G R O W T H P R O P E R T I E S AND G E N E T I C MANIPULATION O F MURINE HEMOPOIET IC S T E M C E L L S by R O B E R T J A M E S P A W L I U K B . S c . H . , The University of A lber ta , 1991 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Med ica l Gene t i cs P rog ramme W e accept t h i ^ h e s i s a s conforming to the required standard T H E U N I V E R S I T Y O F BRIT ISH C O L U M B I A Sep tember , 1996 © Robert J a m e s Pawl iuk, 1996 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 be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) A B S T R A C T T h e d e v e l o p m e n t of recomb inan t retroviral vec to r s ab le to t rans fe r exogenous genet ic material into hemopoiet ic target cel ls has p layed a pivotal role in our current understanding of hemopo ies is and has p layed a p ioneer ing role in the field of gene therapy. However , with the eff iciency of gene transfer to mur ine s tem cel ls only 1 5 % the power of recombinant retroviral gene transfer is currently severe ly compromised by the eff iciency of retroviral infection. To opt imize the utility of recombinant retroviruses, the human C D 2 4 cel l sur face ant igen w a s deve loped a s a dominant se lectab le marker in a retroviral vector to enab le the ident i f icat ion and se lec t ion of retrovirally t ransduced murine bone marrow cel ls , including those with long term in vivo repopulating ability. Fol lowing infection of day 4 5 - F U treated mur ine bone marrow ce l ls and select ion of retrovirally t ransduced ce l ls us ing an an t i -CD24 ant ibody and F luo rescence Act ivated C e l l Sort ing ( F A C S ) , funct ional ana lys is of se lec ted C D 2 4 + ce l ls demonstrated the p resence of hemopoiet ic ce l ls at var ious s tages of development, including in vitro c lonogen ic progenitors, day 12 C F U - S , a n d ce l l s with tot ipotent long- term repopu la t ing po tent ia l . Fu r the r exper iments demonst ra ted the ability to regenerate the hemopo ie t ic s y s t e m s of mye loab la ted recipient mice with cel ls der ived exc lus ive ly from proviral ly marked s tem ce l l s a n d that the t rans fer red C D 2 4 g e n e w a s e x p r e s s e d in va r i ous phenotyp ica l ly def ined popula t ions of ce l ls in v ivo inc luding mar row s tem cel l cand ida tes def ined by the Sca+L in " cel l sur face phenotype. T h u s , C D 2 4 can be uti l ized not only as a selectable marker but a lso as a means to track and phenotype t ransduced ce l ls and their progeny in vitro and in vivo. To provide information on the recovery of hemopoiet ic stem cel ls following bone marrow transplant, irradiated recipient mice were injected with var ious numbers of day 14.5 fetal l iver or day 4 5-F U adul t b o n e mar row es t imated to conta in 10, 100 or 1000 Compe t i t i ve Repopu la t ing Units ( C R U ) . Ana l ys i s of the femoral marrow of pr imary recip ients s h o w e d complete recovery of bone marrow ii cellulari ty and c lonogen ic progenitor content and a near full recovery of day 12 C F U - S n u m b e r s i r respec t i ve of the number or or ig in of the ce l l s ini t ial ly t ransplanted. Whi le the recovery of donor-cel l -der ived C R U w a s incomplete in all c a s e s , fetal l iver w a s marked ly super io r to those f rom adult b o n e mar row. Moreover , proviral integration ana lys is of mice receiv ing retroviral ly t r ansduced C D 2 4 + se lec ted bone marrow ce l l s prov ided ev i dence for a >300-fold c lona l ampl i f ica t ion of a s ing le t r ansduced s tem ce l l . T h e s e s tud ies have p rov ided p rocedures for the se lec t ion , t racking and phenotyping of mur ine bone marrow ce l l s , inc luding those with compet i t ive long term l ympho-mye lo id repopula t ing ability. T h e availabil i ty of such procedures should increase the power of retroviral m a r k i n g s t u d i e s , a n d be a d v a n t a g e o u s in s tud ies a i m e d at the gene t i c man ipu la t ion of hemopo ie t i c s tem ce l l s and their p rogeny, a s wel l a s in the deve lopment of vectors able to opt imize the exp ress ion of t ransferred g e n e s in spec i f i c target ce l ls of interest for use in human gene therapy tr ials. Moreover , t h e s e f ind ings set the s tage for at tempts to e n h a n c e hemopo ie t i c s tem cel l regenerat ion post- t ransplant by the administrat ion of e x o g e n o u s agents or the exp ress ion of intracellular factors that may enhance the regenerat ive potential of s tem cel ls . iii T A B L E O F C O N T E N T S P a g TITLE i ABSTRACT ii T A B L E OF CONTENTS iv LIST OF FIGURES vii LIST OF T A B L E S vii i ABBREVIATIONS ix ACKNOWLEDGMENT xi C H A P T E R 1. Introduction 1.1. The hemopoietic system 1 1.1.1. Overview of hemopoiesis 1 1.2. A s s a y s for early hemopoietic cells 3 1.2.1. In vitro clonogenic progenitors 3 1.2.2. Colony-forming units-spleen (CFU-S ) 5 1.2.3. The hemopoietic stem cell (HSC) 6 1.2.4. Phenotyping and purification of H S C s 9 1.3. Ontogeny of the murine hemopoietic system 13 1.3.1. Development of the hemopoietic system 13 1.3.2. Compar i son of s tem cel ls from fetal liver and adult t issues 14 1.4. Properties of hemopoietic stem cells 16 1.4.1. Cycl ing status 16 1.4.2. Developmental potential and dynamics of H S C s 16 1.4.3. Self-renewal and aging of H S C s 19 1.5. Regulation of hemopoiesis 2 3 1.5.1. Regulation by extracellular factors 2 3 1.5.1.1. Hemopoietic growth factors 2 3 1.5.1.2. The extracellular matrix 2 5 1.5.2. Regulation by intracellular factors 2 7 1.6. Gene t i c manipulat ion of hemopoiet ic cel ls using recombinant retroviruses 2 8 1.6.1. The lifecycle of retroviruses 2 9 1.6.2. Recombinant retroviruses as vectors for gene transduct ion.. 3 3 1.6.2.1. Product ion of helper-free repl ication-defective retroviruses 34 1.6.3. Optimization of retroviral gene transfer 3 6 1.6.3.1. Retroviral infection strategies 3 6 1.6.3.2. Targeting of virions to host cel ls 3 8 1.6.3.3. U s e of selectable markers to increase the utility of recombinant retroviral vectors 4 0 1.6.4. Retroviral vector design 4 0 1.7. Thes is objectives and general strategy 4 4 C H A P T E R 2. Materials and Methods 2.1 . Construct ion of retroviral vectors, virus production and viral assays . . . 4 6 2.1.1. Recombinant retroviral vectors 4 6 2.1.2. Viral packaging and other cell lines 4 7 2.1.3. Generat ion of viral producer cell l ines 4 7 2.1.4. Viral titering and helper virus assay 4 8 iv 2.2. Hemopoiet ic cell culture and assays 4 8 2.2.1. Mice 4 9 2.2.2. Viral infection of bone marrow cell and cell l ines 4 9 2.2.3. In vitro clonogenic progenitor assay 5 0 2.2.4. C F U - S assay 50 2.2.5. Bone marrow transplantation and quantitation of competitive repopulating units (CRU) 5 0 2.3. Molecular analysis 52 2.3.1. Southern blot analysis 52 2.3.2. Antibody staining procedures 5 3 2.3.3. F luorescence activated cell sorting ( F A C S ) 5 5 C H A P T E R 3. Selection of retrovirally transduced hemopoietic cells using CD24 as a marker of gene transfer 3.1. Introduction 57 3.2. Results 58 3.2.1. The C D 2 4 viral vector 58 3.2.2. F A C S select ion of CD24- t ransduced in vitro c lonogenic progenitors and C F U - s p l e e n (CFU-S) 60 3.2.3. Select ion by F A C S of CD24-virus- infected C R U 61 3.3. Discussion 69 C H A P T E R 4. High level reconstitution with preselected hemopoietic cells expressing a transduced gene encoding a cell surface antigen 4.1. Introduction 7 6 4.2. Results 7 7 4.2.1. Viral vectors and experimental design 7 7 4.2.2. The majority of donor-der ived cel ls in recipients contain intact provirus 7 8 4.2.3. Efficient gene transfer to and express ion of C D 2 4 among Sca+Lin" bone marrow stem cell candidates.. . . 8 5 4.3. Discussion 89 C H A P T E R 5. Evidence of both ontogeny and transplant dose regulated expansion of hemopoietic stem cell in vivo 5.1. Introduction 9 3 5.2. Results 94 5.2.1. Overal l experimental design 94 5.2.2. Kinet ics of reconstitution of the terminal compartments 94 5.2.3. Reconstitution of the marrow 9 5 5.2.4. Reconstitution of the marrow C R U compartment 9 7 5.2.5. Regenerat ive ability of a single C R U a s s e s s e d using retroviral marking 99 5.3. Discussion 101 C H A P T E R 6. Discussion 107 C H A P T E R 7 \ References 114 v LIST O F F IGURES C H A P T E R 1 Figure 1.1 Schemat ic representation of the hemopoiet ic hierarchy 2 Figure 1.2. Schemat i c representat ion of the competi t ive repopulat ing unit (CRU) assay. 10 Figure 1.3. Var ious organs in which hemopoies is takes p lace during ontological development 15 Figure 1.4. Schemat ic representat ion of a model to explain H S C behavior following bone marrow transplant 19 Figure 1.5. Schemat ic representation of a typical retrovirus 32 Figure 1.6. Critical features of the retroviral lifecycle 3 3 Figure 1.7. S teps involved in the production of recombinant retrovirus.. 3 5 C H A P T E R 3. Figure 3 .1 . Schemat ic of the JZenCD24 tkneo provirus 59 Figure 3.2. F low cytometric analys is of C D 2 4 express ion of B a / F 3 cel ls infected with the JZenCD24 tkneo retrovirus 60 Figure 3.3. F A C S select ion of C D 2 4 virus-infected in vitro c lonogenic progenitors and day 12 C F U - S 62 Figure 3.4. Select ion of C D 2 4 virus-infected C R U by F A C S 6 5 Figure 3.5. Hemopoiet ic reconstitution from C D 2 4 retrovirus-infected competi t ive repopulating cel ls as a s s e s s e d by Southern blot 6 8 Figure 3.6. Proport ion of recipient mice express ing the transferred C D 2 4 gene at early and late time points posttransplant as assessed by F A C S 71 Figure 3.7. F low cytometric analys is of C D 2 4 express ion in the hemopoiet ic t issues of one repopulated mouse 4 months post transplantation 7 2 C H A P T E R 4. Figure 4 .1 . Viral vectors used and F A C S select ion of retrovirally transduced bone marrow cells 79 Figure 4.2. Detect ion of high levels of intact provirus in recipient mice by Southern blot analysis 8 3 Figure 4.3. A s s e s s m e n t of proviral integration in primary or secondary recipients and day 12 sp leen co lon ies by Southern blot analysis 8 6 Figure 4.4. Express ion of the transferred C D 2 4 gene in primary marrow stem cell candidates as defined by the Sca+L in" cel l sur face phenotype 8 7 Figure 4 .5 . Proport ion of s tem cel l candidates def ined by the Sca+L in " cel l sur face phenotype posit ive for C D 2 4 express ion at 24 weeks post transplant 8 8 C H A P T E R 5. Figure 5.1. Regenerat ion of Ly5.1 donor-der ived cel ls fol lowing the transplantation into secondary recipients 9 8 Figure 5.2. Demonstrat ion of C D 2 4 provirus in bone marrow (B), sp leen (S), and thymus (T) D N A of primary and secondary transplant recipients 101 vi LIST O F T A B L E S C H A P T E R 1 Tab le 1.1. Opt ions in selectable/reporter markers for retroviral vectors 41 C H A P T E R 3 Tab le 3.1. Proviral integration and C D 2 4 express ion on cel ls from individual sp leen co lon ies der ived from sorted and unsorted bone marrow cel ls following C D 2 4 virus infection 6 3 Tab le 3.2. Proviral integration and C D 2 4 express ion on cel ls from competit ively repopulated mice a s s e s s e d 5 weeks post transplant 6 7 C H A P T E R 4 Tab le 4 .1 . F low cytometric analys is of C D 2 4 express ion in var ious hemopoiet ic t issues in recipient mice 24 weeks posttransplant 81 C H A P T E R 5 Tab le 5 .1 . Proport ion of Ly5.1 + peripheral blood cel ls from primary recipients 9 5 Tab le 5.2. Regenerat ion of cellularity, C F C and day 12 C F U - S populat ions in the femurs of primary recipient mice 8 months post transplant 9 6 Tab le 5.3. Expans ion of donor-der ived C R U in primary recipients of fetal liver or adult bone marrow cells 99 vii LIST O F ABBREVIAT IONS 7 A A D 7-amino act inomycin D A D A adenos ine d e a m i n a s e A G M aortic gonada l mesenephros region A I D S acqui red immune def ic iency syndrome A T T C Amer i can Type Culture Col lect ion B F U - E burst forming unit-erythroid B F U - M k burst forming uni t-megakaryocyte B M bone marrow bp base pair B S A bovine serum albumin C A F C cobble stone a rea forming cell c D N A complementary deoxyr ibonucle ic ac id C D cluster designat ion C F U colony forming unit C F U - E co lony forming unit-erythroid C F U - G co lony forming unit-granulocyte C F U - G E M M colony forming unit-granulocyte-erythroid-monocyte-megakaryocy te C F U - G M co lony forming uni t -granulocyte-macrophage C F U - M co lony forming uni t -macrophage C F U - S co lony forming unit-spleen c G y cent iGray C R U competi t ive repopulat ing unit C s ces ium C S calf serum d C T P deoxycyt id ine tr iphosphate D M E M Du lbecco 's Modif ied Eag les Med ium D N A deoxyr ibonuc le ic ac id E C embryonic ca rc inoma cel l E D T A ethylenediaminetetraacet ic ac id E p o erythropoiet in E S embryonic stem cel l F A C S f luorescence act ivated cell sorter F C S fetal calf serum FITC f luorescein- isoth iocyanate F L flk2/flt3 l igand 5 - F U 5-f luorouraci l G 4 1 8 genet ic in G - C S F granulocyte colony stimulating factor G F P green f luorescence protein G M - C S F granulocyte-macrophage colony stimulating factor G 6 P D g lucose 6 phosphate dehydrogenase G p i g lucose phosphate i somerase H B S hepes buffered solution H F Hank 's ba lanced salt solution/2%fetal calf serum H P P - C F C high proliferative potential co lony forming cel l H S A heat stable antigen H S C hemopoiet ic stem cel l viii H X M hypoxanth ine-xanth ine-mycophenol ic ac id ig immunog lobu l in IL inter leukin I R E S internal r ibosomal entry site L F A leukocyte function associat ion LIF leukemia inhibitory factor L T C long term culture L T C - I C long term culture-initiating cel l L T R long terminal repeat M - C S F macrophage colony stimulating factor M D R - 1 multi-drug resistance-1 protein M E S V murine embryonic sa r coma virus M H C major histocompatibi l i ty complex M I P - 1 a macrophage inhibitory p ro te in -1a M o A b monoc lona l ant ibody M o M u L V Moloney murine leukemia virus M P S V myeloprol i ferative s a r c o m a virus M R A - C F U - •S marrow repopulating abil i ty-colony forming unit-spleen M S C V murine stem cel l virus N C S newborn calf serum N H 4 C I ammon ium chlor ide O R F open reading frame P phospha te P B per ipheral b lood P C M V P C C 4 embryonal ca rc inoma ce l l -passaged myeloprol i ferat ive s a r c o m a virus P C R po lymerase chain reaction P G K phosphog lycera te k inase PI propidium iodide R B C red blood cel l Rh rhodamine R N A r ibonucleic ac id R - P E R-phycoerythr in RPMI Roswel l Park Memor ia l Institute RT reverse transcription R U repopulat ing unit S C C M sp leen cel l condi t ioned medium S D s tandard deviat ion S D S sod ium dodecy l sulfate S E M standard error of the mean S S C sod ium chlor ide sod ium citrate T E t r i s -EDTA T G F - p transforming growth factor-p U units V C A M vascu la r cel lular adhes ion molecule V L A very late antigen V S V - G vesicular stomatitis virus G glycoprotein W White spotting mutation W B C white b lood cel l W G A wheat germ agglutinin ix A C K N O W L E D G M E N T S I would like to thank and express my gratitude: to my superv isor Dr. R. Keith Humphr ies for the opportunity to do graduate training at the Terry Fox Laboratory and for his enthusiast ic support and t i reless gu idance throughout this project. to Drs. Conn ie J . E a v e s and Peter Lansdorp for their col laborat ive efforts and many stimulating d iscuss ions . I would a lso like to thank Dr. E a v e s for her invaluable contributions to writing the manuscr ipts descr ibed in this thesis. to the many members of the Humphr ies laboratory for providing an invigorating scientif ic environment in which to study. to Patty Ros ten , Gay le Thornbury, V i s ia Dragowska , M a y a St -Cla i r and F red J e n s o n for expert technical ass is tance. to Drs. Dixie Mager , Muriel Harris and R o s s MacGi l l iv ray for serv ing on my graduate commit tee. to the Medica l Resea rch Counc i l of C a n a d a for f inancial support. to my parents for their years of support and understanding, to Heather Murray for bel ieving in me and to the Dragon Hags for making life fun. x C H A P T E R 1 I N T R O D U C T I O N 1.1 The hemopoietic system 1.1.1 Overview of hemopoiesis H e m o p o i e s i s is the essent ia l , l i felong p r o c e s s whereby mult iple t ypes of h igh ly s p e c i a l i z e d b lood ce l l s a re g e n e r a t e d . T h e s e c e l l s i n c l u d e t h o s e responsib le for carrying out speci f ic funct ions s u c h a s carbon d iox ide and oxygen transport (erythrocytes), b lood clotting (platelets), humora l (B lymphocytes) and cel lu lar (T lymphocytes) immunity as wel l as mount ing phagocy t ic r e s p o n s e s to foreign o rgan isms and their products (g ranu locy tes /monocy tes /macrophages) . In the normal human adult it is est imated that approximately 200 bil l ion erythrocytes (1) and 60 bi l l ion neutrophi l ic l eukocy tes (2) are p r o d u c e d e v e r y d a y . T h i s observat ion has st imulated a great dea l of interest in the ce l ls that are ult imately respons ib le for accommoda t ing this eno rmous dai ly output of ce l l s , and in the m e c h a n i s m s that regulate this p rocess . T h e cel l types ment ioned above c a n be functional ly d iv ided into two distinct groups termed myelo id and lymphoid (Figure 1.1). Dur ing normal adult life myelo id ce l ls are p roduced exc lus ive ly within the bone marrow (3) whi le ce l ls of the lymphoid l i neages are p roduced to vary ing degrees in the bone marrow, sp leen , thymus and lymph nodes . Mature funct ional end ce l ls and their immediate p recursors have a l imited l i fespan and a l imited proliferative capaci ty and hence are not sel f-maintaining. Thus , these ce l ls must be cont inuously replaced from a pool of more primitive proliferating ce l ls . T h e s e ce l ls const i tute a h ierarchy of ce l ls with inc reas ing prol i ferat ive potent ial and w ider differentiative capaci t ies. 1 Figure 1.1. Schemat ic representat ion of the organizat ion of the hemopoiet ic sys tem and s o m e of the a s s a y s used to evaluate H S C s and var ious poo ls of progenitor ce l ls ; H S C , hemopoiet ic s tem cel l ; C R U , competi t ive repopulat ing unit a s s a y ; R U , repopulat ing unit assay ; L T C - I C , ce l ls with the ability to initiate in vitro long-term cul tures; C A F C , in vitro a s s a y for cobb les tone forming a rea ce l ls ; day 12 or day 8 C F U - S , colony-forming-unit sp leen cel ls able to produce sp len ic nodu les 12 or 8 d a y s post inoculat ion respect ive ly ; B last co l . , in vitro blast co lony a s s a y . T h e relat ive potential for proli feration and se l f - renewal of var ious g roups of ce l l s is shown to the right. Ult imately, all ce l ls of both the myeloid and lymphoid l ineages are der ived from cel ls referred to as totipotent stem cel ls which are est imated to compr ise only 0 . 0 1 % of the total marrow compartment. T h e s e cel ls are operat ional ly def ined by their capaci ty to regenerate and sustain both the myeloid and lymphoid a rms of the hemopo ie t i c sys tem for long per iods of t ime fol lowing t ransplant and by their extens ive capaci ty for sel f - renewal , the p rocess of cel lular division result ing in the product ion of daughter ce l l s wh ich are funct ional ly ind is t ingu ishab le f rom the parent ce l l s in te rms of their prol i ferat ive and dif ferent iat ive potent ia l . M a n y important ques t ions remain unanswered regarding the nature and regulat ion of 2 totipotent hemopoie t ic s tem ce l ls . T h e s e include a bas ic unders tand ing of their numbers , b io logical potential, usage over t ime and the g e n e s encod ing extr insic and/or intrinsic factors which are responsib le for the regulation of these biological charac ter is t i cs . Moreover , with an increas ing e m p h a s i s on the deve lopmen t of c l in ica l s t ra teg ies that d e p e n d upon the regenera t ion of H S C n u m b e r s (eg. autograft purging and the genetic therapy of inherited hematologica l d isorders) the a s s e s s m e n t of the sel f -renewal potential of H S C s is of particular interest. T h e ability to manipulate H S C s genet ical ly has provided a powerful tool to begin to add ress the above i ssues . The overal l goal of the work presented in this thes is w a s to deve lop methodologies to increase the utility of gene transfer for the efficient genet ic manipulat ion and tracking of H S C s and to utilize these procedures to def ine more clear ly the sel f - renewal potential of H S C s fol lowing bone marrow transplant. The fol lowing introduction examines the current state of knowledge of the phenotypic and functional propert ies of H S C s as well as the methods avai lable for their genet ic manipulat ion. 1.2. A s s a y s for early hemopoietic cells. 1.2.1. In vitro c lonogenic progenitors T h e ability to culture hemopoiet ic cel ls in vitro has provided a large amount of informat ion on the ce l lu lar organ iza t ion , the prol i ferat ive and dif ferent iat ive potential, and growth factor requirements of cel ls at var ious s tages of hemopoiet ic deve lopment . A s s a y s first descr ibed by Bradley and Metcalf (4), and P luzn ik and S a c h s (5) involved the growth of hemopoiet ic ce l ls in a semi -so l id matrix of agar that a l lowed co lon ies of hemopoiet ic cel ls der ived from single ce l ls to be identif ied and charac te r i zed . Subsequen t a s s a y s for such "c lonogen ic progeni tors" ut i l ized p l a s m a clots or, now most often, methy lce l lu lose to provide semi -so l id med ium. S u c h growth med ium is typical ly supp lemented with nutrients and growth factors required by the dividing cel ls and which have been provided historical ly in part by poorly def ined "condi t ioned" cel l med ium. Today a large number of hemopoie t ic 3 growth fac tors have been pur i f ied and their g e n e s c l o n e d resul t ing in the avai labi l i ty of pure recombinant factors ( reviewed in (6)). T h e vas t majority of c lonogen ic progenitors detectable in a s s a y s of normal bone marrow are of uni- or bipotent potential that are able to give rise to co lon ies cons is t ing of granu locytes a n d m o n o c y t e s / m a c r o p h a g e s (co lony- forming units g r a n u l o c y t e - m a c r o p h a g e ; C F U - G M ) (7), pure granulocytes or monocy tes /macrophages ( C F U - G or C F U - M ) (4, 5), erythrocytes ( B F U - E and C F U - E ) (8), and megakaryocy tes ( B F U - M k ) (9). More recently a s s a y s for cel ls with B, but not T, lymphoid potential have been descr ibed (10). S u c h ce l ls , whi le somet imes p o s s e s s i n g cons iderab le prol i ferative potential y ie ld ing co lon ies of severa l thousands of cel ls , have limited or no capaci ty for self-renewal a s determined by their inability to generate equivalent secondary co lon ies in subsequent replatings (11). T h e s e cel ls a lso appear to be act ively cyc l ing under normal steady-state condit ions in vivo s ince they are highly suscept ib le to killing by cyc le -spec i f i c cytotoxic drugs such as 5-fluorouracil (5-FU) (12). S u c h ce l ls , then, are bel ieved to be relatively late in the hemopoiet ic hierarchy. S u c h in vitro a s s a y s a lso enab le the identi f icat ion of ear l ier progeni tors character ized by their ability to produce colonies consist ing of multiple l ineages (ie. granulocyte / erythrocyte / mac rophage / megakaryocy te from C F U - G E M M ) (13, 14), of great s i ze (eg. high proliferative potential colony- forming ce l ls , H P P - C F C ) (15), and/or ce l ls with a primitive undifferentiated cel lu lar morpho logy (ie. blast co lony forming cel ls) (16-18). C F U - G E M M and blast co lony forming ce l ls are a lso character ized by a capaci ty for sel f-renewal as demonstrated by the ability of s o m e of their c lonal progeny to form secondary and less frequently tertiary mult i - l ineage and blast co lon ies in replate a s s a y s (16, 19). 1.2.2. Colony- forming units-spleen (CFU-S) The first a s s a y avai lable to study hemopoiet ic cel ls shown to p o s s e s s "stem cel l - l ike" propert ies w a s the in v ivo sp leen co lony a s s a y desc r i bed by Til l and M c C u l l o c h in 1961 (20). Th is a s s a y is based on the ability of certain ce l ls ( C F U - S , 4 for c o l o n y fo rming un i t -sp leen) to h o m e to the s p l e e n a n d g row to form m a c r o s c o p i c hemopoiet ic nodu les detectable on the sur face of the sp leen 8-12 days post-transplant. The origin of individual nodules from a s ingle cel l (ie. c lonal in origin) w a s first determined by injecting bone marrow ce l ls from mice harbor ing unique radiat ion- induced ch romosoma l abnormal i t ies (21, 22) , and this w a s later conf i rmed using unique retroviral integration events as markers (23, 24). Individual sp leen co lon ies detected on day 8 post- t ransplant usual ly are restr icted in the types of ce l ls they contain (either erythroid or granulocyt ic but not both (25, 26)) and rarely p roduce daughter co lon ies upon subsequen t ret ransplantat ion (11). T h u s , day 8 C F U - S resemb le uni-potent ial in vitro c lonogen ic progeni tors . In addi t ion, a large proport ion of day 8 C F U - S are a lso sens i t ive to kil l ing by the cyc le-act ive drug 5 - F U (27). However , those C F U - S which result in large co lon ies detected on day 12 are often c o m p o s e d of cel ls of mul t i - l ineages (20), are more resistant to killing by 5 - F U and are often able to generate numerous daughter C F U -S (28). Whether these cel ls p o s s e s s the potential to produce ce l ls of the lymphoid l ineages remains controversial (29-31). A l though sp leen colony- forming ce l ls were once thought to const i tute the most primitive hemopoiet ic compartment s ince they p o s s e s s a large capac i ty for prol i ferat ion, are multi-potential and have signif icant se l f - renewal ability, recent ly the use of counterf low centr i fugal elutriation has been used to show that day 12 C F U - S can be physical ly separa ted from cel ls with long term repopulat ing ability (32). 1.2.3. The hemopoietic stem cell It h a s b e e n p r o p o s e d that the most use fu l , r igorous def in i t ion of a hemopoiet ic stem cel l should be based on such a cell having an in vivo capaci ty for the long term production of all b lood cel l l ineages (33). Ev idence for the ex is tence of ce l ls with such character is t ics has primarily c o m e from transplantat ion mode ls . For example , W u et. a l . (22) detected common ch romosoma l abnormal i t ies in both 5 the myelo id and lymphoid compar tments of myeloabla ted recipient mice that had rece ived a transplant of marrow cel ls from donor mice harbor ing unique radiation i n d u c e d c h r o m o s o m a l abnorma l i t i es . S u b s e q u e n t e v i d e n c e s u g g e s t i n g the ex is tence of totipotent H S C s w a s provided by Nakano et. a l . (34) who per formed m a r r o w t r ansp lan t s b e t w e e n c o n g e n i c s t ra ins of m i c e w h i c h p o s s e s s e d d is t ingu ishable hemoglob in and i soenzyme markers . Moreover , the t ransplant of retrovirally marked adult day 4 5 - F U bone marrow (35) or fetal liver ce l ls (36) into irradiated (35) or genet ical ly anemic 1 W / W V (23) mice has a lso demonst ra ted the ex is tence of l ympho-myelo id repopulat ing s tem ce l ls b a s e d on the detect ion of c o m m o n prov i ra l in tegrants in ce l l s of both the l y m p h o i d a n d m y e l o i d compar tments in the recipient mice. A s desc r ibed in the s u b s e q u e n t sec t ion , hemopo ie t i c s tem ce l ls cannot yet be posit ively identi f ied on the bas i s of any un ique morpho log ica l , phys ica l or cel l su r face charac te r i s t i cs . T h u s , r igorous identif ication of totipotent hemopoiet ic s tem cel ls rel ies on functional a s s a y s b a s e d on the ability of such cel ls to regenerate and susta in the hemopoiet ic sys tems of m y e l o a b l a t e d or gene t i ca l l y a n e m i c ( ie. W / W v ) rec ip ien ts . A var ie ty of t ransplantat ion strategies have been des igned in an effort to quantify hemopoiet ic s tem ce l l s der ived from var ious cel l populat ions. O n e s u c h a p p r o a c h invo lves a s s e s s i n g the su rv iva l of i r rad ia ted rec ip ient m i ce for 30 d a y s fo l low ing t ransplantat ion with limiting numbers of marrow ce l ls (39-41). However , s u c h an a s s a y is compl icated by the potential contribution from residual recipient ce l ls , and the possib i l i t ies that such short term radioprotect ive capac i ty may in part der ive f rom more mature cel l t ypes (42, 43). Moreover , the t ransplant of insuff ic ient numbers of cel ls with short-term radioprotective capaci ty can result in the death of the recipient before cel ls with long term repopulating capaci ty can be read out. 1The W locus, located on chromosome 5, encodes the receptor for Steel factor which is a hemopoietic growth factor shown to support the proliferation of both immature and lineage restricted lymphoid and myeloid progenitor cells in combination with other hemopoietic growth factors (37, 38). Mice harboring mutations at the W locus are characterized by an intrinsic defect in primitive hemopoietic stem cells and thus, the hemopoietic systems of these mice can be replaced with normal wild type hemopoietic cells without the use of conditioning regimens such as irradiation. 6 A s e c o n d , very powerful approach to the quantif ication of hemopoiet ic s tem ce l ls has been the use of the compet i t ive repopulat ion a s s a y first deve loped by Harr ison (44, 45). Th is a s s a y is based upon compar ing the long-term repopulat ing abil i t ies of two separate populat ions of hemopoiet ic cel ls which are d ist inguishable on the bas is of al lel ic d i f ferences in hemoglob in and Gpi-1 i s o e n z y m e markers . O n e of the cel l populat ions (the "competitor") cons is ts of a f ixed number (usually 1-2 x 1fj6 cel ls) of fresh marrow that serves as a standard for repopulat ing potential. Vary ing numbers of a "donor" or "test" source of s tem cel ls are then injected and the m e a n relative contribution of the two populat ions to hemopo ies i s m e a s u r e d . Repopulat ing units (RU) are calculated accord ing to the formula RU=%(C)/ (100-%), where % is the measured percentage of peripheral blood cel ls in the recipient with the donor pheno type , and C is the number of f resh compet i to r mar row ce l l s u s e d / 1 0 5 . E a c h repopulat ing unit represents the repopulat ing ability shown by 1 x 1 0 5 f resh marrow ce l ls from the compet i tor pool . Th is method has a number of advan tages over 30 day surv ival a s s a y s . Rec ip ient m ice are a n a l y z e d at long pe r iods post t ransplant prov id ing a r igorous m e a s u r e of s tem ce l l func t ion . Moreover , the short term hematological rescue of mye loab la ted recipient mice is not dependent upon the cel l populat ion being tested s ince life spar ing d o s e s of donor ce l ls are provided in the compet i tor cel l populat ion. However , this method cannot be used to a s s e s s propert ies of individual H S C s such a s their proliferative capac i ty s ince one cannot dist inguish between differing proliferative capac i t ies and differing numbers of H S C s . T h e Compet i t i ve Repopu la t ing Unit a s s a y deve loped by S z i l v a s s y et. a l . c o m b i n e s l imiting dilution and compet i t ive repopulat ing p rocedu res to quant i fy totipotent repopulat ing s tem cel ls . In its original form, the C R U a s s a y involves the co- inject ion of limiting numbers of male test cel ls a long with a f ixed number (2 x 1 0 5 ) of fema le compet i tor ce l l s into mye loab la ted fema le recipient m ice . T h e compet i tor cel ls are der ived from mice which have undergone two previous rounds 7 of hemopoiet ic transplantat ion and thus, are relatively depleted of ce l ls with long-term repopulat ing capabi l i ty . T h e funct ion of the he lper ce l l s is to ensu re the survival of the myeloablated recipient even when limiting numbers of test ce l ls are t ransp lanted. Rec ip ien ts demonstrat ing repopulat ion by donor H S C s are def ined a s those with a signi f icant contr ibut ion (ie. > 5%) from male ce l l s to both the lymphoid a n d myelo id compar tments as identified by probing D N A extracted from hemopo ie t i c t i ssues with Y c h r o m o s o m e speci f ic s e q u e n c e s . T h e f requency of hemopo ie t i c s tem ce l l s , or compet i t ive repopulat ing units ( C R U ) a s they are referred to in this a s s a y , is determined by a s s e s s i n g the proport ion of recipient mice which meet the repopulat ion criteria descr ibed above when limiting dilution is reached. The f requency of C R U in a test cel l population is subsequent ly ca lcu la ted using P o i s s o n statistics accord ing to: C R U frequency = 1 / number of marrow "test" ce l ls that result in 3 7 % of recipient mice being negative for repopulat ion. The C R U a s s a y has s i nce been modi f ied by the use of normal mar row ce l l s (10^) a s compet i tor cel ls (46) and the use of mouse strains that enab le the identif ication of donor ve rsus recipient cel ls on the bas is of allelic di f ferences at a gene encod ing a peripheral b lood leukocyte cel l surface antigen (Ly5.1 or Ly5.2) (46-48) (see Figure 1.2). T h e ability of this a s s a y to detect totipotent ( lympho-myeloid) repopulat ing s tem cel ls has been demonstrated by the observat ion of c o m m o n reconstitution of l ympho id and mye lo id t i s sues by l imiting numbers of ce l l s , and by retroviral marking procedures (43, 49). O n e of the great advantages of the C R U a s s a y is that it can be used to quantify quickly and accurately the f requency of hemopoiet ic s tem cel ls with long term lympho-myelo id reconstituting ability in any populat ion of test ce l ls (43). S z i l v a s s y and his co l leagues have demonstrated that C R U f requenc ies determined at 10 weeks post transplant were virtually identical to those obtained at later t ime points sugges t ing that this a s s a y c a n be u s e d to obta in accu ra te est imat ions of totipotent H S C numbers in little over two months t ime. 1.2.4. Phenotyping and purification of H S C s 8 T h e f r e q u e n c y of r epopu la t i ng H S C s a m o n g adu l t b o n e m a r r o w mononuc lear cel ls has been est imated to be 1 in 1 0 4 to 1 0 5 (43, 50-53) . Th is low f requency and the fact that the identification of H S C s involves functional a s s a y s (ie. are retrospect ive in nature based upon the long term repopulat ion of irradiated or genet ical ly anemic recipient mice) have hampered direct ana lys is of se l f - renewal and early events assoc ia ted with H S C commitment and differentiation. Major effort has thus been directed towards purification of cel ls with s tem cel l character is t ics . P h y s i c a l , immunologica l and supravital stains have all been emp loyed a lone or in comb ina t i on a s s t ra teg ies to enr ich for H S C s . P h y s i c a l t echn iques (veloci ty sedimentat ion (54), density gradient separat ion (55-57) and counterf low centri fugal elutriation (32, 58)) separate cel ls on the bas is of buoyant densi ty and/or cel l s i ze . Us ing these techn iques , H S C s have been general ly charac te r i zed a s relat ively smal l , low density cel ls with an undifferentiated blast cell- l ike morphology. With the advent of flow cytometry, the goal of isolating purified populat ions of H S C s took a major leap forward. H S C s p o s s e s s medium forward light scatter and low to med ium orthogonal light scatter propert ies indicat ive of intermediate s i ze and low granulari ty respect ively (59). The deve lopment of monoc lona l ant ibodies (MoAb) (60) and the ex is tence of lectins which bind distinct sugar moiet ies on the cel l su r face (ie. Whea t G e r m Agglut in in; W G A ) has enab led the enr ichment of purif ied s tem cel ls cand idates on the bas is of two distinct strategies; posi t ive and negat ive se lect ion procedures. Posi t ive select ion procedures are b a s e d upon the identif ication and se lect ion of ce l ls which exp ress spec i f ic cel l su r face ant igens and/or lect ins. M o A b s di rected against L y - 6 A / E (Sca-1 ) , M H C - c l a s s I mo lecu les (31, 48 , 53 , 59, 61-64), W G A (53, 65-67), Thy-1 and the c-kit receptor have been widely used for the enr ichment of s tem cel ls from adult bone marrow. In addi t ion, 9 neg. test control sample Q . Figure 1.2. S c h e m a t i c representat ion of the l imit ing di lut ion a s s a y for C R U . Limit ing numbers of Ly5.1 + test cel ls der ived from P e p C 3 F 1 donor mice are co -injected along with 1 0 5 Ly5.2 helper cel ls into B 6 C 3 F 1 (Ly5.2) recipients fol lowing 9 5 0 c G y of who le body irradiat ion. Tes t ce l ls are phenotyp ica l ly d is t ingu ishab le from both he lper and surv iv ing e n d o g e n o u s host ce l l s on the b a s i s of a l le l ic di f ferences at the Ly5 locus. The proportion of mice showing > 1% L y 5 . 