly purified progenitor cells plated in semi-solid cultures at very low density have shown that IL-3 (142), GM-CSF (143), and G-CSF (144) all act directly on a variety of target cells, including pluripotent as well as lineage-restricted progenitor cells. Recent investigations of the in vivo biological activities of the H G F s in both primates and m a n have revealed effects consistent with those determined in vitro; intravenous administration of GM-CSF leads to elevations of granulocytic and neutrophilic leukocytes (145), and intravenous administration of Epo leads to a n increase i n hematocrit (146). There are clear h u m a n counterparts of all of the murine H G F s investigated to date and some, like G-CSF (147) and Epo (148). retain two-way biological cross-reactivity between the two species, although many, like GM-CSF and IL-3. do not (149). H G F s are extremely potent molecules, active at picomolar to nanomolar concentrations (150). They are either not detectable or normally present at very low concentrations in the serum, and are also not readily detected in the medium of established L T B M C s (151). However, the recent demonstration of the association of GM-CSF with glycosaminoglycans in the E C M of L T B M C adherent layers (127) illustrates how compartmentalization of H G F s may occur both in vitro and in vivo. For each H G F there appears to have evolved a single specific  - 27 -  class of high affinity cell surface receptor. Receptor densities are relatively low for G-CSF and GM-CSF (152), IL-3 (153) and Epo (154) on all normal hemopoietic cells thus far exarnined, although higher M-CSF receptor levels are found on macrophages (155). Somewhat anomalously, elevated receptor levels for certain growth factors have been found on various cell lines, and this has proven useful for attempts at receptor isolation and characterization (eg. for IL-3 receptor on B6SUtA cells (156)). The distribution of receptors on normal hemopoietic cells is consistent with the known spectrum of biological specificities of the H G F s (147,157,158). Binding of the H G F s to their receptors is essentially irreversible at physiological conditions, and the HGF-receptor complex is rapidly internalized and, i n most cases, degraded (135). Signal transduction appears to involve several complex mechanisms and may be associated with the ATP-dependent phosphorylation of a variety of membrane associated macromolecules, and their subsequent conversion between active and inactive forms, in a growth stimulatory cascade which culminates in the replication of DNA (Reviewed in (159)). Several of the growth factors or their specific cellular receptors that are implicated in the regulation of hemopoiesis appear to be related to known retroviral oncogene products. The receptor for M-CSF is identical to the product of the proto-oncogene c-fms (160); the B-chain of platelet-derived growth factor (PDGF) is identical to the proto-oncogene c-sis (161); and the product of the v-erb-B oncogene represents a constitutively active, truncated form of the epidermal growth factor (EGF) receptor (162). Moreover, in a number of experimental settings, it has recently been demonstrated that the incorporation of a growth factor (eg. GM-CSF) or receptor gene (eg. for EGF) into a retrovirus (where transcription is deregulated by placing it under the control of a promoter i n the retroviral long terminal repeat (LTR)) can confer oncogenic properties onto the gene (162,163). Spontaneous or induced activation of autocrine growth factor production has recently been documented i n a variety of malignant hemopoietic cells of both mouse and human origin (163,164). Interestingly, the genes encoding IL-3. IL-4. IL-5, GM-CSF, CSF-1, and the receptors for CSF-1 (c-fms) and P D G F are all clustered on the distal portion of the long-arm of human chromosome 5 (165,166). This region is often partially  - 28 deleted In several hemopoietic disorders Including a form of myelodysplasia a n d some cases of de novo acute myelogenous leukemia (AML) (165,167), Illustrating the possible existence of complex cytokine linkage groups and potentially deleterious effects of their aberrant regulation. Many HGFs, i n addition to their effects o n hemopoietic cells as single agents, also display a varied array of synergistic activities when present together with one or several other cytokines (168). For example, it has been reported that some H G F s which act alone on one population of target cells to induce their proliferation can act i n combinations on another population of target cells to induce the expression of receptors for H G F s to which these cells are normally unresponsive.  Specific examples of this are as follows. IL-1 synergizes with IL-3 a n d IL-5 to  induce the proliferation of CFU-GEMM and eosinophil progenitors (169). IL-3 can maintain the proliferation of pluripotent hemopoietic cells once they have entered S-phase (61), and synergizes with CSF-1, G-CSF and Epo to increase their effects on the proliferation of more mature progenitors (168). IL-4 (B-cell stimulatory factor-1 (BCSF-1) can induce the proliferation of B and T-cells, but can also synergize with G-CSF, Epo and IL-1 to Induce the proliferation and differentiation of CFU-GEMM, CFU-GM, BFU-E, CFU-E and CFU-Mk respectively (170,171). D>6 (BCSF-2) enhances the proliferation of activated B-cells and CFU-GM (172), but can also synergize with IL-3 to stimulate the proliferation of pluripotent blast cell progenitors (62,173). In addition to the H G F s which act directly on clonogenic cells, there are several factors which are produced i n association with the generation of a n inflammatory response and which do not i n themselves have colony-stimulating activity but can induce the production of H G F s by other cells (eg. fibroblasts, endothelial cells and macrophages). Several examples are IL-1 (174,175), P D G F (176), a n d tumor necrosis factor-a (TNF-a) (177). Even these may be induced; the treatment of endothelial cells and macrophages w i t h endotoxin or phorbol esters results In the production of P D G F which then i n t u r n acts o n various connective tissue cells to activate the production of E C M components, and the secretion of H G F s including GM-CSF (178). This illustrates how complex cascades of regulatory events can be Initiated. TNF-a is produced by  - 29 -  activated monocytes and acts on monocytes and endothelial cells to stimulate the production and release of G-CSF and GM-CSF (179). These studies lend support to the hypothesis that some factors (such as IL-1) whose activities have been described as synergistic may i n fact not reflect true synergy at the target cell level, but rather the activation of other H G F s by accessory cell populations leading to the enhanced proliferation of the target cells studied. S u c h indirect effects are readily produced i n unfractionated target cell populations since these contain significant numbers of accessory cells. T h u s investigations of the biological activities of any factor is likely to be influenced by the degree of purity of the target cell population obtained. The use of defined, highly purified stem and progenitor cell populations i n such studies is thus strongly indicated. In summary, it is clear from the above discussion that the bone marrow is a very heterogeneous hemopoietic organ composed of a variety of clonogenic and stromal cell populations which interact both through physical associations a n d the extracellular release of specific cytokines to regulate the production of blood.  3)  T H E ISOLATION OF HEMOPOIETIC CELLS.  Hemopoietic repopulating stem cells and i n vitro clonogenic progenitors are present i n normal adult mouse marrow at a frequency of approximately 1 i n 1 0 to 10^ cells. Unlike the 4  more mature end cell precursors and all the terminally differentiated cells i n the blood, neither possess any unique morphological characteristics which enable them to be directly distinguished from each other or from the many other cell types present i n bone marrow (eg. medium-sized lymphocytes). Indeed, the identification of hemopoietic cells since their discovery has been retrospective; dependent on the morphology of the progeny they generate i n vitro and i n vivo. The very nature of these assays, however, dictates that once a stem cell is identified it is no longer a stem cell, often having through the generation of it's clone exhausted the very properties which originally made it unique and interesting. Experiments to examine  -30-  the critical events i n the decision between self-renewal and committment, and during the initial stages of differentiation are therefore difficult, and at best cumbersome, to design. Direct analysis of the molecular nature of these early events, and further definition of the properties of cells in the earliest hemopoietic compartments would be greatly facilitated by methods that would allow isolation of highly purified suspensions of biologically homogeneous, primitive hemopoietic cells. The availability of such populations would also greatly facilitate studies of the mode of action of the various hemopoietic growth factors, and the role of cells of the marrow microenvironment in stem cell regulation.  A)  S T E M C E L L PURIFICATION.  Many procedures now exist for the enrichment of both h u m a n and murine hemopoietic stem cells. Historically, attempts at purification were based on exploiting differences i n the physical properties of cells in various stages of differentiation. Common approaches included separation by differences in buoyant density (180), sedimentation velocity (181), and adherence to glass beads or plastic surfaces (182,183). The problem with all of these parameters is that their differences between different early hemopoietic cell populations are insufficient to form the basis of a useful preparitive enrichment strategy when used alone. Additional strategies involving pretreatment of donor animals with cytotoxic drugs (184) or endotoxin (53) prior to marrow collection have helped to provide starting populations that are already more enriched i n the cells of interest. These studies have allowed the general characterization of the more primitive hemopoietic cells (including CFU-S) in normal marrow as relatively small with an undifferentiated "blast" cell morphology. Although none of these techniques used alone has resulted i n a purity or enrichment sufficient for useful preparative applications, some, such as equilibrium density centrifugation, are commonly used today in conjunction with other separation procedures (49). Indeed, the current use of counterflow centrifugal elutriation (CCE) to separate cells on the basis of size (185) reflects the potential of high resolution  - 31 -  methods for separating primitive hemopoietic cells based on their physical properties. Murine CFU-E (186), CFU-GM (184) and CFU-S (187) have all been purified to near homogeneity from various sources using CCE. The development of monoclonal antibodies (188) offered a variety of alternative and complementary methods for cell separation that exploited the features of antibody molecules. These include complement-mediated cytotoxity (34). inimune-rosetting (189), irrmunoadherence (panning) (190), affinity chromatography (191), and more recently immunomagnetic depletion (192) and fluorescence activated cell sorting (49,67,193). Using such procedures most approaches to the purification of cells can be categorized into one of two strategies (Figure 3). In one case, the cells of interest are labelled and hence isolated from the remainder of the mixture. This is referred to as  positive selection and its use requires a  knowledge of the properties (antigenic determinants) unique to the cells of interest. For recovery to be high all the cells of interest need to be phenotypically identical, at least with respect to the expression of the antigen to be labelled. In practice, this is rarely the case since the continuous morphological and functional changes associated with differentiation results in phenotypic heterogeneity among cells with similar biological properties. As yet, no antigens have been identified which are uniquely expressed on all cells of any single primitive hemopoietic cell compartment. A n alternative approach is to label and hence remove all of the cells other than those of interest. This is referred to as  negative selection. For hemopoietic stem cell purification, this  requires a panel of antibodies directed against antigens expressed on hemopoietic precursors and functionally mature end cells but not on stem cells. The advantages of this strategy are that stem cell-specific monoclonal antibodies (or even antigens) need not be identified, and that the purified hemopoietic cells are not coated with any molecules which may influence their behaviour in vivo. For example, labelling of bone marrow cells with conventional indirect antibody complexes is known to reduce the seeding efficiency of CFU-S by approximately  - 32 -  HETEROGENEOUS STARTING POPULATION  Figure 3.  Schematic representation of positive and negative selection approaches to stem (S) cell purification.  - 33 -  two-fold (194). This effect on the In vivo behaviour of positively selected stem cells can also pose practical problems to studies involving the transplantation of limiting numbers of these cells. Although immunological separation procedures have proven to be a significant advance over those based exclusively on physical separation, again the likelihood that primitive hemopoietic cells probably represent a continuum of phenotypic changes not necessarily always co-ordinated i n the same way, makes it unlikely that the presence or absence of a single antigen could allow adequate enrichment of primitive hemopoietic cells from in vivo sources. However, the ability to quantitatively separate cells according to the relative level of cell surface expression of a given antigen, such as was made possible by the development of the fluorescence activated cell sorter (FACS), has offered a significant advance. Figures 4A and B illustrates the principle of flow cytometry and electronic cell sorting using the FACS. Flow cytometers allow one to make rapid and simultaneous, quantitative measurements of multiple parameters of individual cells. In addition these measurements can be combined to select subpopulations of cells with particular combinations of features for sorting. Typically used parameters include size (forward light scatter (FLS)), shape and internal structure (90° or perpendicular (PLS) or orthogonal (OLS) light scatter) (195), cytoplasmic granularity, viability, pigment content (eg. hemoglobin), DNA  and RNA  content, DNA  synthesis, chromatin structure,  redox state, membrane integrity, enzyme activity, endocytosis, surface charge, mitochondrial content, intracellular pH and/or expression of various cell surface molecules labelled with specific fluorochrome-conjugated probes (196). Since the F A C S inspects cells individually prior to sorting, it is possible to isolate defined, highly purified populations of rare cells from very heterogeneous tissues such as bone marrow for analysis of colony-forming or long-term repopulatlng ability. The major disadvantage of the F A C S is, however, also based on this same principle of consecutive analysis and sorting of individual cells. T h u s i n spite of the rapidity with which these processes can be performed, handling of very large numbers of cells (>10^)  Figure 4A.  The major components of the Fluorescence Activated Cell Sorter (FACS).  - 35 -  Figure 4B.  General principles of operation of the FACS. 1)  Sample containing cells (A) is injected by air pressure into sheath fluid (saline) (B) to form a coaxial stream.  2)  Flow toward nozzle outlet (C) compresses stream into an exiting 50 uM fluid jet with cells i n "single-flle".  3)  Viewing through a microscope, laser beam (D) is focussed with lenses (E) mm below nozzle tip.  4)  Ultrasonic nozzle vibrator (F) (40 kHz) forms droplets (40,000/sec) 3 mm tip.  5)  Reflected laser light is focussed, split and filtered as it enters up to five signal detectors (G) connected to a computer. These measure fluorescence at several emmission wavelengths (FL1, FL2, etc.), as well as forward (FLS) (related to cell volume) and orthogonal (OLS) (related to cell shape and internal structure) light scatter. Fluorescent light thus generates an electrical signal that is proportional to the number of fluorescent molecules (antibodies) on each cell.  6)  Multiparameter analysis of each cell as it passes through the laser beam is performed by the computer.  7)  Cells satisfying preselected sort criteria are appropriately charged by a computer-generated signal (H) after travelling from the laser intercept to the end of the stream.  8)  Charged cells are appropriately deflected left or right as they pass through a static electric field of approximately 2000 V (I).  9)  One  or two sorted cell subpopulations are collected (J).  0.25  from  - 36 -  may become prohibitively time consuming or be associated with reduced yields of the cells of interest. The current use of multicolor, multiparameter F A C S either alone or in combination with some of the initial "bulk" separation procedures discussed above has enabled the separation of highly enriched (several 100-fold) populations of primitive murine and rat hemopoietic cells including day 12 CFU-S (49,193) and those responsible for 30 day radioprotection (49), as well as h u m a n marrow cells capable of initiating and maintaining hemopoiesis for at least 5 weeks in L T B M C (117).  B)  DIFFERENTIATION ANTIGENS E X P R E S S E D ON PRIMITIVE HEMOPOIETIC CELLS.  The Increasing, decreasing or transient expression of some antigens with maturation has allowed the isolation of cells In various stages of differentiation by selection of cells in specific regions of the spectrum of antigen expression. Much information has now accumulated  about  particular determinants which have proven useful for both murine and h u m a n stem cell purification.  a)  Theta Antigen (Thy-1).  The murine theta-antigen (Thy-1) was initially identified as a marker specific for T-lymphocytes but has subsequently been shown also to be expressed on neurons, fibroblasts, mammary epithelial cells, and skeletal muscle cells (197). It Is expressed co-dorninantfy in two allelic forms i n the mouse, Thy-1.1 and Thy-1.2 (from the T h y - l and T h y - l alleles, a  b  respectively). These can be distinguished serologically by monoclonal antibodies specifically reactive against one or the other allotype. The Thy-1 glycoprotein is structurally similar to the lmmunoglobins (Ig). It is the size of a n Ig domain (the structural subunit of all Ig peptides) and has strong amino acid homology with the variable regions of Ig molecules (197). Thy-1 is thus  -37 viewed to be a member of the Ig supergene family. Although the function of Thy-1 is unknown, the mitogenic effects of many anti-Thy-1 monoclonal antibodies on murine T-cells (198), and on h u m a n T (199) and B-cell (200) lines rendered Thy-1 positive by transfection with the murine Thy-1.2 gene, suggests a role i n cellular activation which is independent of T-cell receptor (TcR) and other T-cell specific molecules. From its homology with the variable region of Ig, Thy-1 i s thought to play a role i n cell-cell interactions, as do many other Ig-related molecules. Recently, it has been shown that a number of hemopoietic cells with high i n vitro proliferative capacity, such as the HPP-CFC (34), CFU-GEMM (201), and the blast colonyforming cell described b y Nakahata and Ogawa (57), express low levels of the Thy-1 antigen. Cells identified as BFU-E (202), CFU-GM (202) a n d day 9 CFU-S (203), which are believed to represent later stages of hemopoietic cell differentiation, are also weakly Thy-1 positive. Precursor cells from the rat have also been fractionated into different differentiation stages on the basis of varying densities of Thy-1 antigen expression (193). Studies with bone marrow from 5-FU treated mice have further shown that hemopoietic cells vary i n their expression of Thy-1 antigen, according to their i n vitro proliferative capabilities (34). These data suggest a potentially interesting relationship between Thy-1 antigen expression and the differentiation state of primitive hemopoietic stem cells.  b)  Major Histocompatability Complex Class I Antigens (H-2K).  ,  The murine major histocompatability complex (MHC), termed H-2, is located on chromosome 17 and is comprised of four major regions, K, I, S, and D. The Class-I molecules are encoded b y the H-2K and H-2D loci. In association with specific foreign antigens, these molecules are recognized i n immunological reactions by the T3-TcR multimolecular complex on cytotoxic T-cells and constitute the m a i n molecular targets i n the rejection of histoincompatible hemopoietic cells i n allogeneic bone marrow transplantation (204). In addition, Class I antigens stimulate antibody responses by B-cells, and are involved i n restricting the specificity  - 38 -  of cytolytic T-cells reactive against virally infected cells or cells bearing tumor-specific transplantation antigens (205). The H-2K a n d H-2D alloantigens are highly polymorphic glycoproteins with intrachain disulfide bonds typical of Ig supergene family members. They are present o n a wide variety of cell types i n association with P2-microglobulin (205). Since M H C Class-I antigens play a n important role i n reactions mediating the rejection of transplanted hemopoietic cells, several studies have been conducted to determine whether stem cells express M H C Class-I antigens o n their surface. The results have shown that murine pluripotent hemopoietic cells such as the day 13 CFU-S (206) a n d the stem cell responsible for rescuing lethally irradiated recipients i n 30 day survival assays (49) express very high levels of the H-2K antigen. Studies of H-2K expression during ontogeny have shown that these antigens are scarcely detectable o n CFU-S from fetal and newborn tissues but are highly expressed on CFU-S from the bone marrow of four-week old mice (207). These observations have now been extended to reveal a pattern of decreasing expression of H-2K antigen on hemopoietic cells by comparisons of CFU-GEMM a n d more restricted progenitor cell types (208). Expression of H-2K, like that of Thy-1, therefore appears to be related to the differentiated state of the hemopoietic cell i n the developmental hierarchy.  c)  Malor Histocompatability Complex Class II Antigens (la).  The I region of the murine H-2 complex can be further subdivided into distinct "subregions", including I-A, I-B, and I-E (209), which encode the M H C Class II molecules. These molecules are selectively expressed on B-cells and macrophages (I-region alloantigens (la) I-A and I-E), a n d some T-cell subsets. Ia antigens are heterodimers. The I-A locus encodes the A , Ap, and Ep subunits, and the I-E locus encodes the E a  a  subunit. These subunits then  associate o n the cell membrane to form the functional I-A (A^Ap) and I-E (E Ep) molecules. Ia a  molecules stimulate mixed lymphocyte reactions (MLR), and i n association with foreign antigens, constitute the molecular targets of T3-TcR complex-mediated interactions of helper  -39 T-cells with antigen-presenting cells (macrophages) and B-cells. Graft versus host (GvH) reactions following allogeneic bone marrow transplantation are la-restricted (205). This again has provided a rationale for studies of the expression of l a antigens on hemopoietic cells. Early studies of the expression of l a antigens on hemopoietic cells suggested that day 9 CFU-S were la-negative (by both cytotoxicity assays and FACS) (210). However, these experiments relied on the use of a heteroantiserum against l a antigens which was not specific to the (as yet undiscovered) I- subregion gene products. Subsequent cytotoxicity studies with subregion-speciflc antibodies have indicated that CFU-S do express I-E, but not I-A antigens (211). This pattern of antigen expression has, however, also been observed for more mature hemopoietic cells (212) indicating the limited usefulness of I-E expression as a basis for the selective isolation of murine stem cells.  d)  Q a Antigens.  The Q a gene cluster is located on mouse chromosome 17 immediately telomeric to the H-2D  locus of the major histocompatability complex and encodes 9 serologically  distinguishable products, Qa-1 to Qa-9 (213). Although the Q a antigens show considerable structural homology with H-2D, K and L antigens, their restricted tissue distribution (mainly peripheral T-lymphocytes) and limited polymorphism suggest different functions for these antigens. Since Kincade et al. (214) originally demonstrated the presence of Q a antigens o n multipotential hemopoietic cells, it has become apparent that developmentally early hemopoietic progenitors are characterized by the expression of certain Q a determinants which are progressively lost during differentiation (215). Qa-m2 is present at high levels o n megakaryocyte progenitors (216) and day 14 CFU-S, but disappears with maturation to day 8 CFU-S (217). Qa-m7 is also expressed differently on different types of hemopoietic progenitors. Qa-m7 is present at high levels on HPP-CFC and allows distinction of a more primitive subset of these cells dependant on CSF-1 and a synergistic activity (218). Significant increases i n the  - 40 proportion of Qa-m7 positive hemopoietic cells a n d the level of Qa-m7 antigen expression are observed i n 5-FU treated bone marrow. Bertoncello et al. (215) have exploited this observation to isolate a primitive subpopulation of HPP-CFC dependant o n three H G F s (IL-1, IL-3 a n d CSF-1) which appear to co-fractionate with multipotential hemopoietic cells capable of reconstituting lethally irradiated mice.  e)  Wheat Germ Agglutinin (WGA) Binding Sites.  