1 + ce l ls of both lymphoid and myelo id l ineages is determined by sta in ing per iphera l b lood ce l ls with an ant ibody speci f ical ly recogniz ing the Ly5.1 ant igen and ana lys i s by F A C S . The f requency of C R U is then ca lcu la ted us ing P o i s s o n stat ist ics. N B M : normal bone marrow ce l ls obta ined from a unmanipu la ted control m o u s e . S S C : s ide scatter. 10 M o A b AA4.1 has been used for the isolation of H S C s from fetal liver and the yolk s a c of the early mouse embryo (63, 64). Al ternat ively, or in addit ion, negat ive select ion p rocedures can be used to remove ce l ls which are not of interest. H S C s der ived from adult bone marrow have been shown not to exp ress a number of markers character is t ic of later l ineage restr icted ce l l s (68). T h u s , marrow ce l ls exp ress ing s o ca l led l ineage or "L in " markers can be identified and removed using M o A b s recogniz ing B 2 2 0 (B-cel ls) , C D 4 , C D 8 , C D 3 and C D 5 (T-cells), Mac-1 and 15-1.1 (monomyelocyt ic cel ls) , Gr-1 (myeloid cel ls) , and Ter119 and 10-2.2 (erythroid cel ls) (31, 53 , 64, 67-74) . T h e enr ichment of s tem cel ls ach ieved var ies to certain degrees depend ing upon which sets of M o A b s are emp loyed , but general ly, the se lect ion of ce l ls with either the S c a - 1 + L i n - T h y 1 . 1 > o (31, 48 , 75) or S c a - 1 + L i n " W G A + (53, 67) ce l l su r face phenotype have resulted in enr ichment factors of approximately 500 to 1000-fo ld. T h e s e enr ichment factors result in detectable H S C f requencies of 1 in 10 (76) to 1 in 30 (46) s tem cel l cand ida tes . Al though it would s e e m that H S C s have not yet been purif ied to homogenei ty, one must cons ider the possibi l i ty that the long-term repopulat ing a s s a y has a lower limit to the number of t ransplanted H S C s that will reproducibly result in donor reconstitut ion. Th is limit may result from the seed ing eff ic iency of H S C s to the marrow, compet i t ion from H S C s in the compet i tor cel l populat ion or surviv ing endogenous H S C s , or ill def ined regulatory m e c h a n i s m s that direct test H S C s into qu iescence or to differentiate rather than sel f - renew upon arrival in the marrow compartment. Thus , the enr ichment factor stated above may be an underest imat ion of H S C f requencies in purified subpopulat ions. Differential retention of certain f luorescent dyes such a s R h o d a m i n e 123 (Rh123) and Hoechst 33342 has a lso been employed to separate H S C s from more mature progenitors. The retention of these dyes by H S C s tends to be poor (41, 77-80); in the c a s e of R h 1 2 3 this is due to the functioning of a P-glycoprote in pump which actively removes Rh123 from the cell (81). The use of such dyes can enab le 11 an approx imate 500-fold enr ichment of s tem cel l cand ida tes . Interestingly, s o m e phenotyp ic d i f ferences between H S C s der ived from differing onto logica l s o u r c e s have been reported. For examp le , S c a + L i n " subpopu la t ions enr i ched for H S C s der ived from day 14.5 fetal liver stain with the A A 4 . 1 , ant i -Mac, and a n t i - C D 4 5 R B ant ibod ies and retain R h o d a m i n e 123 ( R h 1 2 3 b r ' 9 n t ) whi le adult bone marrow S c a + L i n " ce l l s simi lar ly enr iched for H S C s are A A 4 . 1 " , M a c - 1 " , C D 4 5 R B " and Rh123du l l . D e s p i t e the fact that imp ress i ve a d v a n c e s in the ident i f ica t ion a n d pur i f icat ion of H S C s has been m a d e over the past y e a r s , a su r face marke r spec i f ica l ly e x p r e s s e d on H S C s has not been reported. Moreover , a number of g roups have s h o w n that ce l l su r face pheno type is not a l w a y s an a c c u r a t e predict ion of biological capabi l i ty. Rebe l et. a l . demonst ra ted that a l though the in vitro cul ture of S c a + L i n " W G A + ce l l s for four w e e k s in s e r u m f ree m e d i u m supplemented^/vi th steel factor, IL-3, IL-6 and E p o resulted in >1000-fold inc rease in the number of ce l l s with the start ing su r face pheno type , the n u m b e r of repopulat ing H S C s in these cul tures was 1.3-fold lower than input va lues (46). In addit ion, Spangrude et. a l . demonstrated that fol lowing transplantat ion of recipient mice with 200 T h y - 1 . 1 l o w L i n " S c a + R h - 1 2 3 l o w marrow cel ls , ce l ls with this sur face phenotype were expanded to approximately 1000-fold over input leve ls . Desp i te th is c o n s i d e r a b l e i n c r e a s e in the number of ce l l s with the star t ing su r f ace pheno t ype , t h e s e ce l l s p o s s e s s e d poor reconst i tu t ing abi l i ty; on ly 8 of 8 3 seconda ry transplant recipients (9.6%) receiv ing 5, 10, or 20 R h - 1 2 3 l o w ce l ls or 4 0 0 Thy-1 . l ' 0 W L i n n e 9 L y - 6 A / E + ce l ls obta ined from primary recip ients exhib i ted donor -der i ved ce l l s in per iphera l b lood 12 w e e k s post t ransplant (76). T h u s , functional a s s a y s remain the most reliable and rigorous method to a s s e s s for H S C numbers . 1.3. Ontogeny of the murine hemopoietic system 12 1.3.1. Development of the hemopoietic system During vertebrate ontogeny, hemopo ies is is in a state of cont inuous change in terms of the genes which are exp ressed , the cel lular const i tuents and the site(s) of product ion. For example , during ontogeny hemopo ies is takes p lace sequent ia l ly in the embryon ic yolk s a c , paraaort ic gonada l meseneph ros region ( A G M ) , fetal l iver, sp leen and finally the adult bone marrow. Whi le the mo lecu la r s teps that regulate t hese c h a n g e s are poorly unders tood , recent s tud ies have begun to de l inea te ear ly even ts and pat terns of ce l lu lar migrat ion that const i tu te this deve lopmenta l program. Recent ly there has been a great deal of interest in the use of hemopoiet ic cel ls der ived from developmental ly early sources such a s fetal liver or umbil ical cord blood for use in human bone marrow transplantat ion. Th is interest has been st imulated by the f indings of severa l studies (d iscussed in more detail in sec t ion 1.3.2) hinting that H S C s der ived from ontological ly ear ly s o u r c e s may p o s s e s s a super io r regenerat ive capac i ty as c o m p a r e d to adult bone mar row H S C s . In Chapter 5 of this thesis data are presented demonstrat ing and quantifying the super ior regenerat ive capaci ty of H S C s der ived from fetal liver a s compared to adult bone marrow following transplant. Hemopo ies i s begins in the embryonic yolk s a c at approx imate ly day 7.25. Be tween days 7.5 and 8.5 the yolk s a c is a source of in vitro c lonogenic progenitors (82), B lymphoid precursors (83) and day 12 C F U - S (82) with H S C s capab le of long term lympho-myelo id repopulat ing ability detected by day 11 (82, 84). More recent ly , s imi lar hemopo ie t i c cel l popula t ions have been demons t ra ted to be present within the A G M region with C F U - S activity detectable on day 8 to day 11 (85, 86) and long term repopulating H S C s by day 10 (84). The product ion of b lood cel ls within the yolk s a c and A G M regions begins to decl ine as the fetal liver takes over a s the major site of embryon ic hemopo ies i s . Hemopo ies i s is detectab le in murine fetal liver by day 10 of gestat ion and inc reases between days 10-13 (11). T h e fetal l iver remains the major site of hemopo ies i s until birth whe reupon it 13 d e c r e a s e s rapidly. Erythropoiesis is detectable in the sp leen by day 15 al though by day 17 g ranu lopo ies is p redominates in this o rgan . H e m o p o i e s i s in the sp l een r e a c h e s its peak at approx imate ly 4-8 days fol lowing birth where it d e c r e a s e s steadi ly a s the marrow begins to take over as the predominant site of hemopo ies is for the remainder of the animal 's life. Figure 1.3 shows the contribution of var ious o rgans to hemopo ies is during ontological development. 1.3.2. Compar ison of stem cells from fetal liver and adult t issues T h e relat ionship between fetal liver and adult bone marrow H S C s remains unc lear . A l though H S C s from these two differing populat ions are be l ieved to be ontological ly related, that is der ived from the s a m e initial pool of pr imordial s tem cel ls , fetal liver cel ls display functional character ist ics that differ from those found in the adult marrow. Fetal l iver day 8 C F U - S ce l ls for examp le p o s s e s s a super ior sel f - renewal capaci ty as compared to their adult bone marrow 100 n 1 — i 10D 12D 14D 16D 18D EMBRYONIC AGE BIRTH 1 1 1 7 T I 1 1 r 2D 4D 6D 8D 2Wk 4Wk 6Wk POST NATAL AGE Figure 1.3. Chang ing si tes of hemopo ies is reflected in the relative proport ion of in vitro c lonogenic progenitors cel ls found within the embryonic yolk sac , fetal liver, sp leen and bone marrow during ontogeny and post natally. counterpar ts (87), and fetal l iver ce l ls show a greater compet i t ive repopulat ing ability a s compa red to adult bone marrow ce l ls in in v ivo t ransplantat ion a s s a y s (88). Rebe l et. a l . have shown that limiting numbers of fetal l iver C R U are able to 14 produce a greater output of mature blood cel ls in vivo as compared to adult bone marrow (89). Moreover , it w a s a lso demonst ra ted that when marrow ce l l s from pr imary rec ip ients of l imiting numbers of fetal l iver C R U were in jected into seconda ry recipients, a signif icantly higher percentage of these s e c o n d a r y m ice s h o w e d d o n o r c e l l - d e r i v e d reconst i tu t ion of the i r l y m p h o i d a n d m y e l o i d compar tments as compared to mice that had received marrow ce l ls from primary recipients of similar numbers of adult bone marrow C R U . O n e poss ib le explanat ion for the di f ferences in regenerat ive potential between these two cel l populat ions is that fetal l iver C R U p o s s e s s a greater intr insical ly de te rmined probabi l i ty to undergo se l f - renewal ve rsus differentiation d iv is ions when proliferating within the adult marrow microenv i ronment . Vas i r i et. a l . (90) have recent ly p roposed the theory that the p rogress ive loss of te lomer ic D N A with e a c h round of ce l lu lar div is ion may act as a mitotic c lock and therefore be the molecu lar bas is by which fetal liver ce l ls may be able to undergo a greater number of cel l d iv is ions prior to unde rgo ing s e n e s c e n c e . H o w e v e r , th is exp lana t ion d o e s not p r e c l u d e the possibi l i ty that fetal liver H S C s may maintain di f ferences in cel l cyc le t imes or the number of ce l ls recruited or mainta ined within the microenvi ronment of the adult marrow, or exp ress addit ional intrinsically def ined mo lecu les that may direct the ce l l into favor ing se l f - renewa l v e r s u s dif ferent iat ion d iv i s ions . Intr iguing new ev idence of poss ib le intrinsic determinants of sel f - renewal w a s recently reported by S a u v a g e a u et. a l . in studies of the Homeobox family of transcript ion factors (91). E n g i n e e r e d ove rexp ress ion of H O X B 4 , a gene w h o s e exp ress i on is normal ly restricted to the most primitive adult bone marrow cel ls, was found to result in up to a 50- fo ld i n c r e a s e in the regenera t ion of retroviral ly t r a n s d u c e d H S C s a s compared to marker gene- t ransduced control cel ls . 1.4. Properties of hemopoietic stem cells 1.4.1. Cyc l ing status 15 In contrast to commit ted c lonogen ic progenitors and a large proport ion of day 8 and 12 C F U - S , the vast majority of long term repopulat ing H S C s exist in a state of q u i e s c e n c e within the marrow under normal homeosta t ic cond i t ions a s determined by their ability to survive a single injection of the cyc le-spec i f ic cytotoxic drug 5 - F U (92). However , fol lowing treatment with 5 - F U the most primitive H S C s are recruited into cyc le s ince they become highly sensi t ive to a s e c o n d dose of 5-F U given 3-5 days later (93). 1.4.2. Developmental potential and dynamics of H S C s Al though the ex i s tence of totipotent H S C s is now a wel l d o c u m e n t e d phenomenon , the number of H S C s and their usage in s teady state hemopo ies i s remains unclear . O n e theory that has been put forth to expla in the dynam ics of H S C util ization is the c lonal success ion model p roposed by Kay in 1965 (94). Th is mode l p roposes that hemopo ies i s is mainta ined by the sequent ia l act ivat ion of s tem cel l c lones that proliferate, differentiate and eventual ly b e c o m e exhaus ted . In this mode l H S C s are akin to fuel for hemopo ies is , cont inual ly being act ivated and " c o n s u m e d " from a reserve pool of ce l ls . A signif icant body of da ta from bone marrow transplantat ion studies has provided support for this theory. S tud ies b a s e d upon the t ransplant of var ious mixtures of retrovirally marked or enzymat ica l l y d ist inguishable adult bone marrow or fetal liver s tem cel ls into either irradiated (24, 95) or genet ical ly anemic W / W v mice (36, 96) revealed major f luctuat ions in the contribution of individual H S C s to hemopoies is within the first s ix months fol lowing transplant. During this period numerous H S C c lones were obse rved to contr ibute signif icantly but transiently to hemopoies is , the span of their contribution lasting on average for only a few months. Further ev idence support ing the c lonal s u c c e s s i o n theory w a s a lso obtained from large animal models . Abkowi tz et. a l . prov ided data d e m o n s t r a t i n g la rge c l ona l f l uc tua t ions in c a t s h e t e r o z y g o u s for the X c h r o m o s o m e - l i n k e d e n z y m e g l u c o s e - 6 - p h o s p h a t e d e h y d r o g e n a s e ( G 6 P D ) fo l lowing t ransplants of auto logous marrow (97). However , the an ima ls used in 16 these exper iments were only ana lyzed for relatively short per iods of t ime fol lowing t ransp lan t (ie. 1-5 mon ths or 1-1.5 y e a r s for mur ine a n d fe l ine rec ip ien ts respec t i ve l y ) . S u b s e q u e n t s tud ies us ing retrovira l ly m a r k e d m a r r o w c e l l s sequent ia l ly ana l yzed the c lona l contr ibut ion to the per iphera l b lood of mur ine recipients for longer per iods of t ime (ie. 7-12 months) post transplant (64, 98 , 99). Fo l l ow ing a per iod of c lona l instabi l i ty for up to 6 mon ths post t ransp lant , hemopo ies i s b e c a m e dominated by a smal l number of totipotent s tem cel l c l ones that cont inued to function for long per iods of t ime, often for the remainder of the an imal 's life. Interestingly, the c lonal f luctuations observed in fe l ines t ransplanted with a l logeneic marrow descr ibed by Abkowi tz et. a l . in 1990 (97) were found to stabi l ize when these s a m e recipients were ana lyzed at longer t imes post transplant (3.5-6 years) (100). O n e explanat ion for the c lonal f luctuation initially obse rved fol lowing transplant is that it is due to the util ization of more "mature" s tem cel ls wh ich p o s s e s s a l imited capac i ty for prol i feration and se l f - renewa l . H S C s are be l ieved to be organ ized as a hierarchy of cel ls with decreas ing se l f - renewal and proliferative potential (refer to Figure 1.1) and heterogenei ty with regards to both phys ica l and funct ional propert ies of H S C s have been documen ted . T h u s , more mature H S C s wou ld quickly b e c o m e exhaus ted and eventual ly be rep laced by more primitive H S C s wh ich p o s s e s s a more long- l ived capac i ty for prol i feration a n d se l f - renewa l . A n al ternat ive exp lanat ion of H S C behav io r fo l lowing bone marrow transplant has been provided by Jordan and L e m i s c h k a (98). T h e authors p roposed that the initial per iod of c lonal f luctuation observed w a s the result of an expand ing pool of totipotent H S C s undergoing stochast ic commitment ve rsus self-renewal events. T h e s e events are portrayed in Figure 1.4. Accord ing to this mode l , the c lonal f luctuation observed within the first six months post transplant (depicted in F igure 1.4A) are the c o n s e q u e n c e of ei ther s tochas t i c m e c h a n i s m s or the d e m a n d s of the radiat ion-ablated hemopoiet ic sys tem result ing in the commitment of the H S C s without signi f icant se l f - renewal . T h u s , a l though this c lone initially 17 cont r ibu tes to h e m o p o i e s i s , without se l f - renewal it is unab le to main ta in its numbers and is exhaus ted . S tem cel l c lones which are observed to contr ibute to h e m o p o i e s i s at both short and long per iods post t ransplant are the result of commi tment dec i s ions which occur in paral lel with a signi f icant deg ree of self-renewa l (F igure 1.4B). Al ternat ively, s o m e c l ones may undergo a subs tan t ia l degree of se l f - renewal shortly after transplant. S u c h c lones , a l though they would not be o b s e r v e d to contr ibute to h e m o p o i e s i s dur ing the init ial pe r iod of regenerat ion would ultimately c o m e to dominate hemopo ies i s for lengthy per iods of t ime, possib ly for the remainder of the animal 's life (Figure 1.4C). A. Commitment o • • M L Extinct B. Self-Renewal + Commitment C . Self-Renewal M o / \ > o /\ o o /\ / \ • O • i L / \ ^ M L o o non-O / \ O O contributing / \ / \ o o o o Continued Expansion Continued Expansion • • O O M L Stable • • O O M L Stable Months Post -Transplant 4-6 12-"16 Figure 1.4. M o d e l p r o p o s e d by J o r d a n and L e m i s c h k a (64) to exp la in the behav ior of H S C s at var ious t ime points fol lowing bone marrow transplant. O p e n c i rc les des ignate totipotent H S C s whi le fil led c i rc les represent ce l ls commit ted to either myelo id or lymphoid l ineages. 18 In s u m m a r y , it rema ins unc lear whether the c lona l f luctuat ion init ial ly obse rved fol lowing bone marrow transplant is representat ive of the uti l ization of more mature H S C s with limited proliferative and sel f - renewal capac i t ies , or is the result of s tochast ic mechan isms acting upon an expanding pool of totipotent H S C s . 1.4.3. Self-renewal and aging of H S C s Stud ies using retroviral marking of H S C s have demonst ra ted conv inc ing ly the ability of H S C s to undergo sel f - renewal both in vitro (101) and in v ivo (24, 98, 99) through the detect ion of identical proviral banding patterns in the bone marrow and/or thymus of more than one pr imary or s e c o n d a r y recipient. A l though the studies ment ioned above directly demonstrate that sel f - renewal of H S C s can occur it remains unknown to what extent this process can occur s ince the magni tude of se l f - renewal events have not been documented . Th is has been in part due to the fact that quant i tat ive a s s a y s for measur ing s tem ce l l s with long term in v ivo reconst i tu t ing abil i ty were not ava i lab le until re lat ively recent ly (43, 45) . A s previously stated, the requirement for a functional a s s a y when ana lyz ing long term repopulat ing s tem cel l function is essent ia l s ince a number of groups have shown that cel l sur face phenotype is not an accurate determination of biological capabi l i ty (46, 76). Moreover , the ineff iciency of retroviral marking procedures has m a d e the t racking of individual H S C s in vivo labor ious. Thus , methods which inc rease the utility of recombinant retroviral marking would be advan tageous towards s tud ies a imed at more clearly defining the regenerative potential of H S C s . Chap te r 3 of this thes is desc r ibes the deve lopment of methods to enhance the power of retroviral mark ing procedures. That H S C s have an unlimited ability to replenish their numbers through the p ro cess of se l f - renewal has been brought into quest ion by a number of s tud ies demons t ra t i ng that even a s ing le t ransplantat ion p rocedu re c a n reduce the repopulat ing ability of bone marrow dramatical ly (102-109). For example , using the T 6 c h r o m o s o m e marker sys tem, R o s s et. a l . demonst ra ted that the proport ion of 19 donor cel l mi toses in recipients of bone marrow which had been subjected to one round of transplantat ion w a s only 10% of that detected when normal bone marrow w a s used (102). T h e possibil i ty that the loss in repopulat ing ability of bone marrow fol lowing transplant is due to d a m a g e to the recipients microenv i ronment by the i r radiat ion reg imen w a s exc l uded s ince s imi lar resul ts we re o b s e r v e d us ing genet ical ly anemic W / W v recipient mice. M a u c h and Hel lman (110) eva luated the long-term c o n s e q u e n c e s of transplanting limiting numbers of hemopoiet ic ce l ls and demons t ra ted a direct relat ionship between donor cel l d o s e and day 8 C F U - S recovery and se l f - renewal capac i ty that w a s not ref lected in per iphera l b lood coun ts or bone marrow cel lular i ty. Further, the d e c r e a s e d recovery of C F U - S p a r a m e t e r s d id not c h a n g e with t ime after t ransp lanta t ion s u g g e s t i v e of a permanent loss of marrow regenerat ive capaci ty . Har r ison et. a l . have prov ided da ta s u g g e s t i n g that the reduct ion in H S C n u m b e r a n d funct ion fo l lowing transplant is the result of two mechan isms (111). They calculated that t ransplanted marrow s h o w e d a two-fold reduction in s tem cel l numbers as c o m p a r e d to f resh marrow. Moreover , they found that the average proportion of donor der ived ce l ls in recip ients of prev iously t ransplanted marrow w a s approx imate ly 1 5 % of that of f resh marrow. T h u s , t ransp lanted marrow demons t ra ted a s e v e n - to eight- fold reduct ion in repopulat ing abil i ty as c o m p a r e d to f resh marrow. T h e s e resul ts suggest that not only are transplant recipients compromised in terms of the quantity of s tem cel ls which they p o s s e s s but they are a lso deficient in terms of the quality of s tem ce l l s . O n e exp lanat ion for the loss in the regenerat ive capac i t y of bone marrow fol lowing transplant is the inability of t ransplanted s tem cel ls to reconstitute fully the H S C compar tment to normal (non-transplant) leve ls due to a l imited capaci ty for se l f - renewal . A n alternative hypothesis is that transplantat ion p laces a high deg ree of s t ress on the H S C s to undergo differentiat ive rather than self-renewal events in order to provide sufficient numbers of functionally mature ce l ls to reconsti tute hemopo ies i s in the myeloablated recipient. Another possibi l i ty is that 2 0 H S C regenerat ion fol lowing bone marrow transplant is limited due to the act ion of negat ive regulatory feedback mechan isms in vivo that can limit s tem cel l expans ion prematurely even though the H S C compartment is far from being regenerated back to normal levels. Th is scenar io might occur through the product ion of such factors a s M I P - 1 a or T G F - p that have been shown to dec rease the proportion of primitive hemopoiet ic cel ls in cyc le (112-115). T h e o b s e r v e d dec l ine in repopulat ing capac i t y fo l lowing b o n e mar row t ransplant ra ised the quest ion a s to whether H S C s within the mar row of o ld unmanipu la ted mice might a lso share such a defect. T h e hypothes is , ca l led the generat ion-age hypothesis (116), sugges ted that an entire lifetime d e m a n d on the H S C s might resul t in a poo l of H S C s wh ich w e r e c o m p r o m i s e d in thei r regenerat ive capaci ty s ince they were forced to have undergone a greater number of cel l division events. However , using the competit ive repopulat ion a s s a y Harr ison and his co l l eagues found no d i f ferences in the repopulat ing ability be tween the marrow of young and old donors (117). Bone marrow der ived from mice 2 to 2.5 y e a r s in a g e compe ted equal ly wel l against the compet i tor cel l f ract ion a s d id marrow cel ls der ived from young mice 3 to 6 months of age . Moreover , it w a s a lso s h o w n that bone marrow der ived from old mice compe ted equal ly wel l in ser ia l t ransplantat ion exper iments as those der ived from young mice . For examp le , the proport ion of donor der ived cel ls in recipients of previously t ransplanted marrow der ived from either young or old donor mice was approximately 1 0 % a s compared to f r e s h , non- t ransp lan t mar row. T h u s , a l though da ta f rom mur ine se r i a l t ransplantat ion s tudies have sugges ted that H S C s may have a finite capac i ty for se l f - renewal it s e e m s , us ing the a s s a y s ava i lab le , that this limit is not r eached during the normal life span of the animal . Al though the f indings of numerous studies suggest that the capaci ty of H S C s to undergo se l f - renewal is not unl imited, few data are avai lab le on the extent to which sel f - renewal of H S C s from var ious ontological sources can occur . S u c h data 21 would be valuable in light of the crucial role that H S C s play in a number of cl inical p rocedures including autograft purging, the genet ic therapy of var ious her i table hematologica l d isorders, attempts at ex vivo expans ion of H S C s as wel l as the use of al ternat ive sou rces of t ransplantable s tem cel l populat ions (ie. fetal l iver and umbi l ica l cord blood). Chap te r 5 of this thes is desc r i bes da ta co l lec ted on the regenerat ive capaci ty of H S C s derived from adult bone marrow or fetal liver. 1.5. Regulation of hemopoiesis T h e p rocess by which a smal l number of H S C s cont inuously generate the appropriate number of mature blood cel ls constituting the eight major hemopoiet ic l ineages is a complex, regulated process . Moreover, the entry of mature blood cel ls into the c i rculat ion, their local izat ion to the appropr iate t i ssues a s wel l a s their functional activation is a lso under strict regulation. The known molecu les which are respons ib le for regulating these var ious aspec ts of hemopo ies i s can general ly be div ided into two groups: extracel lular factors (composed of both humora l factors and cel l or matr ix -assoc ia ted factors), and intracellular factors (eg. growth factor receptors and transcript ion factors). A l though there is an abundance of ev idence support ing the role of molecu les belonging to both of these groups in the regulation of h e m o p o i e s i s their relative importance and relat ionship to one another is, at present, unclear. 1.5.1. Regulation by extracellular factors 1.5.1.1. Hemopoietic growth factors T h e most ex tens ive ly cha rac te r i zed group of ext r ins ic fac tors are the hemopoie t ic growth factors. Th is group is c o m p o s e d of the hemopoie t ic co lony-st imulat ing factors ( G - C S F , M - C S F , G M - C S F ) , the inter leukins (IL-1 to IL-17), hemopoiet ic inhibitors (TGF-|3 and MIP-1oc) and "stem cel l " factors (Steel factor and flk2/flt3 l igand). T h e s e factors are glycoproteins of 10-70 ki lodaltons, are able to act at extremely low concentrat ions (ie. 10"6 M), and are produced by a variety of cel l 22 t ypes throughout the body (reviewed in (118) and (38)). To date more than 25 distinct hemopoie t ic growth factors have been identif ied and the g e n e s for most have been c loned . T h e s e factors have been shown to play a role in the surv iva l , pro l i fera t ion a n d di f ferent iat ion of hemopo ie t i c c e l l s at v a r i o u s s t a g e s of deve lopment (119). By far, the majority of data that have been co l lec ted on the funct ion of t hese growth factors has been their ac t ion on the more mature hemopo ie t i c cel l t ypes . M u c h less da ta regarding their ef fects upon H S C s is ava i lab le . However , severa l growth factors have been identified that are involved in the activation and regulation of the proliferation of primitive hemopoiet ic ce l ls and their p rogeny . Entry of qu iescen t blast co lony forming ce l l s into the ce l l c y c l e is regulated by multiple synergist ic factors, including IL-6 (120), granulocyte co lony-st imulat ing factor ( G - C S F ) (121), IL-11 (122, 123), IL-3 (124), and S tee l factor (125). In contrast, factors such as M I P - 1 a a n d T G F - p have been shown to inhibit the entry of primitive murine and human progenitors into the cel l cyc le (113, 126, 127). In addit ion to bringing quiescent primitive hemopoiet ic ce l ls into cyc le , S tee l factor, the l igand for the c-kit receptor, a lso plays a role in the proliferation of these cel ls . Al though Steel factor by itself has little effect upon the proliferation of primitive hemopo ie t i c ce l l s , when comb ined with others factors s u c h a s IL-3, G M - C S F , granulocyte colony-st imulat ing factor ( G - C S F ) , IL-4, IL-1oc, IL-11 and IL-12, S tee l factor has a potent effect upon the proliferation of both immature a n d l ineage-restr icted lymphoid and myelo id ce l ls (128-135). More recent ly a growth factor w h o s e receptor is exp ressed on a more primitive subgroup of hemopoie t ic ce l ls has been desc r i bed . Us ing Northern blot and reverse t ransc r ip tase-po lymerase cha in react ion ( R T - P C R ) ana lys is , express ion of the flt3/flk2 receptor w a s largely conf ined to primitive hemopoiet ic cel l populat ions wh ich inc lude ce l l s with long term lympho-mye lo id repopulat ing ability (136, 137). S imi lar ly , in h u m a n s , the 2 3 flt3/flk2 receptor is e x p r e s s e d on thymic, sp len ic and bone marrow progeni tors pos i t ive for exp ress ion of the C D 3 4 ce l l su r face ant igen, a marke r found on primitive hemopoiet ic cel ls and endothel ial cel ls (138, 139). The flt3/flk2 l igand (FL) (140, 141) is ab le to st imulate the proliferation of primitive mur ine and human hemopo ie t i c progeni tor ce l l s when c o m b i n e d with other hemopo ie t i c growth factors, reminiscent of the effects of Steel factor. C o m b i n e d with Stee l factor a lone, or S tee l f ac to r / I L -3 /GM-CSF , F L increases the proportion of A A 4 . 1 + S c a - 1 + L i n h i g n mur ine fetal l iver ce l ls in cyc le as determined by [3|H]thymidine i nco rpo ra t i on assay (140 ) and the number of in vitro co lon ies in c lonogen ic a s s a y s fo l lowing plating of T h y ' ° S c a - 1 + L i n " s tem cell candidates purified from murine bone marrow (142). Furthermore, when combined with IL-7, F L induces proliferation of immature day 14 mur ine fetal T -ce l l p recursors sugges t i ng a role for F L in l ympho id deve lopment (142). Al though both Steel factor and F L appear to be important early in hemopo ies i s these factors are not essent ia l s ince mice lack ing exp ress ion of S tee l factor, F L or their receptors, due to natural mutat ions or gene knockout , are still v iab le (143, 144). T h e s e observat ions sugges t the ex i s tence of addi t ional unknown stem cel l factors and/or known growth factors able to compensa te for the loss of express ion of Steel factor/FL or their receptors. Wh i l e s o m e growth factors play a role in the deve lopmen t of ce l l s of numerous hemopoiet ic l ineages (ie. IL-3), others appear to be more restr icted. For examp le , erythropoiet in, thrombopoiet in, G - C S F and M - C S F predominant ly act in the d e v e l o p m e n t of e ry th rocy tes , p la te le ts , neut roph i l i c g r a n u l o c y t e s a n d macrophages /monocy tes respectively. Moreover , severa l of the interleukins (ie. IL-2 and IL-7) function predominantly in lymphoid development. 