W G A is a plant lectin which recognizes and binds to sialic acid residues of certain glycosylated cell membrane components. Murine day 12 CFU-S express high densities of WGA-binding sites (53,219). Visser et al. have been the major advocates of the use of fluorescein-conjugated W G A (WGA-FITC) for stem cell purification, and have reported approximately 9-fold enrichment of day 12 CFU-S from normal marrow over that obtained by equilibrium density gradient centrifugation using discontinuous B S A (219). They obtained 40-60 fold overall enrichment of CFU-S sorted i n this way and no deleterious effect o n stem cell homing or radioprotective ability (220). Their subsequent incorporation of WGA-binding into a 3-step purification procedure has enabled a n approximately 130-fold enrichment of day 12 CFU-S (49). In addition, the same population is enriched approximately 180-fold i n the cells responsible for 30 day radioprotection (170 cells required to rescue 5 0 % of irradiated recipients) (49).  f)  Rhodamine-123 (Rh-123) Uptake.  Rh-123 is a supravital, cationic red fluorescent dye that has a relatively high affinity for mitochondrial membranes (221). It is retained to a m u c h greater extent i n cycling cells than i n quiescent cells (222). A l l day 8 CFU-S incorporate high levels of Rh-123; however, recent studies have shown that murine day 12 CFU-S are heterogeneous with respect to their uptake  - 41 of the dye (223). Day 12 CFU-S with a lower affinity for Rh-123 have a greater 30 day radioprotective ability (RPA) and marrow repopulating ability (MRA, measured by reconstitution of marrow CFU-GM and CFU-S numbers 13 days post-transplantation) t h a n those stained brightly with Rh-123 (51). Visser and deVries (224) have recently reported significant (100-200 fold) purification of day 12 CFU-S and more primitive pre-CFU-S i n a population of cells isolated by a 3-step procedure which includes sorting of cells stained with Rh-123. The actual purity and enrichment of cells responsible for RPA is, however, still unclear, due to difficulties i n quantitating these cells. Urilike the majority of CFU-S, it appears that there exist some preCFU-S that initially home preferentially to the bone marrow (17) where they may then divide several times to generate a number of daughter cells, each capable of giving rise to a spleen colony upon transplantation. Until jjmiting dilution experiments are available to establish the CFU-S output by a single MRA  cell, unequivocal assessment of the purity of cells monitered  even using this latter assay will not be possible.  g)  Stem Cell Antigen-1 (Sca-1).  In 1986, A i h u r a et al. (225) produced a monoclonal antibody against a murine pre-T hybridoma which was found to react with several lymphoid cell types, including functional bone marrow progenitors of thymic lymphocytes. The antigen recognized by this antibody has recently been determined to be a member of the Ly-6 family (226) and was designated  Sea-1  (Stem cell antigen-1) because it subdivides mouse bone marrow cells, previously enriched i n CFU-S on the basis of low Thy-1 expression and the absence of a panel of lineage-specific (Lin) markers ( T h y - l L i n " ) (227), into a minor Sca-1 positive (20-30%) and a major Sca-1 negative low  (70-80%o) population (67). T h y - l  l o w  L i n " S c a - l cells represent approximately 0.05% of total  bone marrow and are predominantly i n G  +  0  or Gj-phase. They contain thymocyte precursors,  and are estimated to be 1000-fold enriched i n day 12 CFU-S. Limiting dilution experiments i n lethally irradiated mice demonstrated that approximately 30 such cells are sufficient to allow  - 42 5 0 % of mice to survive for 30 days (67); longer term repopulatlng ability has not yet been reported. Several groups have provided evidence that at least some day 12 CFU-S are distinct from the most primitive stem cells with long-term repopulatlng ability (50,51). The apparent ability of the T h y - l ^ L i n " S c a - l population to form late spleen colonies with unit efficiency o w  +  therefore suggests that the relationship of these cells to those with long-term lympho-myeloid repopulatlng potential is t h u s uncertain.  h)  H u m a n Major Histocompatibility Complex Class II Antigens (HLA-DR).  The h u m a n MHC, termed H L A (human lymphocyte antigens), Is located o n chromosome 6. It encodes three distinct Class II antigens; HLA-DR (equivalent to the murine I-E), DQ(DC) (equivalent to the murine I-A) and DP(SB) (228), which function as restriction elements for T4-positive helper T-cells i n the immune response (205). Compared with HLA-DR expression, HLA-DP molecules are only detected on a proportion of progenitor cells (229); HLA-DQ molecules are expressed at very low levels, if at all on hemopoietic progenitor cells (229). The progenitors of h u m a n CFU-GM express Increasing levels of HLA-DR as they differentiate through more mature progenitor stages (ie. progenitors that generate colonies recognizeable at increasingly earlier times In culture) (230). H u m a n erythroid progenitors display a pattern of HLA-DR expression opposite to that of CFU-GM. BFU-E express high levels while CFU-E are found mostly, but not exclusively, i n fractions sorted on the basis of either low density of HLA-DR or no HLA-DR (231). While h u m a n CFU-GEMM also express HLA-DR (232), studies of the level expression on more primitive hemopoietic cells able to generate BFU-E, CFU-GM and C F U - G E M M i n liquid culture have been conflicting. It now appears, however, that the inability of some researchers (233,234) to detect HLA-DR on these cells was due to the use of complement-dependant cytotoxicity (CDC) as a method of depletion. Using C D C assays only, antigenic determinants with weak expression on hemopoietic stem cells may be overlooked (235). With more sensitive quantitative separation procedures possible with the FACS,  - 43 primitive hemopoietic cells able to generate in vitro clonogenic cells in LTBMC have been clearly shown to express some, albeit low levels, of HLA-DR (117,236). HLA-DR expression has been successfully exploited by several groups to enrich for CFU-GEMM (237), and stem cells able to initiate human LTBMC (117).  i)  Human Cluster of Differentiation 34 Antigen (CD34, My-10).  The monoclonal antibody HPCA-1 (Hemopoietic Progenitor Cell Antigen-1) was raised against the undifferentiated human myeloid leukemia cell line KG-la (238). It recognizes a 115 kD glycoprotein (the My-10 antigen, now identified as part of the CD34 molecule (239)) which is expressed not only on KG-la cells but also on a small subset (1-4% of total) of normal bone marrow cells. Positive marrow cells have mainly a blast morphology and include most of the clonogenic progenitors (240). The My-10 antigen is expressed at relatively low levels on positive cells (approximately 50,000 molecules/cell) (238). My-10 appears to be expressed in a manner similar to the MHC HLA-DR molecule and this has been exploited by several groups in attempts to use dual-color staining and FACS to isolate human hemopoietic stem cells. Lu et al. (237) found >98% of CFU-GM, BFU-E and CFU-GEMM (cloning efficiency 47%) in a fraction of cells sorted on the basis of high My-10 and low HLA-DR Sutherland et al. (117) in the Terry Fox Laboratory have recently described a similar procedure for the purification of human marrow cells able to generate clonogenic progenitors in LTBMCs for >5 weeks. Another monoclonal antibody, 12-8, which appears to recognize a different epitope on the human CD34 molecule, has been used to enrich for primitive hemopoietic cells able to reconstitute hemopoiesis in lethally irradiated babboons (241). Selection of My-10 positive cells would therefore appear to be likely to facilitate the purification of the most primitive human hemopoietic cells.  - 44 4)  THESIS OBJECTIVES AND GENERAL STRATEGY.  As reviewed above, there is now considerable information about the properties of hemopoietic cells detected by short-term in vitro and in vivo assays and of a number of growth factors that regulate the output of their terminally differentiated progeny. However, the properties of the most primitive stem cells which maintain long-term blood cell production in vivo remain poorly understood. One of the features of primitive hemopoietic cell behaviour that is likely to continue to make it difficult to characterize these cells is an inherent heterogeneity in the differentiative and proliferative potential they express. The extent to which such heterogeneity may be attributed to intrinsic differences between individual stem cells, or to extrinsic biological processes that regulate stem cell recruitment in vivo is not known, but is of fundamental importance to our understanding of normal hemopoietic stem cell development. This research project was initiated to address these questions, specifically by attempting to develop procedures for the detection, quantitation and purification of cells with long-term lympho-myeloid repopulatlng potential. It was then anticipated that the functional attributes of these cells could be evaluated by comparing their performance in a variety of in vitro and in vivo assays. In this way I hoped to be able to investigate the extent of overlap between cells detected by current short-term colony assays, and their relationship to the most primitive stem cells detected by their ability to regenerate whole tissues following transplantation. The development of a procedure for the isolation of hemopoietic stem cells with a high capacity for proliferation and differentiation in vivo depends on the availability of an assay which allows quantitation of the number of repopulatlng stem cells present. In the present setting this required a method that allowed such cells to be detected under conditions where they would be transplanted in limiting numbers. It is known that hemopoiesis in transplanted mice can be monoclonal (5,40). However, it is not yet clear how this relates to the number of stem cells with repopulatlng potential present in the original transplant, as such cells may apparently be held in reserve for variable periods of time after transplantation (41,242).  - 45 Therefore my first objective was to try to develop a type of i n vivo limiting dilution assay that might enable the quantitation of totipotent murine hemopoietic repopulating stem cells. Based on the work of Harrison et al. (69) who showed that there may be qualitative differences even amongst stem cells with long-term repopulating ability, I decided to use a competitive transplantation assay. By using sex markers to distinguish repopulation by stem cells of donor, competitor or recipient origin several months later, I hoped to circumvent the critical, and not necessarily valid, assumption that survival endpoints are obviously subject to. The use and validation of this assay for the quantitation of repopulating hemopoietic stem cells is described i n Chapter III and V. The second objective of my research was then to use this assay to develop a procedure for obtaining highly enriched populations of hemopoietic stem cells. I surveyed the known physical properties of primitive hemopoietic cells and analyzed the expression of several hemopoietic "differentiation antigens". By labelling cells with the appropriate  fluorochrome-  conjugated monoclonal antibodies, I hoped that it might be possible to obtain a highly enriched suspension of repopulating stem cells i n a single multi-parameter sort on the FACS. The results of this work are presented i n Chapter III. To determine if the same or different stem cells are responsible for repopulating various lymphoid and myeloid tissues i n vivo, my final objective was to use retroviruses to uniquely mark specific stem cells i n the purified population prior to transplantation. In addition to the use of sex markers to ascertain the donor origin of repopulated tissues, an analysis of retroviral integration patterns would, at least i n theory, enable a determination of the number and relative contribution of inividual stem cell clones to various lymphoid and myeloid lineages. S u c h studies should t h u s make it possible to establish whether the purification procedure developed could be used to isolate stem cells with multilineage repopulating potential.  The  development of this strategy is described i n Chapter IV. The development of procedures for the quantitation and differential isolation of hemopoietic stem cells, as well as strategies for their use as targets for retrovirus-mediated  - 46 gene transfer, represent important first steps towards the realization of the larger goal of understanding the earliest events in normal hemopoietic cell development. The achievement of these objectives has provided a new and solid starting point for additional studies to characterize primitive hemopoietic cells. These are discussed in Chapter VI of this thesis under the title of Summary and Future Directions.  - 47 -  REFERENCES 1.  Moore MAS. Embryologic and phylogenetic development of the hematopoietic system. In: Burkhardt R Advances in the Biosciences 16. Dahlem Workshop on MyelofibrosisOsteosclerosis Syndrome. Oxford: Pergamon Press (1975).  2.  Lichtman MA. The ultrastructure of the hemopoietic environment of the marrow: A review. Exp Hematol 9:391 (1981).  3.  Springer T, Galfre G. Secher DS. Milstein C. Mac-1: A macrophage differentiation antigen identified by monoclonal antibody. Eur J Immunol 9:301 (1979).  4. Metcalf D. The hemopoietic colony stimulating factors. Amsterdam: Elsevier (1984). 5.  Till JE, McCulloch E A Hemopoietic stem cell differentiation. Biochem Biophys Acta 605:431 (1980).  6. Pietrzyk ME, Priestley GV, Wolf NS. Normal cycling patterns of hematopoietic stem cell subpopulations: An assay using long-term In vivo BrdU infusion. Blood 66:1460 (1985). 7.  Rosendaal M, Hodgson GS, Bradley T R Organization of haemopoietic stem cells: The generation-age hypothesis. Cell Tissue Klnet 12:17 (1979).  8. Blackett NM, Millard RE. Different recovery patterns of mouse hemopoietic stem cells in response to cytotoxic agents. J Cell Physiol 89:473 (1976). 9. Metcalf D. Hemopoietic Colonies. In vitro cloning of normal and leukemic cells. Berlin Heidelberg: Springer-Verlag (1977). 10.  Gregory CJ, Henkelman RM. Relationships between early hemopoietic progenitor cells determined by correlation analysis of their numbers In Individual spleen colonies. In: Baum SJ. Ledney GD. Experimental Hematology Today. New York: Springer-Verlag (1977).  11.  Eaves CJ, Humphries RK, Eaves AC. In vitro characterization of erythroid precursor cells and the erythropoietic differentiation process. In: Stamatoyannopoulos G, Nienhuis AW. Cellular and Molecular Regulation of Hemoglobin Switching. New York: Grune and Stratton (1979).  12.  Gregory CJ. Eaves AC. Human marrow cells capable of erythropoietic differentiation in vitro: Definition of three erythroid colony responses. Blood 49:855 (1977).  13.  Pike BL, Robinson WA Human bone marrow colony growth in agar-gel. J Cell Physiol 76:77 (1970).  14.  Fauser A A Messner HA. Identification of megakaryocytes, macrophages, and eosinophils in colonies of human bone marrow containing neutrophilic granulocytes and erythroblasts. Blood 53:1023 (1979).  15.  Dexter TM, Spooncer E. Growth and differentiation in the hemopoietic system. In: Annual Review of Cell Biology, vol 3., (1987).  16.  Metcalf D, Moore MAS. In: Neuberger A, Tatum EL. "Haemopoietic Cells", Frontiers of Biology, vol 24., Amsterdam: North-Holland (1971).  - 48 -  17.  Van Zant G. Studies of hematopoietic stem cells spared by 5-fluorouracil. J Exp Med 159:679 (1984).  18.  Gregory CJ. Eaves AC. Three stages of erythropoietic progenitor cell differentiation distinguished by a number of physical and biologic properties. Blood 51:527 (1978).  19.  Pluznik DH. Sachs L. The cloning of normal mast' cells in tissue culture. J Cell Comp Physiol 66:319 (1966).  20.  Bradley TR, Metcalf D. The growth of mouse bone marrow cells in vitro. Aust J Exp Biol Med Sci 44:287 (1982).  21.  Metcalf D. Merchav S. Wagemaker G. Commitment by GM-CSF or M-CSF of bipotential GM progenitor cells to granulocyte or macrophage formation. In: Baum. Ledney, Thierfelder. Experimental Hematology Today. Basel: Karger (1982).  22.  Humphries RK, Eaves AC, Eaves CJ. Self-renewal of hemopoietic stem cells during mixed colony formation in vitro. Proc Natl Acad Sci USA 78:3629 (1981).  23.  Eaves CJ, Eaves AC. Erythropoiesis. In: Golde DW, Takaku F. Hematopoietic Stem Cells. New York: M Dekker Inc. (1985).  24.  Tepperman AD, Curtis JE, McCulloch EA. Erythropoietic colonies in cultures of human marrow. Blood 44:659 (1974).  25.  Long MW, Gragawski LL, Heffner CH, Boxer LA. Phorbol diesters stimulate the development of an early murine progenitor cell. The burst-forming unit megakaryocyte. J Clin Invest 76:431 (1985).  26.  Metcalf D, MacDonald HR, Odartchenko N, Sordat B. Growth of mouse megakaryocyte colonies in vitro. Proc Natl Acad Sci USA 72:1744 (1975).  27.  Metcalf D, Nossal SJV, Warner NL, Miller JFP, Mandel TE, Layton JE, Gutman GA. Growth of B lymphocyte colonies in vitro. J Exp Med 142:1534 (1975).  28.  Rozenszajn LA, Shoham D, Kalechman I. Clonal proliferation of PHA-stimulated human lymphocytes in soft agar culture. Immunology 29:1041 (1975).  29.  Johnson GR, Metcalf D. Pure and mixed erythroid colony formation in vitro stimulated by spleen conditioned medium with no detectable erythropoietin. Proc Natl Acad Sci USA 74:3879 (1977).  30.  Hara H, Ogawa M. Murine hemopoietic colonies in culture contahiing normoblasts, macrophages and megakaryocytes. Am J Hematol 4:23 (1978).  31.  Messner HA, Izaguirre CA, Jamal N. Identification of T-lymphocytes in human mixed hemopoietic colonies. Blood 58:402 (1981).  32.  Bradley TR, Hodgson GS. Detection of primitive macrophage progenitor cells in mouse bone marrow. Blood 54:1446 (1979).  33.  McNiece I, Kriegler AB, Hapel AJ, Fung MC. Young IG, Bradley TR, Hodgson GS. Recombinant interleukin-3 exhibits synergistic factor activity. Cell Biol Int Rep 8:812 (1984).  - 49 34.  Boswell HS, Wade J r PM, Quesenberry PJ. Thy-1 antigen expression by murine hlgh-proliferative capacity hematopoietic progenitor cells. I. Relation between sensitivity to depletion by Thy-1 antibody and stem cell generation potential; J Immunol 133:2940 (1984).  35.  McNiece IK, Stewart FM, Deacon DM, Temeles DS, Zsebo KM, Clark SC, Quesenberry PJ. Detection of a human CFC with a high proliferative potential. Blood 74:609 (1989).  36.  Schofield R The pluripotent stem cell. Clin Haematol 8:221 (1979).  37.  Till JE, McCulloch EA. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat Res 14:213 (1961).  38.  Wu AM, Till JE, Siminovitch L, McCulloch EA. A cytological study of the capacity for differentiation of normal hemopoietic colony forming cells. J Cell Physiol 69:177 (1967).  39.  Becker AJ, McCulloch EA, Till JE. Cytological demonstration of the clonal nature of spleen colonies derived from transplanted mouse marrow cells. Nature 197:452 (1963).  40.  Dick JE, Magli MC, Huszar D, Phillips RA, Bernstein A Introduction of a selectable gene into primitive stem cells capable of long-term reconstitutlon of the hemopoietic system ofW/W mlce. Cell 42:71 (1985). v  41.  Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell 45:917 (1986).  42.  Curry JL, Trentin JJ. Hemopoietic spleen colony studies: I. Growth and differentiation. Dev Biol 15:395 (1967).  43.  Fowler JH, Wu AM, Till JE, McCulloch EA, Siminovitch L. The cellular composition of hemopoietic spleen colonies. J Cell Physiol 69:65 (1967).  44.  Siminovitch L, McCulloch EA, Till JE. The distribution of colony forming cells among spleen colonies. J Cell Physiol 62:327 (1963).  45.  Magli MC, Iscove NN, Odartchenko N. Transient nature of early haematopoietic spleen colonies. Nature 295:527 (1982).  46.  Hodgson GS, Bradley T R Properties of hematopoietic stem cells surviving 5-fluorouracil treatment: Evidence for a pre-CFU-S cell ?. Nature 281:381 (1979).  47.  Humphries RK, Jacky PB, Dill FJ. Eaves AC. Eaves CJ. CFU-S in individual erythroid colonies derived in vitro from adult mouse marrow. Nature 279:718 (1979).  48.  Johnson GR Nicola NA Characterization of two populations of CFU-S fractionated from mouse fetal liver by fluorescence-activated cell sorting. J Cell Physiol 118:45 (1984).  49.  Visser JWM, Bauman JGJ, Mulder AH, Eliason JF, de Leeuw AM. Isolation of murine pluripotent hemopoietic stem cells. J Exp Med 59:1576 (1984).  50.  Ploemacher RE, Brons NHC. Isolation of hemopoietic stem cell subsets from murine bone marrow: II. Evidence for an early precursor of day-12 CFU-S and cells associated with radioprotective ability. Exp Hematol 16:27 (1988).  - 50 51.  Ploemacher RE, Brons RHC. Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hemopoietic stem cell compartment following irradiation: Evidence for a pre-CFU-S cell. Exp Hematol 17:263 (1989).  52.  Williams DE, L u L, Broxmeyer HE. Characterization of hematopoietic stem a n d progenitor cells. Immunol Res 6:294 (1987).  53.  Ploemacher RE, Brons RHC, Leenen PJM. B u l k enrichment of transplantable hemopoietic stem cell subsets from lipopolysaccharide-stimulated murine spleen. Exp Hematol 15:154 (1987).  54.  Till J E , McCulloch EA. The f-factor of the spleen-colony assay for hemopoietic stem cells. Ser Haematol 5:15 (1972).  55.  Lord BI, Hendry J H . Observations o n the settling and recoverability of transplanted hemopoietic colony-forrning units i n the mouse spleen. Blood 41:409 (1973).  56.  Nakahata T, Ogawa M. Hemopoietic colony-forrning cells i n umbilical cord blood with extensive capability to generate mono- and multipotential hemopoietic progenitors. J C l i n Invest 70:1324 (1982).  57.  Nakahata T, Ogawa M. Identification i n culture of a class of hemopoietic colony-forming units with extensive capability to self-renew and generate multipotential hemopoietic colonies. Proc Natl Acad Sci U S A 79:3843 (1982).  58.  Rowley SD, Sharkis S J , Hattenburg C, Sensenbrenner LL. Culture from h u m a n bone marrow of blast progenitor cells with a n extensive proliferative capacity. Blood 69:804 (1987).  59.  Leary AG, Ogawa M. Blast cell colony assay for umbilical cord blood a n d adult bone marrow progenitors. Blood 69:953 (1987).  60.  S u d a T, S u d a J , Ogawa M. Proliferative kinetics and differentiation of murine blast cell colonies i n culture: Evidence for variable G periods and constant doubling rates of early pluripotent hemopoietic progenitors. J Cell Physiol 117:308 (1983). 0  61.  Leary AG, H i r a i Y , KishimotoT, Clark SC, Ogawa M. Survival of hemopoietic progenitors i n the G period of the cell cycle does not require early hemopoietic regulators. Proc Natl A c a d Sci U S A 86:4535 (1989). 0  62.  Ikebuchi K, Wong GG, C l a r k SC, Ihle J N , Hirai Y, Ogawa M. Interleukin 6 enhancement of interleukin 3-dependent proliferation of multipotential hemopoietic progenitors. Proc Natl Acad Sci U S A 84:9035 (1987).  63.  S u d a T, S u d a J , Ogawa M. Single-cell origin of mouse hemopoietic colonies expressing multiple lineages i n variable combinations. Proc Natl Acad Sci U S A 80:6689 (1983).  64.  Keller G, Holmes W, Phillips RA. Clonal generation of multipotent and unipotent hemopoietic blast cell colonies i n vitro. J Cell Physiol 120:29 (1984).  65.  W u AM, Till J E , Siminovitch L, McCulloch EA. Cytological evidence for a relationship between normal hemopoietic colony-forming cells and cells of the lymphoid system. J Exp Med 127:455 (1968).  - 51 -  66.  Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of mice reconstituted with retrovirus-infected stem cells. EMBO J 6:3955 (1987).  67.  Spangrude GJ, Hetmfeld S, Weissman IL. Purification and characterization of mouse hematopoietic stem cells. Science 241:58 (1988).  68.  Siminovitch L, Till JE, McCulloch EA. Decline in colony forming ability of marrow cells subjected to serial transplantation into irradiated mice. J Cell Comp Physiol 64:23 (1964).  69.  Harrison DE, Astle CM, Delaittre JA. Loss of proliferative capacity in immunohemopoietic stem cells caused by serial transplantation rather than aging. J Exp Med 147:1526 (1978).  70.  Mauch P, Hellman S. Loss of hematopoietic stem cell self-renewal after bone marrow transplantation. Blood 74:872 (1989).  71.  Harrison DE, Astle CM. Loss of stem cell repopulatlng ability upon transplantation. Effects of donor age. cell number, and transplantation procedure. J Exp Med 156:1767 (1982).  72.  Jones RJ, Celano P, Sharkis SJ, Sensenbrenner LL. Two phases of engraftment established by serial bone marrow transplantation in mice. Blood 73:397 (1989).  73.  Harrison DE. Lifesparing ability (in lethally irradiated mice) of W/W with no macroscopic colonies. Radiat Res 52:553 (1972).  74.  McCulloch EA, Till JE. The radiation sensitivity of normal mouse bone marrow cells, determined by quantitative marrow transplantation into irradiated mice. Radiat Res 13:115 (1960).  75.  Lord BI, Hendry JH, Keene JP, Hodgson BW, Xu CX, Rezvani M, Jordan TJ. A comparison of low and high dose-rate radiation for recipient mice in spleen-colony studies. Cell Tissue Kinet 17:323 (1984).  76.  Hendry JH, Potten CS, Roberts NP. The gastrointestinal syndrome and mucosal clonogenic cells: Relationships between target cell sensitivities, LD50 and cell survival, and their modification by antibiotics. Radiat Res 96:100 (1983).  77.  Hodgson GS, Bradley T R Radley JM. The organization of hemopoietic tissue as inferred from the effects of 5-fluorouracil. Exp Hematol 10:26 (1982).  78.  Abramson S, Miller RG, Phillips RA The identification in adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J Exp Med 145:1567(1977).  79.  Lamar EE, Palmer E. Y-encoded, species-specific DNA in mice: Evidence that the Y chromosome exists In two polymorphic forms in inbred strains. Cell 37:171 (1984).  80.  Kadish JL, Basch RS. Hematopoietic thymocyte precursors. III. A population of thymocytes with the capacity to return ("home") to the thymus. Cell Immunol 30:12 (1977).  81.  Mintz B, Cronmiller C, Custer RP. Somatic cell origin of teratocarcinomas. Proc Natl Acad Sci USA 75:2834 (1978). -  v  mouse marrow  - 52 -  82.  Russel ES, McFarland EC. Hemoglobin, comparative molecular biology models for the study of disease. Ann NY Acad Sci 241:25 (1974).  83.  Joyner A, Keller G, Phillips RA, Bernstein A. Retrovirus transfer of a bacterial gene into mouse haematopoietic progenitor cells. Nature 305:556 (1983).  84.  Harrison DE. Do hemopoietic stem cells age ?. In: Monogr Dev Biol, vol 17, BasehKarger (1984).  85.  Russell ES. Hereditary anemias of the mouse; A review for geneticists. Adv Genet 20:357 (1979).  86.  Mintz B, Russell ES. Gene induced embryological modifications of primordial germ cells in the mouse. J Exp Zool 134:207 (1957).  87.  Harrison DE, Astle CM. Population of lymphoid tissues in cured W anemics by donor cells. Transplantation 22:42 (1976).  