1.5.1.2. The extracellular matrix T h e hemopo ie t i c m ic roenv i ronment is c o m p o s e d of a var ie ty of n o n -hemopo ie t i c ce l l s wh ich inc lude f ibroblasts, ad ipocy tes , endothe l ia l ce l l s a n d ret icular ce l l s wh ich together p roduce the ext race l lu lar matr ix. Mo rpho log i ca l 24 s tud ies have demons t ra ted a c lose phys ica l assoc ia t i on be tween s t romal and b lood ce l ls within the marrow cavity (145-147). In addit ion to producing a variety of growth factors (147) ,the above cel l types a lso produce a number of other proteins which have been hypothes ized to play a role in regulating the survival , proliferation and differentiation of hemopoiet ic cel ls by facilitating communicat ion between ce l ls v ia cel l sur face receptors and through the local ized concentrat ion and presentat ion of seques te red growth factors to hemopoiet ic cel ls . Examp les of the different types of prote ins c o m p o s i n g the ext racel lu lar matrix inc lude a number of c o l l a g e n s (Types I, III, IV and V) , g lycoprote ins ( thrombospondin, f ibronect in, hemonec t in , laminins and tenascin) , and g lycosaminog lycans (hyaluronic ac id , heparan sulfate, dermatan sulfate and chondroit in sulfate). G lycopro te ins such as f ibronect in and th rombospond in as wel l a s g l y cosam inog l ycans s u c h a s hya lu ron ic ac id and hepar in sulfate are thought to function in hemopo ies is by act ing upon on primitive hemopo ie t i c p rogen i to rs . Fo r e x a m p l e , f ibronect in a n d t h r o m b o s p o n d i n are adhes ive l igands for a variety of human progenitors including C F U - G E M M , B F U - E , C F U - G M and LTC- IC . (148 , 149). Moreover , hyaluronic ac id , hepar in sul fate and chondroi t in sulfate are involved in the adhes ion of primitive human progeni tors to s t roma in long term cultures in vitro and are thought to enhance hemopo ies is either by binding both primitive progenitors and growth factors that can st imulate them, by concent ra t ing these growth factors at the site of their ce l lu lar receptors , or by e n h a n c i n g the at tachment of primit ive ce l ls to the s t romal feeder layer (150). Interestingly, an inc rease in the product ion of chondroi t in sul fate in long term cul tures has been correlated with an enhancement of day 10 C F U - S and C F U - G M product ion (151). In addit ion, integrins such as LFA-1 and V L A - 4 and the C D 4 4 g lycoprote in wh ich are e x p r e s s e d on C D 3 4 + human bone marrow ce l l s (152) appea r to be important in mediat ing interact ions between primit ive hemopo ie t ic progenitors and s t roma s ince the addit ion of ant ibodies directed against V L A - 4 or C D 4 4 retards lympho- and myelopoies is in long term cultures (153, 154). Further, 2 5 V L A - 4 interactions with f ibronectin, V C A M - 1 and L-select in have been impl icated in mediat ing the in vivo homing of hemopoiet ic progenitors to the marrow and sp leen and the s a m e interactions may be involved in the mobi l izat ion of progenitors from the marrow to the blood (149, 155). The above data strongly suggest that adhes ion mo lecu les and mo lecu les of the ex t race l lu lar matrix p lay a role in the regulat ion of pr imit ive hemopo ie t i c progenitor cel ls (and possib ly totipotent H S C s ) . T h e s e molecu les may funct ion by increas ing the response of progenitors cel ls to var ious hemopoiet ic growth factors s ince thrombospondin has been observed to increase the response of the ce l ls to IL-3 and G M - C S F (156). 1.5.2. Regulation by intracellular factors Signa l t ransduct ion v ia growth factors involves the binding of the growth factor to its appropr iate receptor fo l lowed by d imer izat ion of the receptor and initiation of downs t ream s igna l ing pa thways involv ing a var iety of cy top lasmic in termediates s u c h as protein k i nases and phospha tases and their subs t ra tes . Growth factors are bel ieved to inf luence a variety of cel lular funct ions including cel l surv ival , cyc l ing status, proliferation and possib ly differentiation. A l though many of the d o w n s t r e a m m o l e c u l e s invo lved in these pa thways have not yet b e e n identi f ied, severa l transcript ion factors are now recogn ized as key regulators in hemopo ies i s . Transcr ipt ion factors that appear to be involved in the regulat ion of early hemopoiet ic cel l development and sel f - renewal have recently been identif ied through overexpress ion and gene disruption exper iments (157). Disrupt ion of the G A T A - 2 gene in mice results in a marked reduction in all hemopoiet ic progenitors (158). Moreover , the overexpress ion of H O X B 4 has been shown to be assoc ia ted with an i nc rease in the regenerat ive abil i ty of H S C s fo l lowing bone mar row transplant (91). S o m e factors are more restricted in their patterns of exp ress ion . The SCL /Ta l - 1 and Ikaros transcription factors, for example , are exp ressed in cel ls of the myeloid/erythroid and lymphoid l ineages respect ively (159, 160). A number 2 6 of transcript ion factors that appear to be involved in the deve lopmenta l regulat ion of spec i f ic hemopoiet ic l ineages have recently been identi f ied. T h e transcr ipt ion factors M Z F 1 , N F A T and Oc t -2 /Pax5 are speci f ical ly exp ressed in neutrophi ls, T ce l ls and B cel ls respect ively, while G A T A - 1 is necessa ry for the deve lopment of ce l ls of the erythroid, mast and megakaryocyte l ineages (157, 161, 162). T h u s , a substant ia l amount of data ex is ts sugges t i ng that t ranscr ip t ion factors play a crucia l role in early hemopoiet ic deve lopment . A s a result in tense research efforts are now being focused upon identifying those genes w h o s e activity is regulated by these transcript ion factors and which may e n c o d e the molecu la r factors wh ich are respons ib le for govern ing the se l f - renewa l , prol i ferat ion and commitment of H S C s . 1.6. Genet ic manipulat ion of hemopoiet ic ce l ls us ing recombinant ret roviruses B a s e d upon their capac i ty for highly efficient infect ion and non-tox ic and s tab le integrat ion into the g e n o m e of a wide range of ce l l t ypes , recombinant retroviruses represent the most attractive vehic le for exogenous gene transfer into mammal ian target cel ls . T h e abil i ty to t ransfer e x o g e n o u s g e n e s into H S C s us ing recombinan t retrovi ruses (ie. genet ical ly mark them) has prov ided signif icant insight into the proliferative and differentiative potential of totipotent H S C s and has provided direct e v i d e n c e of the ability of these ce l ls to se l f - renew. In addi t ion, g e n e t ransfer methodo log ies have provided the m e a n s to test crit ically var ious g e n e s encod ing puta t ive regu la tory m o l e c u l e s wh ich may p lay a role in con t ro l l i ng H S C proliferation, differentiation and/or sel f-renewal at the molecular level . Fo r example , the transfer and overexpress ion of a variety of genes encod ing hemopoiet ic growth factors in murine recipients resulted in myeloprol i ferative d isorders s imi lar to those o b s e r v e d in many human leukemia pat ients sugges t i ve that the dys regu la ted 27 exp ress ion of hemopoiet ic growth factors may play a role in human d i s e a s e v ia a u t o s t i m u l a t o r y m e c h a n i s m s (163 -167 ) . In a d d i t i o n , t h e s e e x p e r i m e n t s demonst ra ted conc lus ive ly the role of growth factors in hemopoiet ic deve lopment a n d regulat ion in v ivo , suppor t ing resul ts ob ta ined f rom in vitro c l o n o g e n i c progenitor a s s a y s . Moreover , gene transfer is central to the newly deve loped field of human gene therapy. G e n e therapy became reality in 1990 when a young girl, born with a defect ive vers ion of the gene encod ing adenos ine d e a m i n a s e (ADA) , rece ived inject ions of her own T ce l ls t ransduced with a recombinant retrovirus express ing a normal functioning copy of the A D A gene. S ince then more than 100 c l in ical trials a imed at treating d i s e a s e s ranging from inherited d isorders such a s cyst ic f ibrosis and G a u c h e r s d i sease to cance r and Acqu i red Immune Def ic iency S y n d r o m e (AIDS) have been approved in the U . S . a lone. Unfortunately, desp i te this flurry of activity, the results that have been ach ieved with the current gene t ransfer methodo log ies have been rather d isappoint ing. Th is is due to p rob lems assoc ia ted with the eff iciency of gene transfer protocols, express ion of t ransferred g e n e s a s well as a lack of understanding of the growth and survival requirements of human H S C s . Mark ing studies, the investigation of putative regulatory g e n e s in hemopo ie t i c deve lopment and gene therapy sha re at least one of two major requirements: 1) the efficient and stable transduct ion of the hemopoiet ic s tem cel l and 2) appropr iate express ion of the t ransduced genes in the des i red cel l type of interest. The remainder of this introduction dea ls with the advances that have been made using recombinant retroviruses as a vector for gene transfer. 1.6.1. The lifecycle of retroviruses Retrov i ruses do not exist a s distinct genet ic e lements within their host ce l ls but permanent ly integrate into the genome in the form of a D N A provirus and thus are stably transmitted from parent cel l to progeny ce l l . However , during the course of their l i fecycle, the virus al ternates from an R N A to a D N A intermediate. T h e crit ical features of the retroviral l i fecycle involve the synthes is and packag ing of a 28 genomic R N A copy of the provirus into structural viral proteins; shedd ing of the intact virion particle from the sur face of the host cel l ; entrance of the virion into the target cel l v ia speci f ic cel l sur face receptors; reverse transcript ion of the viral R N A into a doub le -s t randed c D N A copy ; integrat ion of the proviral c D N A into the genomic D N A of its host cel l ; and express ion of the viral genes . In genera l , all retroviruses p o s s e s s three distinct open reading f rames ( O R F ) ca l led gag , pol and env. T h e s e genes encode polyproteins c o m p o s e d of essent ia l structural proteins s u c h as the matrix (MA), caps id (CA) and nuc leocaps id (NC) (gag O R F ) , the enzymes required for the reverse transcription (RT) of viral R N A into c D N A and integration (IN) of the c D N A into the genome of the host cel l (pol O R F ) , and the sur face (SU) and t ransmembrane (TM) proteins respons ib le for ent rance into its target cel l v ia interaction with speci f ic cel l sur face receptors (env O R F ) (reviewed in (168) and (169)). In the proviral form, the gag , pol and env g e n e s are b o u n d e d on ei ther s ide by long terminal repeats ( L T R s ) wh ich con ta in the enhancer and promoter regulatory e lements as well as the transcript ional initiation and po lyadeny la t ion s igna ls . T h e s e e lements are respons ib le for d i rect ing the synthes is of both the cel lular R N A transcripts for the production of the viral proteins a s wel l a s the full length genomic R N A copy of the provirus uti l izing the cel lu lar R N A po lymerase II enzyme. A schemat ic d iagram of a typical murine retrovirus is s h o w n in F igure 1.5. P a c k a g i n g of the genomic R N A into vir ion par t ic les is dependent upon interaction of the viral structural proteins with a speci f ic s e q u e n c e ca l l e d \j/ wh i ch l ies d o w n s t r e a m of the 5' L T R (170). H o w e v e r , add i t iona l s e q u e n c e s present within the downst ream gag gene are now a lso known to be essent ia l for efficient R N A packaging (171, 172). O n c e surrounded in a core of viral protein, the vir ion part icle buds from the cel l obtaining a sur round ing enve lope c o m p o s e d of cel lu lar membrane s tudded with viral enve lope g lycoprote ins . T h e infect ious vir ion part icle subsequent ly ga ins ent rance into a potential target cel l through the interact ion of the viral enve lope proteins with spec i f i c cel l su r face 29 receptors. Severa l of the cell surface receptors which retroviruses have exploi ted to gain entry into target cel ls have recently been identif ied. T h e s e inc lude receptors for ecotropic murine retroviruses (MCAT) (173, 174), fel ine immunodef ic iency virus (CD9) (175, 176), human and s imian immunodef ic iency v i ruses (CD4) (177-179), av ian l eukos i s v i rus type A (Tval) (180), bov ine l eukem ia v i rus (Blvr) (181) amphotropic murine retroviruses (Ram-1) (182, 183) and the c o m m o n receptor for g ibbon ape leukemia virus (184), fel ine leukemia virus group B (185) and s imian sa rcoma-assoc ia ted virus (186) (Glvr-1). Interestingly, the M C A T , Glvr-1 and Ram-1 receptors are transport proteins which perform essent ia l housekeep ing funct ions; M C A T se rves as a cat ionic amino acid transporter of arginine, lysine and ornithine (173 , 187) whi le Ram-1 and Glvr-1 are both s o d i u m - d e p e n d e n t p h o s p h a t e sympor te rs (188). Entry into the target cel l results in the t ransformat ion of the qu iescent enve loped virion into an enzymat ical ly act ive nucleoprotein comp lex that beg ins to reverse t ranscr ibe the genomic R N A template into a doub le s t randed c D N A copy (189, 190). Success fu l integration of the proviral c D N A into the genome of the host cel l has been shown to require that the cel l be act ively cyc l ing (191, 192), with the main barrier to success fu l retroviral integration being the p resence of the nuc lea r m e m b r a n e (193). Upon d isso lu t ion of the nuc lea r m e m b r a n e at p rophase of mitosis the viral D N A enters the nuc leus and inserts itself stably into the g e n o m e of its host through the act ion of the viral e n c o d e d in tegrase wh ich " randomly" n icks the genomic D N A creat ing a 4-6 b a s e pair s taggered cut and l igating it to the s taggered cut viral D N A . A summary of the crit ical s teps in the retroviral l i fecycle are shown in Figure 1.6. 1.6.2. Recombinant retroviruses as vectors for gene transduction At present, severa l hurdles remain to be ove rcome for the sa fe , s table and efficient use of recombinant retroviruses as vectors for gene transfer. T h e s e include the ability to produce retroviral s tocks free of replication competent wild type virus, the ability to introduce the recombinant retrovirus into the target cel l through 30 Figure 1.5. Schemat i c representat ion of a typical retrovirus showing the gag , pol and env g e n e s , sp l i ce accep to r (SA) and sp l ice donor (SD) s i tes , and the \\i packag ing s igna l . T h e L T R is c o m p o s e d of three reg ions; U 3 , R and U 5 . U 3 conta ins two direct repeat 72 bp enhancers and the transcr ipt ional start s igna ls . T h e R N A cap site and polyadenlyat ion site is located at the beginning and end of the R region respectively. 31 Figure 1.6. Cr i t ical features of the retroviral l i fecycle. Infectious v i r ions enter a target cel l through interaction with specif ic receptors on the cel l sur face whereupon the viral R N A is reverse t ranscr ibed into a c D N A copy. The viral c D N A undergoes integration into the genome of the host cell and a genomic length R N A molecu le is p roduced and p a c k a g e d into viral par t ic les c o m p o s e d of v iral s t ructural a n d enzymat ic proteins. The virion buds from the sur face of the host cel l and the cyc le repeats itself. interaction with speci f ic cel lular receptors, and the ability to exp ress the transferred g e n e at high leve ls in the appropr iate ce l l type fo l lowing integrat ion into the genome of the host cel l . 1.6.2.1. Production of helper-free replication-defective retroviruses The ability to produce replication incompetent recombinant retroviruses free of any contaminat ing helper, or replication competent retrovirus is essent ia l to both gene therapy and gene marking studies. The p resence of repl icat ion competent retrovirus can result in the cont inuous production of infectious virus part ic les from target ce l l s produc ing increas ing ly numerous and comp lex mark ing pat terns in 32 c lonal ana lys is , and has been shown to be assoc ia ted with lymphocyt ic leukemia in a non-human primate gene therapy trial (194). The production of replication-defective retroviruses can be ach ieved through the rep lacement of the viral g a g , pol and env g e n e s with a part icular g e n e of interest us ing s tandard genet ic engineer ing procedures and p lasmid const ruc ts . The recombinant retroviral vector is then introduced into a packag ing cel l line using ca l c ium phosphate precipitat ion. A packag ing cel l line is a f ibroblast der ived line into which has been stably introduced the gag , pol and env g e n e s of a wi ld type retrovirus and thus, is the source of the essent ia l retroviral proteins. However , these viral s e q u e n c e s are eng ineered so that although they cont inuously p roduce all of the required viral proteins, their R N A is unable to be p a c k a g e d . T h u s , genomic length R N A mo lecu les der ived from the recombinant retroviral vec tor c o m b i n e s with the viral proteins and buds from the packag ing cel ls resulting in the re lease of infect ious recombinant retroviral part icles capab le of one round of infect ion. Upon entry into a target cel l and integration into the host g e n o m e , the recombinant retrovirus is unable to undergo further rounds of replication due to the a b s e n c e of the required viral genes . The provirus is stably propagated in the target cel l and the progeny of the target cel l and e x p r e s s e s the t ransferred gene of interest (see Figure 1.7). The first packag ing cel ls l ines (\|/2 (195), \ | /AM (196) and P A 3 1 7 (197)) were cons t ruc ted by delet ing approx imate ly 350 nuc leot ides of the 5' region of the retrovirus which conta ined the \|/ packaging signal (\j/2, \|/AM) as well as the 3' L T R and a portion of the 5' L T R (PA317) . However, replication competent v i ruses cou ld be genera ted through rare recombinat ion events dur ing reverse t ranscr ipt ion in which the regions deleted could be corrected for by intact s e q u e n c e s obtained from the recombinant retroviral vector. Today, third generat ion packag ing l ines ( G P + E 8 6 (198, 199) and \ | / C R E (200)) are avai lab le in which the viral g a g , pol and env 3 3 Retroviral Packaging Sequences Recombinant retroviral vector T A R G E T T E D G E N E Figure 1.7. P a c k a g i n g ce l l s l ines cont inual ly p roduce the g a g , pol a n d env proteins but viral R N A is unable to be packaged due to delet ion of the \|/ s i g n a l s e q u e n c e and separat ion of the gag/pol and env g e n e s onto different p lasmids . Introduction of a funct ional recombinant retroviral vector de le ted of viral g e n e s results in the production of genomic length recombinant R N A which are p a c k a g e d into vir ion part ic les and bud from the packag ing ce l l . T h e re lease of in fect ious recombinant retroviral part ic les are subsequen t l y ab le to undergo 1 round of infection and integration into a host target cel l . 34 g e n e s are separa ted onto two separate p lasmid const ructs with gag and pol on one and env on the other. T h e s e l ines require three recombinat ional events for the product ion of helper virus to occur thus greatly decreas ing the l ikel ihood of this event. O n e drawback of using the lines ment ioned above is that the product ion of s tab le , h igh titre viral p roducer ce l ls can take seve ra l w e e k s . M o r e recent ly , ecotropic (able to infect mouse and rat cel ls only) ( B O S C ) and amphotropic (able to infect m o u s e , human a s wel l a s other cel l t ypes) (B ING) p a c k a g i n g ce l l l ines der ived from a highly transfectable 293T cel l line have been produced enabl ing the rapid, transient production of high titre viral supernatants free of helper virus within 72 hours (201) . 1.6.3. Optimization of retroviral gene transfer 1.6.3.1. Retroviral infection strategies T h e eff ic iency of infection of H S C s is dependent upon a large number of factors including their cycl ing status, the addition of exogenous hemopoiet ic growth factors and the method of infection. A s prev ious ly ment ioned, success fu l retroviral infect ion requi res that the target cel l be in a state of active cyc le (191, 192). Unfortunately, the vast majority of H S C s are in a state of q u i e s c e n c e under normal phys io log ic cond i t ions (93). Fol lowing the injection of the cyc le speci f ic cytotoxic drug 5 - F U , however, a larger proport ion of H S C s are actively cycl ing (93) and are more suscept ib le to retroviral infection (202, 203). A l though H S C s can surv ive for a per iod of t ime and be suscep t i b l e to retroviral t ransduct ion in the a b s e n c e of exogenous ly added hemopo ie t ic growth factors (204) , the infectability of day 12 C F U - S was found to be greatly inc reased in the p resence of IL-3 (24, 205) . Moreover , var ious combinat ions of IL-3, IL-6, S tee l factor, IL-1 and leukemia inhibitory factor (LIF) were found to e n h a n c e retrovirus media ted gene transfer into murine as wel l a s human and non-human 3 5 primate H S C s (124, 206-212). In addit ion to bringing cel ls into a state of act ive cel l cyc le growth factors may a lso play a role in increasing the number and/or affinity of receptor mo lecu les on target ce l ls (213). A l though growth factors i nc rease the eff ic iency of gene transduct ion, caut ion must be used s ince the culture of marrow ce l ls in the p resence of growth factors for ex tended per iods of t ime in vitro may impair the engraft ing ability of the H S C s upon transplantat ion into irradiated hosts (214). The method of exposure of the retrovirus to the target marrow ce l ls a lso has a large effect upon the eff iciency of viral t ransfect ion. The most success fu l results using murine bone marrow cel ls as targets have been ach ieved by co-cultur ing the viral p roducer and marrow ce l ls together (215-217). Wh i le ear ly results us ing a non -human pr imate model favored the use of fi ltered viral preparat ions f ree of producer cel ls (218) (219-221), more recent studies have shown that the eff iciency of supernatant infection of primitive human hemopoie t ic ce l ls c a n be i nc reased through the use of an au to logous s t romal support layer dur ing the infect ion procedure (222-224). The stromal layer most likely inc reases the eff ic iency of viral infection v ia the production of var ious defined and/or undef ined growth factors and v ia cel l -cel l mediated interaction. Interestingly, Moritz et. a l . have reported that the p resence of the extracel lular matrix molecule fibronectin can increase the eff iciency of supernatant infection of commit ted progenitor cel ls and primitive ce l ls with long term culture initiating ability (LTC- IC) (225). It has been hypo thes ized that this mo lecu le is s o m e h o w ab le to bring both the target cel l and retrovirus into c l ose proximity thus increasing the probability of retroviral infection. Recent ly , a method by wh ich retroviral supernatant is a l lowed to f low through a porous m e m b r a n e upon which the target cel ls lie has been descr ibed (226). Whi le normal supernatant infection rel ies upon Brownian motion for vir ions to come into contact with a target cel l (a very inefficient p rocess) , the f low-through t ransduct ion sys tem o v e r c o m e s this limitation by "flowing" all virus particles past the target cel ls result ing in a much 36 higher probabil i ty of contact between virus and cel l . Th is sys tem c a n lead to high transfect ion ef f ic iencies even at low viral titre and can be done in the a b s e n c e of po lycat ions such as protamine sulfate or polybrene which have been historical ly uti l ized as methods to increase gene transfer efficiency. 1.6.3.2. Targeting of virions to host cells T h e transduct ion of murine H S C s using ecotropic retroviral vectors carry ing a variety of genes (ie. n e o ^ (23, 35, 204, 205), adenos ine deaminase (215-217), p-globin (227) and g lucocerebros idase (228-230)) has, in genera l , been found to be much more efficient than gene t ransfer into H S C s from larger an ima ls us ing amphot rop ic v i ruses . O n e current limitation to retroviral med ia ted gene transfer into the bone marrow of mice or larger an imals is that the des i red target ce l l , the totipotent H S C , is both rare and largely non-cycl ing (93). Us ing s tandard infection protocols Luskey et. a l . has est imated the eff ic iency of gene transfer to mur ine H S C s to be 2 0 % us ing Sou the rn blot a n a l y s i s of D N A ob ta ined f rom the hemopo ie t i c t i ssues of recipients t ransplanted with retrovirally t r ansduced bone mar row at 4 months post t ransplant as m e a s u r e d by proviral integrat ion and g e n o m e copy number ana lys is (206). Moreover , E inerhand et. a l . have est imated the eff ic iency of gene transfer to marrow repopulating ab i l i t y -CFU-S ( M R A - C F U - S ) which is a cel l that homes to the marrow of the recipients fol lowing transplant and pro l i fe ra tes to p roduce n u m e r o u s day 1 2 - C F U - S ce l l s . Th i r teen d a y s post transplant the recipient mouse is sacr i f iced and its bone marrow used to a s s a y for day 12 C F U - S by injection into secondary recipients. Thus , through the ana lys is of indiv idual day 12 sp leen co lon ies , E ine rhand and his co l l eagues es t imated the eff iciency of gene transfer to these primitive cel ls to be 1 5 % (207). The eff iciency of gene transfer to H S C s der ived from larger an imals such as non-human pr imates (219, 221 , 231), fel ines (232), can ines (233-235) or humans (236, 237) appears to be cons ide rab l y lower. Quant i ta t ive P C R per fo rmed on D N A ob ta ined f rom per ipheral b lood and bone marrow cel ls at a minimum of one year post transplant 37 sugges ted that the eff iciency of gene transfer to H S C s from larger an imals is only 0 . 1 % to 5%. Th is low gene transfer eff iciency may in part be due to the express ion of low levels of the amphotropic receptor Ram-1 (188). Severa l invest igators have at tempted to b y p a s s this prob lem through the use of viral pseudo typ ing wh ich invo lves rep lac ing the amphot rop ic enve lope gene with an env g e n e f rom a different virus whose receptor is expressed at a much higher density on the des i red target ce l l . Comb in ing the Mo loney murine leukemia virus gag and pol g e n e s with the env g e n e der ived from the g ibbon a p e l eukemia v i rus (238-240) or the ves icu la r stomati t is virus G glycoprotein ( V S V - G ) (241, 242) has resul ted in an inc rease in gene transfer to human cel l l ines, commit ted human progenitor ce l ls and L T C - I C . S o m e invest igators have attempted to target retroviral infection to spec i f ic cel l types by creat ing a hybrid env gene encod ing a portion of the viral env protein and a l igand recogn ized by a speci f ic receptor on the sur face of the target ce l l . U s i n g this app roach , retroviruses have been targeted to spec i f i c ce l l s v ia the erythropoietin receptor (243), the low density l ipoprotein receptor (244), ep idermal growth factor receptor (245) as well as major histocompatibi l i ty c l ass I and c l a s s II ant igen receptors (246). Al though this method does show promise, caut ion must be used s ince the fusion of the viral env with a foreign molecu le may result in a non-functional enve lope protein and thus a non-infectious virus particle. 1.6.3.3. U s e of se lec tab le markers to i n c r e a s e the util ity of recombinant retroviral vectors To aid in the identif ication, enr ichment and t racking of t r ansduced target ce l ls a variety of se lectable markers have been incorporated into retroviral vectors. T h e most widely used of these have been genes which confer res is tance to toxic c o m p o u n d s such as neomyc in and hygromyc in . A variety of o thers s u c h a s p-ga lac tos idase and a number of cell surface ant igens have a lso been uti l ized. Tab le 1.1 shows the var ious genes that have been utilized as se lectab le markers to date. 38 The ability to identify and select for retrovirally t ransduced H S C s rapidly and non-toxical ly cou ld enhance the power of marking s tud ies, efforts to a s s e s s the effects of overexpress ing putative H S C regulatory molecu les and current efforts at human gene therapy. However , at the time that the work descr ibed in this thes is w a s ini t iated, the use of g e n e s encod ing cel l su r face an t igens a s se lec tab le markers for the select ion of retrovirally t ransduced in vivo repopulat ing H S C s had not been reported. A s previously ment ioned, the work descr ibed in Chap te r 3 of this thes is descr ibes the novel use of the human cel l sur face ant igen C D 2 4 a s a dominant se lectab le marker in a retroviral vector to enab le the efficient se lect ion of t ransduced murine B M cel ls including those with totipotent long term repopulat ing ability. 1.6.3.4. Retroviral vector design S e v e r a l factors can inf luence the per formance of any part icular retroviral construct including the regulatory e lements used to drive express ion , the number a n d s i z e of t ranscr ip t iona l units, the viral b a c k b o n e u s e d , the d i rec t ion of transcription and the presence or absence of selectable markers. Al though l ineage restricted promoters may be ideal if the transferred gene is to be e x p r e s s e d exc lus ive ly in part icular cel l types (247, 248) , viral L T R s have consistent ly been shown to result in higher levels of gene express ion as compared to a variety of internal promoters of viral or cel lular origin (249, 250). Unfortunately, viral L T R s s u c h a s the Mo loney Mur ine L e u k e m i a V i rus ( M o M u L V ) have been found to be subject to transcript ional shutdown fol lowing long per iods of t ime in v ivo (215, 251) and in primitive cel l types such a s embryonic ca rc inoma (EC) ce l ls (252-254), embryona l s tem (ES) ce l ls (255) and primitive hemopoie t ic cel l l ines (256) poss ib ly due to methylat ion of the regulatory e lements (257). S e v e r a l viral mutants have been isolated that are able to exp ress t ransferred g e n e s at high levels in E C and E S cel ls . The P C M V ( P C C 4 embryonal ca rc inoma ce l l - passaged myelopro l i ferat ive s a r c o m a virus) (258) and the d l587rev v i rus (259) p o s s e s s 39 Table 1.1. Opt ions in Se lec tab le /Repor ter Markers for Retroviral Vec to rs Marker G e n e S i z e in bp* Methods of Quant i f icat ion of Se lec t i on G e n e Exp ress ion at S ing le Ce l l Reso lu t ion n e o m y c i n 0 (260, 261) 8 0 0 h y g r o m y c i n R (262) 1200 puromyc in 7 0 0 methot rexateR(263,264) 7 0 0 cytos ine deaminase 1600 (265) MDR-1 (266) 3 8 0 0 thymidine k inase (267) 1400 p-ga lactos idase (268, 3 ° ° 0 269) a lkal ine phospha tase 4 0 0 0 (270) Leu-1 (271) 2 3 0 0 transferrin receptor (271) 2 8 0 0 MDR-1 (272) 3 8 0 0 truncated nerve growth 1500 factor receptor (273) IL-2 receptor (274) 1400 Heat Stable Ant igen 2 2 8 (HSA) (275) mutated murine prion 1200 protein (276) G r e e n F luo rescence 7 2 0 Protein ( G F P ) (277) C D 2 4 (278) 2 4 0 Thy-1 (279) 4 8 8 drug res is tance no F A C S i m m u n o - b a s e d ( F A C S , panning) y e s 1 y e s * Approx imate s ize of c D N A encompass ing coding region 4 0 var ious delet ions and base pair mutat ions which were found to remove severa l transcript ional b locks located within the L T R and primer binding si tes of the original v i ruses (280-282). It is the removal of these transcript ional b locks that enab les the express ion of transferred genes under the control of these regulatory e lements in both E C and E S cel ls . Recent ly , Hawley et. a l . (283) have p roduced a se r ies of vectors b a s e d upon a Murine S tem Ce l l V i rus ( M S C V ) backbone which comb ines the L T R from P C M V , the 5' untranslated region from the d l587rev v i rus, and a number of convenient c loning si tes. Data will be presented in this Chap te r 4 of this thes is to suggest that the M S C V vectors are able to t ranscr ibe t ransduced g e n e s efficiently in primitive primary hemopoiet ic s tem cel l cand ida tes and their progeny for extended per iods of t ime in vivo. P l a c i n g a gene of interest under the regulatory contro l of an internal p romoter downs t ream of the viral L T R regulatory s e q u e n c e s may resul t in exp ress ion prob lems due to promoter interference between the viral and internal regulatory e lements . Indeed, the express ion of internally dr iven g e n e s has been found to be unpredictable. In s o m e c a s e s efficient exp ress ion from the internal promoter w a s detected (172, 284) while in other c a s e s little or no express ion w a s o b s e r v e d (216, 285-287) . A s a m e a n s to b y p a s s this potent ial p rob lem self-inactivating or "cr ippled" vectors can be used. T h e s e vectors p o s s e s s a delet ion of a portion of the 3' L T R and are des igned such that fol lowing infection of the target cel l the viral enhancer and promoter regulatory e lements are non-funct ional (288-290) . T h e number of transcript ional units present within the retroviral vector can a lso inf luence the eff iciency of transferred gene express ion . Al though s o m e studies have sugges ted that super ior results are obtained using a simpl i f ied vector wh ich con ta ins only one t ranscr ipt ional unit (215, 291) , o thers have found that the inc lus ion of an addi t ional gene encod ing a dominant se lec tab le marker in the vector can be beneficial (249, 292, 293). However, s o m e studies have reported that 41 cel ls se lec ted for on the bas is of express ion of the se lectab le marker can result in the supp ress ion of the adjacent gene , most l ikely due to promoter inter ference between internal and viral regulatory e lements in mult i -gene vectors (274, 294). A solut ion to this problem has been to express both genes from a s ingle regulatory e lement by l inking them together using an internal r ibosomal entry si te ( IRES) e lement. I R E S e lements are discrete fragments of D N A (approximately 500-600-bp in s i z e ) d e r i v e d f rom the 5' u n t r a n s l a t e d r e g i o n s of the po l i o or encepha lomyocard i t i s v i rus (295, 296). I R E S e lements al low r i bosomes to bind internally on the m R N A rather than at the 5' cap site as normal ly occurs and thus, enab les more than one transcript ional unit to be t ranslated from a s ing le m R N A mo lecu le . T h e s e e lements have been shown to link up to three separa te g e n e s thus al lowing the regulation of multiple transcriptional units from a s ingle regulatory e lement (297, 298). 1.7. Thes is objectives and general strategy A l though recombinant re t rov i ruses current ly remain the most eff ic ient me thod for in t roducing e x o g e n o u s genet ic mater ia l into target ce l l s , the low infect ion ef f ic iency of H S C s represents a ser ious hurdle to effect ive retroviral m a r k i n g s t u d i e s a n d g e n e t he rapy . M o r e o v e r , a l t h o u g h g e n e t r ans fe r methodo log ies have provided the crit ical m e a n s to test g e n e s encod ing putative regulatory mo lecu les wh ich may play a role in contro l l ing H S C prol i ferat ion, differentiation and/or sel f -renewal v ia overexpress ion of the transferred gene , at the t ime that the work descr ibed in this thesis w a s initiated, the direct demonstrat ion that any regulatory e lement w a s ab le to dr ive high a n d sus ta i ned leve ls of transferred gene express ion in H S C s had not been ach ieved. The first objective of this thesis work w a s to deve lop procedures that wou ld increase the utility of retroviral gene transfer procedures. T h e s e studies focused on determin ing the feasibi l i ty of uti l izing a c D N A encod ing the human C D 2 4 ce l l sur face antigen as a dominant selectable marker in a retroviral vector to enab le the 4 2 rapid, efficient and non-toxic identification and select ion of retrovirally t ransduced mur ine bone mar row ce l l s , inc lud ing those with tot ipotent long term in v ivo repopulat ing ability (Chapter 3). M y s e c o n d object ive w a s to exploit the se lec t ion protocol d e v e l o p e d in C h a p t e r 3 to demons t ra te the regenera t ion of the hemopo ie t i c s y s t e m s of mye loab la ted recipient mice with cel ls der ived exc lus ive ly from proviral ly marked H S C s and to quantify the levels of express ion of the transferred C D 2 4 marker gene in var ious phenotypical ly def ined populat ions of ce l ls in v ivo including cand ida te H S C s def ined by the Sca+L in" cel l sur face phenotype. The results of these studies (p resen ted in C h a p t e r 4) demonst ra te the use fu lness of the C D 2 4 se lec t i on procedure to enab le effective tracking of the contribution of individual s tem cel ls to the regenerat ion of the H S C and more mature cel l compar tments in myeloab la ted recipient mice fol lowing bone marrow transplant, and to quantify t ransferred gene express ion both in vitro and in vivo. A l though retroviral mark ing s tud ies have prov ided direct ev i dence of the ex is tence of totipotent H S C s , and have provided significant insight into the ability of t hese ce l l s to regenerate the hemopoie t ic s y s t e m s of mye loab la ted t ransplant recipients (ie. their proliferative and differentiative potential), to date quanti tat ive da ta a s s e s s i n g the extent to wh ich these ce l l s can regenera te their numbers fol lowing transplant (ie. self-renew) is limited. The final objective of this work w a s to def ine more c lear ly the regenerat ive (sel f-renewal) capac i ty of H S C s fo l lowing bone marrow transplant. The recovery of totipotent long term repopulat ing s tem cel l numbers in lethally irradiated recipient mice w a s a s s e s s e d a s a funct ion of the number and sou rce (adult bone marrow vs . fetal liver) of ce l l s t ransp lan ted. In add i t ion, us ing the C D 2 4 se lec t ion p rocedure (Chapter 3) the contr ibut ion of individual H S C c lones to the regenerat ion of the H S C compartment w a s a s s e s s e d in s o m e exper iments. The results of this work are presented in Chap te rs 4 and 5. 4 3 C H A P T E R 2 MATERIAL A N D M E T H O D S 2.1. Const ruc t ion of retroviral vectors , v i rus product ion and viral a s s a y s 2.1.1. Recombinant retroviral vectors Exper iments d i s c u s s e d in Chap te rs 3, 4 and 5 uti l ized a retrovirus der ived from the J Z e n l retroviral backbone kindly provided by Dr. S . Cory (Walter and E l i za Ha l l Institute, Me lbourne , Aust ra l ia) . T h e 3' L T R of J Z e n l is der ived f rom the myeloprol i ferat ive s a r c o m a virus (MPSV)(167) . To construct J Z e n C D 2 4 t k n e o , a 310 bp S a l I f ragment contain ing bp 1 to 303 of the publ ished C D 2 4 c D N A s e q u e n c e (299) and e n c o m p a s s i n g the entire 240 bp cod ing region w a s r e m o v e d f rom P A X 1 1 4 (300) and inser ted into the X h o I si te of J Z e n t k n e o us ing s tanda rd procedures. JZen tkneo was constructed by inserting into the H p a I - Hind III s i tes of J Z e n l a 1092 base pair S m a I- Hind III fragment from pTZ19R tkneo that conta ins the n e o R gene l inked to a mutant po l yoma virus enhance r tandem repeat and H e r p e s S i m p l e x virus thymidine k inase gene promoter iso lated f rom p M C I n e o (301). M S C V n e o l R E S C D 2 4 , used in exper iments p resen ted in C h a p t e r 4 , w a s cons t ruc ted us ing the M S C V n e o E B vector (283) (kindly prov ided by Dr. Rober t Hawley , Sunnybrook Heal th S c i e n c e Center , Toronto, O N ) original ly der ived from the M E S C retroviral vector of G r e z et. al . (302) and the L N retroviral vectors of Mil ler et. a l . (303). P G K n e o was removed from M S C V n e o E B by digest ing with Bg l l l /BamHI and religating to create M S C V ( - ) . A 948-bp EcoR I /Xho l f ragment from a previously desc r i bed construct (304) e n c o m p a s s i n g the encepha lomyocard i t i s v i rus internal r ibosomal entry site ( IRES) sequence and the 240-bp cod ing region of the human C D 2 4 cel l sur face ant igen c D N A (299) w a s l igated into an E c o R I / X h o l d iges ted M S C V ( - ) . Last ly, a 847-bp blunted Mlu l /Sa l l fragment der ived from p M C I n e o (301) 4 4 conta in ing a neomyc in res is tance gene w a s inserted by blunt end ligation into the EcoRI site of M S C V I R E S C D 2 4 to create M S C V n e o l R E S C D 2 4 . 2.1.2. Viral packaging and other cell lines T h e ecotropic packag ing cel l line, G P + E - 8 6 (199), and the amphotrop ic cel l l ine, G P + A M 1 2 (198), were used to generate helper- f ree recombinant retrovirus. T h e cel l l ines were maintained in H X M medium c o m p o s e d of Du lbecco ' s Modi f ied E a g l e s M e d i u m ( D M E M ; S t e m C e l l Techno log ies , V a n c o u v e r , Bri t ish Co lumb ia ) , 1 0 % heat - inac t iva ted ( 5 5 ° C for 30 minutes) newborn calf s e r u m ( G i b c o / B R L C a n a d a ; Bur l ington, Ontar io), hypoxanthine (15 mg/ml ; S i g m a C h e m i c a l C o . , St . Lou i s , M O ) , xanth ine (250 mg/ml ; S igma) , and mycopheno l i c ac id (25 mg /m l ; S i g m a ) . T h e IL -3 -dependent mur ine hemopo ie t i c ce l l l ine B a / F 3 (305) w a s ma in ta ined in R P M I with 1 0 % fetal calf s e r u m a n d 5 % m o u s e s p l e e n ce l l condi t ioned med ium ( S C C M ) (S temCe l l Techno log ies ) . A l l ce l ls were cul tured at 3 7 ° C in a humidif ied atmosphere of 5 % C O 2 in air. 2.1.3. Generation of viral producer cell lines Pur i f ied proviral p lasmid D N A was introduced into the G P + A M - 1 2 packag ing ce l l l ine us ing the ca l c ium phospha te ( C a P 0 4 ) t rans fec t ion t echn ique . D N A precipitate w a s formed by combin ing 18 u,g of purif ied D N A , 50 uJ of 2.5 M C a C l 2 a n d d H 2 0 up to 0.5 ml, which w a s slowly added to 0.5 mL of 2 X H B S (50mM H e p e s , 3 M NaCI , 1.5 m M N a 2 H P 0 4 , pH 7.12) while gently bubbl ing with air to mix. T h e solut ion w a s a l lowed to s tand at room temperature for 30 minutes and then added dropwise to 9 ml of medium on 2 x 1 0 5 G P + A M - 1 2 packag ing ce l ls plated in a 100 m m t i ssue culture d ish (Beckton D ick inson , L inco ln Pa rk NJ ) 24 hours previously. After 24 hours the medium was replaced with fresh med ium containing 1 mg/ml (approximately 0.6 mg/ml act ive compound) of the neomyc in ana log G 4 1 8 ( G i b c o / B R L ) . Med ium w a s rep laced every three days . M e d i u m w a s p l aced on d i shes of confluent G 4 1 8 R G P + A M - 1 2 cel ls and 24 hours later the supernatant w a s removed, filtered (0.22u.m filter, Mil l ipore, Bedford, MA) and over layed onto 1 x 1 0 5 4 5 G P + E - 8 6 ecot rop ic viral packag ing ce l ls in the p resence of 7 u,g/ml po lybrene (Sigma). Twenty four hours later the medium was removed and rep laced with fresh med ium containing 1 mg/ml G 4 1 8 . 2.1.4. Viral titering and helper virus assay Vi ra l t i tres we re de te rmined by a s s a y i n g med ium cond i t i oned by v i ra l p roducer cel l l ines for transfer of neomyc in res is tance to N IH-3T3 ce l ls (Amer ican T y p e Cu l tu re Co l l ec t i on [ A T C C ] , Rockv i l l e , M D ) . M e d i u m w a s p l a c e d a top subcon f l uen t 1 0 0 m m d i s h e s of v iral p roducer ce l l s a n d 24 hou rs later the supernatant w a s removed and fi l tered. Var ious di lut ions of the viral supernatant were p laced in a final vo lume of 2 mis and p laced on top of 2 x 1 0 5 3 T 3 cel ls plated in 60 mm t issue culture d ishes (Becton Dickinson) with 7 | ig/ml po lybrene a d d e d . Twenty four hours later the supernatant was replaced with fresh med ium contain ing 1 mg/ml G 4 1 8 . Med ium w a s rep laced every three days . G 4 1 8 R co lon ies were sco red after staining with methy lene blue to derive the number of infect ious virus part ic les carry ing n e o R generated by the viral producers (colony forming units per ml). A s s a y for the p r e s e n c e of he lper v i rus w a s per formed by at tempt ing to serial ly transfer n e o R to 3T3 cel ls (195). Conf luent d ishes of G 4 1 8 R 3 T 3 cel ls were ob ta ined fo l lowing infect ion with 5 mis of f i l tered undi luted viral superna tan t fo l lowed by G 4 1 8 select ion as descr ibed above. M e d i a w a s c h a n g e d on conf luent d ishes of G 4 1 8 R 3T3 cel ls and 24 hours later the medium was removed, filtered and p laced on top of 2 x 1 0 5 3T3 cel ls plated out 24 hours previously. Twenty four hours later the med ium w a s removed and replaced with 5 mis of f resh med ium contain ing 1 mg/ml G 4 1 8 . Co lon ies were subsequent ly scored 2-3 w e e k s later a s desc r ibed above . 2.2. Hemopoietic cell culture and assays 2.2.1. Mice 46 M i c e u s e d in these exper imen ts were 8 to 12 w e e k o ld ( C 5 7 B I / 6 J x C 3 H / H e J ) F 1 ( B 6 C 3 F 1 ) male and female mice bred and mainta ined in the an imal faci l i ty of the Br i t ish C o l u m b i a C a n c e r R e s e a r c h C e n t r e f rom parenta l stra in breeders originally obtained from the J a c k s o n Laborator ies (Bar Harbor, M A ) . M ice used a s bone marrow donors in competi t ive repopulat ion exper iments were 10 to 14 w e e k old male or female (C57B I /6Ly -Pep3b x C 3 H / H e J ) F 1 ( P e p C 3 F 1 ) m ice . B 6 C 3 F 1 and P e p C 3 F 1 mice are phenotypica l ly d is t ingu ishable on the bas i s of a l le l ic d i f fe rences at the Ly5 ce l l su r face ant igen l ocus ; B 6 C 3 F 1 m ice are homozygous Ly5.2 and P e p C 3 F 1 mice are Ly5.1 /Ly5.2 heterozygotes. Al l an ima ls were h o u s e d in micro isolator c a g e s and provided with steri le food and acid i f ied sterile water. 2.2.2. Viral infection of bone marrow cells and cell lines B o n e marrow cel ls from adult male or female B 6 C 3 F 1 or P e p C 3 F 1 mice injected 4 days previously with 5-fluorouracil in sterile phosphate-buf fered sa l ine (5-F U , 150 mg/kg body weight; Ho f fman -LaRoche Ltd. , M i s s i s s a u g a , Ontar io) were f l u s h e d f rom femora l sha f ts with a l p h a m e d i u m a n d 5 % F C S ( S t e m C e l l Techno log ies ) . For co-cul ture infect ions 3 x 1 0 ^ marrow ce l ls were incubated on a conf luent mono layer of irradiated (1500 c G y X-rays) C D 2 4 viral p roducer ce l ls for 24 to 48 hours in med ium c o m p o s e d of D M E M , 1 5 % F C S , 10 ng/ml h u m a n interleukin-6 (IL-6), 6 ng/ml murine IL-3, 100 ng/ml murine Stee l factor, and 7 mg/ml po lybrene (S igma) . A l l growth factors were used as di luted superna tan ts from t rans fec ted C O S ce l l s p repared in the Terry F o x Labora tory . C e l l s u s e d for compet i t ive repopulat ion exper iments were incubated for 48 hours in the a b o v e med ium in the a b s e n c e of polybrene prior to co-culture with viral p roducer ce l ls in an effort to enhance gene transfer to the most primitive ce l ls (124, 217) . Loose ly adherent and non-adherent cel ls were recovered by gent le agitation and wash ing of d i shes with D M E M and 1 5 % F C S . C e l l s were pel le ted, r e s u s p e n d e d in f resh 4 7 culture medium and incubated for a further 48 hours at 3 7 ° C to al low for express ion of the transferred C D 2 4 gene. T h e mur ine ce l l l ine B a / F 3 w a s infected by e x p o s u r e to f i l tered vi ra l superna tan t f rom the J Z e n C D 2 4 t k n e o ecot rop ic viral p roduce r ce l l l ine. V i ra l supernatant w a s supp lemented with 7 u.g/ml polybrene. C e l l s were then plated in methylcel lu lose (descr ibed below) in the presence of G 4 1 8 (1 mg/ml). Independent G 4 1 8 R co lon ies were picked and expanded in the p resence of G 4 1 8 to be used as posit ive controls for F A C S and Southern blot analys is . 2.2.3. In vitro c lonogenic progenitor assay Sor ted and unsorted bone marrow ce l ls were plated in 3 5 m m petri d i shes (Stem Ce l l Techno log ies , Vancouver , Brit ish Co lumbia) in 1.1 ml culture mixtures conta in ing 0 .8% methy lce l lu lose in a lpha med ium supp lemen ted with 3 0 % F C S , 1% bovine se rum albumin (BSA) , 1 0 " 4 M (3-mercaptoethanol, 3 U/ml human urinary erythropoiet in (100,000 uni ts/mg), 2 % S C C M and 1 0 % agar -s t imu la ted human l eukocy te cond i t i oned m e d i u m , al l of wh ich w e r e ob ta ined f rom S t e m C e l l T e c h n o l o g i e s Inc.. C e l l s were plated in the p resence or a b s e n c e of 1.5 mg/ml (approximately 0.9 mg/ml active compound) of G 4 1 8 (G ibco /BRL) and incubated at 3 7 ° C in 5 % C O 2 . Large single and mult i- l ineage co lon ies were sco red after 8-14 days incubation accord ing to standard criteria (19). 2.2.4. C F U - S assay Lethal ly irradiated B 6 C 3 F 1 mice (910-950 c G y , 110 c G y / m i n , 1 3 7 C s y-rays) we re in jected in t ravenous ly with 1 x 1 0 ^ to 1 x 1 0 4 ce l l s f rom ind ica ted ce l l f ract ions. Twe lve days later, an imals were sacr i f iced v ia cerv ica l d is locat ion and wel l isolated macroscop ic sp leen co lon ies individually d issec ted and s u s p e n d e d for flow cytometric and D N A analys is . 2.2.5. Bone marrow transplantation and quantification of C R U In the exper iments desc r i bed in C h a p t e r s 3 and 4 , 1 x 1 0 4 to 4 x 1 0 ^ retroviral ly t ransduced C D 2 4 + se lec ted or unse lec ted bone marrow ce l ls der ived 4 8 f rom P e p C 3 F 1 (Ly5.1/I_y5.2) donors were in t ravenously in jected into i r radiated (137Cs y-rays, 950 c G y , 110 cGy/min) B 6 C 3 F 1 (Ly5.2) recipients with or without a l i fesparing dose of Ly5.2 competi tor bone marrow cel ls; either 1 x 1 0 5 marrow ce l ls from a normal mouse or 2 x 1 0 5 marrow cel ls from a compromised animal (59). The funct ion of these latter ce l ls is to ensure the short term surv ival of the recipient fo l lowing the irradiation procedure . The level of reconsti tut ion of rec ip ients with donor (Ly5.1) cel ls and express ion of the transferred C D 2 4 gene w a s a s s e s s e d at 5 to 32 w e e k s post- t ransplantat ion by f low cytometr ic ana lys i s of per ipheral b lood samp les (50-100 JLLI) obtained by tail vein puncture. For the exper iments descr ibed in Chap te r 5 fetal l ivers were removed from day 14.5 embryos obta ined from t imed mat ings of C57B I /6Ly -Pep3b (Ly5 .1 ) ma le a n d C 3 H / H e J ( L y 5 . 2 ) f ema le m ice . C e l l s we re s u s p e n d e d in a l p h a m e d i u m contain ing 5 % F C S (StemCel l Technolog ies) by repeated gentle aspirat ion through 5 ml, 2 ml and 1 ml pipettes fol lowed by 18- and 2 1 - gauge need les . B o n e marrow ce l l s f rom ma le or fema le P e p C 3 F 1 mice injected 4 days prev ious ly with 5-f luorouraci l ( 5 - F U ; 150 mg/kg body weight) were f lushed from femoral shaf ts with a lpha med ium and 5 % F C S and a single cel l suspens ion similarly obta ined. C e l l s w e r e c o u n t e d us ing a s tandard hemocy tomete r . C e l l surv iva l w a s >98% a s determined by trypan blue exc lus ion . Fetal l iver ce l ls or post 5 - F U bone marrow cel ls from Ly5.1+ donors were injected in combinat ion with a l i fesparing dose of 1 0 5 bone marrow cel ls from normal 8-12 week old B 6 C 3 F 1 (Ly5.2+) mice (59) into the tail ve in of recipient B 6 C 3 F 1 (Ly5.2+) m ice prev ious ly i r radiated with 9 5 0 c G y (110cGy /m in , 1 3 7 C s y-rays). C R U were measured by injecting groups of 7 lethally i rradiated B 6 C 3 F 1 (Ly5.2+) recipients in combinat ion with 1 0 5 syngen ic (Ly5.2+) normal bone marrow ce l ls and a s s e s s i n g the recip ients 16 w e e k s later for the p resence of Ly5.1+ lymphoid and myeloid cel ls in their per ipheral b lood. Rec ip ient mice were cons idered posit ive if >1% of each of myelo id and lymphoid per ipheral b lood cel l populat ions (identified by their unique forward and s ide scat ter prof i les, 49 respectively) demonstrated the donor Ly5.1 cel l sur face phenotype. C R U f requency w a s calculated by determining the number of negative recipients as a function of the number of test ce l ls injected and apply ing P o i s s o n stat ist ics us ing the method of least l ikel ihood (306, 307). 2.3. Molecular analysis 2.3.1. Southern Blot Analysis D N A w a s pur i f ied f rom N a D o d S 0 4 / p r o t e i n a s e K - d i g e s t e d c e l l s by phenol /ch loro form extract ion (308) or us ing the D N A z o l reagent ( C a n a d i a n Li fe Techno log ies , Burl ington, Ontario). D N A was d ia lyzed for 16 hours against 1 x T E (10 m M Tris pH7.5 , 1 m M E D T A pH 8.0) buffer and 10-20 ug d igested with X b a l , Sst l or EcoRI (G ibco /BRL) at 3 7 ° C for 12-16 hours. Fol lowing ethanol precipitat ion, D N A w a s d isso lved in 20 uJ of 1 x T E buffer and separated on a 0 .8% agarose ge l . G e l s were then treated for 35 minutes with Solut ion I (0.5M N a O H , 1.5 NaCI) and for 35 minutes with Solut ion 2 (1M Tr is pH 7.0, 2 M NaCI). D N A was then transferred to a nylon membrane (Zeta-Probe; B i o - R a d Laborator ies, R ichmond C A ) in 10X S S C by s tandard blotting methods . Blots were prehybr id ized at 6 0 ° C for 2 hours in 4 . 4 X S S C , 7 . 5 % formamide, 0 .75% S D S , 1.5mM E D T A , 0 .75% sk im milk, 370 mg/ml of sa lmon spe rm D N A . Blots were then hybr id ized for 20 hours at 6 0 ° C under the s a m e cond i t ions a s above with the inc lus ion of 7 . 5 % dextran sul fate (S igma) . M e m b r a n e s were separate ly probed using a X h o l / S a l l f ragment of p M C I n e o (301) con ta in ing n e o R spec i f ic s e q u e n c e . P robes were labeled with 3 2 p - d C T P (3000 C i / m m o l ; Amersham) by random priming and purified on a S e p h a d e x - G 5 0 co lumn before hybr idizat ion. Membranes were subsequent ly w a s h e d twice at 6 0 ° C for 30 minutes e a c h in 0 .3X S S C , 0 . 1 % S D S and 1 mg/ml of sod ium py rophospha te . Autorad iography was performed with Kodak X A R - 5 film and an intensifying sc reen at - 7 0 ° C for 1-10 days . Membranes were str ipped for re-probing by boi l ing in 1% S D S , and w a s h i n g for 40 minutes . B lo ts were re -probed with the K p n l / M s e l 50 f ragment of p X M ( E R ) - 1 9 0 which re leases the full length erythropoiet in receptor c D N A (kindly provided by A . D 'Andrea, Dana-Farbe r C a n c e r Institute, Bos ton , MA) for an internal control for D N A loading. Densitometr ic ana lys is w a s performed using a phospho- imager with I m a g e Q u a ^ M software (Molecu lar D y n a m i c s , Sunnyva le , C A ) 2.3.2. Ant ibody staining procedures T o a n a l y z e C D 2 4 ce l l su r face ant igen e x p r e s s i o n fo l lowing retroviral infection cel ls were washed once in a lpha medium with 5 % F C S , resuspended (1 to 7 x 1fj6 cel ls/ml) in 0.2 to 0.4 ml of medium condit ioned by hybr idoma 2 . 4 G 2 which secre tes an ant i -mouse IgG F c receptor antibody (309), and incubated on ice for 30 minutes in an effort to reduce non-speci f ic staining. C e l l s were then w a s h e d once with H a n k ' s b a l a n c e d salt so lut ion conta in ing 2 % F C S (HF) . T e t r a m o l e c u l a r comp lexes of monoc lonal ant ibodies were used for the staining procedure (310) by mix ing an t i -CD24 ant ibody 3 2 D 1 2 ( a generous gift from Dr. S Funde rud , O s l o , Norway) , ant i -R phycoerythr in ant ibody ID3 and F(ab')2 f ragments of the anti- IgG ant ibody P 9 (311) at a 1:2:3 molar ratio. S u c h tetrameric comp lexes provide a rapid and f lexible m e a n s of generat ing labeled ant ibody from even smal l quant i t ies of start ing mater ial and are equivalent to directly labeled ant ibody. T h e s e tetrameric ant ibody c o m p l e x e s were used for staining at a final concentrat ion of 0.8 u.g/ml. Ce l l s were incubated on ice for 40 minutes, washed twice with H F , and then sta ined with R-phycoerythr in ( R - P E ) at 2 u.g/ml. After a further 40 minutes on ice, ce l ls were w a s h e d twice with H F and resuspended in H F conta in ing 1 | ig /ml of 7 -amino ac t inomyc in D ( 7 A A D , S igma) to dist inguish dead ce l ls prior to ana lys i s by f low cytometry us ing a F A C S c a n cel l ana lyzer (Becton D ick inson and C o . , S a n J o s e , C A ) . C e l l s used in sort ing exper iments were s ta ined with propid ium iodide (PI, S igma) to dist inguish dead cel ls . Bone marrow cel ls cultured a lone or on G P + E - 8 6 packag ing cel ls were used as negative controls. 51 Repopu la t ion of recipient mice with donor-der ived ce l ls w a s a s s e s s e d by staining peripheral blood cel ls , bone marrow sp leen and thymus cel ls with an F ITC-conjugated anti-Ly5.1 m A b kindly provided by Dr. G . Spang rude (Rocky Mounta in Labora to ry , Hami l ton , MT) and ana l ys i s by f low cytometry . Pe r i phe ra l b lood s a m p l e s were obta ined v ia tail ve in puncture and dep le ted of ery throcytes by incubat ing them for 10 minutes on ice in the p resence of 4 vo lumes of steri le 1M N H 4 C I so lut ion. Thymic and sp len ic cel ls were obta ined by teas ing these o rgans apart and passag ing the ce l ls through an 21 gauge need le in order to obtain a s ingle cel l suspens ion . Bone marrow cel ls were f lushed from femurs and tibia using an 21 gauge need le and a lpha medium and 5 % F C S (Stem Ce l l Techno log ies ) . In s o m e exper iments phenotyp ic ana lys is of Ly5.1 donor der ived per iphera l b lood leukocy tes w a s a c h i e v e d through doub le ant ibody label ing with L y 5 . 1 - F I T C in combina t ion with one of G r - 1 - P E (from hybr idoma R B 6 - 8 C 5 prov ided by Dr. G . Spang rude ) to identify granu locy tes , M a c - 1 - P E (from hybr idoma M 1 / 7 0 , A T C C , Rockv i l le , MD) to identify macrophages , B 2 2 0 - P E (from hybr idoma R A 3 - 6 B 2 , Dr. G . Spangrude) to identify B lymphocytes or L y 1 - P E (from hybr idoma TIB104, A T C C ) to identify T lymphocytes as descr ibed below. Leve ls of express ion of the transferred C D 2 4 gene in ce l ls from repopulated mice were a s s e s s e d by sta in ing ce l ls from hemopoie t ic t i ssues with an t i -CD24 tetrameric ant ibody c o m p l e x e s and R - P E a s d e s c r i b e d a b o v e . C D 2 4 e x p r e s s i o n a m o n g per iphera l b lood l eukocy tes w a s ana l yzed by stain ing per ipheral b lood samp les with a n t i - C D 2 4 / R - P E tet ramers in c o m b i n a t i o n with F I T C l a b e l e d Gr-1 to ident i fy g r a n u l o c y t e s , M a c - 1 for mac rophages , Ly-1 for T cel ls and B220 for B cel ls . C D 2 4 express ion on per ipheral b lood erythrocytes w a s a s s e s s e d by staining per ipheral b lood s a m p l e s prior to exposure to N H 4 C I . The express ion of the transferred C D 2 4 gene on marrow s tem cel l cand ida tes def ined by the S c a + L i n " cel l sur face phenotype w a s a c h i e v e d through mult ip le ant ibody label ing of bone marrow ce l l s with a n t i - C D 2 4 / R - P E 52 tetramers in combinat ion with G r - 1 - F I T C , M a c - 1 - F I T C , B 2 2 0 - F I T C , L y - 1 - F I T C and S c a - 1 - C y - 5 (E13-161.7, Dr. G . Spangrude) . 2.3.3. F A C S sorting Ce l l s were sorted on a FACSta r+ (Beckton Dick inson and C o . , S a n J o s e , C A ) equ ipped with a 5 W argon and a 3 0 m W helium neon laser. Ce l l s were co l lec ted in steri le eppendorf v ials in a lpha medium with 5 0 % F C S . 53 C H A P T E R 3 S E L E C T I O N O F R E T R O V I R A L L Y T R A N S D U C E D HEMOPOIET IC C E L L S USING CD24 A S A M A R K E R O F G E N E T R A N S F E R The results presented in this Chapter have been descr ibed in: Pawl iuk, R., R. Kay , P. M. Lansdorp, and R. K. Humphr ies. 1994. Select ion of retrovirally t ransduced hematopoiet ic cel ls using C D 2 4 as a marker of gene transfer. B lood 84: 2868-2877. Pawl iuk, R., R. Kay , P. M. Lansdorp and R. K. Humphr ies. 1995. C D 2 4 as a marker gene for the select ion and tracking of retrovirally t ransduced stem cel ls . In: Mo lecu lar Bio logy of Hemoglobin Switching. E d s . G . S tamatoyannopou los and A . Nienhuis . Vo l . 2 of the Proceed ings of the 9th Con fe rence on Hemoglob in Swi tch ing, pp. 231-247. 54 3.1. Introduction T h e deve lopment of recombinant retroviruses as vectors for gene transfer h a s p rov ided a power fu l tool to a d d r e s s current ques t i ons regard ing H S C numbers , b io log ica l potent ia l , k inet ics and regulat ion. Mo reove r , recombinan t retroviruses have p layed a pioneer ing role in the field of gene therapy. However , the power of retroviral gene t ransfer is current ly l imited by the poor infect ion eff iciency of H S C s due to their rarity, cycl ing status and paucity of viral receptors. To aid in the identif ication, enr ichment and t racking of t ransduced target cel ls , a variety of selectable markers have been incorporated into retroviral vectors. T h e most widely used of these have been intracellular componen ts which confer res is tance to toxic compounds such as neomyc in (23, 35 , 205), hygromycin (262, 3 1 2 , 3 1 3 ) , c h l o r a m p h e n i c o l (314) , me tho t rexa te (216 , 2 3 4 , 2 6 3 , 3 1 5 ) , mycopheno l i c ac id (316), or var ious chemotherapeut ic agents (266, 317-319) . However , use of these markers in se lect ion protocols carry d i sadvan tages that inc lude non-spec i f ic drug toxicity and difficulties in quantifying exp ress ion levels . T h e bacter ia l (3-galactosidase g e n e ( lacZ) and the h u m a n p lacen ta l a lka l ine phospha tase gene have a lso been emp loyed to se lect t ransduced ce l l s in vitro (268, 269) and a s reporter molecu les both in vitro (268, 270, 320 , 321) and in vivo (322-325) . However , the p resence of an e n d o g e n o u s m a m m a l i a n l y sosoma l (3-ga lac tos i dase and prob lems in ach iev ing adequate levels of exp ress i on of the exogenous p-gal gene have limited its effective use. G e n e s encod ing cel l sur face ant igens have a lso been uti l ized a s se lec tab le markers of gene transfer to f ibroblasts (271, 274, 319) and more recently to human per iphera l b lood l ymphocy tes (326). T h e use of s u c h g e n e s of fers s e v e r a l potentially signif icant advantages including: the rapid and quantitative detect ion of transferred gene express ion in the desired target cell populat ion by f low cytometry; the efficient and non-toxic select ion of t ransduced target ce l ls by F A C S or other immuno-based select ion techniques; and the tracking of t ransduced ce l ls and their 55 progeny both in vitro and in vivo. However , at the t ime that the work descr ibed in this thes is w a s initiated the applicabil i ty of this approach to primitive hemopoiet ic ce l ls had not yet been demonst ra ted. Moreover , the genes which had been used a s se lec tab le markers are relatively large, leaving limited s p a c e in the retroviral vector for other genes of interest. Recen t l y , c D N A s encod ing the human hemopo ie t ic ce l l su r face ant igen C D 2 4 , and its murine homologue the heat stable ant igen (HSA) have been c loned (299, 327). The function of these molecu les is not yet resolved al though C D 2 4 has been assoc ia ted with activation and differentiation events in B ce l ls as wel l a s the oxidat ive burst response in granulocytes (299, 327), and roles for H S A in T cel l deve lopment and activation have been sugges ted (328, 329). Both ant igens are g lycoproteins at tached to the outer sur face of the p lasma membrane by a g lycosy l phosphat idy l inos i to l l ipid a n c h o r and are e x p r e s s e d on mult ip le l i n e a g e s of hemopoiet ic cel ls . The mature pept ides are only 30-35 amino ac ids in s i ze with the entire cod ing region be ing e n c o m p a s s e d within an approx imate 2 4 0 bp D N A f ragment . In addi t ion, the mature C D 2 4 and H S A prote ins sha re only l imited s e q u e n c e homology (57%) with one another and ant ibodies to C D 2 4 and H S A are not c ross reactive (300, 327). T h e s e features of smal l coding s i ze , potential for cel l su r face e x p r e s s i o n on mult ip le hemopo ie t i c l i neages and l imited homo logy , sugges ted to me that H S A and C D 2 4 would be poss ib le cand ida tes for se lec tab le markers of gene transfer to heterologous hemopoiet ic cel ls . In the work presented in this Chap te r I demonstrate the feasibil i ty of util izing C D 2 4 for the identif ication and se lec t ion of retrovirally t ransduced primary ce l ls of the mur ine hemopo ie t i c sys tem including those with long-term lympho-myeloid repopulat ing ability. 3.2. R E S U L T S 3.2.1. The CD24 viral vector 56 To explore the possib le use of C D 2 4 as a selectable cel l sur face marker, the retroviral vector depic ted in Figure 3.1 was const ructed. Th is vector conta ins the minimal 240 bp C D 2 4 c D N A sequence encompass ing the complete cod ing region under the contro l of the M P S V long terminal repeat e n h a n c e r a n d promoter regulatory e lements . For initial feasibi l i ty s tud ies the neomyc in res is tance g e n e under the control of the thymidine k inase gene promoter w a s a lso inc luded in the vector to aid in viral titering and to provide an independent m e a n s to a s s e s s gene t ransfer . A C D 2 4 viral p roducer w a s genera ted us ing the eco t rop ic G P + E - 8 6 packag ing line and had a titre of ~5 x 1 0 5 C F U / m l a s a s s e s s e d by n e o R g e n e transfer to NIH-3T3 cel ls. LTR CD24 tk neo I 1 200 bp Figure 3.1. Schema t i c of the J Z e n C D 2 4 t k n e o provirus. It incorporates a 240-bp portion of the C D 2 4 c D N A encompass ing the complete coding region; a thymidine k inase-neomyc in res is tance casset te (tkneo) from p M C I n e o ; and L T R s e q u e n c e s from the M P S V as descr ibed in Material and Methods in Chapter 2. Transfer of the C D 2 4 gene to hemopoiet ic cel ls w a s initially eva luated in IL-3-dependent murine B a / F 3 cel ls . Approximately 8 0 % of B a / F 3 ce l ls were found to e x p r e s s high levels of sur face C D 2 4 ant igen after 2 days of co-cul t ivat ion with C D 2 4 viral p roducers and a further 5 days of growth in the a b s e n c e of G 4 1 8 se lec t ion (Figure 3.2). Simi lar levels of express ion were detected as ear ly a s 12 hours post infection (data not shown). 