88.  McCulloch EA, Siminovitch L, Till JE. Spleen colony formation in anemic mice of genotype W/W . Science 144:844 (1964). v  89.  Chabot B, Stephenson DA, Chapman VM, Besmer P, Bernstein A. The proto-oncogene c-kit encoding a transmembrane tyrosine kinase receptor maps to the mouse W locus. Nature 335:88 (1988).  90.  Geissler EN, Ryan MA, Housman DE. The dominant-white spotting (W) locus of the mouse encodes the c-kit proto-oncogene. Cell 55:185 (1988).  91.  Boggs DR, Boggs SS, Saxe DS, Gress RA, Confield DR Hematopoietic stem cells with high proliferative potential. Assay of their concentration in marrow by the frequency and duration of cure of W/W mice. J Am Soc Clin Invest 70:242 (1982). v  92.  Nakano T, Waki N, Asai H. Kitamura Y. Effect of 5-fluorouracil on "primitive" hematopoietic stem cells that reconstitute whole erythropoiesis of genetically anemic W/W mice. Blood 73:425 (1989). v  93.  Dexter TM, Allen TD, Lajtha LG. Conditions controlling the proliferation of hemopoietic stem cells in vitro. J Cell Physiol 91:335 (1977).  94.  Bentley SA, Foidart JM. Some properties of marrow derived adherent cells in tissue culture. Blood 56:1006 (1980).  95.  Castro-Malaspina H, Gay RE, Saletan S, Oettgen B, Gay S. Moore MAS. Phenotypic characterization of the adherent cell layer in long-term mouse bone marrow cultures. Blood 58(Suppl l):107a (1981).  96.  Allen TD, Dexter TM. Long-term bone marrow cultures: An ultrastructural review. In: Scanning electron microscopy, SEM, Inc. AMF. Chicago: O'Hare (1983).  97.  Zuckerman KS, Wicha MS. Extracellular matrix production by the adherent cells of long-term murine bone marrow cultures. Blood 61:540 (1983).  98.  Zuckerman KS. Composition and function of the extra-cellular matrix in the stroma of long-term bone marrow cell cultures. In: Wright DG, Greenberger JS. Long-term bone marrow cultures. (1984).  - 53 99.  Zuckerman KS, Rhodes RK. the hematopoietic extracellular matrix. In: Cronklte EP, Dainiak N, McCaffrey RP, Palek J, Quesenberry PJ. Hematopoietic stem cell physiology. New York: Alan R Liss (1985).  100.  Dexter TM, Testa NG. In: Methods in Cell Biology, vol 14. New York: Academic Press (1976).  101.  Bentley SA. Close range cell: Cell interaction required for stem cell maintenance in continuous bone marrow culture. Exp Hematol 9:308 (1981).  102.  Eaves C, Coulombel L, Eaves A. Analysis of hemopoiesis in long-term human marrow cultures. In: Killman SVA. Cronklte EP, Muller-Berat CN. Haemopoietic Stem Cells. Characterization. Proliferation. Regulation. Copenhagen: Munskgaard (1983).  103.  Schofield R, Dexter TM. Studies on the self-renewal ability of CFU-S which have been serially transferred in long-term culture or in vivo. Leuk Res 9:305 (1985).  104.  Harrison DE, Lerner CP, Spooncer E. Erythropoietic repopulatlng ability of stem cells from long-term marrow culture. Blood 69:1021 (1987).  105.  Dorshkind K, Johnson A, Collins L, Keller GM, Phillips RA Generation of purified stromal cell cultures that support lymphoid and myeloid precursors. J Immunol Methods 89:37(1986).  106.  Gartner S, Kaplan HS. Long-term culture of human bone marrow cells. Proc Natl Acad Sci USA 77:4756 (1980).  107.  Greenberg HM, Newburger PE, Parker LM, Novak T. Greenberger JS. Human granulocytes generated in continuous bone marrow culture are physiologically normal. Blood 58:724 (1981).  108.  Coulombel L, Eaves AC, Eaves CJ. Enzymatic treatment of long-term human marrow cultures reveals the preferential location of primitive hemopoietic progenitors In the adherent layer. Blood 62:291 (1983).  109.  Becker AJ, McCulloch EA, Siminovitch L, Till JE. The effect of differing demands for blood cell production on DNA synthesis by hemopoietic colony-forming cells of mice. Blood 26:296 (1965).  110.  Toksoz D, Dexter TM, Lord BI, Wright EG, Lajtha LG. The regulation of hemopoiesis in long-term bone marrow cultures. II. Stimulation and Inhibition of stem cell proliferation. Blood 55:931 (1980).  111.  Cashman J, Eaves AC, Eaves CJ. Regulated proliferation of primitive hematopoietic progenitor cells in long-term human marrow cultures. Blood 66:1002 (1985).  112.  Cashman JD, Eaves AC, Raines EW, Ross R, Eaves CJ. Mechanisms that regulate the cell cycle status of very primitive hematopoietic cells in long-term human marrow cultures. I. Stimulatory role of a variety of mesenchymal cell activators and inhibitory role of TGF-p. Blood 75(Suppl 1):96 (1990).  113.  Dexter TM, Spooncer E, Simmons P, Allen TD. Long-term marrow culture: An overview of techniques and experience. In: Wright DG, Greenberger JS. Long-term bone marrow culture. Kroc Foundation Series, vol 18. New York: Alan R Liss Inc. (1984).  - 54 114.  Sutherland HJ, Lansdorp PM, Henkelman D, Eaves AC. Eaves CJ. Functional characterization of individual human hematopoietic stem cells cultured at limiting dilution on supportive marrow stromal layers. Proc Natl Acad Sci USA (in press) (1990).  115.  Ploemacher RE, Van Der Sluijs JP, Voerman JSA, Brons NHC. An in vitro limiting-dilution assay of long-term repopulating hematopoietic stem cells in the mouse. Blood 74:2755 (1989).  116.  Barnett MJ, Eaves CJ. Phillips GL, Kalousek DK, Klingemann H-G, Lansdorp PM, Reece DE, Shepherd JD, Shaw GJ. Eaves AC. Successful autografting in chronic myeloid leukaemia after maintenance of marrow in culture. Bone Marrow Transplant 4:345 (1989).  117.  Sutherland HJ, Eaves CJ, Eaves AC, Dragowska W, Lansdorp PM. Characterization and partial purification of human marrow cells capable of initiating long-term hematopoiesis in vitro. Blood 74:1563 (1989).  118.  Prchal JT, Throckmorton DW, Caroll AJ, Fuson EW, Gams RA, Prchal JF. A common progenitor for human myeloid and lymphoid cells. Nature 274:590 (1978).  119.  Raskind WH, Fialkow PJ. The use of cell markers in the study of human hematopoietic neoplasia. Adv Cancer Res 49:127 (1987).  120.  Turhan AG, Humphries RK, Phillips GL, Eaves AC, Eaves CJ. Clonal hematopoiesis demonstrated by X-linked DNA polymorphisms after allogeneic bone marrow transplantation. N Engl J Med 320:1655 (1989).  121.  Lichtman MA, Chamberlain JK, Simot W, Santillo PA. Parasinal location of megakaryocytes: A determinant of platelet release. Am J Hematol 4:303 (1978).  122.  Lorimore SA, Wright EG. Stem cell maintenance and commitment to differentiation in long-term cultures of murine marrow obtained from different spatial locations in the femur. Cell Tissue Kinet 20:427 (1987).  123.  Wright EG, Lord BI. Haemopoietic stem cell proliferation: Spatial and temporal considerations. Br J Cancer 53(Suppl VII): 130 (1986).  124.  Tsai S, Patel V. Beaumont E. Lodish HF. Nathan DG. Sieff CA. Differential binding of erythroid and myeloid progenitors to fibroblasts and fibronectin. Blood 69:1587 (1987).  125.  Spooncer E, Heyworth CM, Dunn A, Dexter TM. Self-renewal and differentiation of interleukin-3-dependent multipotent stem cells are modulated by stromal cells and serum factors. Differentiation 31:111 (1986).  126.  Rettenmier CW, Roussel MF, Ashmun RA, Ralph P, Price K, Sherr CJ. Synthesis of membrane-bound colony-stimulating factor 1 (CSF-1) and downmodulation of CSF-1 receptors in NIH 3T3 cells transformed by cotransfection of the human CSF-1 and c-fms (CSF-1 receptor) genes. Mol Cell Biol 7:2378 (1987).  127.  Gordon MY, Riley GP, Watt SM, Greaves MF. Compartrnentallzation of a haematopoietic growth factor (GM-CSF) by grycosaminoglycans in the bone marrow microenvironment. Nature 326:403 (1987).  - 55 128.  Friedenstein AJ, Petrakova KV, Kurolesova AI, Frolova GP. Heterotopic transplants of bone marrow. Analysis of precursor cells for osteogenic and hematopoietic tissues. Transplantation 6:230 (1968).  129.  Whitlock CA, Witte ON. Long-term culture of B lymphocytes and their precursors from murine bone marrow. Proc Natl Acad Sci USA 79:3608 (1982).  130.  Johnson A, Dorshkind K. Stromal cells in myeloid and lymphoid long-term bone marrow cultures can support multiple hemopoietic lineages and modulate their production of hemopoietic growth factors. Blood 68:1348 (1986).  131.  Collins LS, Dorshkind K. A stromal cell line from myeloid longterm bone marrow cultures can support myelopoiesis and B lymphopoiesis. J Immunol 138:1082 (1987).  132.  McCulloch EA, Siminovitch L, Till JE, Russell ES, Bernstein SE. The cellular basis of the genetically determined hemopoietic defect inanaemic mice of genotype Sl/Sl . Blood 26:399 (1965). d  133.  Bernstein SE. Tissue transplantation as an analytic and therapeutic tool In hereditary anemias. Am J Surg 119:448 (1970).  134.  Tavassoli M, Ratzan RJ, Maniatis A, Crosby WH. Regeneration of hemopoietic stroma in anemic mice of S l / S l and W/W genotypes. J Reticuloendothelial Soc 13:518 (1973). d  v  135.  Nicola NA Hemopoietic growth factors and their interactions with specific receptors. J Cell Physiol (Suppl 5):9 (1987).  136.  Stanley E R Jubinsky PT. Factors affecting growth and differentiation of haemopoietic cells in culture. Clin Haematol 13:329 (1984).  137.  Metcalf D. The molecular biology and functions of the colony-stimulating factors. Blood 67:257 (1986).  138.  Eaves AC, Eaves CJ. Erythropoiesis in culture. In: McCulloch EA. Cell Culture Techniques - Clinics in Haematologyvol 13. Eastbourne, England: WB Saunders Co (1984).  139.  Nicola NA, Metcalf D, Matsumoto M, Johnson GR Purification of a factor inducing differentiation in murine myelomonocytic leukaemia cells. J Biol Chem 258:9017 (1983).  140.  Tomonaga M, Golde DW, Gasson JC. Biosynthetic (recombinant) human granulocyte-macrophage colony-stimulating factor: Effect on normal bone marrow and leukemic cell lines. Blood 67:31 (1986).  141.  Hapel AJ, Fung MC, Johnson RM. Young IG, Johnson G, Metcalf D. Biologic properties of molecularly cloned and expressed murine interleukin-3. Blood 65:1453 (1985).  142.  Suda T, Suda J, Ogawa M, Ihle JN. Permissive role of interleukin 3 (IL-3) In proliferation and differentiation of multipotential hemopoietic progenitors in culture. J Cell Physiol 124:182 (1985).  143.  Metcalf D, Johnson G R Burgess AW. Direct stimulation by purified GM-CSF of the proliferation of multipotential and erythroid precursor cells. Blood 55:138 (1980).  granulocyte-macrophage  - 56 -  144.  Metcalf D, Nicola NA. Proliferative effects of purified granulocyte colony-stimulating factor (G-CSF) on normal mouse hemopoietic cells. J Cell Physiol 116:198 (1983).  145.  Vadhan-Raj S, Keating MD, LeMaistre A, Hittelman WN, McCredie K. Trujillo JM, Broxmeyer HE, Henney C, Gutterman JU. Effects of recombinant human granulocyte-macrophage colony stimulating factor in patients with myelodysplastic syndromes. N Engl J Med 317:1545 (1987).  146.  Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW. Correction of the anemia of end-stage renal disease with recombinant human erythropoietin. Results of a combined phase I and II clinical trial. N Engl J Med 316:73 (1987).  147.  Nicola NA, Begley CG, Metcalf D. Identification of the human analogue of a regulator that induces differentiation in murine leukaemic cells. Nature 314:625 (1985).  148.  Eaves CJ. Unpublished observations.  149.  Metcalf D. The granylocyte-macrophage colony-stimulating factors. Science 229:16 (1985).  150.  Metcalf D. The granulocyte-macrophage colony stimulating factors. Cell 43:5 (1985).  151.  Castro-Malaspina H, Gay RE, Resnick G, Kapoor N, Myers P, Chiarieri D, McKenzie S, Broxmeyer HE, Moore MAS. Characterization of human bone marrow fibroblast colony-forming cells (CFU-F) and their progeny. Blood 56:289 (1980).  152.  Nicola NA, Metcalf D. Binding of ^I-labeled granulocyte colony-stimulating factor to normal murine hemopoietic cells. J Cell Physiol 124:313 (1985).  153.  Palaszynski EW, Ihle JN. Evidence for specific receptors for interleukin 3 on lymphokine-dependent cell lines established from long-term bone marrow cultures. J Immunol 132:1872 (1984).  154.  Spivak JL. The mechanism of action of erythropoietin. Int J Cell Cloning 4:139 (1986).  155.  Byrne PV, Guilbert LJ, Stanley E R Distribution of cells bearing receptors for a colony-stimulating factor (CSF-1) in murine tissues. J Cell Biol 91:848 (1981).  156.  Murthy SC, Eaves CJ, Krystal G. A simple three-step purification procedure for interleukin 3 involving absorption to fixed cells. Exp Hematol 17:997 (1989).  157.  Shadduck RK, Pigoli G, Caramatti C, Degliantoni G, Rizzoli V, Porcellini A, Waheed AL, Shiffer L. Identification of hemopoietic cells responsive to colony-stimulating factor by autoradiography. Blood 62:1197 (1983).  158.  Nicola NA, Metcalf D. Binding of iodinated multipotential colony-stimulating factor (interleukin-3) to murine bone marrow cells. J Cell Physiol 128:180 (1986).  159.  Mooibroek MJ, Wang JH. Integration of signal-transduction processes. Biochem Cell Biol 66:557 (1988).  160.  Sacca R Stanley E R Sherr CJ, Retterunier CW. Specific binding of the mononuclear phagocyte colony-stimulating factor CSF-1 to the product of the v-frns oncogene. Proc Natl Acad Sci USA 83:3331 (1986).  12  - 57 -  161.  Waterfield MD, Scrace GT, Whittle N, Stroobant P, Johnsson A, Wasteson A, Westermark B, Heldin C-H, Huang J S , Deuel TF. Platelet-derived growth factor is structurally related to the putative transforming protein p 2 8 of simian sarcoma virus. Nature 304:35 (1983). s i s  162.  Downward J , Yarden Y, Mayes E, Scrace G, Totty N, Stockwell P, Ullrich A, Schlessinger J , Waterfield MD. Close similarity of epidermal growth factor receptor and v-erb-B oncogene protein sequences. Nature 307:521 (1984).  163.  Wheeler EF, Rettenmier CW, Look AT, Sherr CJ. The v-fms oncogene induces factor independence and tumorigenicity i n CSF-1 dependent macrophage cell line. Nature 324:377 (1986).  164.  Young DC, Wagner K, Griffin JD. Constitutive expression of the granulocytemacrophage colony-stimulating factor gene i n acute myeloblastic leukemia. J C l i n Invest 79:100 (1987).  165.  Le Beau MM, Pettenati MJ, Lemons RS, Diaz MO, Westbrook CA, Larson RA, Sherr C J , Rowley JD. Assignment of the GM-CSF, CSF-1, and FMS genes to h u m a n chromosome 5 provides evidence for linkage of a family of genes regulating hematopoiesis and for their involvement in the deletion (5q) i n myeloid disorders. Cold Spring Harbor Symposia on Quantitative Biology, vol LI. p.899 (1986).  166.  Le Beau MM, Lemons RS, Espinosa III R, Larson RA, A r a l N, Rowley JD. Interleukin-4 and interleukin-5 map to h u m a n chromosome 5 i n a region encoding growth factors and receptors and are deleted i n myeloid leukemias with a del(5q). Blood 73:647 (1989).  167.  Le Beau MM, Westbrook CA, Diaz MO, Larson RA, Rowley JD, Gasson J C , Golde DW, Sherr C J . Evidence for the involvement of GM-CSF and FMS in the deletion (5q) i n myeloid disorders. Science 231:984 (1986).  168.  Quesenberry PJ. Synergistic hematopoietic growth factors. Int J Cell Cloning 4:3 (1986).  169.  Warren D J , Moore MAS. Synergism among interleukin 1, interleukin 3, and interleukin 5 i n theproduction of eosinophils from primitive hemopoietic stem cells. J Immunol 140:94 (1988).  170.  Rennick D, Yang G, Muller-Sieburg C, Smith C, A r a l N, Takabe Y, Gemmell L. Interleukin 4 (B-cell stimulatory factor 1) can enhance or antagonize the factor-dependent growth of hemopoietic progenitor cells. Proc Natl A c a d Sci USA 84:6889 (1987).  171.  Peschel C, Paul WE, Ohara J , Green I. Effects of B cell stimulatory factor-1 /Interleukin 4 on hematopoietic progenitor cells. Blood 70:254 (1987).  172.  Hirano T, Yasukawa K, Harada H, Taga T, Watanabe Y, Matsuda T, Kashiwamura S, Nakajima K, Koyoma K, Iwamatsu A, Tsunasawa S, Sakiyama F, M a t s u i H, Takahara Y, Taniguchi T, Kishimoto T. Complementary DNA for a novel h u m a n interleukin (BSF-2) that induces B lymphocytes to produce immunoglobulin. Nature 324:73 (1986).  173.  Ogawa M, C l a r k SC. Synergistic interaction between interleukin-6 and interleukin-3 i n support of stem cell proliferation i n culture. Blood Cells 14:329 (1988).  - 58 -  174.  Kaushansky K, Lin N, Adamson JW. Interleukin 1 stimulates fibroblasts to synthesize granulocyte-macrophage and granulocyte colony-stimulating factors. Mechanism for the hematopoietic response to inflammation. J Clin Invest 81:92 (1988).  175.  Zsebo KM, Yuschenkoff VN, Schiffer S, Chang D, McCall E, Dinarello CA, Brown MA, Altrock B, Bagby J r GC. Vascular endothelial cells and granulopoiesis: Interleukin-1 stimulates release of G-CSF and GM-CSF. Blood 71:99 (1988).  176.  Michalevicz R, Katz F, Stroobant P, Janossy G, Tindle RW, Hoffbrand AV. Platelet-derived growth factor stimulates growth of highly enriched multipotent haemopoietic progenitors. B r J Haematol 63:591 (1986).  177.  Broxmeyer HE, Williams DE, L u L, Cooper S, Anderson SL, Beyer GS, Hoffman R, R u b i n BY. The suppressive influences of h u m a n tumor necrosis factors on bone marrow hematopoietic progenitor cells from normal donors and patients with leukemia: Synergism of tumor necrosis factor and interferon-y. J Immunol 136:4487 (1986).  178.  Delwiche F, Raines E, Powell J , Ross R, Adamson J. Platelet-derived growth factor enhances i n vitro erythropoiesis via stimulation of mesenchymal cells. J Clin Invest 76:137 (1985).  179.  Munker R, Gasson J , Ogawa M, Koeffler HP. Recombinant h u m a n TNF induces production of granulocyte-monocyte colony-stimulating factor. Nature 323:79 (1986).  180.  Bol SJL, Visser JWM, Williams N, V a n den Engh GJ. Physical characterization of haemopoietic progenitor cells by equilibrium density centrifugation. In: Bloemendal H. Cell Separation Methods. Amsterdam: Elsevier/North Holland (1977).  181.  Miller RG. Separation of cells by velocity sedimentation. In: Pain RH, Smith BJ. New Techniques i n Biophysics and Cell Biology, vol 1. London: J o h n Wiley and Sons, Ltd. (1973).  182.  Metcalf D, Moore MAS, Shortman K. Adherence column and bouyant density separation of bone marrow stem cells and more differentiated cells. J Cell Physiol 78:441 (1971).  183.  Kerk DK, Henry EA, Eaves AC, Eaves CJ. Two classes of primitive pluripotent hemopoietic progenitor cells: Separation by adherence. J Cell Physiol 125:127 (1985).  184.  Williams DE, Straneva J E , Shen R-N, Broxmeyer HE. Purification of murine bone-marrow-derived granulocyte-macrophage colony-forming cells. Exp Hematol 15:243 (1987).  185.  Pretlow TG, Pretlow TP. Centrifugal elutriation (counterstreaming centrifugation) of cells. Cell Biophys 4:195 (1979).  186.  Nijhof W, Wierenga PK. Isolation and characterization of the erythroid progenitor cell: CFU-E. J Cell Biol 96:386 (1983).  187.  Nijhof W, Wierenga PK. Isolation of hemopoietic pluripotent stem cells from thiamphenicol-pretreated mice. Exp Cell Res 155:583 (1984).  188.  Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495 (1975).  - 59 189.  Koizumi S, Fine RL, Curt GA, Griffin JD. Enrichment of myeloid progenitor cells from normal human bone marrow using an immune-rosette technique. Exp Hematol 13:560 (1985).  190.  Mouchiroud G, Berthier R, Newton IA, Chapel A. Monoclonal antibodies against human hemopoietic cells and the separation of progenitor cells from bone marrow. Exp Hematol 13:566 (1985).  191.  Matsson P, Pallavicini MG, Summers L. Quantitativeflowcytometric and clonogenic evaluation of glass bead affinity fractionation of antibody-labeled murine bone marrow. J Immunol Methods 105:45 (1987).  192.  Bertoncello I, Bradley TR, Hodgson GS. The concentration and resolution of primitive hemopoietic cells from normal mouse bone marrow by negative selection using monoclonal antibodies and Dynabead monodisperse magnetic microspheres. Exp Hematol 17:171 (1989).  193.  McCarthy KF, Hale ML, Fehnel PL. Purification and analysis of rat hematopoietic stem cells by flow cytometry. Cytometry 8:296 (1987).  194.  Bauman JGJ, Mulder AH, Van Den Engh GJ. Effect of surface antigen labeling on spleen colony formation: Comparison of the indirect immunofluorescence and the biotin-avidin methods. Exp Hematol 13:760 (1985).  195.  Van Den Engh G, Visser J. Light scattering properties of pluripotent and committed haemopoietic stem cells. Acta Haematol 62:289 (1979).  196.  Shapiro HM. Practical Flow Cytometry. 2nd edition New York: Alan R. Liss Inc. (1988).  197.  Pont S. Thy-1: A lymphoid cell subset marker capable of delivering an activation signal to mouse T lymphocytes. Biochimie 69:315 (1987).  198.  Kroczek RA, Gunter KC, Seligmann B, Shevach EM. Induction of T cell activation by monoclonal anti-Thy-1 antibodies. J Immunol 136:4379 (1986).  199.  Gunter KC, Kroczek RA, Shevach EM, Germain RN. Functional expression of the murine Thy-1.2 gene in transfected human T cells. J Exp Med 163:285 (1986).  200.  Kroczek RA, Gunter KC, Germain RN, Shevach EM. Thy-1 functions as a signal transduction molecule in T lymphocytes and transfected B lymphocytes. Nature 322:181 (1986).  201.  Williams DE, Boswell HS, Floyd AD, Broxmeyer HE. Pluripotential hematopoietic stem cells in post-5-fluorouracil murine bone marrow express the Thy-1 antigen. J Immunol 135:1004 (1985).  202.  Miller BA, Lipton JM, Linch DC, Burakoff SJ, Nathan DG. Thy-1 is a differentiation antigen that characterizes immature murine erythroid and myeloid hematopoietic progenitors. J Cell Physiol 123:25 (1985).  203.  Schrader JW, Battye F, Scollay R. Expression of Thy-1 antigen is not limited to T cells in cultures of mouse hemopoietic cells. Proc Natl Acad Sci USA 79:4161 (1982).  204.  Zinkernagel RM, Doherty PC. Advances in Immunology, vol 27. New York: Academic Press (1979).  - 60 205.  Benacerraf B, Unanue E R Textbook of Immunology. Baltimore: Williams and Wilkins (1980).  206.  Mulder AH, Bauman JGJ, Visser JWM, Boersma WJA, Van Den Engh GJ. Separation of spleen colony-forming units and prothymocytes by use of a monoclonal antibody detecting an an H-2K determinant. Cell Immunol 88:401 (1984).  207.  Berridge MV, Ralph SJ, Tan AS, Jeffery K. Changes In cell surface antigens during stem cell ontogeny. Exp Hematol 12:121 (1984).  208.  Okamoto T, Kanamaru A, Hara H, Nagal K. Changes In expression of major histocompatibility complex (MHC) class-I antigen on hematopoietic progenitors during murine development. Exp Hematol 15:190 (1987).  209.  Klein J. The major histocompatibility complex of the mouse. Science 203:516 (1979).  210.  Basch RS, Janossy G, Greaves MF. Murine pluripotential stem cells lack Ia antigen. Nature 270:520 (1977).  211.  Fitchen JH, Ferrone S. Expression of I subregion antigens on murine hematopoietic stem cells. J Supramol Struc 14(Suppl 5): 107 (1981).  212.  Szilvassy SJ. Unpublished observations.  213.  Harris R A Hogarth PM, Penington DG, McKenzie IFC. Qa antigens and their differential distribution on lymphoid, myeloid and stem cells. J Immunogenet 11:265 (1984).  214.  Kincade PW, Flaherty L, Lee G, Watanabe T, Michaelson J. Qa antigen expression on functional lymphoid, myeloid, and stem cells in adult mice. J Immunol 124:2879 (1980).  215.  Bertoncello I, Bartelmez SH, Bradley TR, Hodgson GS. Increased Qa-m7 antigen expression is characteristic of primitive hemopoietic progenitors in regenerating marrow. J Immunol 139:1096 (1987).  216.  Harris RA, Hogarth PM, McKenzie IFC, Penington DG. Differential expression of Qa-m2 alloantigen on murine hemopoietic progenitor cells. Selective enrichment for megakaryocyte progenitors. Exp Hematol 11:527 (1983).  217.  Harris RA, Hogarth PM, Wadeson LJ, Collins P, McKenzie IFC, Penington DG. An antigenic difference between cells forming early and late haematopoietic spleen colonies (CFU-S). Nature 307:638 (1984).  218.  Bertoncello I, Bartelmez SH, Bradley T R Stanley ER, Harris RA, Sandrin MS, Kriegler AB, McNiece IK, Hunter SD, Hodgson GS. Isolation and analysis of primitive hemopoietic progenitor cells on the basis of differential expression of Qa-m7 antigen. J Immunol 136:3219 (1986).  219.  Visser JWM, Bol SJL. A two-step procedure for obtaining 80-fold enriched suspensions of murine pluripotent hemopoietic stem cells. Stem Cells 1:240 (1981).  220.  Visser JWM, Eliason JF. In vivo studies on the regeneration kinetics of enriched populations of haemopoietic spleen colony-forming cells from normal bone marrow. Cell Tissue Kinet 16:385 (1983).  - 61 221.  Darzyrikiewicz Z, Staiano-Coico L, Melamed MR Increased mitochondrial uptake of rhodamine 123 during lymphocyte stimulation. Proc Natl Acad Sci USA 78:2383 (1981).  222.  Cohen RL, Muirhead KA, Gill JE, Waggoner AS, Horan PK. A cyanine dye distinguishes between cycling and non-cycling fibroblasts. Nature 290:593 (1981).  223.  Mulder AH, Visser JWM. Separation and functional analysis of bone marrow cells separated by rhodamine-123 fluorescence. Exp Hematol 15:99 (1987).  224.  Visser JWM, de Vries P. Isolation of spleen-colony forrning cells (CFU-s) using wheat germ agglutinin and rhodamine 123 labeling. Blood Cells 14:369 (1988).  225.  Aihara Y, Buhring H-J, Aihara M, Klein J. An attempt to produce "pre-T" cell hybridomas and to identify their antigens. Eur J Immunol 16:1391 (1986).  226.  Van De Rijn M, Helmfeld S, Spangrude GJ, Weissman IL. Mouse hematopoietic stem-cell antigen Sca-1 is a member of the Ly-6 antigen family. Proc Natl Acad Sci USA 86:4634 (1989).  227.  Muller-Sieburg CE, Whitlock CA, Weissman IL. Isolation of two early B lymphocyte progenitors from mouse marrow: A committed pre-pre-B cell and a clonogenic Thy-l^ hematopoietic stem cell. Cell 44:653 (1986).  0  228.  Albert E. Nomenclature for factors of the HLA system. Hum Immunol 11:117 (1984).  229.  Linen DC, Nadler LM, Luther EA, Lipton JM. Discordant expression of human la-like antigens on hematopoietic progenitor cells. J Immunol 132:2324 (1984).  230.  Sparrow RL, Williams N. The pattern of HLA-DR and HLA-DQ antigen expression on clonable subpopulations of human myeloid progenitor cells. Blood 67:379 (1986).  231.  Robinson J, Sieff C, Delia D, Edwards PAW, Greaves M. Expression of cell-surface HLA-DR HLA-ABC and glycophorin during erythroid differentiation. Nature 289:68 (1981).  232.  Torok-Storb B, Symington FW. Class II MHC antigens and erythropoiesis. In: Solheim BG, Moller E, Ferrone S. HLA class-II antigens. A comprehensive review of structure and function. Berlin: Sprlnger-Verlag (1986).  233.  Moore MAS, Broxmeyer HE, Sheridan APC, Meyers PA, Jacobsen N, Winchester RJ. Continuous human bone marrow culture: la antigen characterization of probable pluripotential stem cells. Blood 55:682 (1980).  234.  Keating A, Powell J, Takahashi M, Singer JW. The generation of human long-term marrow cultures from marrow depleted of la (HLA-DR) positive cells. Blood 64:1159 (1984).  235.  Falkenburg JHF, Vaart-Duinkerken NVD, Veenhof WFJ, Goselink HM, Van Eeden G, Parlevliet J, Jansen J. Complement-dependent cytotoxicity in the analysis of antigenic deterrninants on human hematopoietic progenitor cells with HLA-DR as a model. Exp Hematol 12:817 (1984).  236.  Falkenburg JHF, Fibbe WE, Goselink HM, van Rood JJ, Jansen J. Human hematopoietic progenitor cells in long-term cultures express HLA-DR antigens and lack HLA-DQ antigens. J Exp Med 162:1359 (1985).  - 62 -  237.  Lu L, Walker D, Broxmeyer HE, Hoffman R Hu W, Walker E. Characterization of adult human marrow hematopoietic progenitors highly enriched by two-color cell sorting with MylO and major histocompatibility class II monoclonal antibodies. J Immunol 139:1823 (1987).  238.  Civin CI, Strauss LC. Brovall C, Fackler MJ, Schwartz JF, Shaper JH. Antigenic analysis of hematopoiesis. III. A hematopoietic progenitor cell surface antigen defined by a monoclonal antibody raised against KG-la cells. J Immunol 133:157 (1984).  239.  Boyd AW. Human leukocyte antigens: An update on structure, function and nomenclature. Pathology 19:329 (1987).  240.  Civin CI, Loken MR Cell surface antigens on human marrow cells: Dissection of hematopoietic development using monoclonal antibodies and multiparameter flow cytometry. Int J Cell Cloning 5:267 (1987).  241.  Berenson RJ, Andrews RG, Bensinger WI, Kalamasz D, Knitter G, Buckner CD, Bernstein ID. Antigen CD34 marrow cells engraft lethally irradiated baboons. J Clin Invest 81:951 (1988). +  242.  Mintz B, Anthony K, Litwin S. Monoclonal derivation of mouse myeloid and lymphoid lineages from totipotent hematopoietic stem cells experimentally engrafted in fetal hosts. Proc Natl Acad Sci USA 81:7835 (1984).  - 63 -  C H A P T E R  M A T E R I A L S AND  1)  CELLS.  A)  Animals.  II  METHODS  Six- to 12-week old (C57B1/6J x C S H / H e J l F i (B6C3FJ) male and female mice bred and maintained in the animal facility of the B.C. Cancer Research Centre (Vancouver, B.C., Canada) from parental strain breeders originally obtained from the Jackson Laboratories (Bar Harbor, ME) were used i n all experiments. B 6 C 3 F J mice are homozygous for the Thy-1.2 allele and are of the H-2K /H-2K haplotype. b  B)  k  Preparation of Marrow Cell Suspensions.  Bone marrow cells were obtained from the femurs of either normal male mice, or male mice injected intravenously (IV) 4 days previously with a sterile solution of 5-fluorouracil (5-FU) in phosphate-buffered saline (PBS) at a dose of 150 mg per kg body weight (day 4 5-FU cells). Marrow cells were flushed asepticalfy from the femoral cavity with alpha medium containing 2 % fetal calf serum (FCS) (2%-oc) using a 21 gauge needle and a 3 ml syringe. Clumps of cells were dispersed by gentle aspiration through the needle and the cell suspension was then filtered through a double layer of 20 um nylon mesh. Cells were counted and diluted i n 2%-ct to a concentration appropriate for subsequent plating or injection. In antibody labelling experiments, marrow cells were collected i n Hank's balanced salt solution (HBSS) containing 2 % F C S and 0.02% sodium azide (HFN-buffer). Cells were washed In HFN, resuspended i n  - 64 NH C1-Tris (pH 7.2) for 5 minutes at room temperature to h/se red blood cells, washed once 4  again In HFN, counted and then diluted to a concentration of 5 x 10^ viable cells/ml HFN for immunostaining and subsequent plating or injection.  2)  HEMOPOIETIC CELL PURIFICATION.  A)  Monoclonal Antibodies.  A biotin-conjugated, purified rat anti-Thy-1.2 monoclonal antibody (MoAb) (clone 30-H12, Becton Dickinson, Mountain View, CA) was titrated and used at a dilution of 1:10 in HFN for staining in green. Mouse anti-H-2K MoAb (IgGj) (hybridoma T I B 139, American b  Type Culture Collection) either purified from ascites fluid generated in pristane-primed BALB/c mice or used directly as crude culture supernatant from plateau cultures of hybridoma cells grown in Dulbecco's modified Eagle's medium (DMEM) with 10% FCS and 100 U/ml of highly purified recombinant human IL-6 (kindly provided by L. Aarden, Red Cross Blood Transfusion Laboratory, Amsterdam, The Netherlands), was labelled with R-phycoerythrin (R-PE) by generating tetramolecular antibody complexes (tetramers) with anti-R-PE MoAb's by simple mixing of the component antibodies as previously described (1). In this case, anti-H-2K IgGj b  (approximately 10 pg/ml) was mixed with anti-R-PE IgGj (approximately 500 ug/ml) and rat anti-mouse IgG i (approximately 500 ug/ml) to give a final molar ratio of 1:10:11, respectively, of the different component antibodies. This favours the formation of bispecific tetramers containing both anti-H-2K and anti-R-PE relative to the formation of monospecific anti-H-2K D  x anti-H-2K tetramers. The solution of tetramers was then used as the primary reagent for D  H-2K -specific staining in red. Control cells stained with tetramers containing only R-PE b  antibodies (anti-R-PE x anti-R-PE), prepared by omitting the anti-H-2K reagent, were used to b  measure the level of nonspecific red fluorescence. In some experiments in which sorted cells were subsequently injected for assessment of competitive repopulatlng ability, directly R-PE  D  - 65 conjugated (Fab)2 fragments of purified anti-H-2K were used at 5 ng/ml instead of tetramers D  because this reagant reduced the seeding efficiency of CFU-S to the spleen to a lesser extent than the latter.  B)  Indirect Double-Immunostainlng of Bone Marrow Cells.  For staining, 50 ul to 1.5 ml aliquots of washed cells (at 5 x 10® c/ml) were mixed with an equal volume of either biotinylated anti-Thy-1.2 MoAb, anti-H-2K^ x anti-R-PE tetramers, or both in the case of double-stained samples, incubated for 30 to 45 minutes on ice, washed twice in HFN, resuspended in a solution of fluorescein isothlocyanate (FrTC)-conjugated avidin (5 ug/ml) and/or R-PE (2 ug/ml) in HFN, and then incubated for 30 to 45 minutes more on ice, washed three times in HFN, resuspended to >5 x 10^ cells/ml in HFN, and passed through nylon mesh (20 um) to remove clumps and debris before FACS analysis and sorting. Controls were treated identically substituting HFN or monospecific anti-R-PE tetramers for specific antibodies as required. Cell viability was deterrnined by staining with 0.1% nigrosin and was consistently >90%.  C)  FACS Analysis and Sorting.  Red and green fluorescence, and forward (0.5-13°) and orthogonal (65-115°) light scattering (FLS and OLS) properties were analyzed with a FACS 440 (Becton-Dickinson, Sunnyvale, CA) with the laser at 488 nm (300 mW).  FrTC fluorescence was separated from  R-PE fluorescence with 525/10 nm and 575/26 nm band pass filters in combination with a dichroic mirror (560 nm). For double-stained cells, the signal from the greenfluorescenceinto the detector used to measure red fluorescence was electronically compensated to background levels with cells stained with biotinylated anti-Thy-1.2 and avidin-FrTC only. A logarithmic amplifier was used for all fluorescence signals. The horizontal position of the profiles obtained  - 66 In separate experiments was standardized as follows: For FLS, the peak of the second major population was set at channel 120; for OLS, the major lymphocyte peak was set at channel 30; and for Thy-1.2 fluorescence, the major peak of Thy-1.2-neagatlve cells was set at channel 90. No adjustments were made i n the position of the H-2K fluorescence profile. For analysis, b  2 x 1 0 cells were evaluated per sample. In sorting experiments, cells were sorted at a rate of 4  not more than 2 x 10^ cells/sec, and Inlet and collection tubes were cooled on ice. Sorted cells were collected In 5 0 % F C S i n HBSS. All sorts were performed with the F A C S i n F D E mode, and the abort rate for coincident cells was reproducibly <10%.  3)  ASSAYS.  A)  Methylcellulose Assays.  All sorted and unsorted marrow cells were plated in 35 mm  petri dishes i n 1.1 ml culture  mixtures consisting of 0.8% methylcellulose in alpha medium contaixilng 3 0 % FCS, 1 % bovine serum albumin (BSA), 10" M P-mercaptoethanol, 3 U/ml partially purified h u m a n urinary 4  erythropoietin and 2 % pokeweed mltogen-stimulated mouse spleen cell conditioned medium (PWM-SCCM) (2). For day 4 5-FU marrow cells, 1 0 % agar-stimulated h u m a n leukocyte conditioned medium (LCM) (3) was also added. Unsorted marrow cells from both normal and day 4 5-FU mice were plated at a concentration of 3 x 1 0 cells per dish. Sorted cells were 4  plated at lower concentrations down to 500 cells per dish depending on the degree of enrichment anticipated from preliminary experiments. Cultures were incubated at 37°C in a humidified atmosphere of 5 % CO2 i n air, and colonies were counted in situ after 12 to 14 days (normal marrow) or 21 days (day 4 5-FU marrow). Macroscopic multilineage colonies (termed CFU-GEMM (macro) derived) were scored as previously described (4). All other colonies (>20 cells) present were categorized as follows: Those containing more than four clusters of small erythroblasts but not of macroscopic size are called BFU-E derived (even though many of these  - 67 colonies contained cells of other lineages). Those that did not contain detectable erythroblasts and were composed primarily of granulocytes and macrophages are called CFU-GM derived.  B)  CFU-S Assays.  A l l recipients were female mice administered 800 to 850 cGy total body irradiation (TBI; 124 cGy/min) with a Phillips 250 kVp X-ray machine. This dose was sufficient to completely eliminate endogenous spleen colony formation (to <0.1/spleen). Irradiated animals were injected IV with 0.5 to 50 x 10^ sorted or unsorted, normal or day 4 5-FU marrow cells per mouse, depending on the number of spleen colonies anticipated, and were killed 9 or 12 days later as indicated for macroscopic spleen colony counts (5). Experiments i n which mice were injected with double-stained but unseparated marrow cells showed that the number of spleen colonies produced both 9 and 12 days later was reduced relative to controls by approximately 50%. Data for day 12 spleen colony formation are shown i n Table II. To determine whether staining also reduced the capacity of the injected cells to regenerate hemopoiesis i n the marrow, I compared the femoral content of i n vitro clonogenic progenitors i n mice injected with stained or unstained cells. Twelve days after transplantation of stained cells, the number of clonogenic cells per femur was also decreased to approximately half of the control value (data not shown). B a u m a n n et al. (6) reported a comparable reduction i n detectable CFU-S after standard indirect antibody staining of marrow cells and showed that the reduction could be partially abrogated by pretreating mice with antimacrophage agents such as X-carrageenan (7). Although I was able to confirm this with the double-stained cells (Table II), I chose not to adopt this "correctional" procedure since the effect was only partial and the long-term consequences of such treatments on host stem cell survival and/or reactivation are unknown. The fraction, f, of unstained cells able to colonize the spleen of irradiated but otherwise untreated recipients was assumed to be 1 0 % (a mean of various reported values for f as 3-25% (8,9,10)) and for stained cells was therefore assumed to be 5%.  - 68 -  Table II.  Treatment of Cells Injected  Reduction i n CFU-S Detectable After Double-Staining of Marrow Cells  a  No. of Spleen Colonies (Day 12)  Pretreatment of Recipients *  c  1  Expt. 1  Expt. 2  Expt. 3  None  None  12.8 + 0.8 (5)  15.2 + 1.4 (4)  16.9 + 1.6 (7)  Double-Stained  None  6.7 + 0.7 (3)  5.3 + 0.3 (3)  9.7+1.3 (6)  12.0 + 1.7 (3)  15.0 + 1.7 (6)  8.0 + 0.9 (4)  12.0 + 2.0 (5)  None  X-carrageenan  ND  Double-Stained  X-carrageenan  ND  a b c d  d  A l l mice were injected intravenously with 10^ unstained or double stained normal syngeneic marrow cells, 0.5 mg A.-carrageenan per mouse was injected intraperitoneally immediately following irradiation and approximately 20 hours before transplantation of cells, Values shown represent the mean + S E M of counts from (n) spleens, ND = not done.  - 69 C)  Competitive Repopulation Assays.  Female mice were given 800 to 850 cGy total body X-irradiation as for CFU-S assays. Three to 15 hours after irradiation, recipients were injected IV with male "test" cells and 1 to 2 x 10^ "compromised" marrow cells from female mice that contained approximately normal numbers of CFU-S and i n vitro clonogenic cells (Table III) but had been compromised i n their competitive long-term repopulating ability by having been previously subjected to two cycles of marrow regeneration (11). This was achieved by transplanting 10® female B 6 C 3 F ^ marrow cells/mouse into a group of irradiated (800 to 850 cGy) syngeneic female recipients, and then 5 to 8 weeks later transplanting 10® marrow cells from these primary recipients into each of a second group of irradiated syngeneic female recipients. The secondary recipient mice were then used as donors of compromised marrow cells 1 to 3 months later. In experiments i n which male recipients or male "compromised" marrow cells were used, these were prepared exactly as their female counterparts above. Recipients of mixed transplants of test and compromised marrow cells were killed 1 to 3 months after injection as indicated. Isolated marrow, spleen and thymus cells were pelleted i n PBS and stored at -20°C before  DNA  extraction and Southern analysis using the Y-chromosome-specific probe, pY2. Using this method, it was usually possible to detect as low as a 1 % (and reproducibly, 5%) male contribution to total D N A  Initially, a m i n i m u m of a 2 0 % male contribution was set as the  requirement for a tissue to be scored as positive i n the competitive repopulation assay to ensure that no false positives were included i n the analysis (Chapter III). Subsequently, it was determined that a 5 % criterion did not alter the data analysis so this was used as the m i n i m u m male contribution for positivity (Chapter V). Compromised marrow cells (10^) alone were sufficient to allow injected recipients to survive long-term, but doses up to 10® were ^ s u f f i c i e n t to outcompete the marrow, spleen and thymus repopulating capacity of a graft of 1 0 day 4 5-FU marrow cells (See Chapter V). 4  - 70 -  Table III.  Frequencies of Clonogenic Progenitors i n Unstained Suspensions of Normal and Compromised Marrow Cells  Progenitor Assayed  Frequency (per 1 0  Day 12 C F U - S Day 9 C F U - S CFU-GEMM CFU-GM BFU-E  a b  b  b  D  cells)  a  Normal Donors  Compromised Donors  20 + 3 (9)  14 + 2 (3)  26 + 2 (4)  17 + 2 (3)  13 + 6 (18)  5 + 1 (18)  340 + 20 (18)  390 + 40 (18)  23 + 5 (18)  23 + 5 (18)  Shown are the mean + S E M of values measured in (n) different experiments, Not corrected for seeding efficiency.  - 71 4)  RETROVIRAL MARKING O F PURIFIED M A R R O W CELLS.  A)  Recombinant Retrovirus.  A replication-defective, recombinant retrovirus, T K n e o l 9 , which carries the bacterial gene for neomycin resistance (neo ) under the control of the herpes simplex virus (HSV) 1  thymidine kinase (TK) promoter (Figure 13, Chapter IV) was derived from a myc/neo retrovirus provided by Dr. B. Vennstrom (12,13) by deletion of myc sequences and inversion of the TKneo insert (14). Helper-free T K n e o l 9 viral producer clones were generated i n the psi-2 (¥-2) ecotropic packaging cell line (15) using published procedures (16). The ¥-2 cells were maintained i n D M E M with 1 0 % calf serum. The clone selected produced TKneo 19 virus at a titre of >5 x 10^ per m l as assayed by generation of G418-resistant colonies on NIH 3T3 cells. Virus-containing supernatant from cultures of these cells was found to be negative for the production of helper virus as assessed by attempts to serially transfer TKneo 19 on NIH  3T3  cells (17).  B)  Infection of Purified Marrow Cells.  The infection protocol was based on preliminary studies with unfractionated bone marrow collected 4 days after IV injection of 150 mg/kg 5-FU that yielded gene transfer efficiencies of 7 0 % i n day 12 CFU-S without preselection (Fraser CC, Szilvassy SJ, Eaves C J & Humphries RK, unpublished data). Aliquots of 150 to 2000 purified marrow cells were placed into mlcrocultures containing 0.3 m l of supernatant from logarithmic-phase TKneo 19-producing ¥-2 cell cultures with 4 ug/ml Polybrene, 1 0 % (v/v) PWM-SCCM (18),  and  1 0 % (v/v) agar-stimulated h u m a n L C M (3). After 6 to 8 hours at 37°C, half of the medium was replaced with an equal volume of freshly prepared virus-containing medium, and the cultures were incubated a further 12 to 14 hours. Cells from each well were then collected separately.  - 72 washed, and Injected Into Irradiated (800 to 850 cGy) female B 6 C 3 F } mice (one well per mouse) together with (Chapter IV, experiments 1 and 2), or 2 hours after (Chapter IV, experiment 3) another injection of 2 x 10^ syngeneic "compromised" female marrow cells that had been previously subjected to two cycles of serial marrow transplantation and regeneration (11) as described above.  C)  Spleen Colony Analysis.  Irradiated (800 to 850 cGy) female B 6 C 3 F j mice were injected IV with 5 x 1 0 marrow 4  cells from competitively repopulated mice and sacrificed 12 days later. Well-isolated macroscopic spleen colonies were dissected, pelleted i n PBS, and stored at -20°C prior to DNA extraction.  D)  Separation of Marrow Macrophage and Splenic Lymphocyte Subpopulations.  Suspended marrow cells (2.5 x 10®) were placed for 24 hours at 37°C i n 60 m m  2  tissue  culture dishes containing 5 m l of RPMI 1640 medium with 1 0 % FCS, 1 % PWM-SCCM, and 5 % E M T 6 cell-conditioned medium as a source of growth factors (18,19). Nonadherent cells were removed by washing, and the adherent macrophages were amplified i n the same medium by culture for 1 to 2 weeks. Harvested cells were frozen until DNA extraction. Spleen cells were first fractionated by adding 5 x 1 0 cells i n 1 m l of RPMI 1640 medium 7  with 5 % F C S to a 3 m l nylon wool column, which was then incubated for 1 hour at 37°C prior to elution by extensive washing of the nonadherent fraction (primarily T-lymphocytes) (20). Adherent cells were detached by gently agitating the nylon wool plug for 5 to 10 rnin i n PBS containing 10 m M  EDTA, and B-lymphocytes were isolated by panning of these cells for 1 hour  at 37°C i n 100 mm-diameter plastic dishes (<10^ cells per dish) precoated with unpurified  - 73 rabbit anti-mouse immunoglobulin (21). After washing away most of the nonspecifically bound cells, the adherent B-h/mphocytes were removed and frozen as for other cell samples.  5)  M O L E C U L A R ANALYSES.  A)  Southern Analysis with Y-Specific Probe.  DNA was purified from sodium dodecyl sulfate (SDS)/proteinase K-digested cells by phenol/chloroform extraction (22). After dialysis against I X Tris-EDTA buffer (3 m M mM  Tris, 0.2  EDTA, pH 7.5, TE), 10 pg samples of DNA were digested with PvuII or H i n d l l l at 2 - 5 U/ug  for 4 - 12 hours at 37°C i n the buffer recommended by the manufacturer (Bethesda Research Laboratories (BRL), Gaithersburg, MD). Samples of male and female DNA from normal B 6 C 3 F } mice were used as positive and negative controls. After ethanol precipitation, DNA  was  dissolved i n 20 ul of T E buffer, electrophoresed through a 1 % agarose gel, and transferred to nitrocellulose membranes (Schleicher & Schuell, Mandel Scientific Company, Rockwood, Ontario) (22). Blots were air-dryed and baked under vacuum at 80°C for 2 hours. Membranes were prehybridized for 2 hours at 68°C i n 50 m l of a buffer containing 3 X sodium saline citrate (SSC), 4 X Denhardt's solution and 0.5 mg/ml denatured salmon sperm DNA. conditions were the same except for the inclusion of 0.1% SDS, 3 m M to 0.1 mg/ml denatured salmon sperm DNA.  Hybridization  Tris, and the reduction  Blots of PvuII- or Hindlll-digested DNA were  probed with the pY2 plasmid, which contains a 720 base pair (bp) M b o l fragment of the Y-chromosome from male BALB/c mice cloned into the B a m H I site of pBR322 (23).  pY2 probe  was ^ P - l a b e l l e d to high specific activity by nick-translation with a kit purchased from BRL. After hybridization for 18 to 20 hours at 68°C, filters were washed at a final stringency of 0.1% SDS, 0. I X SSC, and 0.1% sodium pyrophosphate at 65°C. Autoradiography was performed at  - 74 -  -70°C with Kodak XAR-5 film for 24 to 72 hours. Under the conditions used, > 1 % (and reproducibly, 5%) male DNA could be detected.  B)  Southern Analysis with Neo -Speciflc Probe. r  D N A extraction, purification and digestion was performed as described above except for the use of the restriction enzymes B a m H I , H i n d l l l or E c o R I instead of PvuII. Samples of TKneol9-infected mouse spleen colony DNA and uninfected, normal male B 6 C 3 F J DNA were used as positive and negative controls. Electrophoresis was performed as described above, and DNA was transferred to nylon membranes (Zeta-Probe, Bio-Rad Laboratories, Richmond, CA). Blots were prehybridized for 2 hours at 60°C i n 20 m l of a buffer containing 0.9 M NaCl, 1 0 % formamide, 1 % SDS, 2 m M EDTA, 1 % nonfat dried milk a n d 0.5 mg/ml denatured salmon sperm DNA. Hybridization conditions were the same except for the inclusion of 1 0 % dextran sulfate (Sigma Chemical Co., St. Louis, MO). Blots of BamHI-, H i n d l l l - or EcoRI-digested DNA were probed with a 2.3 kbp B a m H I subfragment of the T K n e o l 9 retrovirus containing only the n e o gene a n d the T K promoter [^ P]oligonucleotide-labelled by using the multiprime labelling r  2  method with a kit purchased from Amersham. After hybridization overnight at 60°C, blots were washed and autoradiographed as described above. In some cases, blots were stripped for reprobing by twice gently agitating them for 15 m i n i n 0. I X SSC/0.5% SDS, which was boiled and allowed to cool to room temperature.  C)  Spot Blot Analysis.  5 ug samples of undigested DNA were ethanol precipitated, dissolved i n 300 ul of spotting buffer (60 m M Tris, 0.2 N NaOH, 6 X SSC), heat denatured a n d neutralized with 80 pi 1 M Tris (pH 7.5) prior to spotting directly onto nitrocellulose membranes. Wells were then rinsed with an equivalent volume of 5 X SSC.  5 ug mixtures of male and female D N A i n serially (3-fold)  - 75 -  decreasing ratios from 100% male DNA to 1% and 0% male DNA were prepared and included i n each spot blot as a titration standard. Nitrocellulose membranes were then analysed as described above using the Y-chromosome specific probe. Under the conditions used, >10% male DNA i n such mixtures could be consistently detected.  - 76 REFERENCES 1. Lansdorp PM, Aalberse RC, Bos R, Schutter WG, V a n Bruggen E F J . Cyclic tetramolecular complexes of monoclonal antibodies: a new type of cross-linking reagent. E u r J Immunol 16:679 (1986). 2. Kerk DK, Henry EA, Eaves AC, Eaves C J . Two classes of primitive pluripotent hemopoietic progenitor cells: Separation by adherence. J Cell Physiol 125:127 (1985). 3. Gregory C J , Eaves AC. H u m a n marrow cells capable of erythropoietic differentiation i n vitro: Definition of three erythroid colony responses. Blood 49:855 (1977). 4. Humphries RK, Eaves AC, Eaves CJ. Self-renewal of hemopoietic stem cells during mixed colony formation i n vitro. Proc Natl Acad Sci U S A 78:3629 (1981). 5. Till J E , McCulloch EA. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat Res 14:213 (1961). 6. B a u m a n J G J , Mulder AH, V a n Den Engh GJ. Effect of surface antigen labeling on spleen colony formation: Comparison of the indirect immunofluorescence and the biotin-avidin methods. Exp Hematol 13:760 (1985). 7. Y u n g YP, Cudkowicz G. Abrogation of resistance to foreign bone marrow grafts by carrageenans. II. Studies with the anti-macrophage agents i , k, and X. carrageenans. J Immunol 119:1310 (1977). 8. Metcalf D, Moore MAS. In: Neuberger A, Tatum EL. "Haemopoietic Cells", Frontiers of Biology, vol 24., Amsterdam: North-Holland (1971). 9. Till J E , McCulloch EA. The f-factor of the spleen-colony assay for hemopoietic stem cells. Ser Haematol 5:15 (1972). 10. Lord BI, Hendry J H . Observations on the settling and recoverability of transplanted hemopoietic colony-forrning units i n the mouse spleen. Blood 41:409 (1973). 11. Harrison DE, Astle CM, Delaittre JA. Loss of proliferative capacity i n immunohemopoietic stem cells caused by serial transplantation rather t h a n aging. J Exp M e d 147:1526 (1978). 12. Vennstrom B, K a h n P, A d k i n s B, Enrietto P, Hayman MJ, Graf T, Luciw P. Transformation of mammalian fibroblasts and macrophages i n vitro by a murine retrovirus encoding an avian v-myc oncogene. E M B O J 3:3223 (1984). 13. Wagner EF, Vanek M, Vennstrom B. Transfer of genes into embryonal carcinoma cells by retrovirus infection: efficient expression from an internal promoter. E M B O J 4:663 (1985). 14. Hughes PFD, Eaves C J , Hogge DE, Humphries RK. High-efficiency gene transfer to h u m a n hematopoietic cells maintained i n long-term marrow culture. Blood 74:1915 (1989). 15. M a n n R, Mulligan RC, Baltimore D. Construction of a retrovirus packaging mutant and its use to produce helper-free defective retrovirus. Cell 33:153 (1983).  - 77 16. Miller AD, Trauber DR, Buttimore C. Factors involved in the production of helper virusfree retrovirus vectors. Somatic Cell Mol Genet 12:175 (1986). 17. Cone RD, Mulligan RC. High-efficiency gene transfer into mammalian cells: Generation of helper-free recombinant retrovirus with broad mammalian host range. Proc Natl Acad Sci USA 81:6349 (1984). 18. Murthy SC, Eaves CJ, Krystal G. A simple three-step purification procedure for interleukin 3 involving absorption to fixed cells. Exp Hematol 17:997 (1989). 19. Gregory CJ, Eaves AC. In vitro studies of erythropoietic progenitor cell differentiation. In: Clarkson B, Marks PA, Till JE. Differentiation of Normal and Neoplastic Hematopoietic Cells. New York: Cold Spring Habour Laboratory (1978). 20. Dougherty GJ, Allen CA, Hogg NM. Applications of immunological techniques to the study of tumor-host relationship. In: Weir DM. Handbook of Experimental Immunology, Applications of Immunological Methods in Biomedical Sciences, vol 4. Oxford, UK: Blackwell (1986). 