3.2.2. F A C S se lec t ion of C D 2 4 - t r a n s d u c e d in vitro c l o n o g e n i c progenitors and C F U - s p l e e n (CFU-S) Day 4 5 - F U bone marrow cel ls were co-cult ivated with C D 2 4 viral producers for 24 hours and recovered non-adherent cel ls were cultured for a further 48 hours 57 to al low express ion of the transferred C D 2 4 gene prior to f low cytometr ic ana lys is and cel l sort ing. A s shown in a representative F A C S profile for one exper iment, Uninfected Ba/F3 CD24 virus infected Ba/F3 Log fluorescence Figure 3 .2. F l o w cy tomet r ic a n a l y s i s of C D 2 4 e x p r e s s i o n by B a / F 3 c e l l s previously cocult ivated with C D 2 4 viral producer cel ls and then cul tured for 5 days in the absence of G 4 1 8 . Infected and control (uninfected) cel ls were sta ined with an an t i -CD24 -based tetramolecular ant ibody complex coup led to R - P E a s desc r ibed in Mater ials and Methods (Chapter 2). approximately 1/3 of the bone marrow cel ls recovered after co-cult ivat ion infection without prior growth factor prestimulation were posi t ive for the C D 2 4 cel l sur face antigen (see Figure 3.3 top panel). In 3 exper iments 96±3 .6% of in vitro c lonogen ic progenitors recovered in the CD24+ fraction were G418-res is tant c o m p a r e d to 35 -6 9 % in the unsor ted marrow populat ion (see F igure 3.3 bottom panel ) . S o m e G418- res is tan t progenitors (12-27%) were a lso detec ted in the C D 2 4 " f ract ion, l ikely as a result of over lap between the C D 2 4 + and C D 2 4 " cel l populat ions and the relatively low sorting threshold chosen . C D 2 4 ' a n d C D 2 4 + fract ions were a lso a s s a y e d for day 12 C F U - S . A summary of f indings from 3 exper iments are presented in T a b l e 3 .1 . Al l sp l een co lon ies der ived from the CD24+ fraction (37 of 37 analyzed) were posit ive both for 58 prov i ra l D N A s e q u e n c e s and s ign i f icant leve ls of C D 2 4 e x p r e s s i o n a b o v e background staining (range 5 -79% of cel ls ana lyzed) . The observat ion that not all ce l ls within a colony expressed detectable levels of C D 2 4 is likely a result of s o m e admixture of contaminat ing inter-colony cel ls as well as di f ferences in the absolute leve l of C D 2 4 e x p r e s s i o n on ce l l s within e a c h co lony . In con t ras t , on ly approximately 5 0 % of sp leen co lon ies derived from the unsorted marrow (13 of 30) showed ev idence of gene transfer by Southern blot analys is . Furthermore, only 4 of 13 marked co lon ies were posit ive for express ion of the t ransferred C D 2 4 gene . Surpr is ing ly , a signif icant proport ion of sp leen co lon ies der ived from the C D 2 4 " f ract ion (15 of 26) were a lso found to conta in intact provi rus a l though only 3 e x p r e s s e d detectable levels of C D 2 4 . Thus retroviral ly-transduced day 12 C F U - S can be successfu l ly enr iched based on their immediate express ion of a t ransduced C D 2 4 gene . In addi t ion, the differentiating day 12 progeny of the C F U - S thus se lec ted a lso show maintained express ion of the t ransduced C D 2 4 gene in vivo. 3.2.3. Select ion by F A C S of CD24-virus-infected C R U S u b s e q u e n t exper iments were conduc ted to determine the feasibi l i ty of select ing C D 2 4 t ransduced C R U by F A C S . In an effort to facilitate gene transfer to repopulat ing ce l ls , day 4 5 - F U bone marrow cel ls were prest imulated with growth factors for 48 hours prior to co-culture with C D 2 4 viral producer ce l ls . T h e C D 2 4 exp ress ion profi les of non-adherent cel ls recovered 48 hours after the co-cul ture per iod for 2 exper iments are shown in Figure 3.4. Greater than 5 0 % of ce l ls were C D 2 4 + us ing this infection protocol compared to 3 3 % with no pre-st imulat ion (see Figure 3.3 top panel) . Unsor ted and sorted C D 2 4 - or CD24+ ce l ls (Figure 3.4 expt 1) f rom Ly5.1 donor mice were injected into lethally i rradiated Ly5 .2 recip ients under compet i t ive repopulat ing cond i t ions . In pre l iminary s tud ies inject ion of l imiting numbers of bone marrow cel ls recovered fol lowing the sort ing procedure revea led a compet i t ive repopulat ing cel l f requency of approx imate ly 1/3 x 1 0 4 . Therefore, recipients were transplanted with 1 0 4 or 4 x 1 0 4 unsorted or sorted cel ls 59 Uninfected BM CD24 virus infected BM 1 (r i o u 1 0 Log f luorescence 1 0 io J 1 0 120 expt 1 expt2 expt 3 • unsorted • F r a c l ( C D 2 4 - ) • F rac II (CD24+) Figure 3.3. F A C S select ion of C D 2 4 virus- infected in vitro c lonogen ic progenitors and day 12 C F U - S . (Upper panel) Express ion of C D 2 4 on day 4 5 - F U B M cel ls 48 hours after coculture with C D 2 4 viral producers compared to uninfected day 4 5 - F U B M cel ls . Ce l l s were sta ined with a n t i - C D 2 4 / R - P E tetrameric ant ibody c o m p l e x e s and ana l yzed by flow cytometry. Infected ce l ls were sor ted into C D 2 4 " (I) and C D 2 4 + (II) fractions by F A C S and a s s a y e d for in vitro c lonogen ic progenitors and day 12 C F U - S . (Lower panel) Propor t ion of G418- res i s tan t in vitro c lonogen ic progeni tors in the C D 2 4 " (I), C D 2 4 + (II) and presort B M popula t ions for three independent exper iments . Presor t B M ce l ls were not obta ined in exper iment 3 because of lack of cel ls . No G 4 1 8 resistant c lonogen ic progenitors were observed in uninfected control B M cel ls. Resul ts of ana lys is of day 12 C F U - S - d e r i v e d sp leen co lon ies are presented in Tab le 3.1. 60 Table 3.1. Prov i ra l Integration and C D 2 4 E x p r e s s i o n on C e l l s f rom Individual Sp leen Co lon ies Der ived from Sorted and Unsor ted Bone Marrow C e l l s Fo l lowing C D 2 4 Vi rus Infection. C e l l s T ransp lan ted Total No. C o l o n i e s A n a l y z e d No. Co lon ies Posi t ive for C D 2 4 Express ion No . Co lon ies Posi t ive for Prov i ra l Integration Unsor ted 30 4 13 Frac II (CD24+) 3 7 37 3 7 Frac I (CD24-) 2 6 3 15 Recip ient mice were transplanted with 1 x 1 0 3 to 1 x 1 0 4 bone marrow cel ls , which had been co-cu l tured with C D 2 4 viral p roducer ce l ls , f rom one of e a c h of the presort , C D 2 4 " or CD24+ marrow fractions as descr ibed in the legend to F ig 3.3. We l l iso lated sp leen co lon ies were d issec ted and ana lyzed for C D 2 4 express ion using f low cytometry and for proviral integration using Southern blot ana lys is . T h e table represents data accumulated over 3 separate exper iments. * A sp leen colony was conc luded to be posit ive for C D 2 4 express ion if >5% of the cel ls bound signif icant levels of the an t i -CD24 tetramolecular ant ibody complex . in an effort to minimize the l ikel ihood of C D 2 4 " C R U contributing to the transplant due to con tamina t ion of the C D 2 4 + f ract ion. B a s e d upon the propor t ion of uninfected control ce l ls found within the C D 2 4 + sort ing window (0.7%, F igure 3.4 expt. 1) our calculat ions predict that for every 2 x 104 cel ls sorted less than 0.005 C D 2 4 " (i.e. contaminat ing) C R U would be found within the posi t ive sort w indow due to occu r rences s u c h as non-spec i f i c binding of the C D 2 4 / R P E tet ramer ic ant ibody complex . For recipients transplanted with cel ls from the C D 2 4 + fraction, 10 of 11 were f ound to be recons t i tu ted with p rov i rus -con ta in i ng C R U 5 w e e k s pos t -t ransp lan ta t ion (see T a b l e 3.2). E x p r e s s i o n of C D 2 4 on pe r iphe ra l b lood leukocy tes of these mice ranged from 7-24%. In 3 mice , ev idence of the s a m e provira l integrat ion f ragments in ce l l s of both mye lo id and l ympho id t i s s u e s sugges ted gene transfer to a totipotent repopulating cel l . Retroviral ly- infected C R U were a l so found in the C D 2 4 - fract ion with 8 of 11 mice show ing ev i dence of 61 proviral mark ing in either bone marrow and/or thymus D N A . However , none of t hese 8 recip ients s h o w e d detectable levels of C D 2 4 exp ress ion on per iphera l b lood leukocytes. Of 5 recipients injected with unsorted bone marrow ce l ls , all 5 were repopulated with retrovirally marked cel ls but only 2 showed detectable levels of C D 2 4 express ion in peripheral blood ce l ls . T h e s e results demonst ra te that, a s for day 12 C F U - S C D 2 4 express ion in combinat ion with F A C S can be used for the select ion of retrovirally t ransduced C R U whose progeny maintain express ion of the t ransferred C D 2 4 gene for at least 5 weeks post t ransplantat ion. T h e s e f indings were conf i rmed and extended in a second experiment in which retrovirally infected Ly5.1 b o n e mar row ce l l s we re sor ted into C D 2 4 " a s wel l a s C D 2 4 + ' 0 W , C D 2 4 + m e d , and C D 2 4 + n i g h fractions 48 hours post infection (see F igure 3.4 expt 2). L imit ing numbers (1 x 1 0 4 ) of ce l ls from e a c h of the 4 f ract ions we re then in jected into lethally i rradiated Ly5.2 recip ients under compet i t ive repopulat ing cond i t i ons . B a s e d upon the sort ing w indows c h o s e n (F igure 3.4 expt 2) our ca lcu la t ions suggest that for every 2 x 1 0 4 cel ls sor ted, less than 0 .03, 0.01 and 0.0001 C D 2 4 - C R U were sorted, respect ively in the C D 2 4 + l o w , C D 2 4 + m e d , and C D 2 4 + h i g h windows, due to occur rences such as non-speci f ic ant ibody staining. In t h e s e exper iments , recip ients were ana l yzed for ev i dence of Ly5.1 donor ce l l -de r i ved repopu la t ion a n d C D 2 4 g e n e e x p r e s s i o n at 5 a n d 16 w e e k s post t ransplant, as wel l as for express ion of the n e o R gene and ev idence of proviral marking at 16 weeks post transplant. M i c e t ransp lan ted with C D 2 4 + l o w , C D 2 4 + m e d , or C D 2 4 + h i g h ce l ls all s h o w e d signif icant levels of mult i- l ineage (i.e., lymphoid and myeloid) Ly5.1 donor-ce l l -der ived repopulat ion and all recipients of C D 2 4 + se lec ted marrow ce l ls aga in s h o w e d ev idence of proviral marking in bone marrow and/or thymus D N A (Figure 3.5). In 6 of these mice (mouse m l , C D 2 4 + , 0 W fraction; m o u s e m l , m3 and m4, C D 2 4 + m e d f ract ion; and mice m l and m2, C D 2 4 + h i g n f ract ion; F igu re 3.5) retroviral marking patterns observed by Southern blot ana lys is of bone marrow and 62 Expt 1 Uninfected BM Expt 2 Uninfected BM CD24 virus infected BM CD24 virus infected BM 21% 30% Log fluorescence Log fluorescence Figure 3 .4 . S e l e c t i o n of C D 2 4 v i rus - in fec ted C R U by F A C S . T h e C D 2 4 express ion profi les of day 4 5 - F U uninfected control marrow ce l ls (top panels) and day 4 5 - F U B M cocultured with C D 2 4 viral producer cel ls (bottom panels) 48 hours postinfection are shown for two independent exper iments. C e l l s were s ta ined with a n t i - C D 2 4 / R - P E tetrameric antibody comp lexes and ana lyzed by flow cytometry. In experiment 1, infected cel ls were sorted into C D 2 4 " (I) and C D 2 4 + (II) fract ions and in jected into lethal ly i r radiated recipient mice under compet i t i ve repopu la t ing condit ions at 1 x 1 0 4 to 4 x 1 0 4 cel ls per mouse. M ice were ana lyzed 5 w e e k s post t ransp lanta t ion for both C D 2 4 ce l l su r face e x p r e s s i o n on per iphera l b lood leukocytes and proviral integration in B M and thymus (results are shown in Tab le 3.2). For exper iment 2, B M cel ls in presort, C D 2 4 " , C D 2 4 + l o w , C D 2 4 + m e d and C D 2 4 + n ' 9 n express ing fractions were injected into lethally irradiated recipient mice under competi t ively repopulating condit ions at 1 x 1 0 4 ce l ls per mouse . M ice were ana lyzed 16 w e e k s posttransplantat ion for C D 2 4 cel l sur face exp ress ion on their peripheral blood cel ls and proviral integration in B M and thymus (results are shown in Tab le 3.3 and Figure 3.5). Percen tages of cel ls in the sort w indows are indicated. 6 3 t h y m u s w e r e c lear l y ind icat ive of g e n e t rans fer to a tot ipotent long- te rm repopulat ing ce l l . The degree of Ly5.1 donor cel l repopulat ion a m o n g these mice s t rong ly co r re la ted with the intensi ty of provi ra l mark ing (F igu re 3.5), an observat ion consis tent with gene transfer to all repopulat ing ce l ls in the CD24+ fract ions. Al though C D 2 4 express ion w a s detected in the majority of recipients of C D 2 4 + se lec ted marrow at 5 weeks post transplant, express ion levels were poorly mainta ined at the later 16 week time point ana lyzed in this experiment (Figure 3.6). In this c a s e , in only 2 of 8 mice repopulated with cel ls from the CD24+ f rac t ions (mice m2 and m3 , C D 2 4 + m e d cel ls) were CD24+ per iphera l b lood (4 .3% and 1 7 . 1 % , respect ive ly) or bone marrow ce l ls (3.4% and 4 2 % , respect ive ly) still present. In addit ion, G 4 1 8 R c lonogenic progenitors were detected in the marrow of these two mice (3% and 6 8 % respectively). In the other 6 mice, neither CD24+ nor G 4 1 8 R ce l ls were detected al though provirally marked ce l ls were present in the bone marrow and/or thymus of all of these animals (data not shown). The proportion of recipient mice showing express ion of the transferred C D 2 4 gene on peripheral blood leukocytes a lso dec reased with t ime post transplant in a third exper iment in which recipients were t ransplanted with 1 x 1 0 5 bone marrow cel ls from unsorted or C D 2 4 " or CD24+ sorted fractions. Rec ip ients were a s s e s s e d for C D 2 4 express ion at 8 and 32 weeks post transplant. Whi le the vast majority of rec ip ients t ransp lan ted with C D 2 4 + sor ted marrow s h o w e d e x p r e s s i o n of the t ransferred C D 2 4 gene at 8 w e e k s (14 of 15 mice) only a fraction were found to cont inue to e x p r e s s C D 2 4 when ana l yzed at 32 w e e k s (3 of 13) desp i te the detect ion of intact provirus in all m ice at this t ime point. Rep resen ta t i ve f low cytometric profiles for 1 of these mice are shown in Figure 3.7. O n e of 8 mice transplanted with unselected marrow showed the p resence of C D 2 4 on peripheral blood leukocytes at 8 weeks post transplant al though this mouse was found to be negat ive for C D 2 4 exp ress ion at the later t ime point. N o m ice t ransp lanted with C D 2 4 " sorted cel ls demonstrated C D 2 4 express ion at either of 64 Table 3.2. Proviral Integration and C D 2 4 Express ion on Ce l l s from Compet i t ive ly Repopu la ted Mice A s s e s s e d 5 W e e k s Post -Transplantat ion (Expt. 1). C e l l s T ransp lan ted Unsor ted Sort Fraction I (CD24- ) Sort Fract ion II (CD24+) Reconst i tut ion with Retroviral ly Marked C e l l s * 5/5 8/11 10/11 C D 2 4 Ce l l S u r f a c e Express ion in P . B . C e l l s (>5%)+. 2/5 0/11 10/11 S h o w n are the number of m ice found to be posi t ive for reconst i tu t ion with retroviral ly marked ce l ls or C D 2 4 cel l sur face exp ress ion on per iphera l b lood (P.B.) leukocytes over the total ana lyzed at 5 weeks post transplantat ion. * Retroviral marking w a s a s s e s s e d by Southern blot ana lys is of bone marrow and thymus from e a c h transplanted recipient. Blots were separate ly probed with 3 2 P labeled f ragments of the n e o R gene and C D 2 4 c D N A with identical results. * Express ion of the C D 2 4 antigen on peripheral blood leukocytes w a s a s s e s s e d by sta in ing per ipheral b lood samp les depleted of erythrocytes with a n t i - C D 2 4 / R - P E tetrameric antibody complexes and analys is by flow cytometry. 65 M U b unsorted CD24 C D 2 4 ' l o w CD24*m '» J CD24 ' h | 9 h contrl m l m2 ml m2 m3 ml ml m2 m3 m4 ml m2 m3 B T B T B T B T Figure 3 .5. Hemopoie t ic reconstitution from C D 2 4 retrovirus-infected compet i t ive repopulat ing cel ls as a s s e s s e d by Southern blot ana lys is of proviral integration in B M (B) and thymus (T) 16 weeks posttransplantation. Recip ients received 1 x 1 0 4 ce l ls of C D 2 4 virus- infected marrow from presort, C D 2 4 " , C D 2 4 + l o w , C D 2 4 + m e d , or C D 2 4 + n ' g h fract ions as shown for Expt 2 Figure 3.4. Individual recipient mice are labe led as m1-m4. D N A (20 |ig) from each t issue samp le w a s d igested with E c o R l , an enzyme that cuts once within the C D 2 4 provirus sequence . S h o w n are results of a blot probed with a 3 2 P - l a b e l e d fragment of the n e o R gene ; ident ical resul ts were obse rved us ing the C D 2 4 c D N A as a probe. T h e posi t ive control represents D N A obta ined from a retrovirally infected B a / F 3 c lone harbor ing two cop ies of provirus. the t ime points ana lyzed in this experiment (see Figure 3.6). Addi t ional analys is of bone marrow and thymus D N A from these mice revealed no gross rearrangements in proviral structure to account for the lack of expression in these recipients. Lack of C D 2 4 express ion also did not appear to be a result of promoter interference s ince G 4 1 8 R c lonogen ic progenitors were detected only in those mice showing C D 2 4 express ion on per ipheral b lood leukocytes. The proportion of G 4 1 8 R c lonogen ic progeni tors tended to correlate with the proport ion of C D 2 4 + per iphera l b lood leukocytes in recipient mice, ranging from 3 % to 68%. 66 3.3. D i s c u s s i o n In the work presented in this Chapter , I have tested the utility of a c D N A encod ing C D 2 4 , a human cel l sur face ant igen, for the post- infect ion se lec t ion of hemopo ie t i c ce l l s t r ansduced with a retrovirus con ta in ing this c D N A . F A C S ana lys is in combinat ion with functional studies revealed that under the condi t ions u s e d , B a / F 3 ce l ls , a factor-dependent hemopoiet ic cel l l ine, a s wel l a s pr imary marrow in vitro c lonogenic progenitors, day 12 C F U - S and , most signif icantly, the ear l ies t ce l l s c a p a b l e of compet i t i ve long-term hemopo ie t i c repopu la t ion al l e x p r e s s e d C D 2 4 within 48 hours of terminat ion of the infect ion p rocedu re . Southern blot ana lys is of bone marrow and thymus cel ls from mice compet i t ively repopulated with limiting numbers of se lec ted C D 2 4 + ce l ls demonst ra ted proviral integration in virtually all an ima ls in which donor ce l ls were de tec ted . B e c a u s e long-term repopulat ing cel ls are such rare cel ls and have been difficult to purify to homogene i t y , it has been difficult to a n a l y z e the divers i ty in p roper t ies and behavior of individual cel ls of this type, particularly fol lowing their transplantat ion in v ivo. T h e ability to obtain, prior to transplant, a populat ion of ce l ls that are 1 0 0 % proviral ly marked shou ld now great ly e n h a n c e the power of s tud ies a i m e d at address ing such quest ions. In addit ion this type of strategy shou ld be useful a s a precl inical model for the development of more effective gene transfer st rategies to long-term human repopulating s tem cel ls for use in gene therapy. T h e recovery of compet i t ive repopulat ing ce l l s in the top 8 % of C D 2 4 exp ress ing ce l ls prov ides indirect ev idence that the M P S V L T R e n h a n c e r and promoter are ab le to dr ive high level gene exp ress i on in the most pr imit ive hemopoie t ic ce l ls present in adult marrow t issue. The use of s u c h a se lec tab le marker shou ld be useful for optimizing vectors to ach ieve high and susta ined levels of transferred gene express ion in very primitive hemopoiet ic cel ls and ultimately for s tud ies a imed at the genet ic manipulat ion of s tem cel l behav ior . A n interest ing f inding of this study w a s that C D 2 4 express ion obse rved in primitive retrovirally 67 infected cel ls at the time of select ion was also maintained in their progeny al though this did d e c r e a s e with t ime after t ransplantat ion. The most dramat ic ev idence of con t i nued C D 2 4 e x p r e s s i o n w a s s e e n at the level of C F U - S ; 1 0 0 % of the retrovirally-infected day 12 C F U - S in the sorted C D 2 4 + fraction gave rise to sp leen co lon ies wh ich were a lso posit ive for C D 2 4 exp ress ion . In contrast, desp i te the detec t ion of intact provi rus in approx imate ly half of the C F U - S in the initial (unsorted) or C D 2 4 " fract ion, only a minority of these y ie lded co lon ies of ce l l s express ing detectable levels of C D 2 4 . O n e explanat ion for this observat ion is that the e x p r e s s i o n of the t ransfer red C D 2 4 gene is integrat ion si te dependen t . Similar ly, most mice transplanted with limiting numbers of sorted C D 2 4 + C R U were found to have C D 2 4 exp ress ing per iphera l b lood leukocy tes 5-6 w e e k s post t ransp lanta t ion (expt. 1), w h e r e a s s u c h ce l ls were not o b s e r v e d in a n i m a l s repopulated with C R U from the C D 2 4 " fraction despite the p resence of provirus in the marrow and/or thymus of a number of these. Sus ta ined express ion of C D 2 4 in rec ip ients of initially se lec ted C D 2 4 + C R U w a s a lso obse rved 4 months post-t ransplantat ion al though in a smal l proportion of mice despi te the pers is tence of retrovirally marked ce l ls in myelo id and/or lymphoid t i ssues in al l . Nei ther g ross rear rangement in proviral structure (as a s s e s s e d by Southern blot ana lys is ) or promoter interference between the C D 2 4 and n e o R g e n e s could account for the lack of C D 2 4 express ion observed in these mice. The loss of C D 2 4 express ion in s o m e long term repopu la ted mice t ransp lan ted with C D 2 4 + se lec ted ce l l s is sugges t ive of a shutdown of exogenous promoter activity in vivo, a p h e n o m e n o n which has been previously reported (215, 251). S u c h shutdown of promoter activity may be related to the methylation status of the regulatory e lements (257, 330 , 331). However , others have reported the cont inued express ion of transferred g e n e s such a s h u m a n C D 8 (291) and the human g lucoce reb ros idase g e n e (332) for long per iods post transplantat ion. It is important to note, however, that in these studies recipients were purposely t ransplanted with only one or few s tem cel ls to enab le 68 Figure 3.6. Proport ion of recipient mice found to be exp ress ing the t ransferred C D 2 4 gene at early and late time points post transplant. Ly5.2 recipient mice were transplanted with 1 x 1 0 4 to 1 x 1fj5 retrovirally infected Ly5.1 B M cel ls from the sort fract ions indicated. Erythrocyte-depleted per ipheral b lood s a m p l e s were tested 5-32 weeks post transplantation with an t i -CD24 tetrameric ant ibody c o m p l e x e s / R - P E and analys is by flow cytometry. An ima ls were scored posit ive if > 2 % of ce l ls were C D 2 4 + . The total number of an ima ls tes ted is shown in pa ren theses a n d the proport ion posi t ive for C D 2 4 exp ress ion ind icated in percent . Repopu la t i on of recip ients with Ly5.1 donor-der ived ce l ls w a s a s s e s s e d by s ta in ing with F I T C -conjugated Ly5.1 antibody and analys is by flow cytometry. 69 Control mouse Mouse repopulated with CD24 virus infected marrow Peripheral blood mononuclear cells Peripheral red blood ceils Bone marrow Spleen Thymus Log fluorescence Figure 3.7. F low cytometr ic ana lys i s of C D 2 4 e x p r e s s i o n in the hemopo ie t i c t i ssues of a mouse repopulated with C D 2 4 retrovirus-infected B M 4 months post t ransp lanta t ion . C e l l s were s ta ined with a n t i - C D 2 4 / R - P E te t ramer ic ant ibody c o m p l e x e s . To a s s e s s C D 2 4 exp ress ion on red b lood ce l l s s a m p l e s were not depleted of erythrocytes prior to the staining procedure. 70 the a n a l y s i s of g e n e e x p r e s s i o n at the c lona l level w h e r e a s in the s tud ies ment ioned above recipients were not t ransplanted at limiting di lut ion. My results would suggest that at least for L T R control led gene express ion the phenomenon of p romote r shu tdown may be more w i d e s p r e a d than p rev ious ly a p p r e c i a t e d . Moreover , it may be that mice repopulated with limiting numbers of C D 2 4 + se lec ted ce l ls cont inue to express the transferred C D 2 4 gene but at levels below that which c a n be detected with F A C S . B e c a u s e of the e a s e and sensit iv i ty of methods for moni tor ing t r ansduced C D 2 4 exp ress ion in per iphera l b lood ce l l s , this vec tor construct should be well sui ted for further studies of vector modif icat ions that may abroga te promoter shutdown in primitive hemopoie t ic ce l ls and their long-term progeny. No g ross abnormal i t ies in hemopo ies is were observed in mice exp ress ing high leve ls of C D 2 4 fol lowing transplantat ion with C D 2 4 v i rus- in fected marrow c o m p a r e d to normal control an ima ls . Further, sus ta ined exp ress ion in recipient mice at least 4 months post transplantat ion demonst ra tes that the use of s u c h a foreign ant igen as a retroviral marker is compat ib le with long term exp ress ion in hemopoiet ical ly reconstituted lethally irradiated recipients. T h e work p resen ted in this Chap te r has demons t ra ted the feasibi l i ty of uti l izing the C D 2 4 cel l sur face ant igen as a se lec tab le marker and a reporter mo lecu le in primary hemopoiet ic cel ls including the most primitive e lements of this sys tem. Th is technique should prove useful as a method for increas ing the power of retroviral mark ing s tud ies, rapidly test ing var ious retroviral infect ion protocols and the identif ication of regulatory e lements which opt imize gene exp ress ion in primitive hemopoiet ic stem cel ls or other target cel ls of interest. 71 C H A P T E R 4 HIGH L E V E L RECONSTITUTION WITH P R E S E L E C T E D HEMOPOIET IC C E L L S E X P R E S S I N G A T R A N S D U C E D G E N E ENCODING A C E L L S U R F A C E A N T I G E N . The results presented in this Chapter have been descr ibed in: Pawl iuk, R., and R. K. Humphr ies. High level reconstitution with preselected hemopoiet ic cel ls express ing a t ransduced gene encoding a cel l sur face ant igen. Manuscr ip t in preparat ion. 72 4 . 1 . Introduction G e n e therapy represents an attractive strategy for the treatment of var ious human heri table d isorders , c a n c e r and Acqu i red Immune Def ic iency S y n d r o m e (A IDS) . S u c c e s s f u l gene therapy of hemato log ica l d isorders requi res that two major goa ls be met: (1) efficient and stable transduct ion of the hemopoie t ic s tem cel l ( H S C ) and (2) appropriate susta ined express ion of t ransduced gene(s) in the ce l ls of interest. A l though signif icant progress has been made towards ach iev ing these goa ls , efficient high level retroviral gene transfer to long term repopulat ing s tem cel ls remains a chal lenge, likely due to a multiplicity of factors including low levels of viral receptors and the largely quiescent nature of H S C s . T o o v e r c o m e these di f f icul t ies, a number of s t ra teg ies invo lv ing the incorpora t ion of se lec tab le marker g e n e s into retroviral vec to rs h a v e b e e n deve loped to enab le the enr ichment of t ransduced target ce l ls . The most widely used of these have been those which confer res is tance to toxic compounds such a s neomyc in (23, 35 , 205) or hygromyc in (262, 312) . T h e ana l ys i s of mur ine rec ip ients engraf ted with bone marrow ce l ls sub jec ted to drug se lec t ion ei ther before (215, 227 , 260) or after (266) transplant has shown that the proport ion of proviral ly marked cel ls in the recipient can be increased using this strategy. More recently, a number of c D N A s encod ing cel l sur face ant igens, including the human low affinity nerve growth factor receptor (273, 326), Thy-1 (279), and M D R - 1 (multi-drug res is tance-1) (272), have been uti l ized as dominant se lec tab le marke rs . R i c h a r d s o n et. a l . used a c D N A encod ing M D R - 1 as a se lec tab le marke r to demonst ra te that preselect ion of retrovirally t ransduced midgestat ional fetal l iver ce l l s l eads to an i nc rease in the proport ion of c i rcu la t ing per iphera l b lood leukocytes express ing M D R - 1 as compared to mice receiv ing unse lec ted marrow (333). H o w e v e r , desp i te the potent ial that t h e s e p r o c e d u r e s p o s s e s s , their e f fec t iveness for ach iev ing high level hemopoiet ic reconsti tut ion with exc lus ive ly provi ra l ly m a r k e d ce l l s h a s not yet been fully exp lo red . M o r e o v e r , desp i te 73 i nc reas ing interest in genet ica l ly manipu la t ing H S C behav io r , little is known regarding regulatory e lements which maximize the express ion of transferred genes in these ce l l s . T h u s , methods enab l ing the efficient se lec t i on , a s wel l a s the tracking and quantif ication of transferred gene express ion in primitive hemopoiet ic ce l ls and their progeny for extended periods of t ime in vivo are required. I have examined these i ssues in a sys tem utilizing the human cel l sur face ant igen C D 2 4 a s a dominant se lec tab le marker in combinat ion with F A C S . In Chap te r 3 I demonstrated that retrovirally t ransduced totipotent in vivo repopulat ing s tem cel ls cou ld be se lected for on the bas is of C D 2 4 express ion within 48 hours of terminat ion of the infection protocol (278). In the work descr ibed in this Chap te r , th is C D 2 4 s e l e c t i o n a p p r o a c h is s h o w n to e n a b l e the a lmos t e x c l u s i v e regenerat ion of hemopo ies is in myeloablated recipient mice with provirally marked cel ls . The contribution of individual H S C s to hemopo ies is w a s ana lyzed by proviral integration analys is . Moreover , persistent express ion of the transferred C D 2 4 gene w a s o b s e r v e d in a s igni f icant proport ion of S c a + L i n " bone mar row ce l l s (a subpopula t ion known to be enr iched for cel ls with long term in v ivo repopulat ing abil ity), per ipheral b lood leukocytes, red b lood ce l ls , sp l een , thymus and who le bone marrow ce l ls for a minimum of 6 months post transplant. Final ly, this study revea ls intr iguing ev idence of vec to r -based d i f fe rences in the abil i ty to dr ive persistent express ion of transferred genes in vivo. 4.2. Results 4.2.1. Viral vectors and experimental design In the cou rse of these exper iments two vectors b a s e d on different viral backbones (Figure 4.1 upper panel) were emp loyed . In both vectors , the minimal 240bp cod ing region of the C D 2 4 c D N A was p laced under the control of the viral L T R regulatory e lements. Both vectors a lso contained a c D N A encod ing res is tance to the neomyc in analog G 4 1 8 . In the J Z e n C D 2 4 t k n e o vector (hereafter referred to a s J Z e n C D 2 4 ) the neomyc in res is tance casse t te is dr iven from the thymid ine 74 k inase g e n e promoter , whi le in M S C V n e o l R E S C D 2 4 (hereafter referred to a s M S C V C D 2 4 ) both the neomycin resistance and C D 2 4 c D N A s are dr iven from the viral L T R due to the inclusion of an internal r ibosomal entry site der ived from the 5' untranslated region of the encepha lomyocard i t i s v i rus. Both the J Z e n C D 2 4 and M S C V C D 2 4 viral p roducers y ie lded titres of approx imate ly 5 x 1 0 5 C F U / m l a s a s s e s s e d by transfer of n e o R to NIH-3T3 ce l ls with no helper v i rus detec ted in either line. In an effort to facilitate gene transfer to repopulating s tem cel ls , day 4 5-F U B M cel ls were prestimulated with growth factors for 48 hours prior to 48 hours of co-cul ture with irradiated viral producer cel ls. Ce l l s recovered from co-cul ture were s u b s e q u e n t l y s ta ined for C D 2 4 e x p r e s s i o n a s de ta i led in C h a p t e r 2. T h e exp ress ion profi les and sort thresholds for 1 exper iment are shown in F igure 4.1 (lower panel) . Fol lowing F A C S select ion, 4 x 1 0 5 C D 2 4 + cel ls per recipient were t ransplanted into a total of 14 myeloablated mice (7 recipients per viral construct). Th i s t ransplant dose w a s est imated to contain 12 ± 4 compet i t ive repopulat ing units ( C R U ) as determined from prev ious exper iments carr ied out to determine C R U f requenc ies in the populat ion of ce l ls recovered fol lowing the infection and select ion procedure (data not shown). 4.2.2. The majority of donor-derived cells in recipients contain intact provi rus Six months post t ransplant, the average proport ion of t ransplant der ived (Ly5.1+) cel ls in recipients was 5 8 % for bone marrow (range 2 2 % to 68%), 8 5 % for per iphera l b lood mononuc lea r ce l ls (range 4 7 % to 95%) , and 9 3 % for thymus (range 4 7 % to 100%)(see Tab le 4.1). Thus , the vast majority of m ice s h o w e d virtually comple te reconsti tut ion of bone marrow and thymus with Ly5.1 donor de r i ved ce l l s (taking into accoun t that approx imate ly 3 0 % of the mar row is c o m p o s e d of differentiating erythroid progenitors and their progeny wh ich do not exp ress the Ly5.1 antigen). Express ion of the t ransduced C D 2 4 gene was detected 75 VIRAL V E C T O R S AND F A C S SELECTION JZenCD24tkneo LTR CD24 TK NEO =1 3.0 Kb Sstl MSCVneolRESCD24 LTR NEO IRES CD24 - LTR 2.5 Kb Sstl 4 C D 2 4 - P E Figure 4.1. V i ra l vec tors used and F A C S se lec t ion of retroviral ly t r ansduced bone mar row ce l l s . (Upper panel ) S c h e m a t i c of the J Z e n C D 2 4 t k n e o a n d M S G V n e o l R E S C D 2 4 provirus. J Z e n G D 2 4 t k n e o incorporates a 240-bp portion of the c D N A e n c o m p a s s i n g the comp le te cod ing reg ion ; a thymid ine k i n a s e -neomyc in res is tance casset te (tkneo) from p N C I n e o ; and L T R s e q u e n c e s from M P S V . M S C V n e o l R E S C D 2 4 conta ins the nep gene from p M C I n e o ; the cod ing region of the C D 2 4 c D N A ; an internal r i bosoma l entry s e q u e n c e f rom the encepha lomyoca rd i t i s v i rus; and L T R s e q u e n c e s f rom M S C V . (Lower panel ) E x p r e s s i o n of C D 2 4 on day 4 5 - F U B M ce l l s 4 8 hours after co-cu l tu re with J Z e n C D 2 4 or M S C V C D 2 4 viral producers compared to uninfected day 4 5 - F U B M cel ls . Ce l l s were sta ined with a n t i - C D 2 4 / R - P E tetrameric ant ibody comp lexes and a n a l y z e d by f low cytometry . C D 2 4 + ce l l s were sor ted a n d t ransp lan ted into myeloablated recipient mice at 4 x 105 C D 2 4 + c e l l s per recipient. 76 on per ipheral b lood leukocytes, red blood cel ls and whole bone marrow ce l ls for all recip ients and on thymocytes for all but 2 mice. The proport ion of t ransplant der ived ce l ls express ing C D 2 4 (ie. L y 5 . 1 + C D 2 4 + cel ls) var ied with the l ineage of the cel l type tested, and as a function of the virus used (Table 4.1). T h e greatest proport ion of C D 2 4 + ce l ls was observed among B M , per ipheral b lood leukocytes and red b lood cel ls , with the lowest proportion found in thymus. For example , 7 4 % of whole B M , 9 0 % of red blood ce l ls and 5 8 % of per ipheral b lood leukocy tes in recipient m5, t ransplanted with M S C V C D 2 4 t ransduced marrow, were posit ive for C D 2 4 express ion compared to only 1 8 % of thymocytes (Table 4.1). M ice receiv ing M S C V C D 2 4 t ransduced B M showed some 6-fold higher proportion of C D 2 4 + cel ls for all cel l types tested as compared to mice transplanted with J Z e n C D 2 4 infected mar row (Tab le 4.1) . A s an independent a s s e s s m e n t of g e n e t ransfer , the proport ion of G 4 1 8 res is tance c lonogen ic progeni tors in the bone mar row of recip ients w a s determined and found to correlate with the proport ion of C D 2 4 + cel ls in total bone marrow (Table 4.1) St r ingent sor t ing th resho lds dep ic ted in F igure 4.1 (< 0 . 