21. Chan P-Y, Takei F. Molecular cloning and characterization of a novel murine T cell surface antigen, YE1/48. J Immunol 142:1727 (1989). 22. Maniatis T, Fritsch P, Sambrook J. Molecular Cloning: A Laboratory Manual. New York: Cold Spring Harbor (1982). 23. Lamar EE, Palmer E. Y-encoded, species-specific DNA in mice: Evidence that the Y chromosome exists in two polymorphic forms in inbred strains. Cell 37:171 (1984).  - 78 -  C H A P T E R  III  ISOLATION IN A S I N G L E S T E P O F A H I G H L Y E N R I C H E D M U R I N E H E M O P O I E T I C S T E M C E L L P O P U L A T I O N W I T H C O M P E T I T I V E L O N G - T E R M R E P O P U L A T L N G ABILITY  1)  INTRODUCTION.  The functional properties of the most primitive hemopoietic cells, defined by their capacity for long-term blood cell production in vivo, are poorly understood. Transplantation experiments have clearly shown that such cells persist i n the marrow throughout adult life (1) and constitute a population with a very slow rate of turnover (2,3). Direct analysis of the clonal progeny of individual stem cells (4), as well as indirect analyses of the progeny of stem cells carrying unique chromosomal (5) or retroviral (6-8) markers, have revealed some evidence of heterogeneity i n the differentiation capacity as well as the proliferative potential of normal hemopoietic cells with in vivo repopulatlng ability. However, the extent to which such heterogeneity may be attributed to Intrinsic cellular differences as opposed to stochastic or biologic processes that regulate stem cell recruitment in vivo is not known. A possible approach to the resolution of these questions Is to obtain pure populations of hemopoietic stem cells whose functional attributes can then be evaluated with a variety of methodologies. Three key considerations to such an approach are the choice of assay used to define the stem cell population to be purified, the cellular characteristics used to select this population, and the method used for cell separation. In the present study, I introduced the use of a stringent, competitive long-term repopulation assay to ensure that the isolation procedure was selective for the most primitive type of hemopoietic stem cell. This assay involved transplanting irradiated female recipient mice with limiting numbers of separated (or  - 79 -  unseparated) populations of marrow cells from male donors together with a much larger number of "compromised" marrow cells from female mice. The compromised cells contained an almost normal frequency of CFU-S and in vitro clonogenic cells, but their competitive long-term repopulatlng ability had been markedly reduced by two previous cycles of successive marrow transplantation and regeneration (9). The presence of male cells i n the lymphoid and myeloid tissues of the reconstituted female recipients was then assessed 1 to 3 months later by Southern or spot blot analysis of extracted DNA  using a Y-chromosome-specific probe (10).  This approach is based on the hypothesis that a population of hemopoietic stem cells more primitive than day 12 CFU-S may  not begin to proliferate and differentiate for several weeks  after transplantation and would therefore not be detected in a transplantation assay unless sufficient short-term repopulatlng cells were also transplanted. The injection of >10^ compromised female marrow cells thus serves the dual purpose of ensuring survival of recipients Injected with limiting numbers of purified stem cells, and providing a competitive pressure to identify a class, of stem cells with a greater capacity for long-term repopulation. The characteristics used for stem cell purification were forward and orthogonal lightscattering properties (FLS and OLS), and Thy-1 and H-2K  antigen expression. Recent studies  have suggested that low levels of Thy-1 are expressed on primitive hemopoietic cells and that this level decreases with differentiation (11-13). Comparisons of the level of class I major histocompatibility complex (MHC  H-2K)  antigen on different hemopoietic cell types have also  suggested a progressive decrease of this antigen with differentiation (14,15). To develop a simple purification procedure that could be used with a single laser FACS, I devised a cell labelling technique In which Thy-1-positive cells are indirectly stained with one fluorochrome (fluorescien isothiocyanate, FITC) by a biotin-avidin complex, and H-2K-positive cells are indirectly stained with a second fluorochrome (R-phycoerythrin, R-PE) by a tetramolecular antibody complex (16). The tetramolecular complex consists of two different mouse IgGj monoclonal antibodies (MoAbs; in this case, one being directed against the H-2K antigen, the other against R-PE) linked together by two identical rat MoAbs specific for an  D  - 80 -  epitope bilaterally expressed on the Fc region of mouse IgGj. The advantages of tetramolecular antibody complexes for flow cytometry are that they do not require the use of purified antibodies or fluorochromes, and conjugates can be formed instantly simply by mixing of the antibodies desired (16.17). Sorting experiments were undertaken with suspensions of normal marrow as well as marrow from mice that had been treated four days previously with 5-fluorouracil (5-FU). The latter were included because 5-FU selectively decreases the number of mature hemopoietic cells present (18,19), and I therefore anticipated that marrow from 5-FU-treated mice might serve as a n already enriched starting population.  2)  RESULTS.  A)  Light-Scatter Properties, Thy-1, and H-2K Antigen Expression of In Vitro Clonogenic Cells.  Representative F L S profiles for normal and day 4 5-FU marrow (total nucleated cells) are shown i n Figures 5A and B, respectively. The corresponding distributions of CFU-GEMM, CFU-GM and BFU-E are shown i n Figures 5C and D. CFU-GEMM i n normal marrow were maximally enriched in the fraction isolated between channels 121 and 150 (eg. 2.7-fold i n the experiment shown i n Figure 5C). Corresponding enrichments of CFU-GM and BFU-E i n the same fraction were 1.6-fold and 3.0-fold. A s expected, day 4 5-FU marrow showed a depletion of larger cells (Figure 5B), an increased frequency of CFU-GEMM, and a decreased frequency of CFU-GM and BFU-E (Figure 5E). All three of these progenitor types showed a higher F L S than their counterparts i n normal marrow. A s a result, optimal enrichment of CFU-GEMM from day 4 5-FU marrow was obtained i n the fraction gated between channels 151 and 255 (4.1-fold  0  Normal B M  00  80  o  =  ~  i i  60-  1  40  5 ^ ^ 3 3  20  O O CD  1  i  3  Li_ L L LL.  i j  o  T ™!™— i -  0  40  80  120  i  i  160  200  1—  Forward Scatter (Channel No.)  Figure 5.  0  40  80  120  160  200  Forward Scatter (Channel No.)  Normal B M / D a y 4 5 - F U B M  Representative FLS profiles of normal and day 4 5-FU marrow cells. (A and B) Total nucleated cells. (C and D) Frequencies of in vitro clonogenic cells in sequential fractions (30 through 60, 61 through 90. 91 through 120, 121 through 150, 151 through 255). (E) Frequencies of in vitro clonogenic cells in the corresponding unseparated starting populations.  - 82 -  in the experiment shown). Enrichment of day 4 5-FU CFU-GM in this fraction was similar (3.9-fold). The OLS properties of CFU-GEMM, CFU-GM, and BFU-E in normal and day 4 5-FU marrow were similarly analyzed (Figure 6). Maximum enrichment of CFU-GEMM and CFU-GM (4.2-fold and 2.4-fold, respectively, in the experiment shown) from normal marrow was obtained in the region between channels 46 and 90. Maximum enrichment of BFU-E was observed in a window of cells with slightly lower OLS (4.1-fold in the region between channels 31 and 60). Day 4 5-FU marrow was also, as expected, depleted of cells with higher OLS properties (Figure 6B). In vitro clonogenic cells in day 4 5-FU marrow showed a higher OLS than those in normal marrow and were maximally enriched in the fraction gated between channels 61 and 120. In the experiment shown, CFU-GEMM were enriched 2.0-fold in this region. Staining of normal marrow cells with biotinylated anti-Thy-1.2 followed by avidin-FITC revealed a small population including both weakly and strongly fluorescent cells (~8% of the total marrow population, Figure 7A). The proportion of such cells in day 4 5-FU marrow was higher (-24%, Figure 7B), facilitating better resolution of the subpopulation expressing low levels of Thy-1.2. May-Grunwald-GIemsa staining of the sorted cells showed that most cells expressing low levels of Thy-1.2 antigen had a blast cell morphology (ie. had a high nucleus:cytoplasm ratio, and pseudopodia). Conversely, most in the peak of strongly Thy-1.2-positive cells appeared to be lymphocytes (ie. were small, and had a lower nucleus:cytoplasm ratio). In vitro clonogenic cells in both normal and day 4 5-FU marrow were maximally enriched in the weakly Thy-1.2-positive fraction distributed between channels 121 and 150 (Figures 7C and D). Sorting of cells in this region gave an enrichment of CFU-GEMM from normal and day 4 5-FU marrow of 6.9-fold and 2.3-fold, respectively, in the experiment shown. CFU-GM and BFU-E were also maximally enriched in this fraction, although thelr distributlons extended further into the negative region. This is consistent with the reported decreased expression of Thy-1 on the more mature classes of murine hemopoietic progenitors.  80 60 40 20 -  T  1  1  r  0 40 80 120 160 200 Orthogonal Scatter (Channel No.)  C) o =  Q O O  4)  o  li. U.  60 40 -  40  O  20  "  I I  O O  i"  20 -  t  —K 0  40  i i —H [—•  80  ,  1  120 160 200  i '  Orthogonal Scatter (Channel No.)  Figure 6.  Day 4 5-FU BM  80 -  1 L  60 "  0)  °„ u  1 ; •  i  in  O) U) - i "  D)  Normal BM  ! i _ -  0 40 80 120 160 200 Orthogonal Scatter (Channel No.)  o-  0  1  40  80  1-'  120 160 200  Orthogonal Scatter (Channel No )  Representative O L S profiles of n o r m a l a n d day 4 5-FU m a r r o w cells. (A a n d B) T o t a l n u c l e a t e d cells. (C a n d D) F r e q u e n c i e s of In vitro c l o n o g e n i c cells i n s e q u e n t i a l f r a c t i o n s (5 t h r o u g h 30. 3 1 t h r o u g h 45. 4 6 t h r o u g h 60. 61 t h r o u g h 90. 91 t h r o u g h 120. 121 t h r o u g h 150, 151 t h r o u g h 255).  s  T 160  200  40  1 80  1 120  r ~ 160  r 200  T h y 1.2 Fluorescence Intensity  T h y 1.2 Fkjorescence Intensity  (channel number)  (channel number)  oo  3 3 D u_ UL LL O O CD  ~i  160  200  Thy 1.2 Fluorescence Intensity (channel number)  Figure 7.  0  40  80  1 120  r 160  200  Thy 1.2 Fluorescence Intensity (channel number)  Representative Thy-1.2 fluorescence profiles of normal and day 4 5-FU marrow cells. (A and B) Total nucleated cells; control fluorescence profiles of normal ( ) and day 4 5-FU marrow cells (—) stained with H F N + avidin-FITC. Profiles of cells stained with biotinylated antl-Thy-1.2 m A b and avidin-FITC (—). (C a n d D) Frequencies of i n vitro clonogenic cells i n sequential fractions (20 through 90, 91 through 120, 121 through 150, 151 through 255).  - 85 -  H-2KP expression was measured by incubating cells with preformed bispecific tetramolecular antibody complexes containing an anti-H-2K antibody coupled to an anti-R-PE D  antibody followed by stalriing with R-PE as described in the Materials and Methods. The resultant red fluorescence profile was compared with that obtained by staining cells with control tetramers monospecific for R-PE.  These control profiles indicated a low but significant  degree of nonspecific staining, probably due primarily to Fc-receptor-mediated binding. Nevertheless, i n comparison, antl-H-2K^ x anti-R-PE tetramers gave a m u c h stronger staining of all cells, as shown by the marked shift of the entire profile (Figure 8A). The staining characteristics of normal and day 4 5-FU marrow cells with respect to H-2KP antigen expression were indistinguishable. Cells were sorted from increasingly selective fractions representing increasingly higher levels of H - 2 K ° antigen expression. Results from a representative experiment are shown in Figures 8B and C. Although only 1 5 % and 3 % of all CFU-GEMM i n normal and day 4  5-FU  marrow, respectively, were found in the fraction contajjriing the 2 % of cells expressing the highest levels of H-2K , maximum enrichment of CFU-GEMM (7.7-fold and 1.7-fold i n the b  experiment shown i n Figure 8) was obtained i n this fraction. In contrast, CFU-GM and BFU-E numbers were not significantly enriched in this fraction, although some selection i n favor of CFU-GM of higher proliferative potential (i.e. yielding colonies contaiiiing >5,000 cells) was noted.  B)  Characterization of Cells with Competitive Long-Term Repopulating Ability.  E a c h of the four types of analyses performed using assays for clonogenic cells to identify primitive hemopoietic cells was repeated using the competitive repopulation assay as follows: 10  4  male cells from each sorted fraction were injected with 2 x 10^ compromised female  marrow cells (obtained as described in the Materials and Methods) into each of two irradiated female recipients. Five weeks later, the repopulated marrows, spleens and thymuses of the  oo-  A) Normal BM or Day 4 5-FU BM  A l  0  1  40  '  80  120  I  r  160  200  H - 2 K Fluorescence Intensity b  B)  in  O  (channel number) Normal B M  _  100  j  80  ~  60-  CO 0)  oj  40UJ  5  20  O O uJ 3 3 3  i  U- LL U_  O O CD  20 10 5 2 Upper % H - 2 K Fluorescence Intensity b  Figure 8.  20  10  5  2  Upper % H - 2 K Fluorescence Intensity b  Representative H - 2 K fluorescence profiles of normal and day 4 5-FU marrow cells. (A) Total nucleated cells in day 4 5-FU marrow; unstained cells (—) (ie. incubated with H F N + R-PE or H F N only); non-specifically stained cells ( ) (ie. after incubation with monospecific anti-R-PE x anti-R-PE control tetramers followed by R-PE); specifically stained cells (—) (ie. after incubation with bispecific a n t i - H - 2 K x anti-R-PE tetramers followed by R-PE). Identical profiles were obtained with normal marrow cells and are not shown separately. (B and C) Frequencies of in vitro clonogenic cells in the 2 % to 2 0 % of cells showing the highest level of H - 2 K specific fluorescence. D  B  T  - 87 -  recipients were individually analyzed for the presence of male cells. The results obtained with the various fractions of the F L S profile of day 4 5-FU marrow showed that cells with competitive long-term repopulating ability were present i n all fractions above channel 90, whereas fractions of smaller cells (less than channel 90) were relatively depleted of this activity. The F L S fractionation results for normal marrow were inconclusive. Similar analysis of the OLS profile of day 4 5-FU marrow showed that competitive long-term myeloid and lymphoid repopulating cells were maximally enriched i n the fraction gated between channels 61 and 150. For normal marrow, these cells, like in vitro clonogenic cells, appeared to have lower OLS properties and were concentrated i n the window between channels 31 and 90. Competitive repopulating hemopoietic cells in both normal and day 4 5-FU marrow were concentrated i n the fraction corresponding to low Thy-1.2 expression (channels 121 through 150, although some of these cells do appear to be Thy-1.2 negative, channels 91 through 120), and i n the fraction corresponding to the 2 % to 5 % of cells expressing the highest levels of H-2K . A n b  example of the data obtained i n these experiments is shown for the O L S analysis (Figure 9).  C)  Enrichment of Primitive Hemopoietic Cells After Four Parameter Sorting.  Normal and day 4 5-FU marrow cells were then sorted by gating all four parameters simultaneously as follows: intermediate to high F L S (channels 91 through 255), intermediate OLS (channels 31 through 90 for normal marrow and channels 61 through 150 for day 4 5-FU marrow), low Thy-1.2 (channels 121 through 150), and high H-2K (top 2 to 5%). b  Approximately 0.1% of the starting marrow cells (both normal and day 4 5-FU) were found i n the window defined by these gates, and from 1.5 x 1 0 cells, 4-6 x 10^ cells could be routinely 7  isolated i n a 1.5 to 2 hour sort. A typical two dimensional (Thy-1.2 versus H-2K ) contour plot b  of the distribution of day 4 5-FU marrow cells relative to the sort window is shown i n Figure 10. A s shown i n Table IV, all clonogenic cells were enriched in the sorted population. For day 4 5-FU marrow, enrichment and recovery values for CFU-GEMM and CFU-S (both day 9 and  - 88 -  b  S  t  5-30 31-45 CD  46-60  O CO CD  61-90 91-120  o  121-150 151-255 5-255  100  30 10  3  1  % Male  Figure 9.  Spot blot analysis of DNA from tissues of two female mice 5 weeks after transplantation of IO male day 4 5-FU marrow cells sorted from the same regions of the OLS profile shown in Figure 6D, together with 2 x 10^ 4  compromised female marrow cells (data from one sorting experiment), (b. bone marrow; s, spleen; t, thymus). Bracket (right) indicates channels gated in  subsequent four parameter sorts of day 4 5-FU marrow cells. Control mixtures of male and female DNA hybridized to the Y-specific probe, pY2, are shown for reference.  - 89 -  Figure 10.  Contour plot of Thy-1.2 (green fluorescence) vs. H-2K" (red fluorescence) of double stained day 4 5-FU marrow cells after gating for FLS and OLS as described In the text. Green and red fluorescence gates used for four parameter sorts (—). Scale of fluorescence Intensity is in arbitrary units.  Table IV. In Vitro and In Vivo Assayable Clonogenic Cell Content of Marrow Cells Isolated by Four Parameter Sorting*  Source of Cells  BFU-E  CFU-GM  CFU-GEMM  Day 9 CFU-S  b  Day 12 CFU-S  Normal BM Frequency (% purity) Enrichment {-fold) Recovery (%)  0.4 + 0.1 (5) 18 + 4 5+1  6.6 + 0.9(5) 19 + 3  1.3 + 0.1(3)  14 + 6(5)  9+1(3)  100 + 28  53 + 23  46 + 9  5+ 1  27 + 8  14 + 6  12 + 2  3 + 1 (4)  4 + 1 (3)  10 + 3 (5)  23 + 3 (5)  89 + 44  93 + 38  75 + 27  24 + 8  25+ 10  20 + 7  Dav 4 5-FU BM Frequency (% purity)  0.01 + 0.003 (3)  Enrichment (-fold)  2.0 + 0.6  32 + 14  Recovery (%)  0.4 + 0.2  9+4  a b  Gate selection described in the text. Shown are the mean + SEM of values from (n) different experiments. Assuming a seeding efficiency to the spleen, f, of 10% for unstained cells and 5% for antibody-labelled cells (see Chapter II).  b  - 91 day 12) were similar. For normal marrow, both types of CFU-S showed somewhat lower enrichment and recovery values, suggesting that the gates used i n this case were not optimal for selection of these cells. The most likely explanation for the lower enrichment of CFU-S is that the FLS window used was defined exclusively on the basis of preliminary CFU-GEMM enrichment data and excluded cells i n the low FLS region i n which other investigators have shown some of the CFU-S to be present (20,21). Because the gates used gave better enrichment of more primitive cells (i.e. day 12 CFU-S) from day 4 5-FU marrow relative to normal marrow, I focused on the former population to define more quantitatively the extent to which competitive long-term repopulatlng cells had been purified. In an Initial series of six experiments, 500 sorted male day 4 5-FU marrow cells were injected together with 10^ compromised female marrow cells into irradiated female recipients. Subsequent analysis showed that male cells constituted a readily detectable (typically >80%) proportion of the cells i n the repopulated marrow, spleen and thymus i n all of >20 s u c h mice killed 30 to 90 days after transplantation. Results for six mice from two experiments are shown i n Figure 11. Overall, the proportion of cells i n the repopulated myeloid and lymphoid tissues that originated from the 500 sorted male cells did show some variation both w i t h i n and between mice i n different experiments. To obtain a more precise measure of the frequency of repopulatlng cells i n the sorted day 4 5-FU marrow cell population, groups of irradiated mice were injected with 2 x 10^ compromised female marrow cells and decreasing numbers of sorted male cells, and then killed 35 days later for analysis. The results, shown In Figure 12, indicate that 100 sorted male day 4 5-FU marrow cells were sufficient to generate >20%  of the cells present i n the  repopulated marrow of 5 0 % of the recipients, whereas 1,450 unstained, unseparated day 4 5-FU cells were required to achieve the same result. From these data, the frequency of competitive long-term repopulating stem cells i n the sorted day 4 5-FU marrow population was calculated to be 1 i n 170 cells ( 9 5 % confidence limits: 1 i n 90 to 1 i n 300) (22). By comparison, the frequency of competitively repopulating cells i n the unstained, unseparated cell suspension  - 92 -  Figure 11.  Demonstration by Southern analysis of male DNA i n bone marrow (b), spleen (s), and thymus (t) of female mice transplanted with 500 double-stained male day 4 5-FU marrow cells isolated by four-parameter sorting (as described i n the text) together with 10^ compromised female marrow cells. Recipients were sacrificed 30 (A,B,D,E) or 90 (C,F) days after transplantation. Pure male and female marrow cell DNA samples hybridized to the Y-specific probe, pY2, are shown at right.  - 93 -  CO  10°  10  10  1  2  10  3  10  4  Number of Male Cells Injected/Mouse (with 2 x 10 Compromised Female Cells) 5  Figure 12.  Percent of recipients (8 animals per group) showing >20% repopulation of the marrow with male cells 35 days after transplantation of varying numbers of male day 4 5-FU marrow cells before (O ) or after ( •) staining and four-parameter sorting, together with 2 x 10^ compromised female marrow cells. The proportion of male DNA was determined by Southern analysis of 10 |ig Pvu II digested marrow cell DNA from each recipient using pY2. "Corrected" frequency for the same sorted cells assuming a 5 0 % loss i n vivo due to antibody staining (—) (Materials and Methods, Table II).  - 94 -  was 1 i n 2,300 cells ( 9 5 % confidence limits: 1 In 1,100 to 1 in 4,600). Assuming a 5 0 % loss of stem cells i n vivo due to the staining procedure alone (Table II, Chapter II), the enrichment obtained is ~28-fold. The recovery of competitive repopulatlng cells based on these data was calculated to be 7.4%.  3)  DISCUSSION.  In this study, I first analyzed i n vitro clonogenic cells present i n normal marrow and marrow from 5-FU-treated mice with respect to four unique parameters that c a n be discriminated using a single-laser FACS. Cells capable of generating macroscopically visible multilineage colonies (called CFU-GEMM) were maximally enriched (100-fold) i n a fraction of normal marrow cells that showed intermediate to high F L S and low to intermediate O L S ("the blast window"), and that bound low levels of anti-Thy-1 antibodies and very high levels of anti-H-2K antibodies. Previous studies of these progenitors of large mixed colonies present in normal marrow and marrow from 5-FU-treated mice have shown that some of these cells have a significant capacity for self-renewal and for generating day 9 CFU-S (19,23,24), suggesting overlap with cells detected as day 9 or day 12 CFU-S. This finding is further supported by the present finding that both types of CFU-S were purified i n fractions maximally enriched in CFU-GEMM. However, i n the case of normal marrow, both recovery and enrichment of day 9 and day 12 CFU-S i n the sorted population were somewhat lower than for CFU-GEMM, suggesting that the cells defined by these operational criteria may include mutually exclusive subpopulations. Other investigators have suggested that day 12 CFU-S in normal marrow are heterogeneous with respect to their uptake of rhodamine-123 and that this allows distinction of a subset of cells able to promote the 30-day survival of lethally irradiated mice (25). Some CFU-S also exhibit a lower F L S (20,21) than which was selected here. These reports and the present finding that individual CFU-S differ widely with respect to the parameters evaluated provide additional support for the concept of heterogeneity i n the CFU-S population.  - 95 -  The concentration of primitive cells in the marrow of mice pretreated with 5-FU was twoto three-fold higher than that i n normal marrow. Because of this and the increased power of CFU-GEMM light-scattering properties to predict the distribution of more primitive hemopoietic cells in 5-FU-treated marrow, the day 12 CFU-S content of sorted day 4 5-FU marrow was significantly higher than that of sorted normal marrow (Table IV). A s reported by other researchers (26), labelling of cells with antibody resulted in an apparent two-fold reduction in stem cell numbers detected by in vivo assays (Materials and Methods, Table II). Assuming that antibody staining resulted in a seeding efficiency of 5 % for CFU-S, one in four cells i n the sorted day 4 5-FU marrow population is mtrinsically capable of forrning a macroscopic spleen colony visible 12 days after injection. Long-term repopulation of mice could be reproducibly achieved with as few as 500 stained, sorted male day 4 5-FU marrow cells despite the requirement to compete with a compromised but protective graft of female marrow cells. Furthermore, engraftment by male cells could be demonstrated in a proportion of mice receiving as few as 50 sorted cells (Figure 12). Limiting-dilution analysis indicated that the frequency of long-term repopulating cells detectable in the stringent competitive assay was approximately one in 170 of the stained, sorted day 4 5-FU marrow cell population. However, this calculation assumes no cell loss during the transplantation process and a 1 0 0 % efficiency of activation of biologically comparable stem cells i n the presence of the cotransplanted compromised marrow cells. Assessment of either of these assumptions is difficult, but both may lead to a considerable underestimation of the actual content of stem cells with competitive long-term repopulating potential. With respect to the former, I already showed a 5 0 % loss of CFU-S due to staining alone (Materials and Methods, Table II), mdicating a real purity of at least one in 85 cells. Little is known about the mechanisms that determine which and how many stem cells are recruited in a transplant recipient, but this may be influenced by the total number of cells injected, resulting i n a relatively greater competitive pressure in recipients injected with decreasing numbers of test cells and a fixed number of compromised cells.  - 96 -  Nevertheless, the large discrepancy between the recovery and purity of the day 12 CFU-S and cells detected i n the competitive repopulation assay suggests that these are not identical populations even i n day 4 5-FU marrow. Since the compromised marrow used as a source of competing cells contained near normal numbers of day 12 CFU-S (Table III, Chapter II), i n some situations the capacity for forming day 12 spleen colonies clearly can be biologically separated from the properties that confer a high level of competitive repopulating potential. The results i n Chapter V show, using appropriate genetic markers, that the compromised cells can contribute detectably, although at low levels, to the regenerated hemopoietic populations. Additional experiments to compare the frequencies of purified repopulating cells assayed under different conditions and at later times should help to clarify further the extent of heterogeneity that may  exist among cells with stem cell potential and to devise additional strategies to permit  their differential isolation and detection. These results show that a population of very primitive hemopoietic cells can be obtained by a simple, single-step isolation procedure, although this population is probably not yet as homogeneous as that which can be obtained by more complicated procedures that may require a dual-laser F A C S (21,25). Although we, like Spangrude et al (21), demonstrated the presence of both lymphoid and myeloid repopulating cells i n the cell suspensions isolated, direct evidence that these are dual functions of a single stem cell type has not yet been provided. Experiments to address this question using retrovirus-mediated gene transfer to mark day 4 5-FU marrow cells sorted as described here are outlined i n Chapter IV. The results indicate that the competitive repopulation assay detects such a cell. Extension of this approach should therefore be an important avenue to further analysis of hemopoietic cells with stem cell properties.  - 97 REFERENCES 1.  Harrison DE. Long-term erythropoietic repopulating ability of old, young, and fetal stem cells. J Exp Med 157:1496 (1983).  2.  Becker AJ, McCulloch EA, Sirninovitch L, Till JE. The effect of differing demands for blood cell production on DNA synthesis by hemopoietic colony-forming cells of mice. Blood 26:296 (1965).  3.  Pietrzyk ME, Priestley GV, Wolf NS. Normal cycling patterns of hematopoietic stem cell subpopulations: An assay using long-term in vivo BrdU infusion. Blood 66:1460 (1985).  4.  Magli MC, Iscove NN, Odartchenko N. Transient nature of early haematopoietic spleen colonies. Nature 295:527 (1982).  5.  Abramson S, Miller RG, Phillips RA. The identification in adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J Exp Med 145:1567(1977).  6.  Dick JE, Magli MC, Huszar D, Phillips RA, Bernstein A. Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system ofW/W mice. Cell 42:71 (1985). v  7.  Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell 45:917 (1986).  8.  Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of mice reconstituted with retrovrjrus-infected stem cells. EMBO J 6:3955 (1987).  9.  Harrison DE, Astle CM, Delaittre JA. Loss of proliferative capacity in immunohemopoietic stem cells caused by serial transplantation rather than aging. J Exp Med 147:1526 (1978).  10.  Lamar EE, Palmer E. Y-encoded, species-specific DNA in mice: Evidence that the Y chromosome exists in two polymorphic forms in inbred strains. Cell 37:171 (1984).  11.  Boswell HS, Wade Jr PM, Quesenberry PJ. Thy-1 antigen expression by murine high-proliferative capacity hematopoietic progenitor cells. I. Relation between sensitivity to depletion by Thy-1 antibody and stem cell generation potential. J Immunol 133:2940 (1984).  12.  Williams DE, Boswell HS. Floyd AD, Broxmeyer HE. Pluripotential hematopoietic stem cells in post-5-fluorouracil murine bone marrow express the Thy-1 antigen. J Immunol 135:1004(1985).  13.  Berman JW, Basch RS. Thy-1 antigen expression by murine hematopoietic precursor cells. Exp Hematol 13:1152 (1985).  14.  Mulder AH, Bauman JGJ, Visser JWM, Boersma WJA, Van Den Engh GJ. Separation of spleen colony-forrning units and prothymocytes by use of a monoclonal antibody detecting an an H-2K determinant. Cell Immunol 88:401 (1984).  15.  Okamoto T, Kanamaru A, Hara H, Nagai K. Changes in expression of major histocompatibility complex (MHC) class-I antigen on hematopoietic progenitors during murine development. Exp Hematol 15:190 (1987).  - 98 16.  Lansdorp PM, Aalberse RC, Bos R, Schutter WG, Van Bruggen EFJ. Cyclic tetramolecular complexes of monoclonal antibodies: a new type of cross-linking reagent. Eur J Immunol 16:679 (1986).  17.  Wognum AW, Thomas TE, Lansdorp PM. Use of tetrameric antibody complexes to stain cells for flow cytometry. Cytometry 8:366 (1987).  18.  Eaves AC, Bruce WR Endotoxin-induced sensitivity of hematopoietic stem cells to chemotherapeutic agents. Ser Haemat V(Suppl 2):64 (1972).  19.  Kerk DK, Henry EA, Eaves AC, Eaves CJ. Two classes of primitive pluripotent hemopoietic progenitor cells: Separation by adherence. J Cell Physiol 125:127 (1985).  20.  Van Den Engh G, Visser J. Light scattering properties of pluripotent and committed haemopoietic stem cells. Acta Haematol 62:289 (1979).  21.  Spangrude GJ, Heimfeld S, Weissman IL. Purification and characterization of mouse hematopoietic stem cells. Science 241:58 (1988).  22.  Taswell C. Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis. J Immunol 126:1614 (1981).  23.  Humphries RK. Eaves AC, Eaves CJ. Self-renewal of hemopoietic stem cells during mixed colony formation in vitro. Proc Natl Acad Sci USA 78:3629 (1981).  24.  Eaves CJ, Humphries RK, Eaves AC. Self-renewal of hemopoietic stem cells: Evidence for stochastic regulatory processes. In: Stamatoyannopoulos G, Nienhuis AW. Hemoglobins in Development and Differentiation. New York: Alan R Liss, Inc (1981).  25.  Visser JWM, de Vries P. Isolation of spleen-colony forming cells (CFU-s) using wheat germ agglutinin and rhodamine 123 labeling. Blood Cells 14:369 (1988).  26.  Bauman JGJ, Mulder AH, Van Den Engh GJ. Effect of surface antigen labeling on spleen colony formation: Comparison of the indirect immunofluorescence and the biotln-avidin methods. Exp Hematol 13:760 (1985).  - 99 -  C H A P T E R  RETROVTRUS-MEDIATED  IV  G E N E T R A N S F E R T O PURIFIED H E M O P O I E T I C S T E M C E L L S WITH  L O N G - T E R M LYMPHO-MYELOPOIETIC REPOPULATING  1)  ABILITY  INTRODUCTION.  The biological properties of the most primitive hemopoietic cells capable of long-term blood cell production in vivo are not well defined. Analysis of the clonal progeny of mouse marrow cells carrying unique chromosomal (1) or retroviral (2-4) markers has provided strong evidence of the presence i n normal adult bone marrow of stem cells that are individually capable of regenerating and mamtajjning both lymphoid and myeloid systems for many weeks after transplantation. The existence of h u m a n hemopoietic stem cells with these potentialities has also been indicated recently by the demonstration of clonal populations of mature blood cells of multiple lineages i n normal bone marrow transplant recipients (5). Progeny analyses have further suggested the existence of hemopoietic stem cells that can function in vivo for extended periods of time but that express more restricted differentiation potentialities or that are less competitive in generating mature progeny of a particular lineage (3,4,6), raising the possibility of considerable heterogeneity at this level. Further characterization of the most primitive blood cell elements would be greatly facilitated by the development of procedures for their purification and subsequent use i n lmeage-mapping studies. Recent reports have suggested the feasibility of obtajjning highly purified populations of mouse marrow stem cells with in vivo repopulating ability, although the extent to which these may be detected i n assays for more mature hemopoietic cells such as CFU-S (colony-forming unit-spleen: cells capable of generating macroscopic spleen colonies visible 9-14 days after transplantation) (7) appears to  - 100 -  differ markedly according to the procedure used (8.9). Moreover, direct evidence that current purification procedures selectively enrich for stem cells with lymphopoietic as well as myelopoietic potential has not yet been reported. The experiments i n Chapter III describe the development of a procedure for the single-step isolation of a population of primitive hemopoietic stem cells from 5-fluorouracil (5-FU)-treated mouse bone marrow; 1 i n 4 of these cells are day 12 CFU-S, and at least 1 i n 85 are capable of competitive long-term marrow repopulation when cotransplanted into lethalry irradiated recipients with 2 x 1 0 ^ twice serially transplanted marrow cells. In the present study I have used this purification procedure in combination with retrovirus-mediated gene transfer to demonstrate the long-term lympho-myelopoietic repopulating ability of individual purified stem cells.  2)  RESULTS.  A)  Transfer of the Neo Gene to Purified Stem Cells. r  In three experiments, a total of 37 mice were transplanted with 150 to 2000 purified, male marrow cells that had been exposed to the recombinant retrovirus TKneo 19. To ensure activation of the most primitive stem cells in the purified population and to facilitate short-term survival of these animals, each recipient was also injected with 2 x 1 0 ^ "compromised" female marrow cells prepared as described i n Chapter II. In all of the 28 recipients that were evaluated for this study, male cells were determined by Southern analysis to comprise more than 2 0 % of the cells in at least one of the various hemopoietic tissues analyzed. However, as noted previously (Chapter III), the proportion of male cells often varied considerably between different tissues of the same animal. Results for peripheral blood leukocytes from 19 mice (Experiment 2) sampled 35 days after transplantation are shown i n Figure 13. Analysis of the same DNA with a n e o gene-specific probe revealed that 16 of these 19 animals had circulating r  - 101 -  V V  neo -probe r  D  V  V  1?  H | M  V  'V- V V V  «*  V  --//EH  Figure 13.  •—j  *V>n>n> n> V  • *#  fli  B  TKneol9  V  1?  ¥  U  w £  -2.3kb  B  Neo  ITJ—M~ EH  Southern analysis of DNA from peripheral blood leukocytes isolated from 19 female mice 35 days after transplantation with 300 (mice 2.1 - 2.15) or 500 (mice 2.16 - 2.19) TKneo 19-tafected, purified, male day 4 5-FU marrow cells together with 2 x 10^ compromised female marrow ceils. (Top) Hybridization with the Y-specific probe, pY2. Uninfected normal male and female DNA samples are shown as controls. E a c h lane was loaded with approximately 5 pg of B a m HI-digested DNA. (Middle) Identical blot reprobed with the neo -specific probe. (Bottom) Structure of the TKneo 19 provirus with B a m H I (B), Hind III (H), and Eco R l (E) restriction sites; the dashed line represents flanking genomic sequences. r  - 102 -  neo -positive leukocytes. Assessment of marrow, spleen and thymus cell DNA r  from 8 other  mice (Experiment 1) sacrificed on day 35 after transplantation showed 3 animals to be both male and neo -positive (Table V). Another mouse (Experiment 3), sacrificed on day 140 after r  transplantation, was found to contain male and neo -positive cells i n the spleen but not i n the r  marrow or thymus (Table V). Gene transfer to purified repopulating stem cells was thus detected i n 20 of the 28 (71%) recipients analyzed. The intensity of the neo -specific signal r  varied considerably among these reconstituted animals, consistent with a variable proportion of marked male stem cells contributing to hemopoiesis at any given time. To obtain definitive evidence that retrovirally marked cells were derived from the transplanted purified male stem cells, marrow cells from three primary mice (1.7, 2.4 and  2.10)  i n which neo -positive cells were detected were used to generate macroscopic spleen colonies r  (day 12) i n secondary female recipients. These were then mdividually excised for DNA  analysis.  Results for seven spleen colonies generated from the marrows of mice 2.4 and 2.10 sacrificed 49 days after the initial transplant of purified stem cells are shown i n Figure 14. Overall, 10 (5 from mouse 2.4, and 5 from mouse 1.7) of 16 day 12 spleen colonies analyzed with pY2 were male and therefore must have originated from a stem cell with competitive long-term repopulating ability i n the original purified population. The detection of some female spleen colonies is consistent with the male-female chimerism seen i n many recipients of the type of mixed transplants used here. Five of 10 male spleen colonies also contained unrearranged provirus with an integration pattern (data not shown) identical to that of the donor marrow. No new integration sites were detected and the n e o gene was not detected i n any of the other r  spleen colonies analyzed (male or female).  B)  Analysis of Individual Neo -Marked Clones. r  To determine the number and distribution of unique proviral integration sites i n different lymphoid and myeloid cell lineages, DNA was digested separately with H i n d l l l and E c o R I  Table V. Summary of 12 Mice Transplanted with Retrovirally Marked Purified Male Repopulating Stem Cells.  No. Purified Cells  Time of Assessment  Transplanted  (days post-Tx)  PB  Mouse  0  Proportion of Male Cells  Retroviral Fragment Size (Kbp)  bm  spl  thy  bm  spl  thy  5.2  1.3  1000  35  ND  +  ++  +++  -  5.2  1.4  250  35  ND  ++  +  -  5.8 8.9  5.8 8.9  2.4  300  35 49  +++ ND  +++  +++  +++  8.1 10.9  8.1 10.9  8.1 10.9  2.10  300  1.7  2000  d d  35 49  +++ ND  +  +++  ++  -  7.1 8.8  7.1 8.8  61 98 121  +++ +++ ND  +++  +++  +++  7.2  7.2  7.2  +++  -  -  7.1  -  3.10  150  140  ND  2.5  300  35 144  ++ ND  +++  ++  +++  -  5.8  35 144  +++ ND  +++  +++  +++  -  12.8  ND  35 144  ++ ND  ++  +++  +  ND  21.0  ND  35 196  ++ ND  +  +++  ND  -  2.6  300  2.13  300  2.8  300  2.15  2.18  300  x  500  6.1  6.1  6.8  6.8 5.6 7.2  35 196  ++ ND  ++  +++  +  5.6 7.2  5.6 7.2  35 196  ++ ND  +  +++  +++  -  10.2 7.1  a All mice were co-transplanted with purified male day 4 5-FU marrow cells and 2 x 10 compromised female marrow cells. b Estimated proportion of male cells: - denotes 100% female; + denotes <10% male; ++ denotes 10-80% male; +++ denotes >80% male. c Determined after digestion with Hindlll or EcoRI (mouse 2.4). All integrations also verified by analysis of DNA digested with EcoRI or Hind III (mouse 2.4). d Weakly detected in thymus on original autoradiogram. Abbreviations: Tx, transplantation; PB, peripheral blood; bm, bone marrow; spl, spleen; thy, thymus; ND, not determined. 5  c  - 104 -  mouse 2.10  mouse 2.4  n r  1 2  3  4  Y-probe  neo -probe r  Figure 14.  S o u t l i e m analysis of DNA from individual day 12 spleen colonies generated from the marrows of mice 2.4 and 2.10 sacrificed 49 days after transplantation. Uninfected normal male and female DNA samples are shown as controls. (Upper) Pvu II-digested DNA (lOug per lane) hybridized to the Y-specific probe. (Lower) B a m HI-digested DNA (30 ug per lane) hybridized to the neo -specific probe. r  - 105 (enzymes that cut only once within the retroviral sequence, Figure 13). Of the 20 neo -positlve r  mice identified from all three experiments, 12 were selected for this type of analysis on the basis of an initial demonstration of a relatively high proportion of male and retrovlrally marked cells. Results for 6 mice are shown in Figures 15 and 16 and are summarized together with the results for the other 6 mice In Table V. For convenience they are presented according to the interval between the time of transplantation and the time of sacrifice. All mice showed the presence of some retrovlrally marked cells at the time of sacrifice in at least one tissue; when only one tissue was involved (3 mice only), this was always the spleen. Although the variable content of male cells in many mice indicated that hemopoietic reconstltution was typically oligoclonal, only one or two integration sites were detected in neo -positlve tissues. This type of r  preliminary analysis cannot distinguish between single clones marked by two retroviral integration events and two prominent clones each marked by a single unique integration event; however, it is obvious that the conditions used enabled the assessment of the competitive repopulating potential of individual retrovirus-infected stem cells. The first 2 mice shown In Figure 15 (Experiment 1, mice 1.3 and 1.4) were sacrificed 35 days after transplantation. In both, the proportion of male cells in the marrow, spleen and thymus was different in each tissue, but in each case was correlated with the intensity of neo -specific hybridization. In mouse 1.3, a single 5.2 Kbp provlral fragment was found in r  both the spleen and thymus but was not detectable in the marrow. Because of the smear present In the bone marrow lane, however, the presence of neo -specific hybridization to r  fragments >5 Kbp cannot be ruled out.  DNA  In mouse 1.4, a 5.8 Kbp and an 8.9 Kbp provlral  fragment were present at the same level in the marrow and spleen but were not detectable in the thymus. Mouse 2.4 (Experiment 2) was sacrificed 49 days after transplantation. At that time the marrow, spleen and thymus were all predominantly male. Common 8.1 Kbp and 10.9 Kbp provlral fragments were clearly evident in marrow and spleen DNA analyzed by EcoRI digestion. These were also detectable In the thymus but at very low levels (not reproduced from the original autoradiogram) in Figure 15. The presence of a common pair of bands in marrow  - 106 -  35dPostTx  49dPostTx  144dPostTx  196dPostTx  mouse 1.3 mouse 1.4  mouse 2.4  mouse 2.5  mouse 2.15  F i g u r e 15.  Southern analysis of DNA from the bone marrow (b), spleen (s), and thymus (t) of five female mice sacrificed 35-196 days (35d-196d) after transplantation (post Tx) with 1,000 (mouse 1.3), 250 (mouse 1.4), or 300 (mice 2.4, 2.5, and 2.15) TKneo 19-infected, purified, male day 4 5-FU marrow cells together with 2 x 10^ compromised female marrow cells. (Upper) Hind III (mice 1.3, 1.4, 2.5, and 2.15)- or Eco Rl (mouse 2.4)-digested DNA (30 pg per lane) hybridized to the neo -specific probe. (Lower) Pvu II-digested DNA (10 ug per lane) hybridized to the Y-specific probe. Provlral fragment sizes are described in Table V. r  - 107 -  121dpostTx m o u s e 1.7  F i g u r e 16.  Presence of a common retrovirally marked clone in bone marrow (bm), spleen (spl), thymus (thy), lymph node (LN) and separately isolated marrow macrophage (bm mac), splenic B (spl B), and T (spl T) lymphocytes from mouse 1.7 sacrificed 121 days (12 Id) after transplantation (post Tx). (Upper) Hind Ill-digested DNA (lanes bm, spl, and thy: 25 ug per lane; lanes LN, bm mac, spl B, and spl T: 5 ug per lane) hybridized to the neo -specific probe. (Lower) Identical blot reprobed with the Y-specific probe. Uninfected normal male DNA (25 ug) is shown as a control. r  - 108 and spleen and seen weakly In thymus was verified by analysis with H i n d III (data not shown). Mouse 1.7 (Experiment 1) was sacrificed 121 days after transplantation, and the marrow, spleen and thymus were also all found to be predominantly male. A single 7.2 Kbp proviral fragment was seen i n each of these tissues as well as i n lymph node cells and i n separately isolated bone marrow macrophages, splenic B-lymphocytes and splenic T-lymphocytes which were also predominantly male (Figure 16). Mouse 2.5 (Experiment 2) was sacrificed on day 144 after transplantation. The marrow, spleen and thymus all contained a high, albeit variable, proportion of male cells, but a single 5.8 Kbp proviral fragment was detected i n only the spleen of this animal. Mouse 2.15 (Experiment 2) was sacrificed on day 196 after transplantation. This animal contained decreasing proportions of male cells i n the spleen, marrow and thymus; however, a 5.6 Kbp and a 7.2 Kbp proviral fragment were evident i n all three tissues. Overall, at least 3 of the 12 mice had clearly been repopulated by a stem cell with lymphoid as well as myeloid differentiation potential, since they showed the same integration pattern i n both marrow and thymus (mice 1.7, 2.4 and 2.15) at the time of sacrifice.  3) DISCUSSION.  I have developed a simple procedure that allows small numbers of purified hemopoietic stem cells from mouse bone marrow to be stably and efficiently infected with a neo -containing r  retrovirus without any apparent effect on their potential for subsequent proliferation and differentiation i n vivo. A l l female recipients were injected with 2 x 10^ "compromised" female marrow cells i n addition to small numbers of purified male marrow cells that had been incubated overnight with Trtoeol9-containing medium. A l l 28 such mice analyzed, mcluding some that were injected with, at most, 150 purified marrow cells (of which 1 i n 85 are estimated to be capable of competitive long-term repopulation. Chapter III) showed subsequent reconstitution of their hemopoietic tissues by cells i n the purified (male) population. In 20 of these mice (71%), proviral DNA was also detected i n the regenerated hemopoietic cells.  - 109 Interestingly, i n every case this was seen i n cells from the peripheral blood or the spleen regardless of the time when the mice were evaluated (1-6 months after transplantation), but not necessarily i n the marrow or thymus. These findings clearly show that the cells isolated by the purification procedure used are capable of sustaining hemopoiesis for extensive periods of time after transplantation, even i n the presence of a competing graft that could, if injected alone, reconstitute these same animals (Chapter V). In one mouse the neo -containing restriction fragment demonstrated i n DNA r  from  marrow, spleen and thymus 121 days after transplantation was also identified i n purified subpopulations of marrow macrophages and splenic T- and B-lymphocytes, t h u s establishing the multilineage differentiation potential of the parent stem cell. Since all of these populations were also predominantly male, the cell from which the marked clone arose must have been present i n the purified, day 4 5-FU-treated marrow cell population. Similar results were obtained i n 2 other mice. In each of these, myeloid (marrow) as well as lymphoid (thymus) tissues contained the same two provlral Inserts. The equivalent intensity of the two bands (relative to one another) i n all tissues suggested that a single clone consisting of cells containing a double integration was present, rather t h a n two clones each having retrovirus integrated at a single unique site. Evidence of clones with double integrations was also obtained i n 4 other mice, although the tissue distribution of these clones at the time of analysis was not as broad. This does not necessarily mean that the differentiation potential of the original stem cells i n these latter animals was more restricted, since multiple parameters as yet not understood likely influence the commitment and amplification of individual lymphomyelopoietic stem cells transplanted i n vivo. Moreover, the cell turnover kinetics of different lineages may prevent simultaneous detection of multilineal clones as their contribution to a particular lineage modulates with time (4). Thus, although simultaneous involvement of lymphoid and myeloid cells was not obtained i n 9 of the 12 mice repopulated with neo -positive r  clones, it is quite possible that many, or even all, were derived from lympho-myelopoietic stem cells. More extensive longitudinal studies of purified subpopulations from individual mice  - 110 -  Injected with limiting numbers of purified stem cells will be necessary to characterize and quantitate more precisely the cell types present i n this purified population. The present experimental system thus appears well suited for further studies of purified hemopoietic stem cells with retroviral markers to identify individual clones regenerated i n irradiated recipients. However, it should be noted that under the conditions used, reconstitution of the entire hemopoietic system was usually oligoclonal. Many of the recipients exhibited male-female chimerism in at least one tissue, and the prevalence of a marked clone did not always correlate with the proportion of male cells. This situation appears to correspond well to that obtained i n the clinical setting of allogeneic bone marrow transplantation, where monoclonal reconstitution may be encountered but appears to be relatively uncommon (5,10). A s yet purified h u m a n stem cells have not been transplanted as an alternative to whole or T-lymphocyte-depleted marrow for clinical purposes. The present findings suggest the theoretical feasibility of such an approach. They also highlight the potential of using gene transfer to purified h u m a n hemopoietic stem cells for analyzing the value and importance of various hemopoietic cell subpopulations in h u m a n marrow for marrow rescue and their candidacy as targets for gene therapy.  - Ill -  REFERENCES 1. Abramson S. Miller RG, Phillips RA. The identification In adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J Exp Med 145:1567 (1977). 2. Dick JE, Magli MC, Huszar D, Phillips RA, Bernstein A. Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/W mice. Cell 42:71 (1985). v  3. Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell 45:917 (1986). 4. Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of mice reconstituted with retrovirus-infected stem cells. EMBO J 6:3955 (1987). 5. Turhan AG, Humphries RK, Phillips GL, Eaves AC, Eaves CJ. Clonal hematopoiesis demonstrated by X-linked DNA polymorphisms after allogeneic bone marrow transplantation. N Engl J Med 320:1655 (1989). 6. Barker JE, Braun J, McFarland-Starr EC. Erythrocyte replacement precedes leukocyte replacement during repopulation of W/W mice with limiting dilutions of+/+ donor marrow cells. Proc Natl Acad Sci USA 85:7332 (1988). v  7. Till JE, McCulloch EA. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat Res 14:213 (1961). 8. Spangrude GJ, Heimfeld S, Weissman IL. Purification and characterization of mouse hematopoietic stem cells. Science 241:58 (1988). 9. Ploemacher RE, Brons RHC. Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hemopoietic stem cell compartment following irradiation: Evidence for a pre-CFU-S cell. Exp Hematol 17:263 (1989). 