2 % of n o n -t r a n s d u c e d contro l bone mar row ce l l s posit ive) were c h o s e n in an effort to max im ize the probabil i ty that only retrovirally t ransduced s tem ce l ls exp ress ing s igni f icant leve ls of C D 2 4 wou ld be iso la ted. Desp i te this, on a v e r a g e only a proport ion of transplant der ived cel ls were found to exp ress C D 2 4 , sugges t ive of p romoter shu tdown or the ineff icient se lec t ion of t r a n s d u c e d s t e m ce l l s . To discr iminate between these possibi l i t ies Southern blot ana lys is w a s carr ied out to a s s e s s the p resence of intact provirus in the B M , sp leen and thymus of recipient mice. A s shown in Figure 4.2, high levels of intact provirus were detectable in the hemopo ie t i c t i ssues of f ive representat ive recip ients of J Z e n C D 2 4 t r ansduced mar row. C o m p a r a b l e resul ts we re o b s e r v e d for rec ip ien ts of M S C V C D 2 4 t ransduced B M (data not shown). Densi tometr ic ana lys is of the s igna l intensit ies ob ta ined w h e n m e m b r a n e s were sequent ia l ly p robed with a f ragment of the 7 7 Table 4.1. Flow Cytometric Analaysis of CD24 Expression in Various Hematopoietic Tissues in Recipients of JZenCD24 or MSCVCD24 Virus-Infected Marrow Assessed 24 Weeks Post-Transplant. %G418 R Whole BM Peripheral Thymus Peripheral Blood BM Prog. RBCs Leukocytes Cells % CD24+/ % CD24+ % CD24 + / Transplanted Mouse % Ly5.1 + Ly5.1+* % CD24+1 %Ly5.1 + Ly5.1 + %Ly5.1 + Ly5.1 + JZenCD24 ml 64 8 7 98 0 93 4 6 m2 66 18 3 98 1 94 2 13 m3 58 16 6 98 3 93 4 11 m4 57 5 4 96 0 82 4 3 m5 66 12 2 99 3 91 2 5 m6 67 21 15 99 1 90 8 24 m7 67 25 13 99 7 92 10 34 62 + 2 15 + 3 7 ± 3 98 + 0.5 2+1 91 +2 5+ 1 14 ± 6 MSCVCD24 ml 29 52 25 99 8 70 14 56 m2 68 85 44 98 21 81 33 90 m3 22 41 23 31 10 47 28 40 m4 64 30 29 93 15 92 5 64 m5 61 74 90 100 18 95 58 87 m6 66 88 41 99 20 90 29 90 52+ 10 62 ± 1 1 42 ± 13 87 ± 1 4 15 ± 4 79 ± 9 28 ± 9 71+ 10 Repopulation of recipients with Ly5.1 donor-derived cells was assessed by staining erylhrocyte-depleted peripheral blood samples with FITC-conjugated Ly5.1 antibody and analysis by flow cytometry. Samples were also tested for expression of the transferred CD24 gene by staining cells with anti-CD24 tetrameric antibody complexes/R-PE and analysis by flow cytometry. CD24 expression on bone marrow stem cell candidates was achieved by combining anti-CD24 tetrameric antibody complexes with a cocktail of antibodies recognizing lineage specific antigens as explained in Chapter 2. The proportion of G418R clonogenic progenitors in BM was determined as described in Material and Methods. * The expression of CD24 in various hemopoietic tissues is given as the proportion of cells with the Ly5.1+ cell surface phenotype expressing CD24. +Ly5.1 is not expressed on red blood cells (RBCs). Therefore, the % CD24+ values given for RBC represents the % of total RBCs. neomyc in res is tance gene to detect recombinant provirus and the erythropoiet in receptor gene as an endogenous control showed that the ave rage proviral copy number per haploid genome w a s 2.6 (range 1.2 to 4.2) and 2.5 (range 0.6 to 6.5) for rec ip ients of J Z e n C D 2 4 and M S C V C D 2 4 infected B M respect ive ly . T h e s e f indings are consis tent with high level regenerat ion with proviral ly marked ce l l s and suggest that C D 2 4 express ion in only a fraction of transplant der ived (Ly5 .1 + ) ce l l s is most l ikely the result of down regulat ion of the t ransferred C D 2 4 g e n e rather than the absence or gross alteration of recombinant provirus. To further a s s e s s the degree of reconstitution with provirally marked ce l ls , femoral B M from primary recipients was t ransplanted into secondary recipients to generate day 12 C F U - S . Of 19 and 20 individual sp leen co lon ies generated from the B M of recipients of J Z e n C D 2 4 and M S C V C D 2 4 infected marrow, respect ively, all were found to contain intact provirus. Th is finding strongly sugges ts that at the t ime of sac r i f i ce essen t i a l l y all h e m o p o i e s i s w a s de r i ved f rom retrovira l ly t r ansduced s tem ce l l s . Integration ana lys i s of hemopo ie t i c t i s s u e s of pr imary transplant recipients and of day 12 sp leen co lon ies generated from these mice are presented in Figure 4 .3 . Two of the se lec ted mice mani fested different patterns of c lonal reconstitut ion. Ana lys i s of splenic and thymic D N A obtained from a primary recipient of J Z e n C D 2 4 t ransduced marrow (top panel) shows a complex pattern of p rov i ra l mark ing indicat ive of po lyc lonal reconst i tut ion. T h e ana lys i s of day 12 sp leen co lon ies generated from the B M of this animal is consistent with at least 4 c lona l integration patterns be ing detected and hence , is sugges t i ve of mult iple H S C s act ively contributing to hemopo ies is at the time of sacr i f ice. In contrast, the a n a l y s i s of B M and thymic D N A ob ta ined f rom a rec ip ient of M S C V C D 2 4 t r a n s d u c e d mar row s h o w e d a pattern of prov i ra l mark ing cons i s ten t with monoc lona l reconst i tut ion (Figure 4 .3 , bottom panel ) . T h e ident ical pattern of proviral mark ing among all day 12 sp leen co lon ies generated from the B M of this mouse further supports this conc lus ion. 79 neg. pos. Ctrl m l m2 m3 m4 m5 ^ B S T B S T B S T B S T B S T - 10 ^ » f # ^ — 3.0 Kb Figure 4 .2 . Detect ion of high levels of intact provirus in the bone marrow (B), sp leen (S), and thymus (T) of recipient mice t ransplanted with 4 x 1 0 ^ C D 2 4 + se lec ted , J Z e n C D 2 4 virus infected bone marrow cel ls at 24 weeks post transplant by Southern blot ana lys is . Individual recipient mice are labeled as m1-m5 . D N A (10 jig) from each t issue samp le was d igested with Sst I, an e n z y m e which cuts once within each proviral L T R . Shown are the results of a blot probed with a 3 2 p . labe led f ragment of the n e o R gene (top) and a fragment of the erythropoiet in receptor gene as an endogenous control for D N A loading (bottom). T h e posit ive control represents D N A obtained from a retrovirally infected B a / F 3 c lone harboring two cop ies of provirus. 80 T h u s , despi te the transplant of multiple t ransduced H S C s , long term hemopo ies i s w a s assoc ia ted with monoclonal s tem cel l activity. T h e s e di f ferences in the pattern of hemopoiet ic reconstitution, however, were not related to the viral vector used . B M from primary recipients of C D 2 4 + se lec ted ce l ls w a s a lso t ransplanted into secondary myeloabla ted recipients in an effort to a s s e s s the regenerat ion of t r a n s d u c e d long term repopu la t ing s tem ce l l s . S e c o n d a r y rec ip ien ts w e r e sacr i f i ced 5 months post transplant and D N A from B M and thymus a n a l y z e d by Southern blot. The p resence of recombinant provirus was detected in both B M and thymus of all seconda ry recip ients (Figure 4.3) demonst ra t ing that the femora l mar row of the pr imary recipient had been regenera ted with H S C s that were retroviral ly t r ansduced . Moreover , in seve ra l i ns tances the pattern of proviral marking observed in the hemopoiet ic t i ssues of secondary recipients w a s identical to that observed among day 12 sp leen co lon ies. For example , an identical pattern of proviral marking was observed between mouse D and sp leen co lon ies 2 and 5 (top panel) ; mouse A B M with sp leen co lon ies 6-8 (top panel) ; and mice F-J with co lon ies 1-9 (bottom panel) . Th is observat ion demonst ra tes that hemopo ies i s in the primary recipients w a s being susta ined by totipotent cel ls with the capac i ty for long term hemopo ie t i c reconst i tu t ion. In addi t ion F igure 4 .3 p rov ides c l ea r ev idence of se l f - renewal of t ransduced H S C s within the femora l marrow of the pr imary recipient . T h e most str ik ing ev idence of se l f - renewal is s e e n a m o n g secondary recipients shown in the bottom panel . Al l 5 mice ( M S C V C D 2 4 F-J) show the s a m e pattern of proviral marking in B M and thymic D N A sugges t ing that the H S C s respons ib le for regenerat ing these mice were der ived from a s tem cel l that had undergone a minimum of 4 sel f -renewal division events in the femoral marrow of the primary recipient. Toge the r , t hese resul ts demons t ra te the utility of the C D 2 4 se lec t i on approach to enab le the regenerat ion of the hemopoiet ic sys tems of mye loab la ted recipient mice to very high levels with provirally marked cel ls , and as a m e a n s to 81 t rack the se l f - renewal of individual H S C s and their contr ibut ion to hemopo ies i s fol lowing transplant. 4.2.3. Gene transfer to and expression of CD24 among Sca+Lin" bone marrow stem cell candidates T o obtain more direct ev idence of express ion of the transferred C D 2 4 gene in primit ive hemopoie t ic ce l ls , C D 2 4 express ion w a s ana l yzed in the S c a + L i n " subpopula t ion of B M , a populat ion known to be enr iched for ce l ls with long term repopulat ing ability. Ana lys i s w a s carr ied out 48 hours fol lowing termination of the infection procedure, and in reconstituted mice 6 months post transplant. Ant ibody s ta in ing per fo rmed 48 hours after retroviral infect ion s h o w e d that both the M S C V C D 2 4 and J Z e n C D 2 4 vec tors were ab le to dr ive high leve ls of C D 2 4 exp ress ion in the majority of Sca+L in " cel ls (75% vs . 6 6 % respect ively) (Figure 4.4). No dif ference in the absolute level of C D 2 4 express ion among ce l ls infected with either virus was detected. At 6 months post transplant, the proportion of regenerated S c a + L i n " ce l ls exp ress ing the t ransferred C D 2 4 gene var ied as a funct ion of the vector u s e d . Wh i le 8 1 % (range 53%-97%) of the regenerated S c a + L i n " ce l ls in recipients of M S C V C D 2 4 t ransduced marrow were posi t ive for C D 2 4 e x p r e s s i o n , only 2 2 % ( range 12%-45%) of C D 2 4 + S c a + L i n " ce l l s were de tec ted in rec ip ients of J Z e n C D 2 4 t r ansduced mar row (F igure 4.5). T h e s e resul ts cont ras t with the essent ia l ly comple te reconstitution with t ransduced ce l ls observed in the B M for both vectors . The express ion of C D 2 4 w a s a lso detected in secondary transplant recipients, al though aga in , vector related di f ferences in the proportion of Sca+L in " ce l l s exp ress ing C D 2 4 were detected. The average proport ion of S c a + L i n " ce l ls e x p r e s s i n g C D 2 4 w a s 5 9 % (range 3 9 % - 8 4 % ) , and 8 % (range 0%-27%) for recipients of M S C V C D 2 4 and J Z e n C D 2 4 infected marrow respect ively. 82 JZenCD24tkneo pos. 2° neg. Ctrl. -|0 day 12 spleen col. A B C D ctrl S T 1 2 3 4 5 6 7 8 B T B T B S B T ¥ — 15Kb — 8.8Kb — 5.8Kb — 4.8Kb — 3.0Kb MSCVneolRESCD24 pos. 2 ° neg. Ctrl. 10 day 12 spleen col. F G H I J ctrl B T 1 2 3 4 5 6 7 8 9 B T B T B T B T B T — 7.2Kb — 4.5Kb Figure 4 .3 . A s s e s s m e n t of proviral integration in bone marrow (B), sp leen (S), and/or thymus (T) in primary or secondary recipients and day 12 sp leen co lon ies generated from the femoral marrow of se lected primary recipients by Southern blot ana l ys i s . Pr imary recip ients rece ived 4 x 1 0 5 C D 2 4 + se lec ted J Z e n C D 2 4 or M S C V C D 2 4 infected bone marrow ce l ls and were sacr i f iced at 24 w e e k s post t ransplant . Femora l bone marrow from se lec ted primary mice were used a s a s o u r c e of donor ce l l s for the generat ion of day 12 sp l een co lon ies and for t ransplant into secondary recipients. For the ana lys is of day 12 sp leen co lon ies lethally i rradiated recipient mice were t ransplanted with 1 x 1 0 5 femora l bone marrow ce l ls . Twe lve days later an imals were sacr i f iced v ia cerv ica l d is locat ion and indiv idual sp leen co lon ies d i ssec ted and D N A puri f ied for Sou the rn blot ana l ys i s . Lethal ly i rradiated seconda ry t ransplant recip ients rece ived 2 x 10® femoral bone marrow cel ls . Mice were sacr i f iced 5 months later and D N A extracted from bone marrow, sp leen and thymus. Individual recipient mice are labeled as 1 ° or 2 ° (A-J) whi le individual day 12 sp leen co lon ies are numerical ly labeled (1-9). D N A (10 u.g) w a s d igested with E c o R I , an e n z y m e which cuts once within the proviral s e q u e n c e . S h o w n are the results of two separa te blots (top; J Z e n C D 2 4 ; bottom; M S C V C D 2 4 ) probed with a 3 2 P labeled fragment of the n e o R gene . The posit ive control is D N A obtained from a retrovirally infected B a / F 3 c lone harboring two cop ies of the J Z e n C D 2 4 t k n e o provirus. The negat ive control is D N A obtained from the sp leen of a normal unmanipulated control mouse. 83 48 hrs post inf 24 wks post tx v.-f»" • h 0 50 100 150 200 250 FSC-H Control bm - b UJ 0. DC ° • Q o o - c.E w 0) + (0 o • a o a. x UJ 1 ° J Z e n 1 ° M S C V 2 ° JZen 2 ° M S C V CD24 CD24 CD24 CD24 Figure 4 . 5 . S h o w s the propor t ion of s tem ce l l c a n d i d a t e s de f i ned by the S c a + L i n " cel l sur face phenotype posit ive for C D 2 4 exp ress ion at 24 w e e k s post t ransplant in rec ip ients of J Z e n C D 2 4 or M S C V C D 2 4 C D 2 4 + s e l e c t e d bone marrow. Femora l bone marrow ce l ls from individual recipient mice were s ta ined with S c a - 1 - C y 5 , a n t i - C D 2 4 t e t r a m e r i c / R - P E an t i body c o m p l e x e s a n d a combinat ion of an t i -Gr -1 -F ITC, an t i -Mac -1 -F ITC, an t i -Ly -1 -F ITC and B 2 2 0 - F I T C recogniz ing g ranu locy tes /macrophages , T and B l ymphocy tes and a n a l y z e d by flow cytometry. M ice represented by sol id c i rc les were used a s marrow donors into secondary recipients. Hor izonta l l ines show the ave rage proport ion of S c a + L i n " C D 2 4 + stem cel l candidates in primary and secondary recipients of J Z e n or M S C V t ransduced bone marrow. 85 4 . 3 . D i s c u s s i o n A key f inding of the work descr ibed in this Chap te r w a s that preselect ion of t ransduced cel ls based upon the transfer and express ion of a c D N A encod ing the C D 2 4 cel l sur face ant igen enab led the reconst i tut ion of recipient m ice a lmost exc lus i ve l y with proviral ly marked ce l ls . Desp i te the relat ively sma l l t ransplant d o s e , reconst i tut ion of rec ip ients with t ransplant de r i ved ( L y 5 . 1 + ) ce l l s w a s essent ia l ly comple te (Table 4.1). Southern blot ana lys is of D N A from B M , sp leen and thymus revealed the p resence of high levels of recombinant provirus (Figure 4.2) , with proviral copy numbers per haplo id g e n o m e >1 in a lmost all m i ce . Moreover , the f inding that all day 12 sp leen co lon ies con ta ined intact provi rus (F igure 4.3) further suppor ts the c o n c l u s i o n that al l H S C s cont r ibut ing to hemopo ies is in these an imals were t ransduced. Whi le recipient mice were t ransplanted with numerous t ransduced H S C s , the number of s tem cel ls found to contribute to long term hemopo ies is w a s found to vary. In s o m e , hemopo ie t i c reconst i tut ion was essent ia l l y monoc lona l whi le in o thers a number of i ndependent c l o n e s contr ibuted to h e m o p o i e s i s . T h e s e d i f fe rences sugges t a s tochas t i c p h e n o m e n o n attr ibutable to e i ther s e e d i n g e f f i c ienc ies that are < 1 0 0 % and/or dec i s ions relat ing to se l f - renewa l v e r s u s differentiation of the stem cell upon arrival in the marrow. It shou ld be poss ib le to explore these i ssues in more detail us ing the C D 2 4 select ion approach to provide a populat ion of H S C s in which all are retrovirally marked. T h e recovery of long term repopulating stem cel ls within the C D 2 4 + se lec ted fraction indicates that both the M P S V (JZen) and M S C V L T R regulatory e lements can drive signif icant levels of gene express ion immediately post infect ion. Th is is further suppor ted by the detect ion of C D 2 4 express ion on the majority of Sca+L in " subpopulat ion of cel ls fol lowing the infection protocol (Figure 4.4). Moreover , C D 2 4 express ion persisted on regenerated Sca+L in" cel ls in both primary and secondary transplant recipients (Figure 4.5). Whi le both J Z e n and M S C V based vectors gave 86 simi lar absolute levels of express ion and proport ions of t ransduced ce l ls fol lowing infect ion, M S C V proved super io r to J Z e n in t ransplant rec ip ients desp i te the detect ion of roughly simi lar levels of intact provirus. Th is w a s a lso evident in the levels of C D 2 4 express ion observed among more mature cel ls in the B M , thymus and per iphery (Table 4.1). T h e s e da ta sugges t that the J Z e n L T R regulatory e lements are more suscept ib le to suppress ion/shutdown (215, 251), pe rhaps due to the methylat ion of the promoter and/or enhance r e lements (257, 331) . T h e M S C V vector (283) comb ines the L T R from P C M V ( P C C 4 embryona l ca r c i noma c e l l - p a s s a g e d myeloprol i ferat ive s a r c o m a virus) (258) and the 5' unt rans la ted region of the dl587rev virus (259), both v i ruses isolated as mutants able to exp ress t ransferred g e n e s in embryon ic ca r c i noma (EC) and embryon ic s tem (ES) ce l l l ines. T h e s e retroviral mutants p o s s e s s var ious delet ions and b a s e pair c h a n g e s be l ieved to remove potential b locks to transcript ion located within the L T R and primer binding si tes of the original v i ruses (252, 280-282, 334). My studies provide st rong ev idence that these c h a n g e s may similarly have important c o n s e q u e n c e s for the e x p r e s s i o n of t r ansduced g e n e s in H S C s and their p rogeny in v ivo . D i f fe rences in the exp ress ion of the transferred C D 2 4 gene w a s a lso detec ted a m o n g ce l l s of different l i neages . The proport ion of C D 2 4 + t hymocy tes w a s signi f icant ly lower than that obse rved a m o n g ce l ls of either the mye lo id (bone marrow) or erythroid l ineage (Table 4.1). Further s tudies using markers s u c h a s C D 2 4 may facilitate the identification of regulatory e lements to further increase the levels and pers is tence of transferred gene express ion in hemopoiet ic s tem cel ls or other target cel ls of interest. In conc lus ion , these studies demonstrate the feasibi l i ty of us ing the C D 2 4 ce l l su r face ant igen in combina t ion with F A C S to enab le vir tual ly comp le te regenera t ion of the hemopo ie t ic sys tems of mye loab la ted recipient m ice with proviral ly marked ce l ls . Th is approach prov ides a powerful strategy to track the behavior of individual H S C s in vivo, and as a method to detect and quantify levels 87 and pers is tence of gene express ion in var ious phenotypical ly def ined populat ions of cel ls in vivo. Efforts to compare the levels of express ion of gene express ion from al ternat ive retroviral vectors such as M F G , and to extend this app roach to the human setting are currently in progress . 88 C H A P T E R 5 E V I D E N C E O F B O T H O N T O G E N Y A N D T R A N S P L A N T D O S E R E G U L A T E D EXPANSION O F HEMOPOIETIC S T E M C E L L S IN VIVO The results presented in this Chapter have been descr ibed in: Pawl iuk, R., C . J . E a v e s and R. K. Humphr ies. 1996. Ev idence of both ontogeny and transplant dose regulated expans ion of hematopoiet ic s tem cel ls in vivo. B lood in p ress . 89 5 . 1 . Introduction There is an increas ing emphas i s on the deve lopment of c l in ical s t rategies that depend on the regenerat ive potential of t ransplantable H S C s . T h e s e inc lude efforts d i rected at autograft purging, gene transfer ( including at tempts to se lec t t ransduced cel l populat ions) ex vivo expans ion and the use of alternative sou rces of t ransp lantab le hemopoie t ic cel l populat ions (such as fetal l iver and umbi l ical co rd b lood) . At the s a m e t ime, the mo lecu la r m e c h a n i s m s that de f ine the regenerat ive potential of H S C , or that regulate its express ion , part icularly in v ivo, remain poorly unders tood. Ser ia l transplantat ion s tudies have prov ided ev idence that the se l f - renewa l capac i t y of H S C s may be finite or at least sub ject to exhaus t ion (105, 107, 109). In addit ion, comparat ive s tud ies of the prol i ferat ive act iv i t ies of fetal l iver and adult bone marrow ce l ls have shown that, under the s a m e condi t ions, these may display di f ferences suggest ing intrinsically determined di f ferences in regenerat ive potential. (87, 88, 335). However , none of these studies has rel ied on the use of direct measuremen ts of H S C numbers , at least in part b e c a u s e quantitative a s s a y s for cel ls with long-term repopulat ing ability have only recently become avai lable (43, 45). The critical importance of such a s s a y s is further u n d e r s c o r e d by a growing body of e v i d e n c e indicat ing that the ce l l su r face phenotype is not necessar i ly a reliable indicator of retained s tem cel l funct ion, nor is the p r e s e n c e of terminal ce l l s a rel iable ref lect ion of the s i ze of the H S C compar tment (53, 76, 111). T h u s , parameters that may affect the k inet ics and ul t imate extent of regenera t ion of the H S C compar tmen t ob ta ined after the transplantat ion of different numbers or sources of hemopoiet ic cel ls have not been def ined. To add ress these quest ions, I used a limiting dilution a s s a y for ce l ls with long- term compet i t i ve repopula t ing abil i ty to m e a s u r e H S C regenera t ion in mye loab la ted recipients of different numbers of H S C s from adult bone marrow or fetal liver. 90 5.2. Results 5.2.1. Overall experimental design The purpose of these exper iments was to examine and compare the kinetics of recovery of every level of hemopo ie t i c cel l deve lopmen t in mye loab la ted rec ip ients a s a funct ion of both the C R U content and the origin of the ce l l s t ransp lanted. Donor ce l ls were obta ined from either day 14.5 fetal l ivers or the bone marrow of adult mice injected 4 days previously with 150 mg/kg of 5 - F U and g roups of mice then injected with a range of cel l numbers es t imated to conta in 1000, 100 or 10 C R U , ie. 10%, 1% and 0 . 1 % of the total marrow C R U content of an ave rage untreated adult mouse (43) or 90%, 9% or 0 .9% of the C R U content of a s ingle day 14.5 fetal liver (335). Al l mice a lso received 1 0 5 bone marrow cel ls from an unmanipu la ted adult Ly5.2 B 6 C 3 F 1 donor contain ing an est imated compet ing graft of 10 C R U (53). Rec ip ien ts of these grafts were sacr i f i ced 8-12 months fo l lowing t ransp lanta t ion for a s s e s s m e n t of the test graft ( L y 5 . 1 + ) - d e r i v e d contr ibut ion both to their reconst i tuted per iphera l b lood ce l l s a s wel l a s their marrow C F C , C F U - S and C R U populat ions. 5.2.2. Kinetics of reconstitution of the terminal compartments Flow cytometr ic ana lys is of mature cel ls in the per ipheral b lood 8 months post t ransplant revea led ex tens ive reconstitut ion of both lymphoid and mye lo id compar tments with L y 5 . 1 + test cel ls for all transplant groups (see Tab le 5.1). The proport ion of L y 5 . 1 + ce l ls contributing to both of these mature compar tments w a s highly cons is tent be tween exper iments for e a c h t ransplant d o s e and m a d e up nearly all (> 80%) of the cel ls in recipients of fetal liver cel ls and in recipients of all but the lowest transplant dose of adult bone marrow cel ls (containing 10 C R U ) . In these , the proportion of peripheral blood cel ls that were Ly5.1 + w a s slightly lower, albeit still approximately 5 0 % of the total, reflecting their origin from an equivalent proportion (also - 5 0 % ) of all the C R U transplanted. 91 Table 5.1. Proport ion of L y 5 . 1 + Per ipheral B lood Ce l l s from Pr imary Rec ip ients Transp lanted with Vary ing Numbers of Ly5.1 + Syngen ic Adult B o n e Marrow- or Fetal L iver-Der ived C R U Source and Transplant D o s e Estimated No. C R U Transplanted* % L y 5 . 1 + P B cells Expt. 1 Expt. 2 B M 2 x 1 0 6 B M 2 x 1 0 5 B M 2 x 1 0 4 1000 100 10 9 1 + 2 (5) 77 + 4 (5) 42 ± 7 (13) 91 ± 2 (3) 81 ± 3 (3) 59 + 8 (13) F L 1.7 x 1 0 7 F L 1 .7x10® F L 1 . 7 x 1 0 5 1000 100 10 91 ± 1 (5) 94 ± 0 (3) 79 ± 4 (5) 92 + 0 (2) 89 ± 1 (2) 82 ± 4 (2) V a l u e s shown are the mean ± S D (number of mice analyzed) of the proport ion of L y 5 . 1 + ce l l s in the c i rculat ing W B C populat ions present in pr imary t ransplant recipients ana lyzed 8 months post transplantation in two individual exper iments. * The est imat ion of number of C R U transplanted is der ived from a control va lue of 1/2000 (95% conf idence interval: 1 in 1300 to 1 in 5700) for adult 5 - F U B M (43) and 1/17000 (95% conf idence interval: 1 in 11500 to 1 in 26,000) for fetal l iver (335). 5.2.3. Reconstitution of the marrow T w o mice per transplant group were chosen for further ana lys is of the level of test t ransplant-der ived cel ls in the marrow 8 months post-transplant. In an effort to detect the max imum levels of reconstitution attained by this source , recipients in wh ich at least 8 0 % of the per ipheral b lood ce l ls were Ly5.1+ were s e l e c t e d . In addit ion, the bone marrow cel ls of one of the recipients were a lso s ta ined with the Ly5.1 ant ibody and upon F A C S ana lys is were found, like the b lood , to conta in >70% Ly5.1+ (ie. test transplant-derived) cel ls (data not shown). T h e total mar row cel lu lar i ty and C F C numbers in all pa i rs of pr imary recipients, irrespective of the number or origin of the cel ls initially t ransplanted, had regenera ted to levels equivalent to those found in unmanipu la ted a g e - m a t c h e d control mice (Table 5.2). In contrast, for day 12 C F U - S , this w a s true only for the recipients of the highest transplant dose of fetal liver cel ls. For recipients of 92 Table 5.2. Regeneration of Cellularity, Clonogenic Progenitor and Day 12 C F U - S Content in Femoral Marrow of Mice Transplanted with Various Numbers of Adult BM or Fetal Liver C R U (8 Months Post Transplant) Transplant Sou rce Est imated No . C R U Transp lan ted Total nucleated cel ls/ femur (±SD) C F U C/femur x 1 0 2 ( ± S D ) Day 12 C F U -S/ femur x 1 0 2 (±SD) Adult B M 10 2.1 ±0.7 x 1 0 7 382 ± 6 1 . 0 10.1 +2 .4 100 2.3 + 0.1 x 1 0 7 504 ± 1 0 1 16.1 ± 1 . 8 1000 2 . 5 ± 0 . 5 x 1 0 7 475 + 160 17 + 1.5 Fetal L iver 10 2.1 ± 0 . 5 x 1 0 7 330 + 98.7 11.8 ± 2 . 4 100 2 . 3 ± 0 x 1 0 7 393 ± 17.9 14.3 ± 1.0 1000 2 . 4 ± 0 . 7 x 1 0 7 394 ± 37.4 21.6 ± 1.0 Normal Contro l 2.3 + O. lx 1 0 7 4 7 6 ± 1 3 8 24.4 ± 2.6 Bone marrow cel ls from primary recipients (2 mice per group) 8 months post transplant were counted using a standard hemocytometer and plated in methylcel lulose at 2.7 x 1 0 4 cel ls/ml and colonies scored 12 days later. Regenerat ion of day 12 C F U - S w a s a s s e s s e d by transplanting 7.5 x 1 0 4 -1 x 1 0 5 marrow cel ls from pooled marrow from two primary recipients into 5 irradiated (950cGy) B 6 C 3 F 1 mice/group. An imals were sacr i f iced 12 days later and macroscop ic sp leen colonies counted. Normal control animals were 6-8 month old unmanipulated B 6 C 3 F 1 mice. S e e Tab le 5.1. for ca lcu lated C R U f requencies in unmanipulated adult bone marrow and fetal liver. For each exper iment 2 mice showing the highest proportion of L y 5 . 1 + cel ls in the peripheral b lood were chosen a s donors for secondary transplantat ion. The average proportion of L y 5 . 1 + cel ls in the peripheral blood of chosen mice for each group was ; adult marrow 10 C R U : 8 1 % , 100 C R U : 8 5 % , 1000 C R U : 9 4 % , fetal liver 10 C R U : 8 2 % , 100 C R U : 9 4 % , 1000 C R U : 92%. all other sou rces or numbers of test cel ls , recovery of day 12 C F U - S numbers w a s incomplete ranging from 4 5 % to 8 5 % of normal va lues (Table 5.2). A l though there w a s a trend towards a greater recovery of this more primitive compar tment in rec ip ients of h igher initial t ransplant d o s e s , the actual d i f ferences be tween the groups were not statistically significant (p < 0.05). 5.2.4. Reconstitution of the marrow C R U compartment T o c o m p a r e the a c c o m p a n y i n g level of regenera t ion of L y 5 . 1 + donor -der ived s tem cel ls in the marrow of these s a m e pairs of primary recipients, C R U f requenc ies and hence numbers were a lso determined. The p resence of Ly5.1 + myelo id and lymphoid ce l ls in the peripheral b lood of the seconda ry C R U a s s a y recip ients w a s eva luated after 16 w e e k s and used to der ive the C R U numbers shown in Figure 5.1. In most c a s e s , the Ly5.1+ C R U population had regenerated to a level cor responding to only a smal l proportion of the normal C R U populat ion and the levels ach ieved correlated posit ively with the original dose of L y 5 . 1 + C R U used to reconstitute the primary recipients. Interestingly, the level of C R U regenerat ion was.consis tent ly higher (p < 0.01 to < 0.1) for fetal liver transplants as compared to marrow t ransplants containing the s a m e original number of C R U . Thus , on a per input C R U bas is , C R U ampl i f icat ion in the fetal l iver t ransplants w a s 10X more effective than with primary bone marrow transplants and , only in primary recipients of 1000 C R U of fetal l iver origin, did the t ransplanted C R U recover to a normal s i zed populat ion within 8 months post-transplant. E f fec ts of the t ransplant d o s e a s wel l a s the t i ssue or igin are more dramatical ly revealed by using the same data to calculate the extent of L y 5 . 1 + C R U ampli f icat ion in each experimental situation tested. The results of such calculat ions are shown in Tab le 5.3. It can be s e e n that for both fetal l iver and adult bone marrow transplants, C R U expans ion w a s inversely related to the initial number of C R U transplanted although the extent of C R U expans ion for input fetal 94 100 N o r m a l T 0 > O O 0 rr o z D C O o 10-1-Adult BM Fetal Liver 1 10 100 1000 10000 Number of C R U Transplanted Figure 5 . 1 . R e g e n e r a t i o n of L y 5 . 1 + d o n o r - d e r i v e d c e l l s f o l l ow ing the t ransp lanta t ion into s e c o n d a r y rec ip ients of bone mar row ce l l s f rom pr imary rec ip ients original ly t ransplanted with 10, 100 or 1000 fetal l iver or adult bone marrow C R U 8 months previously. C R U numbers were determined by measur ing the f requency of C R U by limiting dilution ana lys is in s e c o n d a r y rec ip ients and ca lcu lat ing the number of C R U in the marrow of primary transplant recipients a s compared to an unmanipulated control mouse based upon the assumpt ion that an average adult mouse has approximately 200 million marrow cel ls. Fo r each transplant dose C R U regenerat ion by fetal liver w a s signif icantly greater than adult bone marrow; 10 C R U : p < 0.01, 100 C R U : p < 0.1, 1000 C R U : p < 0.01. For adult bone marrow the 10 C R U transplant dose is signif icantly lower than both the 100 and 1000 C R U dose at p < 0 .01. For fetal l iver the 10 C R U d o s e is signif icantly lower than both the 100 and 1000 C R U dose ; 100 C R U : p < 0.1, 1000 C R U : p < 0.01. The 100 C R U dose is significantly lower than the 1000 C R U dose at p < 0 . 1 . The est imation of the number of C R U transplanted is derived from a control va lue of 1/2000 (95% con f idence interval: 1 in 1300 to 1 in 5700) for adult 5 - F U bone marrow (43) and 1/17,000 (95% conf idence interval: 1 in 11,500 to 1 in 26,000) for fetal liver (335). 95 Table 5.3. Expans ion of Donor-Der ived C R U in Pr imary Rec ip ients of Feta l L iver or Adult Bone Marrow Ce l l s Transp lant S o u r c e Est imated No. C R U Transp lan ted * Est imated No. Bone Marrow C R U Post E x p a n s i o n (range + S E M ) f Fo ld Expans ion Adult bm 10 100 (75-150) 10 100 1500 (1000-2100) 15 1000 1800 (1300-2500) 2 Fetal Liver 10 1500 (1000-2100) 148 100 3100 (2300-4200) 31 1000 6200 (4500-8500) 6 Unt ransp lan ted - 10000 (6300-14800) -Cont ro l * S e e T a b l e 5.1 for ca lcu la ted C R U f requenc ies in unman ipu la ted adult bone marrow and fetal liver. f R e s u l t s are exp ressed as the number of C R U per mouse based on the est imate that 2 x 1 0 7 femora l marrow ce l l s const i tutes approx imate ly 1 0 % of the total hemopoiet ic populat ion of the mouse . l iver C R U w a s greater than that of their counterpar ts in bone mar row. T h e max imum C R U expans ion observed was 150-fold for fetal liver C R U by compar ison to a max imum C R U expans ion of only 15-fold for adult bone marrow transplants. 5.2.5. Regenerative ability of a single C R U a s s e s s e d using retroviral mark ing To gain further insight into C R U regenerat ion I used retroviral mark ing for c lona l ana lys i s of se l f - renewal and lympho-myelo id reconst i tut ion. 5 -FU- t rea ted Ly5.1 marrow ce l ls were co-cul tured with cel ls producing a retrovirus conta in ing the cod ing region of the human C D 2 4 cel l sur face ant igen and were then s ta ined with an an t i -CD24 ant ibody and C D 2 4 + cel ls se lec ted by F A C S 48 hours post-infect ion a s descr ibed in Chap te r 2; Mater ial and Methods . 10$ of these Ly5.1 + ce l ls (est imated to contain 3.0 + 1.0 C R U ) were transplanted into Ly5.2+ recipients. 96 Al l of 13 such recipients showed detectable levels of Ly5.1+ (donor cel l -der ived) repopulat ion with va lues ranging from 3 -32% Ly5.1+ per ipheral b lood leukocytes (10/13 myelo id/ lymphoid repopulat ion; 3/13 lymphoid-restr icted repopulat ion). T h e hemopoiet ic t issues of all of these mice also showed the p resence of intact provirus at 11 months post transplant but in only 3 w a s C D 2 4 express ion detectable in the per ipheral b lood leukocytes (where va lues of 2 -20% C D 2 4 + were measured) . O n e of these recipients showing 3 2 % donor-der ived Ly5.1+ ce l ls , of wh ich 6 3 % were C D 2 4 + , w a s chosen for further study to quantify the c lonal regenerat ion of Ly5.1 + C R U . Southern blot analys is of the bone marrow, sp leen and thymus of this mouse s h o w e d ident ical proviral banding patterns and band intensit ies in all of t hese t i s s u e s cons is ten t with the repopulat ion of this pr imary recipient by a s ing le t ransduced totipotent repopulating stem cel l . W h e n the marrow ce l ls of this m o u s e were then a s s a y e d for their content of Ly5.1+ C R U , 0.2 ± 0.05 Ly5.1+ C R U per 1 0 5 marrow cel ls or 50 Ly5.1+ C R U per femur were detected. Th is represents 3 .7% of the C R U populat ion in the femur of a normal adult B 6 C 3 F 1 mouse . S e v e n t e e n of the 19 s e c o n d a r y t ransplant rec ip ients who s h o w e d the p r e s e n c e of Ly5.1 + peripheral b lood cel ls were a lso posit ive for C D 2 4 express ion , with va lues ranging from 1.5% to 1 4 . 2 % C D 2 4 + per iphera l b lood leukocy tes . Str ik ingly, the s a m e proviral banding pattern seen in the primary animal was a lso observed exc lus ive ly in the hemopoiet ic t issues of all of the secondary recipients who were reconsti tuted with Ly5.1+ cel ls (Figure 5.2). Th is observat ion indicates a 370-fold ampli f icat ion of the or ig ina l t r a n s d u c e d C R U dur ing the per iod of 11 mon ths after it w a s transplanted into the primary recipient. This result provides formal ev idence of C R U se l f - renewal in v ivo and ex tends the prev ious f indings indicat ing the extent to which this can occur. 5 . 