10. Nash R. Storb R, Neiman P. Polyclonal reconstitution of human marrow after allogeneic bone marrow transplantation. Blood 72:2031 (1988).  - 112 -  C H A P T E R  V  VALIDATION O F A NEW A S S A Y F O R T H E L Y M P H O - M Y E L O I D H E M O P O I E T I C S T E M C E L L USING A C O M P E T I T I V E R E P O P U L A T I O N S T R A T E G Y  1)  INTRODUCTION.  The most primitive cell i n the hemopoietic system appears to be one that is capable of regenerating and maintaining all lymphoid and myeloid lineages for many months after transplantation (1-5). Nevertheless, despite more t h a n two decades of research to characterize this cell, there is still no quantitative assay that can be used to measure it exclusively. This situation probably reflects our relatively recent appreciation of the fact that the cells present i n the hemopoietic tissues of lethally irradiated mice at different times after transplantation of normal marrow cells are derived from different types of precursors (6) and, particularly at later times, cannot necessarily be assumed to be of donor origin (7). For example, it now seems likely that most, if not all, spleen colonies detectable up to 14 days after transplantation are derived from cells (CFU-S, colony-forming unit-spleen (8)) that are not capable of long-term (> 4 weeks) hemopoietic reconstitution (9). Two approaches to the development of longer term assays have therefore been used. The first and technically simplest has been to measure the frequency of cells able to confer protection against a lethal dose of radiation (10). This method has the advantage of being quantitative, but fails to take into consideration the possibility that death may result from the early absence i n the recipient of adequate numbers of cells capable of rapid, albeit transient, production of mature blood cells, even i n the presence of more primitive cells whose mature progeny may not appear until later times. Another approach has been to measure after two weeks the regeneration of new CFU-S i n the recipient's marrow.  - 113 thereby extending v i a a second short-term In vivo assay, the total time of hemopoietic regeneration to 4 weeks (11). However, the technical complexity of this latter double transplant assay, and potential limitations of its sensitivity, suggest that derivation of stem cell frequencies by limiting dilution analysis using this approach may be difficult. Moreover, i n the absence of adjunct procedures to establish the origin of the cells eventually regenerated i n either of these methods, calculated stem cell frequencies may significantly overestimate the actual content of injected long-term repopulating cells. The need for a practical assay that detects donor-derived cells with long-term lymphomyeloid repopulating ability and that allows such cells to be readily quantitated is thus strongly indicated. In Chapter III, I introduced the use of a 5 week competitive repopulation assay i n which lethally irradiated female mice are transplanted with limiting numbers of syngeneic, but male, "test" cells together with a large number (1 - 2 x 10^) of "compromised" (twice serially transplanted) marrow cells from syngeneic female mice. This latter population of cells contains a near normal frequency of CFU-S and various i n vitro clonogenic cells and can alone rescue recipients from the lethal effects of the radiation. However, because they have been subjected to two previous cycles of marrow transplantation and regeneration, such cells exhibit a markedly reduced competitive long-term repopulating ability (12). Their co-injection thus serves the dual purpose of ensuring the survival of the recipients and of providing a selective pressure to identify a class of stem cells with a high capacity for competitive long-term repopulation. The frequency of competitive repopulating units (CRU) i n the male test cell suspension c a n then be calculated using Poisson statistics to analyze test cell repopulated recipients injected with limiting numbers of test cells (13). Using this approach, I have identified and isolated a subpopulation of day 4 post-5-FU murine marrow cells that is highly enriched i n C R U ( > 5 0 % of mice injected with as few as 100 of these cells show donor regeneration of the majority of their marrow cells 5 weeks later, Chapter III). In addition, i n experiments described i n Chapter IV, I used retroviral marking to directly demonstrate that at least some of the cells i n this purified population are lympho-myeloid stem cells.  - 114 In the present study, I have undertaken a more complete characterization of several parameters of the competitive repopulation assay itself. Specifically, I have established the importance of genotype analysis to identify the test cell origin of cells i n regenerated hemopoietic tissues even i n this completely syngeneic system. I have also compared the effect of varying the time of tissue analysis from 5 weeks to 10 weeks post-transplant, and the effect of analyzing recipient marrow or thymus on the calculated frequency of C R U i n various cell suspensions. The results suggest that both 5 and 10 week endpoints may detect the same stem cell population, each member of which has the potential to differentiate along both lymphoid and myeloid lineages.  2)  RESULTS.  A)  Origin of Regenerated Cells i n the Competitive Long-Term Repopulation Assay.  Figure 17 outlines diagrammatically the competitive long-term repopulation assay procedure. In a n initial series of experiments, I found that 10^ compromised marrow cells alone are sufficient to allow lethally irradiated syngeneic recipients to survive long-term ( > 30 days), although 10® such cells are insufficient to out-compete the marrow or thymus repopulating ability of 1 0 day 4 5-FU marrow cells (14). To quantitate more precisely the 4  long-term repopulating potential of the compromised cell population, groups of lethally irradiated female mice were injected with 2 x 10^ male compromised cells and then sacrificed 10 weeks later. A s illustrated i n Figure 18A (Group 1), Southern analysis of marrow and thymus with a Y-specific probe revealed the presence of readily detectable ( > 5%) donor-derived (i.e. compromised cell) progeny i n both myeloid and T-lymphoid lineages of most mice analyzed (10 of 17 = 5 9 % , 3 experiments). To determine the relative contribution of reactivated surviving host stem cells under these same conditions, the identical experiment was performed using female compromised cells and lethally irradiated male mice. A s  "compromised" mouse  F i g u r e 17.  Schematic representation of the competitive long-term repopulation assay. Lethally irradiated female mice are injected with limiting numbers of male "test" marrow cells together with 1 - 2 x 10^ female "compromised" marrow cells. The proportion of male cells in the competitively repopulated marrow and thymus 5-10 weeks later is determined by Southern blot analysis using a Y-specific probe.  - 116 -  1  2  MALE C O M P CELLS  MALE HOSTS  (female hosts)  (female comp cells)  O (f  MARROW  THYMUS  Figure 18A. Southern analyses of bone marrow and thymus cells from lethally irradiated mice transplanted 10 weeks previously with 2 x 10 compromised female cells. Group 1: male compromised cells, female hosts (9 mice); Group 2: female compromised cells, male hosts (10 mice). The black box designates lanes that were not loaded with any DNA (because of insufficient numbers of cells in the thymus of these animals). Pure male and female marrow cell DNA controls hybridized to the Y-specific probe, pY2, are shown in the upper right panel. 5  - 117 illustrated i n Figure 18A (Group 2), i n many mice (10 of 13 = 7 7 % , 2 experiments) endogenous stem cells were able to out-compete 2 x 1 0 ^ compromised marrow cells and contribute significantly to the long-term reconstitution of both the marrow and thymus. Similar results were also obtained i n the marrows of 5 of 6 (= 8 3 % , 1 experiment) irradiated male mice injected with 10® compromised female marrow cells (data not shown). I next evaluated the repopulating potential of both the compromised cells and the host when these were assessed i n the context of the competitive repopulation assay using day 4 5-FU marrow cells as a potential source of cells with a high content of C R U (Chapter III). In these experiments, irradiated recipients were injected with 1 0 day 4 5-FU cells and 2 x 10^ 4  compromised cells. Male mice were used only for the component being evaluated so that the entire experiment was performed i n triplicate a total of three times. Representative results are shown i n Figure 18B. They indicate that although 1 0 day 4 5-FU cells are more competitive 4  i n reconstituting the marrow and thymus (i.e. 18 of 18 mice = 1 0 0 % contained > 5 % of cells derived from the day 4 5-FU cell suspension (Group 3)), i n some mice (3 of 15 = 2 0 % , 3 experiments), contributions to long-term hemopoiesis from cells i n the compromised population were, nevertheless, also detectable (Group 4). Similarly, readily detectable contributions from endogenous stem cells i n the irradiated host were also found i n the majority of mice (Group 5, 9 of 14 = 6 4 % , 3 experiments).  B)  Selection of endpoint time.  I next undertook a series of experiments to examine the effect of varying the time after transplantation prior to assessment of tissue repopulation by donor test cells on the frequency of C R U measured by limiting dilution analysis. In these experiments, groups of female recipients were co-transplanted with varying numbers of male test cells and a constant number of 2 x 10^ compromised female cells. Comparisons of 5 and 10 week data were made for 3 different test cell suspensions: normal marrow, day 4 5-FU marrow and the highly enriched  - 118 -  5  4  3  MALE C O M P CELLS  MALE HOSTS  (female c o m p cells)  (female 5-FU cells)  (female 5-FU cells)  ( f e m a l e hosts)  ( f e m a l e hosts)  (female c o m p cells)  M A L E 5-FU  CELLS  MARROW  THYMUS  Figure 18B.  Southern analyses of bone marrow and thymus cells from lethally irradiated mice transplanted 10 weeks previously with l O day 4 5-FU cells and 2 x 10^ compromised marrow cells. Group 3: male 5-FU cells, female compromised cells, female hosts (10 mice); Group 4: female 5-FU cells, male compromised cells, female hosts (8 mice); Group 5: female 5-FU cells, female compromised cells, male hosts (9 mice). Pure male and female DNA controls hybridized to the Y-specific probe, pY2, are shown in Figure 18A. 4  - 119 subpopulation of day 4 5-FU marrow obtained by F A C S selection as described i n Chapter III. For each test cell suspension, all available data were pooled and then the frequency of C R U calculated as described i n Chapter III. Figure 19 shows the results that were obtained for day 4 5-FU marrow based on analyses of marrow repopulation i n recipients sacrificed 10 weeks after transplantation. The frequency of C R U derived from these data is 1 per 1,200 day 4 5-FU cells ( 9 5 % confidence limits: 1 per 780 cells to 1 per 2,000 cells). This frequency is similar to the value obtained when recipients were analyzed 5 weeks earlier i n separate experiments (Chapter III and Table VI). Similarly, the frequencies of C R U measured i n normal marrow or i n the sorted day 4 5-FU marrow subpopulation also appeared relatively Insensitive to the time of recipient marrow assessment after 5 weeks, although the frequency of C R U In these test cell suspensions differed by a factor of 30 to 200-fold.  C)  Comparison of Thymus Versus Marrow Analyses.  In Chapter IV, I used retrovlral-marklng of purified repopulating stem cells to show that the competitive repopulation assay can, and frequently does, detect a stem cell with lymphoid as well as myeloid repopulating potential. If this were a consistent selective feature of this assay, one would expect to measure similar C R U frequencies irrespective of whether marrow or thymus repopulation were assessed, given that sufficient time for both to have occurred h a d elapsed. In each of the experiments described above, both thymus and marrow from the recipients were therefore analyzed for the presence of male "test" cells and the results of C R U determinations compared. A representative set of data for day 4 5-FU marrow based on analysis of thymus repopulation after 10 weeks is shown i n Figure 20, and a summary of all derived C R U frequencies for the different test cell suspensions is included i n Table VI. Again, the tissue evaluated, like extension of the time interval prior to tissue evaluation from 5 weeks to 10 weeks, had no consistent effect on the calculated frequency of C R U i n the different cell suspensions. A s a further test of this, I analyzed all 170 mice injected with limiting numbers  - 120 -  Number of Test Cells Per Mouse  Figure 19.  Percent of recipients (8 animals per group) showing > 5 % repopulation of the marrow with male test cells 10 weeks after transplantation of varying numbers of male day 4 5-FU marrow cells together with 2 x 10^ female compromised marrow cells. The proportion of male DNA was determined by Southern analysis of marrow cell DNA from each recipient using pY2. Open and closed circles designate the data from two separate experiments.  - 121 -  Table VI.  Comparison of the Frequency of C R U In Different Marrow Cell Populations Assessed Using Different E n d p o i n t s a  5 Weeks Post-Transplantation Cells  BM  Normal B M  10,000 (6,200 - 16,000)  Day 4 5-FU B M  2,700 (1,300 - 5,700)  Sorted Day 4 5-FU B M D  65 (40 - 120)  Thy 10,000 (6,000 - 16,400) 1,300 (540 - 3,100) 125 (65 - 250)  10 weeks Post-Transplantation BM 3,600 (1,600 - 8,000)  Thy 15,700 (7,200 - 34.300)  1,200 (780 - 2,000)  1,300 (830 - 2,000)  125 (65 - 240)  75 (440 - 135)  a  Determined by limiting dilution analyses (see legend to Figure 19). Values shown are the r e c i p r o c a l of the C R U frequency w i t h t h e corresponding 9 5 % confidence l i m i t s defined b y ± 2 S E (shown i n brackets) based on 1 to 3 pooled experiments (20 to 3 0 mice per experiment).  b  C R U frequency i n sorted cells i s compensated for the two-fold r e d u c t i o n i n seeding efficiency due to antibody coating of the cells as noted previously (Table II).  - 122 -  10  10  2  10  3  10  4  10  5  Number of Test Cells Per Mouse  Figure 20.  Percent of recipients (8 animals per group) showing >5% repopulation of the thymus with male test cells 10 weeks after transplantation of varying numbers of male day 4 5-FU marrow cells together with 2 x 10^ female compromised marrow cells. The proportion of male DNA in the thymus of each recipient was determined using pY2. Open and closed squares designate data for mice from the same two experiments shown in Figure 19 and the dashed line represents the curve from Figure 19, shown for comparison.  - 123 of male cells where both marrow and thymus results were available to determine whether the presence (or absence) of detectable male progeny i n these two tissues were independent of one another. The results of this analysis (Table VII) for each of the three types of male test cells used (normal marrow, day 4 5-FU marrow and sorted day 4 5-FU marrow) showed that i n each case repopulation of the marrow and repopulation of the thymus were not independent (p < 0.005, G-test of independence). A n association between the repopulation of the two tissues using this test was also apparent when all mice were evaluated together regardless of the source of transplanted male test cells (p < 0.001).  3)  DISCUSSION.  A simple i n vivo assay for a class of primitive hemopoietic stem cells with competitive long-term repopulating ability was described. Using limiting dilution analysis techniques, an estimate of the relative frequency of this stem cell i n various test cell suspensions can be obtained. S u c h estimates are of necessity underestimates since the detection efficiency of the assay procedure is not known, but is almost certain to be less than one. I have therefore suggested the term competitive repopulating unit (CRU) to distinguish between the true number of competitive repopulating cells i n a given suspension and the presumably m i n i m u m number detectable using the i n vivo assay described here, as first suggested when the term CFU-S was introduced for the cells measured by the spleen colony assay (8). A s for any assay, it was important to establish the optimal input - output interval and to investigate the sensitivity of the results to variations i n this interval. Initially, I chose 5 weeks as a m i n i m u m time to detect the progeny of cells more primitive t h a n CFU-S. This was based on previous work by Hodgson et al. (11) suggesting that there exists a cell population i n murine marrow that differs from CFU-S and which can be detected by measuring the number of day 13 CFU-S produced i n the marrow of primary irradiated recipients 13 days after transplantation. In the present study, I compared the frequency of CRU  i n normal marrow, marrow from  Table VII.  Test of Independence of Marrow and Thymus Repopulation by Male Test Cells When Co-Injected With 2 x 1 0 ^ Compromised Female Marrow Cells*  Cells  BM+ Thy+  BM+ Thy-  BMThy+  BMThy-  P-value  Normal BM  14  9  7  24  <0.005  Day 4 5-FU B M  26  2  3  24  <0.001  Sorted Day 4 5-FU B M  16  5  3  37  <0.001  TOTAL  56  16  13  85  <0.001  D  a  Values shown are the number of mice i n each category using > 5 % male cells as the criterion for positivity. Only mice in groups where the likelihood of repopulation by male cells was > 0 % but < 1 0 0 % (from experiments used to derive the values shown in Table VI) were included i n the analysis shown here.  b  Probability that repopulation of the marrow is independent of the repopulation of the thymus (G-test of independence).  - 125 5-FU-treated mice, and a highly purified subpopulation of day 4 5-FU marrow using either 5 or 10 week endpoints and found no evidence that the additional 5 weeks made any difference. This is the result that would be expected if, after 5 weeks under the conditions used (i.e. under conditions of competitive repopulation), the composition of the recipient's marrow already reflected the proliferative activity of the most primitive hemopoietic cells present i n the recipient immediately post-transplant. A s a further test of this hypothesis, I compared the C R U frequency derived from assessment of marrow versus thymus repopulation data. The results appear to be independent of the tissue evaluated between 5 and 10 weeks after transplantation. This provides further support for the view that either tissue endpoint detects the proliferative and differentiative activity of the same initial stem cell population. Since the marrow of these animals contains primarily myeloid cells and the thymus contains almost exclusively T-cells at the times assessed, this implies detection of an initial stem cell type with the capacity to regenerate both types of progeny. Additional evidence for this is indicated by the significant association found between test cell contributions to repopulation of the marrow and thymus of individual mice, i n spite of considerable variation i n the levels attained i n either. Some variation i n the contribution even of single totipotent cells to different tissues was noted i n the experiments described i n Chapters III and IV and is not surprising given the likelihood that different myeloid and lymphoid compartments have different rates of turnover. In addition, some cells with the potential for generating both lymphoid and myeloid progeny may not actually express both due to the mechanisms that determine lineage-commitment  of totipotent cells in vivo.  Although the molecular nature of these mechanisms are still completely obscure, evidence that stem cell differentiation i n vivo appears stochastic has been well documented (17,18). S u c h a mechanism predicts that a proportion of totipotent cells will not express their full differentiation potential i n vivo because they happen to become lineage restricted within their first division.  - 126 It is interesting to note that the frequency determined here for CRU  i n normal B 6 C 3 F J  marrow (approximately 1 per 1 0 cells) is similar to that previously estimated by other 4  endpoints related to long-term repopulation of the hemopoietic system; ie. rescuing lethally irradiated mice (10), curing W/W  v  mice (15) or analysis of the number of clones contributing to  hemopoiesis after 6 weeks (16). Although some of these latter assays are less cumbersome to perform t h a n the competitive long-term repopulation assay described here, they may not be specific for the most primitive cells of donor origin. Indeed, the use of lethally irradiated mice as transplant recipients may require the presence i n the graft of relatively mature hemopoietic cells which play an early supportive role i n engraftment and may, i n the absence of adjunct procedures to distinguish short and long-term reconstitution phases, diminish the specificity of this system for the most primitive stem cells. The use of W/W  v  recipients can circumvent this  problem to some extent, although endogenous stem cells i n such mice can exert significant competitive pressure thus preventing the seeding and proliferation of those i n the graft. The present use of acutely irradiated normal recipients, co-transplanted short-term repopulating cells and sex-markers to distinguish the proliferation of the most primitive donor stem cells from those which function primarily at earlier times after transplantation provides a sensitive and specific i n vivo system for the quantitation of the former. This competitive assay is particularly useful for the evaluation of limiting numbers of highly purified stem cells which will be depleted of short-term repopulating cells required to rescue lethally irradiated mice, and which, i n W/W  v  mice may be prevented from effective penetration of occupied stem cell niches.  Several reports have suggested the feasibility of obtaining highly purified populations of mouse marrow enriched for stem cells with i n vivo repopulating ability (19-21). These purification strategies have proven useful i n studying the relationships among the cells i n the most primitive hemopoietic cell compartments. However, marked differences i n the extent to which these procedures also co-purify CFU-S, and the possibility that the populations isolated may be subdivided further (9,21), suggests that biologic homogeneity has not yet been achieved. Application of the assay described here to measure and compare C R U frequencies i n  - 127 -  purified cell populations should facilitate the development of additional procedures for achieving this goal as a n essential step towards the ultimate characterization of lymphomyeloid repopulating cells at the molecular level. Similarly, assessment of C R U i n other tissues (e.g. spleen, blood, yolk sac and fetal liver) or i n long-term marrow cultures, will provide important information about the size of this stem cell population during ontogeny a n d following genetic, biologic (with growth factors) or pharmacologic manipulation i n vitro. Hemopoietic stem cells able to regenerate a detectable and therefore significant proportion of both the marrow a n d thymus were also found to be present i n the suspension of compromised marrow cells and i n lethally irradiated recipients. The contribution of such stem cells i n the compromised marrow inoculum to the long-term reconstitution of hemopoiesis was largely out-competed when 20 to 100-fold fewer unseparated day 4 5-FU cells were co-transplanted. However, the contribution of residual host stem cells to the long-term reconstitution of mice rescued with a graft of compromised marrow cells, although minor, remained detectable i n many mice even when highly competitive day 4 5-FU cells were cotransplanted at numbers sufficient to dominate the regenerated tissues of most animals. These data are consistent with the concept that the majority of stem cells i n the compromised population are qualitatively reduced i n their repopulating ability by comparison to normal marrow, whereas the chief effect of the acute irradiation given to assay recipients is to greatly reduce, but not completely eliminate, the number of highly competitive repopulating stem cells. These data also imply that the long-term regeneration of hemopoiesis obtained i n the assay described is commonly polyclonal with some contributions of progeny from both sources of transplanted cells as well as from the host. This is i n apparent contrast to the monoclonal pattern of repopulation reported by several groups after primary transplantation of small numbers of effective marrow cells (3,4,22,23). However, none of these latter studies examined specifically the contribution of residual host stem cells to hemopoietic recovery and many have relied o n the use of W/W  v  rather than acutely irradiated normal mice as recipients. The  present findings t h u s highlight the likelihood of residual host stem cell activity even i n  - 128 "lethally" irradiated (8 - 8.5 Gy) but otherwise normal syngeneic recipients, and encourage the investigation of additional procedures that may allow such host cells to be selectively recruited or suppressed.  - 129 REFERENCES  1. W u AM, Till J E , Siminovitch L, McCulloch EA. Cytological evidence for a relationship between normal hemopoietic colony-forming cells and cells of the lymphoid system. J Exp Med 127:455 (1968). 2. Abramson S, Miller RG, Phillips RA. The identification i n adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J E x p Med 145:1567 (1977). 3. D i c k J E , Magli MC, Huszar D, Phillips RA, Bernstein A. Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/W mice. Cell 42:71 (1985). v  4. Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell 45:917 (1986). 5. Keller G, Paige C, Gilboa E, Wagner EF. Expression of a foreign gene i n myeloid and lymphoid cells derived from multipotent haematopoietic precursors. Nature 318:149 (1985). 6. Magli MC, Iscove NN, Odartchenko N. Transient nature of early haematopoietic spleen colonies. Nature 295:527 (1982). 7. Jones R J , Celano P, Sharkis S J , Sensenbrenner LL. Two phases of engraftment established by serial bone marrow transplantation i n mice. Blood 73:397 (1989). 8. Till J E , McCulloch EA. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat Res 14:213 (1961). 