3 . D i s c u s s i o n P r e v i o u s s tud ies have ind ica ted that the capac i t y of mos t pr imi t ive hemopoiet ic cel ls capab le of regenerat ing the entire sys tem cannot be mainta ined 97 on ser ia l transfer. In this Chap te r I showed that this may be at least partial ly attributable to a common failure of the C R U compartment to be fully regenerated Primary Secondary pos neg. Recipient Recipients ctrl. ctrl. 1 2A 2B 2C 2D 2E 2F 2G 2H 21 b s b s b s t b s t b s t b s t b s t b s t b s t b s — 6.6 Kb — 4.0 Kb — 2.2 Kb F i g u r e 5.2. Demonstrat ion of C D 2 4 provirus in bone marrow (B), sp leen (S), and thymus (T) D N A from primary and secondary t ransplant rec ip ients. Rec ip ien ts r e c e i v e d 10® C D 2 4 + se lec ted ce l ls in combinat ion with 1 0 5 normal mar row compet i to r ce l l s . Individual m ice are labe led as 1 for the pr imary rec ip ient (sacr i f i ced at 11 mon ths post - t ransplant ) , or 2A-I for s e c o n d a r y rec ip ien ts (sacri f iced at 13 weeks post-transplant). D N A (15 u.g) from each t issue sample was d igested with E c o R l , an enzyme that cuts once within the C D 2 4 proviral sequence . S h o w n are results of a blot probed with a 3 2 p - | a D e | e c | fragment of the n e o R gene. The posit ive control represents D N A obtained from C D 2 4 viral producer cel ls which contain > 10 proviral cop ies . even 8 months post-transplant in spite of, indeed perhaps because of, a complete recovery of later cel l types including day 12 C F U - S as well as ce l ls detectable as C F C . Interestingly, there was , nevertheless, a quantitative relationship between the extent of amplif ication seen in donor C R U numbers and both the s i ze of the initial transplant and its ontological source. 98 It has been suggested by severa l authors that t ransplantable s tem cel ls may fail to regenerate the s tem cel l compar tment to normal (non-transplant) leve ls b e c a u s e of an inherent ly l imited capac i t y for se l f - renewa l (103, 105 , 109). A c c o r d i n g to such a mode l , the abso lu te extent of C R U ampl i f icat ion wou ld be ant ic ipated to dec rease as the number of s tem cel ls t ransplanted w a s d e c r e a s e d s ince the number of s tem cel ls with the highest sel f - renewal potential wou ld a lso d e c r e a s e proportionately. O n the other hand, it is poss ib le that the transplantat ion of smal le r numbers of marrow ce l ls might p lace a higher "s t ress" on the sys tem result ing in the production of stimuli that could favor differentiation rather than self-r enewa l r e s p o n s e s . E v i d e n c e of d e c r e a s e d s t e m ce l l r egene ra t i on under condi t ions that support their proliferation both in vitro (336), in utero (11) and after bone mar row t ransplantat ion (102, 103, 105, 109, 117) have been repor ted. However , s u c h s tud ies do not necessar i l y measu re the capac i t i es of the ce l l s tested but rather their response under a given set of molecular ly undef ined and poorly understood environmental condit ions. The present studies, which have used a quantitative a s s a y to provide a direct measurement of the s ize of the regenerated totipotent, t ransplantable s tem cel l compartment indicate that a higher degree of s tem cel l ampl i f icat ion is obtained fol lowing the transplant of smal le r numbers of s tem cel ls a s compa red to the transplant of larger numbers of s tem ce l l s , even though this is insufficient to ach ieve a comparab le level of regenerat ion of the s tem cel l populat ion by compar i son to its s i ze in unperturbed an imals (Table 5.3 and Figure 5.1). Thus , the extent to which either fetal liver or adult bone marrow s tem cel ls express their full regenerat ive potential var ies accord ing to how many of them (and/or a c c o m p a n y i n g marrow cel ls) are t ransplanted and this d e c r e a s e s with innoculum s ize . A poss ib le explanat ion for this f inding would be the act ivat ion of negat ive feedback regulatory mechan isms in vivo that can limit s tem cel l expans ion premature ly , pe rhaps v ia the product ion by mature hemopo ie t i c ce l l s of s u c h factors a s M I P - 1 a and T G F - 3 that may se lect ive ly d e c r e a s e the proport ion of 99 primit ive hemopo ie t i c ce l l s in cyc le (112-115 , 337) s u c h fac tors might thus attenuate or even terminate C R U expans ion even though the numbers of these cel ls might still be significantly below the "normal" level. It is interesting to specu la te that this effect might be promoted by the co-transplantat ion of large numbers of mature hemopo ie t i c ce l ls or their immedia te p recursors . S u c h a possib i l i ty is consis tent with the observat ion that the max imum degree of C R U expans ion w a s obse rved with the smal lest transplant dose . Accord ing ly , it wou ld be ant ic ipated that s tem ce l l e x p a n s i o n might be e n h a n c e d w h e n pur i f ied s tem ce l l s a re t ransplanted to reduce the number of mature cel ls present during the initial s tages of engraftment. It is important to note that in these exper iments only the regenerat ion of donor L y 5 . 1 + C R U were quanti f ied. My results clearly show that the t ransplanted L y 5 . 1 + C R U were unab le to reconst i tute the C R U compar tmen t of pr imary t ransplant recipients to levels found in normal adult mice. However , the extent to which 10 Ly5.2+ C R U present in the competitor cell population or those surviving in the host might have contributed to regeneration of the total C R U compartment is not known. It s e e m s highly unlikely that a large reserve of inactive L y 5 . 2 + C R U would have been present in the primary recipients s ince the vast majority of all the cel ls in the marrow and the peripheral blood of these mice were L y 5 . 1 + . Resu l t s from many prev ious studies have indicated that the regenerat ive behavior of hemopoiet ic cel ls in a transplant setting is a function of their ontological state. Thus , over 20 years ago, it was found that a transplant of fetal l iver ce l l s wou ld outcompete adult bone marrow following their combined transplantat ion into recipient m ice (88). Simi lar ly , 10 years later, it w a s shown that day 8 C F U - S der ived from fetal liver p o s s e s s a greater capaci ty for sel f - renewal as compared to their adult bone marrow counterparts (87). Recent ly, Rebe l et. a l . (335) have shown that limiting numbers of fetal liver C R U are able to p roduce a greater output of mature b lood ce l ls in vivo as compared to adult bone marrow. Moreover , when 100 marrow ce l ls from primary recipients of limiting numbers of fetal l iver C R U were injected into secondary recipients, a signif icantly higher percentage of these mice s h o w e d donor-der ived reconstitution of their lymphoid and myelo id compar tments a s c o m p a r e d to mice that had rece ived marrow ce l ls from primary recip ients of s imi lar numbers of adult bone marrow C R U . T h e s e results are thus a lso highly suggest ive of a greater regenerat ive capabil i ty of fetal liver H S C by compar i son to adult bone marrow. However , because of their des ign, none of these studies cou ld d iscr iminate between quantitative di f ferences in the sel f - renewal of t ransplantable H S C and poss ib le di f ferences in the extent of proliferation ach ieved by their more differentiated progeny. By address ing this quest ion here, it has been poss ib le to es tab l i sh that fetal l iver C R U are indeed super ior to their adult bone mar row counterparts both in terms of the relative and absolute numbers of C R U they will regenerate under similar condit ions. A poss ib le explanat ion for these dif ferences is that fetal liver C R U p o s s e s s a greater intrinsically regulated probability for sel f - renewal when st imulated to divide in the microenv i ronment of the post- t ransplant mye loab la ted m o u s e . Intriguing ev idence of candidate genes that may be involved in the intrinsic control of self-renewal w a s recently reported by S a u v a g e a u et. a l . from studies of the effects of H o x g e n e s on hemopo ies i s in v ivo (91). T h e s e s h o w e d that ove rexp ress ion of H O X B 4 , w h o s e express ion is normal ly restricted to the most primitive adult bone marrow ce l ls in the adult (91) resulted in a 50-fold increase in the regenerat ion of t r ansduced C R U a s c o m p a r e d to neo- t ransduced control ce l l s . Howeve r , the possibi l i ty that the genes exp ressed within fetal liver and adult bone marrow H S C inf luence this se l f - renewal probabil i ty does not prec lude other m e c h a n i s m s that might inf luence the observed dif ferences in the rates of fetal liver and adult marrow C R U ampl i f icat ion. For example , these may a lso exhibit d i f ferences in the t ime requi red to transit one comp le te cel l cyc le (338) or in the proport ion of ce l l s 101 recrui ted or mainta ined within the microenvi ronment of the marrow of the adult mouse . T h e use of recombinant re t rov i ruses a s genet ic tags h a s b e e n u s e d extensively to track the proliferative and differentiative behavior of individual s tem cel l c l ones both in vitro (49) and in vivo (98, 99). In the present study I uti l ized a recombinan t retrovirus conta in ing the cod ing region of the h u m a n C D 2 4 ce l l sur face ant igen to confirm the totipotentiality of the C R U detected using the C R U a s s a y , and to aid in the verif ication and quantif ication of the degree of expans ion exh ib i ted by indiv idual C R U in v ivo. On ly one proviral band ing pattern w a s de tec ted in bone marrow and thymic D N A in the pr imary and al l s e c o n d a r y t ransplant recipients (Figure 5.2), highly suggest ive that the regenerat ion of the s tem cel l compar tment in the primary recipient w a s monoc lona l in nature. Th i s c lone regenerated the C R U compartment to 3 .7% of normal level , comparab le to that observed in recipients of 10 adult bone marrow C R U . This represents a C R U expans ion of 370-fold, which was higher than that observed for any transplant dose of fetal l iver or adult bone marrow (Table 5.3). Di f ferences in es t imates of C R U expans ion at the level of the whole population versus individual c lones likely reflect he te rogene i t y a m o n g s t the regenera t i ve act iv i ty d i s p l a y e d by b io log ica l l y equiva lent individual H S C responding to prol i ferative st imuli (19, 89 , 99 , 101 , 339). However , at the s a m e time the present results confirm the ability of totipotent repopulat ing s tem cel ls to undergo sel f - renewal d iv is ions during c lonal expans ion in vivo and demonstrate the enormous regenerat ive potential that s o m e s tem cel ls may therefore p o s s e s s . T h e resul ts of this study have important impl icat ions for bone mar row transplantat ion efforts. A sufficiently quantitatively or qualitatively impaired pool of hemopoiet ic stem cel ls could inf luence the longevity of a patient's graft, and further, might reduce to lerance to cytotoxic agents or other c i r cums tances wh ich would impose a proliferative demand on the stem cel l pool . Th is study thus highlights the 102 impor tance of opt imiz ing s tem cel l numbers in bone mar row t ransp lan ts and sugges ts potential c o n s e q u e n c e s of transplant ing different numbers or sou rces of s tem cel ls in protocols seek ing to rescue, or, alternatively, to genet ical ly modify the mar row. T h e s e s tud ies a l so set the s tage for a t tempts to e n h a n c e C R U regenerat ion post- t ransplant by the administrat ion of e x o g e n o u s agents or the e x p r e s s i o n of t ransduced intracel lular factors that may e n h a n c e the apparent regenera t ive potent ial of s tem ce l ls e x p r e s s e d under a de f ined exper imen ta l condit ion in vivo. 103 C H A P T E R 6 D I S C U S S I O N The use of recombinant retroviruses to transfer exogenous genes into H S C s h a s p rov ided s ign i f icant insight into the o rgan iza t ion a n d regulat ion of the hemopoie t ic sys tem and has p layed a central role in the emerg ing field of gene therapy. However , many important quest ions remain unanswered regard ing the nature and regulation of totipotent H S C s , including a bas ic understanding of their usage over t ime, their potential for se l f - renewal and prol i feration, and the g e n e s e n c o d i n g ex t race l lu la r and /o r in t racel lu lar fac tors wh ich are r espons ib l e for regulat ing t h e s e b io log ica l charac ter is t i cs . A l though the abil i ty to genet ica l l y manipulate H S C s using recombinant retroviruses provides a powerful tool to begin to add ress these issues , the poor infection eff iciency to H S C s remains an obstac le. T h e overal l goal of the work presented in this thesis was to develop methodolog ies to e n h a n c e the utility of current gene transfer protocols for the efficient genet ic manipulat ion and tracking of H S C s and to utilize these procedures to more clearly def ine the sel f - renewal potential of H S C s following bone marrow transplant. Work descr ibed in Chapter 3 demonstrates the use of the C D 2 4 cel l sur face ant igen a s a dominant se lec tab le marker in a recombinant retroviral vec tor in combinat ion with F A C S to enab le the rapid and non-toxic se lect ion of retrovirally t ransduced mur ine bone marrow ce l ls including in vitro c lonogen ic progeni tors, day 12 C F U - S and totipotent repopulat ing s tem ce l ls . The p resence of totipotent repopulat ing s tem cel ls within the CD24+ fraction (Chapters 3 and 4) demonst ra tes that the M P S V and M S C V L T R regulatory e lements are able to drive h igh- level gene exp ress ion in the most primitive hemopoiet ic ce l ls present in adult marrow t i s sue . T h e abil i ty of t hese e l emen ts to dr ive g e n e e x p r e s s i o n in pr imit ive hemopoie t ic ce l ls w a s a lso directly conf i rmed in s tud ies desc r ibed in Chap te r 4 with the ana lys is of C D 2 4 express ion among a S c a + L i n " subpopu la t ion of bone marrow ce l ls . T h e s e observat ions have a number of important impl icat ions. T h e 104 abil i ty to ach ieve high and sus ta ined levels of t ransferred g e n e exp ress i on in pr imit ive hemopo ie t i c ce l l s is essen t ia l for s tud ies a i m e d at tes t ing , through ove rexp ress ion , g e n e s encod ing mo lecu les which may be involved in regulat ing the surv ival , mobility, cyc l ing, proliferation, differentiation, or sel f - renewal of H S C s . Moreover , the ability to obtain a populat ion of cel ls in wh ich 1 0 0 % are proviral ly marked should enhance the detection of phenotypes which may be assoc ia ted with the o v e r e x p r e s s i o n of potential putat ive s tem cel l regulatory m o l e c u l e s . T h e overexpress ion of H o x B 4 has been shown to increase the regenerat ive ability of H S C s fol lowing bone marrow transplant (91). Addit ional mo lecu les which would be of interest to test using this procedure include the l igand for the flt3/flk2 receptor, s i nce the exp ress ion of this receptor is restricted to primitive hemopo ie t ic ce l l s (137), and V L A - 4 whose interactions with V C A M - 1 , f ibronectin and L-select in have b e e n imp l i ca ted in med ia t ing the in v ivo homing of pr imi t ive hemopo ie t i c progenitors to the marrow and sp leen (155). O n e intriguing finding descr ibed in Chapter 4 of this thesis w a s the detect ion of vector related d i f ferences in the pers is tence of exp ress ion of the t ransferred C D 2 4 gene fol lowing ex tended per iods of t ime in v ivo. A greater proport ion of C D 2 4 + lymphocytes, g ranu locy tes /macrophages, erythrocytes and Sca+L in " s tem cel l cand ida tes were observed in the recipients of M S C V C D 2 4 t ransduced bone marrow a s compa red to J Z e n C D 2 4 t ransduced marrow despi te the detect ion of roughly equivalent levels of intact recombinant provirus. O n e explanat ion for this is shutdown of the viral (JZen) L T R regulatory e lements in vivo, a phenomenon which h a s b e e n p rev ious ly repor ted (215, 251) , and wh ich may be the result of methylat ion of the promoter and/or enhance r e lements (257, 331) . In addi t ion, d i f ferences in the express ion of the transferred C D 2 4 gene were detected a m o n g ce l ls of different l ineages. The proportion of C D 2 4 + thymocytes w a s signif icant ly lower than that obse rved a m o n g ce l ls of either the mye lo id (bone marrow) or erythroid l ineage (Chapter 4). Together, these results suggest that despi te efforts to 105 const ruc t / ident i fy vec to rs ab le to dr ive high a n d pers is ten t l eve ls of g e n e express ion in all which contain recombinant provirus ( irrespective of cel l l ineage), this goal has not yet been real ized. Never the less, the ef fect iveness of C D 2 4 a s a m e a n s to rapidly quantify the levels of transferred gene express ion attainable from different vectors sugges ts that this marker would be useful in efforts to identify regulatory e lements that max imize the levels and pers is tence of t ransferred gene exp ress ion in H S C s or any other target cel l of interest. Recent ly , a novel vector ca l led M F G (Richard Mul l igan, Whi tehead Institute for B iomedica l R e s e a r c h , M.I.T., C a m b r i d g e , MA) has been shown to provide higher leve ls of t ransfer red gene express ion than Mo loney murine leukemia virus based vectors (340). In addit ion, Dr. Dona ld Kohn and his co l l eagues (Chi ldrens Hospi ta l , L o s A n g e l e s , C A ) are current ly at tempt ing to so l ve the prob lem of promoter shu tdown in v ivo by identifying and removing potential methylat ion s i tes present in vec tors b a s e d on the Mo loney murine leukemia virus. It would be interesting to use C D 2 4 to quantify and compare the levels and pers is tence of gene express ion obtainable from these vectors as compared to J Z e n and M S C V . C D 2 4 would a lso be of use in efforts to rapidly sc reen alternative retroviral infect ion protoco ls involv ing the use of newly d i scove red hemopo ie t i c growth factors (eg. flt3/flk2 l igand), or newly deve loped strategies such as those involving the use of f ibronectin (descr ibed in detail on page 37-38) or the "f low-through" viral t ransduct ion sys tem (descr ibed on page 38). A s an examp le of this, Dr. C r a i g J o r d a n (Somat ix , A l a m e d a , C A ) is currently us ing C D 2 4 a s a marker of gene transfer in combinat ion with F A C S in an effort to correlate the cyc l ing status of primitive human hemopoiet ic ce l ls with the eff ic iency of gene transfer (personal communica t ion) . The select ion approach descr ibed and util ized in Chap te rs 3 and 4 has now been extended to the human sett ing. Us ing retroviral vectors contain ing H S A , the mur ine homo logue of C D 2 4 , in combinat ion with F A C S , C o n n e a l l y et. a l . h a s 106 demons t ra ted the abil i ty to enr ich for retroviral ly t r ansduced primit ive h u m a n hemopoiet ic cel ls , including L T C - I C (275). Efforts to determine whether t ransduced ce l ls capab le of repopulat ing immunodefic ient mice can be enr iched for using this technique are currently underway. In col laborative efforts with Dr. Stefan Kar l ssons ' lab (National Institutes of Heal th, Be thesda , MD) the select ion approach deve loped in Chap te r 3 has been appl ied to test the therapeutic potential of vectors a imed at the genet ic therapy of G a u c h e r s d i sease (a human autosomal recess ive l ysosomal s to rage d isorder c a u s e d by a def ic iency of the e n z y m e g lucoce reb ros idase ) . A retroviral vector conta in ing the c D N A encod ing H S A a s a se lec tab le marker in combinat ion with a therapeut ic g lucocerebros idase (GC) c D N A w a s const ruc ted and used to infect t ransformed B cell l ines from G a u c h e r patients. Whi le retrovirally t ransduced, unselected B cel ls showed a slight increase in the level of G C enzyme a s compared to unt ransduced B cel ls , F A C S select ion of H S A + B cel ls e n h a n c e d G C e n z y m e levels to more than 5-fold over unt ransduced cel l levels, and even to leve ls more than 2-fold over those in normal n o n - G a u c h e r ce l l s (341). S u c h strategies based on this approach may enable the implantation of t ransduced ce l ls to an ima ls or patients, al though the use of cel l sur face ant igens a s markers for human gene therapy remains controvers ia l . Never the less , this app roach shou ld p r o v e use fu l in p rec l i n i ca l s t u d i e s (eg . us i ng n o n - h u m a n p r i m a t e s or immunocompromised mice) a imed at optimizing the therapeutic potential of vectors for use in human gene therapy trials. The observat ion that the C D 2 4 cel l sur face ant igen can be exp ressed on red blood cel ls sugges ts potential for this mo lecu le a s a se lec tab le marker in vectors a imed at the genet ic t reatment of s i ck le cel l a n e m i a and t h a l a s s e m i a s . In co l laborat ive efforts with Dr. Ph i l i ppe L e B o u l c h (M.I.T., Bos ton , MA) , Dr. Conn ie E a v e s (Terry Fox Laboratory, Vancouver , B.C.) and Dr. R o n a l d Nage l (Albert E ins te in C o l l e g e of Med ic i ne , Bronx , N Y ) C D 2 4 is current ly be ing emp loyed in retroviral constructs conta in ing a human 8/(3-globin 107 fusion gene des igned for anti-sickl ing propert ies with the goal of test ing these in a (3s (sickle) t ransgenic mouse model . T h e low eff iciency of gene transfer to H S C s has made efforts to track the ut i l izat ion, prol i ferat ion and se l f - renewa l of ind iv idual H S C s us ing retroviral marking techn iques extremely laborious. The ability to obtain, prior to transplant, a populat ion of ce l ls in which 100% are provirally marked would inc rease the power and e a s e of retroviral marking studies a imed at address ing these i ssues . Resu l ts presented in Chapte r 4 demonstrate that the C D 2 4 select ion approach can enab le the isolation of a population of hemopoiet ic target cel ls in which virtually 1 0 0 % are provi ra l ly m a r k e d and thus, a l lows one to t rack the behav io r of aM. H S C s t ransplanted. The power of this approach to track individual H S C s is demonst ra ted in F igure 4 .3 . T h e transplant of exc lus ive ly provirally marked H S C s enab led the determinat ion of whether hemopo ies i s was polyc lonal (top panel) or monoc lona l (bottom panel) , and a lso a l lowed the accurate determinat ion of the number and relat ive contr ibut ion of each H S C to hemopo ies is in the pr imary recipient at the t ime of sacr i f ice. The dynamics of H S C util ization under homeosta t ic condi t ions rema ins un reso l ved . O n e app roach to add ress i ng this ques t ion wou ld be to sequent ia l l y s a m p l e the per iphera l b lood of mice t ransp lanted with retroviral ly t ransduced H S C s obtained fol lowing the C D 2 4 select ion procedure. S tud ies us ing sequen t i a l a n a l y s i s of per iphera l b lood to t rack ind iv idua l H S C s in m i ce t ransp lan ted with retroviral ly t r a n s d u c e d (non-se lec ted) mar row h a v e b e e n descr ibed (98, 99). Alternatively, a procedure has been descr ibed whereby femoral marrow can be sequent ia l ly obtained from a mouse by inserting a need le into the knee joint of an anes the t i zed m o u s e (342). Suf f ic ient quant i t ies of D N A for Southern blot ana lys is is obtained by using the extracted marrow to generate day 12 C F U - S in secondary recipient mice. The use of this procedure in combinat ion with the C D 2 4 se lec t ion app roach would enab le the contr ibut ion of indiv idual H S C s to hemopo ies is to be t racked, theoretically, for the entire life of the recipient 108 m o u s e a n d shou ld prov ide insight into the number and pe rs i s tence of H S C s contributing to hemopo ies is under homeostat ic condit ions. T h e results of transplant s tudies presented in Chapte r 5 reveal the inability of H S C numbers to recover to normal levels fol lowing the transplant of even large numbers of adult bone marrow or fetal liver s tem ce l ls . T h e s e results may have re levance to cl in ical bone marrow transplantat ion s tudies. A l though c l in ical bone mar row t ransplantat ion has been per formed for only 20 to 25 yea rs , today it const i tutes an major e lement of therapy for the treatment of numerous forms of c a n c e r a s we l l a s her i tab le hema to log i ca l d i so rde rs . Indeed , o v e r 6 ,000 au to logous bone marrow transplants were performed in North A m e r i c a in 1994 a lone (personal communica t ion : International B o n e Marrow Transp lant Regist ry) . Might the poor recover of s tem cell numbers following bone marrow transplant be a contributing factor to marrow fai lure? It will be important to monitor the recipients of bone marrow grafts for long per iods to attempt to determine whether poor s tem cel l recovery p lays a role in bone marrow failure and subsequent death of the patient. Moreover , the poor recovery of H S C numbers fol lowing bone marrow transplant shou ld be taken into considerat ion when attempting human gene therapy, which may require the select ion of t ransduced stem cel ls, as well as the transplant of fetal l iver or umbil ical cord blood H S C s s ince both may require the transplant of modest numbers of cel ls into the patient. The ability to transplant H S C s in which 100% are provirally marked provided intriguing insight into the regenerat ive capaci ty of H S C s fol lowing bone marrow transplant (Chapter 5). Provi ra l integration ana lys is of mice receiv ing retrovirally t ransduced, C D 2 4 + se lec ted , bone marrow cel ls provided ev idence for a >300-fold c lona l ampl i f icat ion of a s ingle t ransduced stem cel l . A number of ques t ions are ra ised by this f inding. Do all H S C s p o s s e s s such a capaci ty for regenerat ion? If so , what factors or c i rcumstances will enable this potential to be fully real ized fol lowing t ransplant? T h e s e f indings set the stage for attempts to max imize the recovery of 109 H S C s numbers fol lowing bone marrow transplant through the adminis t rat ion of e x o g e n o u s agen ts s u c h a s hemopo ie t ic growth factors or the e x p r e s s i o n of t r a n s d u c e d g e n e s encod ing intracel lu lar factors . It wou ld be interest ingly to determine whether the administration of the "stem cel l " factors such as Stee l factor and/or F L fol lowing bone marrow transplant may increase the level of regenerat ion of s tem cel l numbers. Alternatively, such an exper imental strategy cou ld be uti l ized to test the role that particular genes , found to be exp ressed in primitive hemopoiet ic ce l l s , may play in regulat ing the se l f - renewal or commi tment of t hese ce l l s by ove rexpress ion us ing recombinant retroviral vectors comb ined with the use of a dominant se lectab le marker gene such as C D 2 4 . In summary these studies have provided methods which enhance the utility of current gene transfer protocols and have for the first t ime provided quantitative da ta regard ing the regenerat ion of H S C numbers fo l lowing the t ransplant of var ious numbers of H S C s from var ious ontological sou rces . T h e s e studies set the s tage for future exper iments des igned to provide further insight into the behav ior and biological potential of H S C s as well as studies a imed at deve lop ing improved vectors for use in gene therapy. 110 C H A P T E R 7 R E F E R E N C E S 1. Ers lev, A . J . 1983. Product ion of erythrocytes. In Hematology. McGraw-H i l l , N e w York. 2. Dancey , J .T. , K.A. Deubel le iss, L.A. Harker, and C . A . F inch. 1976. Neutrolphil kinetics in man. J . C l in . Invest. 58:705. 3. L ichtman, M.A. 1981. The ultrastructure of the hemopoiet ic environment of the marrow. Exp . Hemato l . 9:391. 4. Bradley, T .R . , and D. Metcalf. 1966. The growth of mouse bone marrow cel ls in vitro. Aust . J . Exp . Biol . M e d . S c i . 44:287. 5. Pluznik, D.H. , and L. S a c h s . 1965. The cloning of normal mast cel ls in t issue culture. J . Ce l l . C o m p . Phys io l . 66:319. 6. E a v e s , C . J . 1995. In vitro a s s a y s for quantitating hemopoiet ic progenitor cel ls . In In Wi l l iams Hematology. W . J . Wi l lam, editor. M c G r a w Hill inc. 7. Metcalf, D., S . Merchav, and G . Wagemaker . 1982. Commitment by G M -C S F or M - C S F of bipotential G M progenitor cel ls to granulocyte or macrophage formation. In Exper imental Hematology Today. L. B a u m , Thierfelder, editor. B a s e l : Karger. 8. Gregory , C . J . , and A . C . E a v e s . 1977. Human marrow cel ls capab le of erythropoietic differentiation in vitro: Definition of three erythroid co lony responses . B lood 49:855. 9. Metcalf, D., H.R. MacDona ld , N. Odar tchenko, and B. Sordat. 1975. Growth of mouse megakaryocyte colonies in vitro. Proc . Natl. A c a d . S c i . U S A 72 :1744 . 10. S u d a , T., S . O k a d a , J . S u d a , Y . Miura, M. Ito, T. Sudo , S . Hayash i , S . N ish ikawa, and H. Nakauch i . 1989. A stimulatory effect of recombinant murine interleukin-7 (IL-7) on B-cel l colony formation and an inhibitory effect of IL-1a. B lood 74:1936. 11. Metcalf, D., and M . A . S . Moore. 1971. Hematopoiet ic cel ls. In Frontiers of Biology. A . Neuberger and E.L. Tatum, editors, Ams te rdam, North Hol land. 448 . 12. V a n Zant, G . 1984. Studies of hematopoiet ic s tem cel ls spared by 5-f luorouraci l . J . Exp . M e d . 159:679. 13. Fauser , A .A . , and H.A. Messner . 1979. Identification of megakaryocy tes , mac rophages and eosinophi ls in co lon ies of human bone marrow containing neutrophil ic granulocytes and erythroblasts. B lood 53 :1023 . 111 14. J o h n s o n , G . R . , and D. Metcalf. 1977. Pure and mixed erythroid co lony formation in vitro st imulated by sp leen condit ioned medium with no detectable erythropoietin. Proc . Natl . A c a d . Sc i U S A 74:3879. 15. Bradley, T .R. , and G . S . Hodgson . 1979. Detection of primitive macrophage progenitor cel ls in mouse bone marrow. B lood 54 :1446. 16. Nakahata , T., and M. O g a w a . 1982. Identification in culture of a c l ass of hemopoiet ic colony-forming units with extensive capabil i ty to sel f - renew and generate multipotential hemopoiet ic co lon ies. Proc . Nat l . A c a d . S c i . U S A 79 :3843. 17. Leary, A . G . , and M. O g a w a . 1987. Blast cell co lony a s s a y for umbil ical cord b lood and adult bone marrow progenitors. B lood 69 :953. 18. Rowley , S . D . , S . J . Shark is , C . Hattenburg, and L.L. Sensenbrenner . 1987. Cul ture from human bone marrow of blast progenitor cel ls with an extensive proliferative capaci ty. B lood 69:804. 19. Humphr ies, R.K., A . C . E a v e s , and C . J . E a v e s . 1981. Sel f - renewal of hemopoiet ic stem cel ls during mixed colony formation in vitro. Proc . Nat l . A c a d . S c i . U S A 78:3629. 20 . Til l , J . E . , and E.A. McCu l l och . 1961. A direct measurement of the radiation sensitivity of normal mouse bone marrow cel ls. Radiat ion R e s e a r c h 14:213. 2 1 . Becker , A . , E.A. McCu l l och , and J . E . Til l. 1963. Cytological demonstrat ion of the c lonal nature of sp leen co lon ies der ived from transplanted mouse marrow cel ls . Nature 197:452. 22 . W u , A . M . , L. Siminovi tch, J . E . Til l, and E.A. McCu l l och . 1968. Ev idence for a relationship between mouse hemopoiet ic s tem cel ls and ce l ls forming co lon ies in culture. Proc . Natl. A c a d . S c i . U S A 59:1209. 23 . Dick, J . E . , M .C . Magl i , D. Huszar , R.A. Phil l ips, and A . Bernste in. 1985. Introduction of a se lectable gene into primitive s tem cel ls capab le of long term reconstitution of the hemopoiet ic sys tem of W / W v mice. Ce l l 42 :71 . 24. Lemischka , I.R., D.H. Raulet, and R . C . Mul l igan. 1986. Developmenta l potential and dynamic behavior of hematopoiet ic stem cel ls . Ce l l 45 :917. 25 . Curry, J .L . , and J . J . Trentin. 1967. Hemopoiet ic sp leen colony studies. I. Growth and differentiation. Deve l . Bio l . 15:395. 26 . Magl i , M . C , N .N . Iscove, and V . Odar tchenko. 1982. Transient nature of early haematopoiet ic sp leen co lon ies. Nature 295:527. 27 . Hodgson , G . S . , and T .R . Bradley. 1979. Propert ies of haematopoiet ic s tem cel ls surviving 5-fluorouracil treatment: ev idence for a p r e - C F U - S cel l? Nature 281 :381 . 112 28. Siminovi tch, L , E.A. McCu l loch , and J . E . Til l. 1963. The distribution of colony-forming cel ls among sp leen colonies. J . Ce l l . C o m p . Phys io l . 62 :327 . 29 . W u , C , and M. L iu. 1984. Character ist ics and differentiation of sp leen colony-forming cel ls from bone marrow. International Journa l of Ce l l C lon ing 2:69. 30 . Mulder, A . H . , J . W . M . V isser , and G . J . V a n den Engh . 1985. Thymus regenerat ion by bone marrow cel l suspens ions differing in the potential to form early and late sp leen colonies. Exp . Hemato l . 13:768. 3 1 . Spangrude , G . J . , S . Heimfeld, and I.L. W e i s s m a n . 1988. Purif ication and character izat ion of mouse hematopoiet ic s tem cel ls . S c i e n c e 241:58. 32 . J o n e s , R . J . , J . E . Wagner , P. Ce lano , M .S . Z icha , and S . J . Shark is . 1990. Separat ion of pluripotent haematopoiet ic stem cel ls from sp leen colony-forming cel ls . Nature 347:188. 33 . Orl ic, D., and D .M. Bodine. 1994. What def ines a pluripotent hematopoiet ic s tem cel l ( P H S C ) : will the real P H S C p lease stand up! B lood 84 :3991 . 34. Nakano , T., N. Wak i , H. A s a i , and Y . Ki tamura. 1987. Long-term monoc lona l reconsitution of erythropoiesis in genetical ly anemic W / W v mice by injection of 5-fluorouracil-treated bone marrow cel ls of Pgk-1a /Pgk-1b mice. B lood 70:1758. 35 . Kel ler, G . , C . Pa ige , E. G i lboa , and E .F . Wagner . 1985. Express ion of a foreign gene in myeloid and lymphoid cel ls der ived from multipotent haematopoiet ic precursors. Nature 318:149. 36 . C a p e l , B., R. Hawley, L. Covarrub ias, T. Hawley, and B. Mintz. 1989. C lona l contributions of smal l numbers of retrovirally marked hematopoiet ic s tem cel ls engrafted in unirradiated neonatal W / W v mice. Proc . Natl . A c a d . S c i . U S A 86:4564. 37 . Mart in, F .H . , S .V . Suggs , K .E . Langley, and et.al. 1990. Pr imary structure and functional express ion of rat and human stem cell factor D N A s . Ce l l 63 :203. 38 . C o z e , C M . 1994. G lossary of cytokines. In Cytok ines and Growth Factors , vol . 