9. Ploemacher RE, Brons RHC. Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hemopoietic stem cell compartment following irradiation: Evidence for a pre-CFU-S cell. Exp Hematol 17:263 (1989). 10. McCulloch EA, Till J E . The radiation sensitivity of normal mouse bone marrow cells, determined by quantitative marrow transplantation into irradiated mice. Radiat Res 13:115 (1960). 11. Hodgson GS, Bradley TR, Radley J M . The organization of hemopoietic tissue as inferred from the effects of 5-fluorouracil. Exp Hematol 10:26 (1982). 12. Harrison DE, Astle CM, Delaittre JA. Loss of proliferative capacity i n immunohemopoietic stem cells caused by serial transplantation rather t h a n aging. J Exp Med 147:1526 (1978). 13. Taswell C. Limiting dilution assays for the determination of immunocompetent ceil frequencies. I. Data analysis. J Immunol 126:1614 (1981). 14. Eaves C J , Sutherland HJ, Szilvassy SJ, Fraser CC, T u r h a n AG, Humphries RK, Eaves AC, Lansdorp PM. Phenotypic and functional characterization of primitive hemopoietic stem cells. In: Proceedings of a n International Symposium on Molecular Biology of Hematopoiesis (in press) (1989).  - 130 15. Boggs DR, Boggs SS, Saxe DS, Gress RA, Confield D R Hematopoietic stem cells with high proliferative potential. Assay of their concentration i n marrow by the frequency and duration of cure of W/W mice. J A m Soc C l i n Invest 70:242 (1982). v  16. Micklem HS, Lennon J E , Ansell JD, Gray RA. Numbers and dispersion of repopulating hematopoietic cell clones i n radiation chimeras as functions of injected ceil dose. Exp Hematol 15:251 (1987). 17. Till J E , McCulloch EA, Siminovitch L. A stochastic model of stem cell proliferation, based on the growth of spleen colony-forming cells. Proc Natl Acad Sci USA 51:29 (1964). 18. Gregory C J , Henkelman RM. Relationships between early hemopoietic progenitor cells determined by correlation analysis of their numbers i n individual spleen colonies. In: B a u m S J , Ledney GD. Experimental Hematology Today. New York: Springer-Verlag (1977). 19. Visser JWM, B a u m a n J G J , Mulder AH, Eliason JF, de Leeuw AM. Isolation of murine pluripotent hemopoietic stem cells. J Exp Med 59:1576 (1984). 20. Spangrude G J , Heimfeld S, Weissman IL. Purification and characterization of mouse hematopoietic stem cells. Science 241:58 (1988). 21. Visser JWM, de Vries P. Isolation of spleen-colony forming cells (CFU-s) using wheat germ agglutinin and rhodamine 123 labeling. Blood Cells 14:369 (1988). 22. Snodgrass R Keller G. Clonal fluctuation within the haematopoietic system of mice reconstituted with retrovirus-infected stem cells. E M B O J 6:3955 (1987). 23. Capel B, Hawley R, Covarrubias L, Hawley T, Mintz B. Clonal contributions of small numbers of retrovlrally marked hematopoietic stem cells engrafted i n unirradiated neonatal W/W mice. Proc Natl Acad Sci USA 86:4564 (1989). v  - 131 -  C H A P T E R  VI  S U M M A R Y A N D F U T U R E DIRECTIONS  Hemopoiesis i n both mouse and m a n is characterized by a spectrum of primitive hemopoietic cell populations cUstinguished by differences in the differentiation potential they express (1-3). The extremes of this spectrum are represented on the one hand by a group of committed, lineage-restricted or unipotential progenitor cells (which are the immediate precursors of the morphologically recognizeable, terminally differentiated cells of the various myeloid a n d lymphoid lineages), a n d on the other by a class of totipotent stem cells capable of extensive self-renewal and able to regenerate and maintain both myeloid and lymphoid systems for very long periods following transplantation. Evidence that these two extremes are separate and distinct populations (the former being derived from the latter) has come primarily from cell separation experiments i n combination with analyses of the progeny of individual cells generated either i n vivo using unique genetic markers (4) or i n vitro i n semi-solid culture systems that support the proliferation a n d differentiation of single or multiple lineages of cells i n isolated colonies (reviewed i n (5)). Over the last thirty years, a variety of assays for primitive hemopoietic cells with multilineage differentiation potential have been described. The assay for CFU-S (6) was the first s u c h assay a n d has been widely used to quantitate a pluripotent hemopoietic stem cell capable of extensive proliferation and some self-renewal i n vivo. Although it has been reported that spleen colony-forrning cells generating late-appearing colonies (day 12 CFU-S) can rescue mice from lethal doses of irradiation (7), recent reports of the differential purification of day 12 CFU-S a n d more primitive cells (termed pre-CFU-S) whose frequency more accurately predicts marrow repopulating ability (8) suggests that these are likely to be largely separable cell types. Future characterization of the most primitive cells in the hemopoietic system and the  - 132 mechanisms that regulate their behaviour will thus require new assays specific for these latter cells. In the mouse, the capacity of a marrow cell population to reconstitute lymphomyelopoiesis upon transplantation into lethally irradiated or mutant, hemopoieticalfy compromised hosts has t h u s recently come to be recognized as the only reliable means of identifying the most primitive hemopoietic cells with multilineage differentiation potential. A combination of appropriate genetic markers have been used to show that such cells c a n individually regenerate and maintain both lymphoid and myeloid systems for many months after transplantation (2,4,9,10). These studies have also revealed some evidence of heterogeneity i n the differentiative as well as the proliferative potential of normal hemopoietic cells with i n vivo repopulating potential. However, the extent to which this heterogeneity may be attributed to intrinsic genetic differences among individual stem cells, as opposed to stochastic or biologic processes that regulate stem cell recruitment i n vivo is not known. Moreover, because of the retrospective nature of progeny analyses, the properties of the original precursor of interest are not readily accessible to study. This requires the availability of purified repopulating stem cells identified as such prior to transplantation. A major goal of my research was thus to develop a procedure that might ultimately enable murine lympho-myeloid repopulating stem cells to be obtained i n pure form. Three key considerations to such a n approach were the choice of assay used to define the stem cell population to be purified, the cellular characteristics used to select this population, and the method used for cell separation. The first studies I undertook showed that it was possible to use a relatively simple cell sorting procedure to obtain a very small subpopulation of day 4 post-5-FU marrow cells that was significantly enriched i n hemopoietic cells detectable by several short-term i n vitro and i n vivo assays. This population was also enriched i n cells with marrow reconstituting ability (Chapter III). The procedure I developed was based on preliminary characterization of these cells with respect to their forward and orthogonal lightscattering properties, a n d their surface expression of the Thy-1 and Class I M H C (H-2K)  - 133 antigens. Although several groups had reported the expression of these antigens on relatively mature hemopoietic cells (11-14), very little was known about their expression on the most primitive stem cells with repopulating potential. The studies I completed confirmed the existence of considerable functional overlap among the hemopoietic cells evaluated with respect to these four parameters. However, by appropriate gating, I was able to sort a population which was enriched up to 100-fold l n CFU-GEMM, and i n which 1 i n 4 cells was a day 12 CFU-S. To determine the extent to which this procedure copurifies a primitive stem cell with long-term lympho-myeloid repopulating potential, I devised a competitive repopulation assay i n which male "test" cells were transplanted into lethally irradiated female mice together with a "compromised" female marrow cell population with normal short-term, but diminished longterm, repopulating ability. The use of the compromised cells ensured the survival of recipients transplanted with limiting numbers of test cells, and also provided a competitive pressure to ensure that regeneration from test stem cells would only occur if these had superior repopulating potential. Because the test cells were obtained from male donors, their progeny could be specifically detected using a Y-specific probe. Assessment of DNA  from recipient  tissues following transplantation of the sorted day 4 5-FU marrow population under these conditions revealed that they were able to repopulate all hemopoietic tissues upon transplantation. Quantitation of this potential by limiting dilution analysis indicated that at least 1 i n 85 sorted cells was capable of competitive long-term marrow repopulation and that an enrichment of ~30-fold over unseparated day 4 5-FU marrow had been achieved. To formally establish that lympho-myeloid stem cells had been isolated i n this sorted population, I undertook experiments determine if this could be demonstrated using retroviral marking techniques. The approach was to uniquely mark the sorted stem cells prior to transplantation and then analyze the contribution of individual clones to the repopulation of multiple lineages after varying periods of time. The availability of a psi-2 producer cell line liberating high titres of an ecotropic, helper-free recombinant retrovirus carrying the n e o gene R  - 134-  facilitated the development of an i n vitro infection protocol i n which small numbers of sorted cells were incubated overnight with vims-containing supernatants. Assessment of the progeny of sorted day 4 5-FU marrow cells that had been marked i n this way prior to transplantation i n the competitive repopulation assay revealed the presence of uniquely marked donor (male) cells i n several lymphoid and myeloid lineages of 3 of 12 mice analyzed 35 to 196 days posttransplantation (Chapter IV). These studies demonstrated that it was possible to use retroviruses to effect gene transfer to small numbers of highly purified repopulating hemopoietic stem cells. The detection i n some mice of male lymphoid and myeloid cells containing common retroviral integration sites also directly demonstrated that at least some stem cells i n the purified population were totipotent. Clones with more restricted tissue distributions at the time of analysis were also detected. M y final objective was to investigate whether the competitive repopulation assay was i n fact able to serve as a procedure for the exclusive quantitation of long-term lympho-myeloid repopulating stem cells. Experiments i n which male mice were used uniquely either for the compromised or recipient components of this assay system demonstrated that stem cells from both these sources are able to contribute significantly to the repopulation of marrow and thymus, particularly at later times (10 weeks) after transplantation, although both were largely out-competed b y even minimal numbers (<10 ) of cotransplanted day 4 5-FU marrow cells. 4  Because recipient survival is not a limiting factor i n the sensitivity of the repopulation assay, I was able to use a limiting dilution approach to the quantitation of stem cells with competitive repopulating ability i n a large number of marrow cell suspensions. The results of these experiments revealed that the frequency of competitive repopulating units (CRU) calculated using this method was independent of whether recipient marrow or thymus was analyzed for its content of donor (ie. male) cells (Chapter V). This suggested that either tissue endpoint could be used to define a common stem cell with dual lympho-myelopoietic repopulating potential (provided that sufficient time (>5 weeks) had elapsed for both to have occurred). This  - 135 conclusion was further supported by the finding that within individual mice the degree of repopulation of marrow and thymus by donor cells was similar. A n understanding of hemopoietic repopulation i n the murine system is required for the potential application of stem cell transplantation to the treatment of hemopoietic malignancies i n man.  While the experiments described i n this thesis have provided m u c h information about  the characteristics and biology of normal murine hemopoietic stem cells, they have also raised a number of critical issues for future investigation. For example, although the marrow cell population isolated by the four-parameter sorting procedure described here is clearly enriched i n lympho-myeloid repopulating stem cells, it seems unlikely that total purity is achieved as a result. The addition of further or alternative enrichment steps may facilitate the separation of transplantable hemopoietic cells from those able to generate colonies i n vitro, and allow a determination of whether these are i n fact completely distinct. The inclusion of Rhodamine123 uptake (15) i n a five-parameter sorting procedure appears presently be the most promising approach to this end. However, the nature of hemopoietic stem cell recruitment i n vivo makes an evaluation of the real purity of repopulating cells difficult. Until more is known about the seeding efficiency and recruitment of long-term repopulating stem cells after transplantation, i n vivo assays will always be accompanied by some uncertainty and provide only minimal estimates of stem cell frequency. Based on earlier discussions of the importance of local interactions between hemopoietic and stromal cells in the maintenance and activation of the former, it follows that determinants expressed on the hemopoietic cell membrane must play some role i n these associations. In addition to facilitating the separation of otherwise indistinguishable hemopoietic cell subpopulations, an understanding of the composition of the hemopoietic stem cell membrane may also provide important insights into the mechanism of their regulation. Why is the Thy-1 antigen expressed on hemopoietic stem cells ? Do stem cell surface Class I MHC  antigens  function to mediate interactions between these cells and others in the marrow ? Is this a mechanism for the host T-cell-mediated graft failure which sometimes follows allogeneic bone  - 136 marrow transplantation (16,17) ? These are important questions which can be addressed using purified stem cells as targets for the deletion of genes encoding cell surface antigens by homologous recombination (18), for membrane solubilzation or immunoprecipitation studies, or molecular analyses using subtraction hybridization. At present, the primary limitation to all these strategies is the availability of large numbers of biologically homogeneous stem cells. The development of the competitive repopulation assay for the detection and quantitation of lympho-myeloid repopulating stem cells has also raised a number of interesting questions about factors that regulate primitive hemopoietic cell activation after their transplantation. What is the role of the "compromised" marrow cells i n short-term engraftment ? To what extent does serial transplantation qualitatively versus quantitatively alter the most primitive stem cell type ? These are particularly important questions i n view of growing evidence that reconstitution of hemopoiesis following marrow transplantation occurs i n two phases with relatively mature progenitors that do not have extensive self-renewal capacity providing initial, transient engraftment, and primitive totipotent stem cells ultimately sustaining long-term hemopoiesis (19,20). Despite their average reduction i n competitive long-term repopulating potential, stem cells i n the compromised population can contribute to hemopoiesis i n some mice at later times after transplantation. What is the frequency of CRU i n the compromised population ? This can be addressed by transplanting limiting numbers of male compromised cells to compete against their female counterparts for the reconstitution of female mice i n the limiting dilution form of the competitive repopulation assay. How  does the ratio of  "compromised" to "test" marrow cells influence the determined frequency of C R U ? In its present form, all recipients i n the competitive repopulation assay are injected with 2 x 10^ compromised cells and variable numbers of test cells. In mice recieving very few test cells, stem cells may be subject to increasingly greater competitive pressure which may  ultimately  prevent their detection. It would be very interesting to undertake experiments similar to those described here where a constant compromised:test cell ratio were used, and then to vary this ratio to determine the effect on the distribution of positive test cell responses from which  CRU  - 137 frequencies are derived. Indeed, a more thorough analysis of the effect of varying a number of arbitrarily selected thresholds (ie. the time of recipient analysis, minimal proportion of male cells required for a positive response) is propitious. The extent of residual functional host stem cell activity i n "lethally" irradiated mice was a somewhat surprising finding. This potential of the host for auto-repopulation has significant implications for the development of future strategies for clinical bone marrow transplantation i n man. Using a n approach similar to that described here, the usual ablative therapy might be followed by transplantation of transiently reconstituting supportive populations, a n d subsequent selective activation of endogenous normal stem cells which would regenerate and maintain long-term hemopoiesis. A significant advantage might then be the ability to use T-depleted grafts from histoincompatible donors. Conversely, it might be possible to consider the development of strategies to correct the uncontrolled proliferation of leukemic stem cells i n situ by suppression with specific factors following immediate ablation of the more mature components of the malignant clone with cycle-specific cytotoxic agents. The lack of an i n vivo assay for h u m a n repopulating stem cells has made a direct analysis of the most primitive h u m a n hemopoietic cells difficult. However, stem cells with the capacity for long-term repopulation are maintained i n murine and probably also h u m a n longterm bone marrow cultures (LTBMCs) (21,22). The use of the competitive repopulation assay to determine the extent to which C R U are also maintained i n murine L T B M C may thus allow a determination of its relationship to both murine and h u m a n long-term culture initiating cells (23), and help to validate the use of L T B M C initiation as a n assay for h u m a n stem cells (24). If C R U are present i n murine LTBMC, the ability to enrich for such cells using the sorting procedure described (with or without adjunct selective steps) provides a powerful opportunity to manipulate repopulating stem cells i n vitro, either by exposure to specific hemopoietic growth factors or by selective activation or suppression of genes which play a role i n stem cell development.  - 138 The potential for retxovirus-mediated gene transfer to defined, highly purified populations of lympho-myeloid repopulating stem cells offers perhaps the most exciting avenue of future research into the regulation of primitive hemopoietic cell development. In view of the diversity of lymphomas and leukemias, the effect of specific oncogenes o n the subsequent development of multipotential hemopoietic stem cells is of particular interest to pursue. S u c h studies may allow the isolation of permanent stem cell lines which can be maintained continously i n vitro, but retain their capacity to differentiate along several lymphoid and myeloid lineages. Another class of genes particularly Important to analyze for their effects on stem cells will be regulatory genes with a putative role i n lineage commitment. Examples of such genes include certain homeobox genes, zinc finger genes, or helix-loop-helix genes, some of which themselves have recently been implicated as hemopoietic oncogenes (25,26). Because very little is known about the growth requirements of totipotent stem cells and the optimal conditions which favor their high frequency infection by recombinant retroviruses, such studies may require a definition of culture conditions that will permit stem cell cycling and self-renewal. The increasing availability of purified hemopoietic growth factors will be of particular benefit to these studies.  In summary, my studies have contributed to a description of the phenotype of long-term lympho-myeloid repopulating stem cells enabling their isolation and characterization before transplantation. The development of the competitive repopulation assay, combined with the use of retrovirally-marked stem cells, is crucial to the detection, and hence quantitation, of limiting numbers of cells with this potential. Further studies should make it possible to determine their relationship to murine a n d hence to h u m a n hemopoietic cells that are detected by their ability to initiate sustained hemopoiesis i n long-term marrow cultures. The procedures described here should also facilitate future studies of the role of carrier/accessory cell populations i n the regulation of hemopoietic stem cell activation i n vitro a n d i n vivo.  - 139 REFERENCES 1. W u AM, Till J E , Sirninovitch L, McCulloch EA. Cytological evidence for a relationship between normal hemopoietic colony-forming cells and cells of the lymphoid system. J Exp Med 127:455 (1968). 2. Abramson S, Miller RG, Phillips RA. The identification i n adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J Exp Med 145:1567 (1977). 3. Till J E , McCulloch EA. Hemopoietic stem cell differentiation. Biochem Biophys A c t a 605:431 (1980). 4. D i c k J E , Magli MC, Huszar D, Phillips RA, Bernstein A. Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/W mice. Cell 42:71 (1985). v  5. Metcalf D. Hemopoietic Colonies. In vitro cloning of normal and leukemic cells. Berlin Heidelberg: Springer-Verlag (1977). 6. Till J E , McCulloch EA. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat Res 14:213 (1961). 7. Visser JWM, B a u m a n J G J , Mulder AH, Eliason JF, de Leeuw AM. Isolation of murine pluripotent hemopoietic stem cells. J Exp Med 59:1576 (1984). 8. Ploemacher RE, Brons RHC. Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hemopoietic stem cell compartment following irradiation: Evidence for a pre-CFU-S cell. Exp Hematol 17:263 (1989). 9. Lemischka IR, Raulet DH, Mulligan RC. Developmental potential and dynamic behavior of hematopoietic stem cells. Cell 45:917 (1986). 10. Snodgrass R, Keller G. Clonal fluctuation within the haematopoietic system of mice reconstituted with retrovirus-infected stem cells. E M B O J 6:3955 (1987). 11. Boswell HS, Wade J r PM, Quesenberry PJ. Thy-1 antigen expression by murine high-proliferative capacity hematopoietic progenitor cells. I. Relation between sensitivity to depletion by Thy-1 antibody and stem cell generation potential. J Immunol 133:2940 (1984). 12. Williams DE, Boswell HS, Floyd AD, Broxmeyer HE. Pluripotential hematopoietic stem cells i n post-5-fluorouracil murine bone marrow express the Thy-1 antigen. J Immunol 135:1004 (1985). 13. Mulder AH, B a u m a n J G J , Visser JWM, Boersma WJA, V a n D e n E n g h GJ. Separation of spleen colony-forming units and prothymocytes by use of a monoclonal antibody detecting a n a n H-2K determinant. Cell Immunol 88:401 (1984). 14. Okamoto T, Kanamaru A, H a r a H, Nagai K. Changes i n expression of major histocompatibility complex (MHC) class-I antigen on hematopoietic progenitors during murine development. Exp Hematol 15:190 (1987). 15. Visser JWM, de Vries P. Isolation of spleen-colony forming cells (CFU-s) using wheat germ agglutinin and rhodamine 123 labeling. Blood Cells 14:369 (1988).  - 140 -  16. B u t t u r i n l A, Seeger RC, Gale RP. Recipient immune-competent T lymphocytes can survive intensive conditioning for bone marrow transplantation. Blood 68:954 (1986). 17. Vriesendorp HM. Engraftment of hemopoietic cells. In: van B e k k u m DW & Lowenberg B. Bone Marrow Transplantation: Mechanisms and Clinical Practice. New York and Basel:Marcel Dekker Inc., p. 73 (1985). 18. Smithies O, Gregg RG, Boggs SS, Koralewski MA, Kucherlapati RS. Insertion of DNA sequences into the h u m a n chromosomal p-globin locus by homologous recombination. Nature 317:230 (1985). 19. Jones R J , Celano P. Sharkis S J , Sensenbrenner LL. Two phases of engraftment established by serial bone marrow transplantation i n mice. Blood 73:397 (1989). 20. Jones R J , Wagner J E , Sharkis S J , Celano P. Separation of pluripotent hematopoietic stem cells (PHSC) from multipotential progenitors (CFU-S) [Abstract]. Blood 74(Suppl l):114a (1989). 21. Fraser C, Eaves C J , Szilvassy S, Humphries RK. Use of retroviral marking to demonstrate hemopoietic stem cells with lympho-myeloid repopulatlng ability i n long-term murine marrow cultures [Abstract]. Blood 74(Suppl l):113a (1989). 22. Barnett MJ, Eaves C J , Phillips GL, Kalousek DK, Klingemann H-G, Lansdorp PM, Reece DE, Shepherd JD, Shaw G J , Eaves AC. Successful autografting i n chronic myeloid leukaemia after maintenance of marrow In culture. Bone Marrow Transplant 4:345 (1989). 23. Sutherland H J , Eaves C J , Eaves AC, Dragowska W, Lansdorp PM. Characterization and partial purification of h u m a n marrow cells capable of initiating long-term hematopoiesis i n vitro. Blood 74:1563 (1989). 24. Sutherland H J , Lansdorp PM, Henkelman D, Eaves AC, Eaves C J . Functional characterization of individual h u m a n hematopoietic stem cells cultured at limiting dilution on supportive marrow stromal layers. Proc Natl Acad Sci USA : (1990). 25. Mellentln JD, Smith SD, Cleary ML. Lyl-1, a novel gene altered by chromosomal translocation i n T cell leukemia, codes for a protein with a helix-loop-helix DNA binding motif. Cell 58:77 (1989). 26. Mellentln JD, Murre C, Donlon TA, McCaw PS, Smith SD, Carroll A J , McDonald ME, Baltimore D, Cleary ML. The gene for enhancer binding proteins E12/E47 lies at the t(l;19) breakpoint i n acute leukemias. Science 246:379 (1989).  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100545/manifest

Comment

Related Items