7:1. M. Brenner, editor. Bail l iere Tindal l , London. 17. 39 . McCu l l och , E.A. , and J . E . Til l. 1960. The radiation sensitivity of normal mouse bone marrow cel ls, determined by quantitative marrow transplantation into irradiated mice. Radiat ion R e s e a r c h 13:115. 40 . J o n e s , R . J . , P. Ce lano , S . J . Shark is , and L.L. Sensenbrenner . 1989. Two p h a s e s of engraftment establ ished by serial bone marrow transplantat ion in mice. B lood 73:397. 113 4 1 . Spangrude , G . J . , and G . R . Johnson . 1990. Rest ing and act ivated subsets of mouse multipotent hematopiet ic stem cel ls. Proc . Natl . A c a d . S c i . U S A 87 :7433 . 42 . Brecher , G . , S . Neben , M. Y e e , J . Bull is, and E . P . Cronki te. 1988. Pluripotential s tem cel ls with normal or reduced self renewal survive lethal irradiation. Exp . Hemato l . 16:627. 4 3 . Sz i l vassy , S . J . , R.K. Humphr ies, P . M . Lansdorp, A . C . E a v e s , and C . J . E a v e s . 1990. Quantitative assay for totipotent reconstituting hematopoiet ic stem cel ls by a competit ive repopulation strategy. P roc . Natl . A c a d . S c i . U S A 87:8736. 44. Harr ison, D .E . 1980. Compet i t ive repopulation: a new a s s a y for long-term stem cel l functional capaci ty. B lood 55:77. 45 . Harr ison, D.E. , C .T . Jordan, R.K. Zhong , and C M . Ast le. 1993. Primitive hemopoiet ic stem cel ls: direct a s s a y of most productive populat ions by competi t ive repopulation with s imple binomial , correlat ion and covar iance calculat ions. Exp . Hemato l . 21:206. 46 . Rebe l , V.I., W . Dragowska, C . J . Eaves , R.K. Humphries, and P . M . Lansdorp . 1994. Amplif icat ion of S c a - 1 + L in" W G A + cel ls in se rum-free cultures containing Steel Factor, interleukin-6, and erythropoietin with maintenance of cel ls with long-term in vivo reconstituting potential. B lood 83 :128 . 47 . Sz i l vassy , S . J . , and S . Cory . 1993. Phenotypic and functional character izat ion of competit ive long-term repopulating hematopoiet ic s tem cel ls enr iched from 5-fluorouracil-treated murine marrow. B lood 81:2310. 48 . Spangrude , G . J . , and R. Scol lay. 1990. A simplif ied method for enr ichment of mouse hematopoiet ic stem cel ls. Exp . Hemato l . 18:920. 49 . Fraser , C . C . , S . J . Sz i l vassy , C . J . Eaves , and R.K. Humphr ies. 1992. Proliferation of totipotent hematopoiet ic stem cel ls in vitro with retention of long-term competit ive in vivo reconstituting ability. Proc . Natl . A c a d . S c i . U S A 89:1968. 50 . Boggs , D.R., S . Boggs , D. Safe , L.A. G r a s s , and D.R. Canf ie ld . 1982. Hematopoiet ic stem cel ls with high proliferative potential. A s s a y of their concentrat ion in marrow by the frequency and duration of cure of W / W v mice. J . C l in . Invest. 70:242. 5 1 . Harr ison, D.E. , C M . Ast le , and C . Lerner. 1988. Number and cont inuous proliferative pattern of transplanted primitive immunohematopoiet ic s tem cel ls . Proc . Natl . A c a d . S c i . 85:822. 52 . V isser , J . W . M . , and P. de Vr ies . 1988. Isolation of sp leen-co lony forming cel ls ( C F U - S ) using wheat germ agglutinin and rhodamine 123 label ing. B lood Ce l l s 14:369. 114 53. Rebe l , V.I., W . Dragowska, C . J . Eaves , R.K. Humphr ies, and P . M . Lansdorp . 1994. Amplif icat ion of Sca-1+ L in- W G A + cel ls in serum-free cultures containing steel factor, interleukin-6, and erythropoietin with maintenance of cel ls with long-term in vivo reconstituting potential. B lood 83 :128 . 54. Worton, R . G . , E.A. McCu l loch , and J . E . Til l. 1969. Phys ica l separat ion of hemopoiet ic s tem cel ls differing in their capaci ty for sel f - renewal . J . Exp . M e d . 130:91. 55 . Turner, R .W.A. , L. Siminovi tch, E.A. McCu l loch , and J . E . Til l. 1967. Density gradient centifugation of hemopoeit ic colony forming cel ls . J . C e l l . Phys io l . 69 :873 . 56 . Moore , M .A .S . , T .A . McNei l l , and J . S . Haski l l . 1970. Density distribution ana lys is of in vivo and in vitro colony forming cel ls in developing fetal liver. J . C e l l . Phys io l . 75:181. 57 . Haski l l , J . S . , and M . A . S . Moore. 1970. Two d imensional cel l separat ion: compar ison of embryonic and adult haemopoiet ic s tem cel ls . Nature 226 :853 . 58. Orl ic, D., and D .M. Bodine. 1992. Pluipotent hematopoiet ic s tem cel ls of low and high density can repopulate W / W v mice. Exp .Hemato l . 20 :1291 . 59 . Sz i l vassy , S . J . , P . M . Lansdorp, R.K. Humphr ies, A . C . E a v e s , and C . J . E a v e s . 1989. Isolation in a single step of a highly enr iched murine hematopoiet ic stem cel l population with competit ive long term repopulat ing ability. B lood 74:930. 60. Kohler, G . , and C . Milstein. 1975. Cont inuous cultures of fused cel ls secret ing ant ibody of predef ined specif icity. Nature 256:495. 6 1 . V isser , J . W . M . , J . G . J . Bauman , A . H . Mulder, J . F . E l iason, and A . M . deLeeuw. 1984. Isolation of murine pluripotent hemopoiet ic s tem cel ls . J . Exp . M e d . 59:1576. 62 . Jurec ic , R., T. van Nguyen, and J . W . Belmont. 1993. Enr ichment and functional character izat ion of Sca -1+WGA+, L i n -WGA+ , L in -Sca+, and L in-S c a - 1 + W G A + bone marrow cel ls from mice with an Ly -6a haplotype. B lood 82 :2673 . 63 . Huang , H., and R. Auerbach . 1993. Identification and character izat ion of hematopoiet ic stem cel ls from the yolk sac of the early mouse embryo. Proc . Natl . A c a d . S c i . U S A 90:10110. 64. Jo rdan , C.T . , J . P . M c K e a r n , and I.R. Lemischka . 1990. Cel lu lar and developmenta l propert ies of fetal hematopoiet ic s tem cel ls . Ce l l 61 :953. 115 65 . P loemacher , R .E . , J . C . M . van der Loo, C . A . J , van Beurden, and M . R . M . Baert. 1993. Wheat germ agglutinin affinity of murine hemopoiet ic s tem cell subpopulat ions is an inverse function of their long term repopulat ing ability in vitro and in vivo. Leukemia 7:120. 66 . Down, J . D . , and R . E . P loemacher . 1993. Transient and permanent engraftment potential of murine hematopoiet ic s tem cel l subsets : differential effects of host condit ioning with g a m m a radiation and cytotoxic drugs. Exp . Hemato l . 21:913. 67. Jurec ic , R., N T . V a n , and J . W . Belmont. 1993. Enr ichment and functional character izat ion of S c a - 1 + W G A + , L i n " W G A + , L i n " S c a - 1 + , and L i n - S c a -1 + W G A + bone marrow cel ls from mice with an Ly -6a haplotype. B lood 82 :2673 . 68 . Morr ison, S . J . , and I.L. W e i s s m a n . 1994. The long-term repopulating subset of hematopoiet ic stem cel ls is deterministic and isolable by phenotype. Immunity 1:661. 69 . F leming, W . H . , E . J . A lpern, N. Uch ida, K. Ikuta, G . J . Spangrude, and I.L. W e i s s m a n . 1993. Funct ional heterogeneity is assoc ia ted with the cel l cyc le status of murine hematopoiet ic stem cel ls. J . Ce l l . B io l . 122:897. 70 . Uch ida , N., H.L. Agui la , W . H . F leming, L. Jerabek, and I.L. W e i s s m a n . 1994. Rap id and susta ined hematopoiet ic recovery in lethally irradiated mice transplanted with purified Thy-1.1 loL in-Sca+ hematopoiet ic s tem cel ls . B lood 83:3758. 7 1 . Ikuta, K., T. K ina, I. MacNe i l , N. Uchida, B. Peault , Y . Ch ien , and I.L. W e i s s m a n . 1990. A developmental switch in thymic lymphocyte maturation potential occurs at the level of hematopoiet ic s tem cel ls . Ce l l 62 :863. 72 . L i , C .L . , and G . R . Johnson . 1992. Long-term hemopoiet ic repopulat ion by Thy-1lo, L in- , Ly6A/E+ cel ls. Exp . Hemato l . 20:1309. 73 . Co rso , A . , M . G . C . Hogeweg-Platenburg, P. de Vr ies , and J . W . M . V isser . 1993. A protocol for the enrichment of different types of C F U - S from fetal mouse liver. Haemato log ica 78:5. 74. Morr ison, S . J . , E. L a g a s s e , and I.L. W e i s s m a n . 1994. Demonstrat ion that Thylo subsets of mouse bone marrow that express high levels of l ineage markers are not significant hematopoiet ic progenitors. B lood 83 :3480. 75 . Spangrude, G . J . , and D .M. Brooks. 1993. Mouse strain variability in the express ion of the hematopoiet ic stem cell antigen L y - 6 A / E by bone marrow cel ls . B lood 82:3327. 76. Spangrude, G . J . , D .M. Brooks, and D.B. Tumas . 1995. Long-term repopulat ion of irradiated mice with limiting numbers of purified hematopoiet ic stem cel ls: In vivo expans ion of s tem cell phenotype but not function. B lood 85:1006. 116 77. P loemacher , R .E . , and R . H . C . Brons. 1989. Separat ion of C F U - S from primitive cel ls responsib le for reconstitution of the bone marrow hemopoiet ic stem cell compartment following irradiation : Ev idence for a p r e - C F U - S cel l . Exp . Hemato l . 17:263. 78 . McAl ister, I., N .S . Wolf, M .E . Pietrzyk, and P . S . Rabinovi tch. 1990. Transplantat ion of hematopoeit ic stem cel ls obtained by a comb ined dye method fractionation of murine bone marrow. B lood 75:1240. 79 . L i , C .L . , and G . R . Johnson . 1992. Rhodamine123 reveals heterogeneity within murine Lin", S c a - 1 + hemopoiet ic stem cel ls. J . Exp . M e d . 175:1443. 80 . Wolf, N .S. , A . Kone , G .V . Priestley, and H. Barte lmez. 1993. In vivo and in vitro character izat ion of long-term repopulating primitive hematopoiet ic ce l ls isolated by sequent ia l Hoechst 33342-rhodamine 123 F A C S select ion. Exp . Hemato l . 21:614. 8 1 . Chaudhary , P . M . , and I.B. Ron inson. 1991. Express ion and activity of P-glycoprotein, a multidrug efflux pump, in human hematopoiet ic s tem cel ls . Ce l l 66 12:1. 82 . Metcalf, D., and M . A . S . Moore. 1971. Haemopoiet ic Ce l l s . Frontiers of Biology, eds . A . Neuberger and E.L. Tatum, 24. North-Hol land Publ ish ing C o m p a n y , Amste rdam, London. 550 pp. 83 . C u m a n o , A . , C . Furlonger, and C . J . Pa ige . 1993. Differentiation and character izat ion of B-cel l precursors detected in the yolk s a c and embyro body of embryos beginning at the 10-to 12-somite stage. Proc . Natl . A c a d . S c i . U S A 90:6429. 84. Medv insky, A .L . , N . L Samoy l ina , A . M . Muller, and E.A. Dz ierzak. 1993. A n early pre-l iver intra-embryonic source of C F U - S in the developing mouse . Nature 364:64. 85 . Medv insky, A .L . , N.L. Samoy l ina , A . M . Muller, and E.A. Dzierzak. 1993. A n early pre-liver intraembryonic source of C F U - S in the developing mouse . Nature 364:64. 86. Muller, A . M . , A . Mevinsky, J . Strouboulis, F. Grosve ld , and E. Dz ierzak. 1994. Development of hematopoiet ic stem cell activity in the mouse embryo. Immunity 1:291. 87 . Zuca l i , J . R . 1982. Sel f - renewal and differentiation capaci ty of bone marrow and fetal liver stem cel ls. Br. J . Haemato l . 52:295. 88. Mick lem, H.S. , C . E . Ford, E . P . Evans , D.A. O g d e n , and D.S. Papwor th . 1972. Compet i t ive in vivo proliferation of foetal and adult haematopoiet ic cel ls in lethally irradiated mice. J . Ce l l . Phys io l . 79 :293. 117 89. Rebe l , R.I., C . L Miller, C . J . Eaves , and P . M . Lansdorp. In press. The repopulat ion potential of fetal liver hematopoiet ic s tem cel ls in mice exceeds that of their adult bone marrow counterparts. B lood . 90 . Vasi r i , H., W . Dragowska, R . C . Al lsopp, T . E . Thomas , C . B . Harley, and P . M . Lansdorp . 1994. Ev idence for a mitotic c lock in human hematopoiet ic s tem cel ls : loss of telomeric D N A with age. Proc . Natl . A c a d . S c i . U S A 91:9857. 9 1 . S a u v a g e a u , G . , U. Thorsteindottir, C . J . E a v e s , H.T. Lawrence, C . La rgman, P . M . Lansdorp , and R.K. Humphr ies. 1995. Overexpress ion of H O X B 4 in hematopoiet ic cel ls c a u s e s the select ive expans ion of more primitive populat ions in vitro and in vivo. G e n e s Deve l . 9:1753. 92 . Lerner, C , and D.E. Harr ison. 1990. 5-flourouracil spares hemopoiet ic s tern cel ls responsib le for long term repopulation. Exp . Hemato l . 18:114. 93 . Harr ison, D.E. , and C P . Lerner. 1991. Most primitive hematopoiet ic s tem cel ls are st imulated to cyc le rapidly after treatment with 5-f luorouracil . B lood 78:1237. 94. Kay , H . E . M . 1965. How many cell generat ions? Lancet 2:418. 95 . Mick lem, M.S. , and J . F . Loutit. 1966. T issue grafting and radiation. A c a d e m i c press Inc., New York. 96. Mintz, B., K. Anthony, and S . Litwin. 1984. Monoc lona l derivation of mouse myeloid and lymphoid l ineages from totipotent hematopoiet ic s tem cel ls experimental ly engrafted from fetal hosts. Proc . Natl . A c a d . S c i . U S A 81 :7835 . 97. Abkowi tz , J .L . , M.L. Linenberger, M.A. Newton, G . H . Shel ton, R.L. Ott, and P. Guttorp. 1990. Ev idence for the maintenance of hematopoies is in a large animal by the sequent ial activation of s tem cell c lones . Proc . Nat l . A c a d . S c i . U S A 8 7 : 9 0 6 2 . 98 . Jo rdan , C .T . , and I.R. Lemischka . 1990. C lona l and systemic ana lys is of long-term hematopoies is in the mouse. G e n e s Deve l . 4 :220. 99 . Kel ler, G . , and R. Snodgrass . 1990. Life span of multipotential hematopoiet ic stem cel ls in vivo. J . Exp . M e d . 171:1407. 100. Abkowitz, J .L . , M.T. Persik, G . H . Shel ton, R.L. Ott, V . Kik levich, S . N . Cat l in, and P. Guttorp. 1995. Behavior of hematopoiet ic stem cel ls in a large animal . Proc . Natl . A c a d . S c i . U S A 92:2031. 101 . Fraser , C . C , C . J . E a v e s , S . J . Sz i l vassy , and R.K. Humphr ies. 1990. Expans ion in vitro of retrovirally marked totipotent hematopoiet ic s tem cel ls . B lood 76 :1071. 102. R o s s , E . A . M . , N. Anderson , and H.S. Mick lem. 1982. Ser ia l deplet ion and regenerat ion of the murine hematopoiet ic sys tem. Implications for 118 hematopoiet ic organizat ion andthe study of cel lular ag ing. Journa l of Exp . M e d . 155:432. 103. Harr ison, D.E. , C M . Ast le, and J . A . DeLaittre. 1978. Loss of proliferative capaci ty in immunohemopoiet ic stem cel ls c a u s e d by ser ial transplantat ion rather than ageing. J . Exp . M e d . 147:1526. 104. O g d e n , D.A., and H.S. Mick lem. 1976. The fate of serial ly t ransplanted bone marrow cel l populat ions from young and old donors. Transplantat ion 22 :287 . 105. Siminovi tch, L., J . E . Til l, and E.A. McCu l l och . 1964. Decl ine in colony-forming ability of marrow cel ls subjected to ser ial transplantation into irradiated mice. J . Ce l l . C o m p . Phys io l . 64:23. 106. Cudkow icz , G . , A . C Upton, and G . M . Shearer . 1982. Lymphocyte content and proliferative capaci ty of serial ly transplanted mouse bone marrow. Nature 201:165. 107. Harr ison, D .E . 1972. Normal function of transplanted mouse eyrythrocyte precursors for 21 months beyond donor life span . Nature N e w Biology 237 :220 . 108. Harr ison, D.E. 1973. Normal production of erythrocytes by mouse marrow cont inuous for 73 months. Proc . Natl. A c a d . S c i . U S A 70:3184. 109. He l lman, S . , L E . Botnick, E . C Hannon, and R . M . Vigneul le. 1978. Proliferative capaci ty of murine hematopoiet ic stem cel ls. Proc . Nat l . A c a d . S c i . U S A 75:490. 110. M a u c h , P. , and S . Hel lman. 1989. Loss of hematopoiet ic s tem cel l self-renewal after bone marrow transplantation. B lood 74 2:872. 111 . Harr ison, D.E. , M. Stone, and C M . Ast le. 1990. Effects of transplantation on the primitive immunohematopoiet ic stem cel l . J . Exp . M e d . 3:119. 112. C a s h m a n , J .D . , A . C . E a v e s , and C . J . E a v e s . 1992. Granulocyte-macrophage colony-st imulat ing factor modulat ion of the inhibitory effect of transforming growth factor-b on normal and leukemic human hematopoiet ic progenitor cel ls . Leukemia 6:886. 113. E a v e s , C . J . , J .D . C a h m a n , R . J . Kay, G . J . Dougherty, T. Otsuka, L.A. Gaboury , D .E . Hogge, P . M . Lansdorp, A . C . E a v e s , and R.K. Humphr ies. 1991. Mechan i sms that regulate the cell cyc le status of very primitive hematopoiet ic cel ls in long-term human marrow cultures. II. Ana lys i s of posit ive and negative regulators produced by stromal cel ls within the adherent layer. B lood 78:110. 114. Ruscett i , F.W., S . E . J a c o b s e n , R . M . Birchenal l , H.E. Broxmeyer, G .L . Enge lmann , C . Dubois, and J . R . Kel ler. 1991. Ro le of transforming growth factor-betal in regulation of hematopoies is . A n n . Ny. A c a d . S c i . 31 :31 . 119 115. Ruscett i , F.W., C M . Dubois, S . E . J a c o b s e n , and J . R . Kel ler. 1992. Transforming growth factor beta and interleukin-1: a paradigm for oppos ing regulation of haemopo ies is , vol . 5:3. Bail l iere Tindal l , London . 703 . 116. Rosendaa l , M. , G . S . Hodgson , and T. R. Bradley. 1979. Organizat ion of haemopoiet ic stem cel ls: the generat ion age hypothesis. Ce l l and T i ssue Kinet ics 12:17. 117. Harr ison, D.E. , and C M . Ast le . 1982. Loss of stem cell repopulating ability upon transplantation. J . Exp . M e d . 156:1767. 118. Metcalf, D., and N.A. Nico la. 1995. The hemopoiet ic colony-st imulat ing factors: from biology to cl inical appl icat ions. Cambr idge Universi ty P r e s s , Cambr idge . 338 pp. 119. Metcalf, D. 1993. Hematopoiet ic regulators: redundancy or subtlety? B lood 82 :3515 . 120. Ikebuchi, K., G . G . W o n g , S . C Clark, J . N . Ihle, Y . Hirai, and M. O g a w a . 1987. lnterleukin-6 enhancement of the inter leukin-3-dependent proliferation of multipotential hemopoiet ic progenitors. Proc . Natl . A c a d . S c i . U S A 84:9035. 121 . Ikebuchi, K., S . C . Clark, J . N . Ihle, L M . S o u z a , and M. O g a w a . 1988. Granu locy te colony-st imulat ing factor enhances interleukin 3-dependent proliferation of multipotential hemopoiet ic progenitors. P roc . Natl . A c a d . S c i . U S A 85:3445. 122. Musash i , M. , Y . - C Y a n g , S . R . Pau l , S . C . Clark, T. Sudo , and M. O g a w a . 1991. Direct and synergist ic effects of interleukin 11 on murine hemopo ies is in culture. Proc . Natl. A c a d . S c i . U S A 88:765. 123. Musash i , M., S . C . Clark, T. Sudo , D . L Urdal , and M. O g a w a . 1991. Synerg is t ic interactions between interleukin-11 and interleukin-4 in support of proliferation of primitive hematopoiet ic progenitors of mice. B lood 78:1148. 124. Bod ine , D.M. , S . Kar lsson , and A . W . Nienhuis. 1989. Combinat ion of interleukins 3 and 6 preserves stem cell function in culture and e n h a n c e s retrovirus-mediated gene transfer into hematopoiet ic stem cel ls . Proc . Nat l . A c a d . S c i . U S A 86:8897. 125. Tsuj i , K., K . M . Zsebo , and M. O g a w a . 1991. Enhancement of murine blast cel l co lony formation in culture by recombinant rat s tem cel l factor, l igand fore-ki t . B lood 78:1223. 126. Kel ler, J . R . , C . Mantel , G . K . S ing, L.R. Ell ingsworth, S .K . Ruscett i , and F.W. Ruscett i . 1988. Transforming growth factor b1 select ively regulates early murine hematopoiet ic progenitors and inhibits the growth of IL-3-dependent myeloid leukemia cell l ines. J . Exp . M e d . 168:737. 120 127. C a s h m a n , J .D . , A . C . Eaves , E.W. Ra ines , R. R o s s , and C . J . E a v e s . 1990. Mechan i sms that regulate the cell cyc le status of very primitive hematopoiet ic cel ls in long-term human marrow cultures. I. St imulatory role of a variety of mesenchymal cell activators and inhibitory role of T G F - b . B lood 75:96. 128. Ka tayama, N., S . C . Clark, and M. O g a w a . 1993. Growth factor requirement for survival in cel l -cycle dormancy of primitive murine lymphohematopoiet ic progenitors. B lood 81:610. 129. De Vr ies, P., K.A. Brase l , J . R . E isenman, A . R . Alpert, and D.E. Wi l l iams. 1991. The effect of recombinant mast cell growth factor on purified murine hematopoiet ic stem cel ls. J . Exp . M e d . 173:1205. 130. Migl iaccio, G . , A . R . Migl iaccio, J . Val insky, K. Langley, K. Z s e b o , J . W . M . V isser , and J . W . A d a m s o n . 1991. S tem cell factor induces proliferation and differentiation of highly enr iched murine hematopoiet ic cel ls . Proc . Natl . A c a d . S c i . U S A 88:7420. 131 . P loemacher , R . E . , P .L . van Soest , H. Voorwinden, and A . Boudewi jn. 1993. lnterleukin-12 synerg izes with interleukin-3 and Stee l factor to enhance recovery of murine hemopoiet ic stem cel ls in liquid culture. Leukemia 7 :1381 . 132. Brandt, J . E . , K. Bhal la , and R. Hoffman. 1994. Effects of interleukin-3 and c-kit l igand on the survival of var ious c l a s s e s s of human hematopoiet ic progenitor cel ls . B lood 83:1507. 133. Metcalf, D., and N.A. Nico la. 1991. Direct proliferative act ions of s tem cel l factor on murine bone marrow cel ls in vitro: Effects of combinat ion with colony-st imulat ing factors. Proc . Natl. A c a d . S c i . U S A 88:6239. 134. Bernste in, I.D., R . G . Andrews, and K . M . Zsebo . 1991. Recombinant human stem cell factor enhances the formation of co lon ies by C D 3 4 + and CD34+l in- cel ls , and the generat ion of colony-forming cel l progeny from CD34+l in - cel ls cultured with interleukin-3, granulocyte colony-st imulat ing factor, or granulocyte-macrophage colony-st imulat ing factor. B lood 77 :2316 . 135. Broxmeyer , H.E. , L. Lu , G . Hangoc, S . Cooper , P . C . Hendr ie, J . A . Ledbetter, M. X iao , D.E. Wi l l iams, and F.W. S h e n . 1991. C D 4 5 cel l sur face ant igens are l inked to stimulation of early human myeloid progenitor cel ls by interleukin 3 (IL-3), granulocyte/macrophage colony-st imulat ing factor ( G M - C S F ) , a G M - C S F / I L - 3 fusion protein, and mast cel l growth factor (a c-kit l igand). J . Exp . M e d . 174:447. 136. Rosnet , O. , S . Marchetto, O . deLapeyr iere, and D. B i rnbaum. 1991. Mur ine flt3, a gene encoding a novel tyrosine k inase receptor of the P D G F R / C S F 1 R family. Oncogene 58: 1641. 121 137. Matthews, W. , C .T . Jordan, G . W . Wiegand , D. Pardol l , and I.R. Lemischka . 1991. A receptor tyrosine k inase specif ic to hematopoiet ic s tem and progenitor cel l -enr iched populat ions. Ce l l 65 :1143. 138. Rosnet , O. , C . Schiff, M . -J . Pebusque , S . Marchetto, C . Tonnel le , Y . Toiron, F. Birg, and D. B i rnbaum. 1993. Human F L T 3 / F L K 2 gene: c D N A cloning and express ion in hematopoiet ic cel ls. B lood 82:1110. 139. Sma l l , D., M. Levenste in, E. K im, C . Carow, S . Amin , P. Rockwel l , L. Witte, C . Burrow, M.Z. Ratajczak, A . M . Gewirtz, and C.I. C iv in . 1994. S T K - 1 , the human homologue of Flk-2/Fl t -3, is select ively exp ressed in C D 3 4 + human bone marrow cel ls and is involved in the proliferation of early progenitor/stem cel ls. Proc . Natl. A c a d . S c i . U S A 91:459. 140. Lyman , S .D . , L. J a m e s , T. Vanden Bos , P. de Vr ies , and et.al. 1993. Molecu lar c loning of a l igand for the flt3/flk-2 tyrosine k inase receptor: A proliferative factor for primitive hematopoiet ic cel ls . Ce l l 75 :1157. 141 . Lyman , S .D . , L. J a m e s , L. Johnson , K. Brase l , P. De Vr ies , S . S . Escobar , H. Downey, R.R. Splett, M .P . Beckmann , and H.J . M c K e n n a . 1994. C lon ing of the human homologue of the murine flt3 l igand: A growth factor for early hematopoiet ic progenitor cel ls. B lood 83:2795. 142. Hannum, C , J . Cu lpepper , D. Campbe l l , T. M c C l a n a h a n , S . Zurawsk i , J . F . B a z a n , R. Kaste le in, S . Hudak, J . Wagner , J . Mattson, S . M e n o n , A . Shanafel t , M. Muench , G . Kelner, R. Namikawa, D. Rennick, M . - G . Roncaro lo , A . Zlotnik, O . Rosnet , P. Dubreui l , D. B i rnbaum, and F. Lee . 1994. L igand for F L T 3 / F L K 2 receptor tyrosine k inase regulates growth of haematopoiet ic stem cel ls and is encoded by variant R N A s . Nature 368 :643 . 143. Russe l l , E . S . 1979. Hereditary anemias of the mouse : A review for genet ic ists. A d v a n c e s in Genet ics 20:357. 144. Mackarehtsch ian, K., J . D . Hardin, K.A. Moore, S . Boast , S . P . Goff, and I.R. Lemischka . 1995. Targeted disruption of the flk2/flt3 gene leads to def ic iencies in primitive hematopoiet ic progenitors. Immunity 3:147. 145. Tavasso l i , M. , and M. Shak la i . 1979. A b s e n c e of tight junctions in endothel ium of marrow s inuses : possib le s igni f icance for marrow cel l egress . Br. J . Haemato l . 41:303. 146. W e i s s , L. 1976. The hematopoiet ic microenvironment of the bone marrow: A n ultrastructural study of the stroma in rats. Anat. Rec . 186:161. 147. L ichtman, M.A. 1981. The ultrastructure of the hemopoiet ic environment of the marrow: A review. Exp . Hematol . 9 :391. 148. Cou lombe l , L., M .H . Vuillet, C . Leroy, and G . Tchern ia . 1988. L ineage- and stage-speci f ic adhes ion of human hematopoiet ic progenitor ce l ls to extracel lular matr ices from marrow fibroblasts. B lood 71:329. 122 149. Wi l l iams, D.A., M. R ios , C . S tephens, and V . P . Pate l . 1991. Fibronect in and V L A - 4 in haematopoiet ic stem cel l -microenvironment interactions. Nature 352 :438 . 150. Gal lagher , J .T. , E. Spooncer , and T . M . Dexter. 1983. Ro le of the cel lular matrix in haemopo ies is l . Synthes is of g lycoaminog lycans by mouse bone marrow cel l cultures. Proc . Natl. A c a d . S c i . U S A 86:4654. 151 . Spooncer , E. , J .T . Gal lagher, F. Kr izsa , and T . M . Dexter. 1983. Regulat ion of haemopo ies is in long-term bone marrow cultures. IV. G lycosaminog lycans synthesis and the stimulation of haemopo ies is by b-D-Xy los ides. J . Ce l l . B io l . 96:510. 152. L iesve ld , J .L . , J . M . Winslow, K .E . Frediani , D.H. R y a n , and C . N . Abboud . 1993. Express ion of integrins and examinat ion of their adhes ive function in normal and leukemic hematopoiet ic cel ls. B lood 81:112. 153. Miyake, K., I.L. W e i s s m a n , J . S . Greenberger , and P .W. K incade . 1991. Ev idence for a role of the integrin V L A - 4 in lymphopoies is . Journa l of Exp . M e d . 173:599. 154. Miyake, K., K . L Medina, S.I. Hayash i , S . Ono , T. Hamaoka , and P .W. K incade . 1990. Monoc lona l ant ibodies tp P g p - 1 / C D 4 4 block lymphopoies is in long-term bone marrow cultures. J . Exp . M e d . 171:477. 155. Papayannopou lou , T., C . Craddock , B. Nakamoto, G . V . Priest ley, and N .S . Wolf. 1995. The V L A 4 / V C A M - 1 adhes ion pathway def ines contrast ing mechan i sms of lodgement of t ransplanted murine hemopoiet ic progenitors between bone marrow and sp leen. Proc . Nat l . A c a d . S c i . U S A 92:9647. 156. Long, M.W., R. Briddell, A . W . Walter, E. Bruno, and R. Hoffman. 1992. Human hematopoiet ic s tem cel l adherence to cytokines and matrix molecu les . J . C l in . Invest. 90:251. 157. N e s s , S .A . , and J . D . Enge l . 1994. Vintage reds and whites: combinator ia l transcription factor utilization in hematopoiet ic differentiation. Current Opin ion in Genet i cs and Development 4:718. 158. Tsa i , F.-Y., G . Keller, F .C . Kuo, M.J . We iss , J . C h e n , M. Rosenblatt, F.W. Alt, and S . H . Orkin. 1994. A n early hematopoiet ic defect in mice lacking the transcription factor G A T A - 2 . Nature 371:221. 159. Ork in , S . H . 1995. Transcript ion factors and hematopoiet ic development . J Bio l C h e m 270:4955. 160. Bockamp, E. -O. , F. McLaughl in , A . Murrell, and A . R . G r e e n . 1994. T ranscription factors and the regulation of haemopo ies is : l essons from G A T A and S C L proteins. B i o E s s a y s 16:481. 123 161 . W e i s s , M.J . , and S . H . Orkin. 1995. G A T A transcription factors: key regulators of hematopoies is . Exp . Hemato l . 23:99. 162. Urbanek, P., Z . -Q . W a n g , I. Fetka, E .F . Wagner , and M. Bussl inger. 1994. Comple te block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking P a x 5 / B S A P . Ce l l 79 :901 . 163. W o n g , M . C , S . -W. Chung , C E . Dunbar, D.M. Bodine, S . Ruscett i , and A . Nienhuis. 1989. Retrovirus-mediated transfer and express ion of the lnterleukin-3 gene in mouse hematopoiet ic cel ls results in a myeloprol i ferative disorder. Molec . Ce l l . B io l . 9:798. 164. Hawley, R . G . , A . Z . C . Fong , B.F. Burns, and T . S . Hawley. 1992. Transplantab le myeloproli ferative d i sease induced in mice by an interleukin- retrovirus. J . Exp . M e d . 176:1149. 165. C h a n g , J . M . , and G . R . Johnson . 1991. Effects on sp leen colony-forming unit sel f -renewal after retroviral-mediated gene transfer of mult i-colony-stimulating factor, granulocyte-macrophage colony-st imulat ing factor, or granulocyte colony-st imulat ing factor. Exp . Hemato l . 19:602. 166. Hawley, T . S . , B .F . Burns, and R . G . Hawley. 1991. Leukocytos is in mice fol lowing long-term reconstitution with genetical ly- modif ied bone marrow ce l ls constituit ively express ing interleukin 1 a lpha or interleukin 6. L e u k e m i a R e s e a r c h 15:659. 167. Johnson , G .R . , T . J . G o n d a , D. Metcalf, I.K. Har iharan, and S . Cory . 1989. A lethal myeloproli ferative syndrome in mice transplanted with bone marrow cel ls infected with a retrovirus express ing granulocyte-macrophage co lony stimulating factor. E M B O : 4 4 1 . 168. Va rmus , H.E. , and R. Swanst rom. 1984. Repl icat ion of retroviruses. In R N A tumor v i ruses. J . Coff in, editor, Co ld Spr ing Harbor, New York: C o l d Spr ing Harbor Laboratory. 169. Va rmus , H.E. , and P. Brown. 1989. Retroviruses. In Mobi le D N A . M. Howe and D. Berg , editors, Washington, D. C . 53. 170. Mann , R., and D. Balt imore. 1985. Varying position of a retrovirus packag ing sequence results in encapsidat ion of both unspl iced and sp l iced R N A s . J . Virol . 54:401. 171 . Bender , M.A., T .D . Palmer, R .E . Ge l inas , and A . D . Miller. 1987. Ev idence that the packag ing signal of moloney murine leukemia virus extends into the gag region. J . Virol . 61:1639. 172. Armentano, D., S . - F . Y u , P.W. Kantoff, R.T. Von , W . F . Anderson , and E. G i lboa . 1987. Effect of internal viral sequences on the utility of retroviral vectors. J . Virol. 61:1647. 124 173. W a n g , H., M P . Kavanaugh , R.A. North, and D. Kabat . 1991. Ce l l -sur face receptor for ecotropic murine retroviruses is a bas ic amino-ac id transporter. Nature 352:729. 174. K im, J .W. , E.I. C l o s s , L .M. Albritton, and J . M . Cunn ingham. 1991. Transport of cationic amino ac ids by the mouse ecotropic retrovirus receptor. Nature 352 :725 . 175. Hos ie , M .J . , B . J . Willett, T . H . Dunsford, O. Jarrett, and J . C . Nei l . 1993. A monoc lona l antibody which b locks infection with fel ine immunodef ic iency virus identifies a possib le non -CD4 receptor. J . Virol . 67:1667. 176. Willett, B . J . , M .J . Hosie, O . Jarrett, and J . C . Nei l . 1994. Identification of a putitive cel lular receptor for feline immunodef ic iency virus as the fel ine homologue of C D 9 . Immunol. 81:228. 177. Dalg le ish, A . G . , P . C . Bever ley, P .R. C lapham, D.H. Crawford, M.F. G r e a v e s , and R.A. We i ss . 1984. The C D 4 (T4) antigen is an essent ia l component of the receptor for the A I D S retrovirus. Nature (London) 312 :763 . 178. K la tzmann, D., E. Champagne , S . Chamaret , J . Gruest , D. Guetard , T. Hercend , J . C . G luckman , and L. Montagnire. 1984. T- lymphocyte T4 molecule behaves as the receptor for human retrovirus L A V . Nature (London) 312:767. 179. Sattentau, Q . J . , P .R . C lapham, R.A. We iss , P.c. Bever ley, L. Montagnire, M.F . A lha lab i , J . C . G luckmann , and D. K la tzmann. 1988. The human and s imian immunodef ic iency v i ruses HIV-1, HIV-2, and SIV interact with similar epi topes on their cel lular receptor, the C D 4 molecule. A I D S 2 :101. 180. You ing , J .A . , P. Bates, and H.E. Varmus. 1993. Isolation of a ch icken gene that confers suseptibil ity to infection by subgroup A avian leukosis and s a r c o m a v i ruses. J . Virol . 67 :1811. 181 . B a n , J . , D. Portetelle, C . Altaner, B. Horion, D. Mi lan, V . Krchnak, A . Burny, and R. Ket tmann. 1993. Isolation and character izat ion pf a 2.3-k i lobase-pair c D N A fragment encoding the binding domain of the bovine leukemia virus cel l receptor. J . Virol . 67:1050. 182. Miller, D .G . , R .H . Edwards , and A . D . Miller. 1994. C lon ing of the cel lular receptor for amphotropic murine retroviruses reveals homology to that for g ibbon ape leukemia virus. Proc . Natl. A c a d . S c i . U S A 91:78. 183. V a n Zeij l , M. , S .V . Johann , E. C l o s s , J . Cunn ingham, R. Eddy, T .B . S h o w s , and B. O 'hara. 1994. A human amphotropic retrovirus receptor is a second member of the gibbon ape leukemia virus receptor family. P roc . Natl . A c a d . S c i . U S A 91:1168. 184. O 'hara, B., S .V . Johann , H P . Klinger, D .G . Blair, H. Rub inson , K . J . Dunn, P. S a s s , S . M . Vitek, and T. Robins. 1990. Character izat ion of a human gene 125 conferr ing sensitivity to infection by gibbon ape leukemia virus. Ce l l Growth and Differentiation 1:119. 185. Takeuch i , Y . , R . G . Vi le, G . S impson , B. O 'Hara , M.K. Col l ins, and R.A. W e i s s . 1992. Fel ine leukemia virus subgroup B uses the s a m e cel l sur face receptor as gibbon ape leukemia virus. J . Virol . 66 :1219. 186. Sommerfel t , M.A., and R.A. We iss . 1990. Receptor interference groups of 20 retroviruses plating on human cel ls. Virol . 176:58. 187. Albritton, L. M. , L. Tseng , D. Scadden and J . M Cunn ingham. 1989. A putative murine ecotropic retrovirus receptor gene encodes a multiple membrane-spann ing protein and confers suseptabi l i ty to virus infection. Ce l l 57: 659. 188. Kavanaugh , M.P. , D .G . Miller, W . Zhang , W. Law, S .L . Kozak , D. Kabat , and A . D . Miller. 1994. Cel l -sur face receptors for g ibbon ape leukemia virus and amphotropic murine retrovirus are inducible sod ium-dependent phosphate symporters. Proc . Natl. A c a d . S c i . U S A 91:7071. 189. Pangan iban , A T . , and D. Fiore. 1988. Ordered interstrand and intrastrand D N A transfer during reverse transcription. Sc i ence 241:1046. 190. H u , W . - S . , and H.M. Temin . 1990. Retroviral recombinat ion and reverse transcript ion. S c i e n c e 250:1227. 191 . Miller, D .G. , M.A. A d a m , and A . D . Miller. 1990. G e n e transfer by retrovirus vectors occurs only in cel ls that are actively replicating at the t ime of infection. Molec . Ce l l . B io l . 10:4239. 192. Springett, G . M . , R . C . Mown, S . Anderson, R . M . B laese , and A . W . F. 1989. Infection eff iciency of T lymphocytes with amphotropic retroviral vectors is cel l cyc le dependent. J . Virol . 63:3865. 193. Roe , T.Y. , T . C . Reynolds, G . Y u , and P .O. Brown. 1993. Integration of murine leukemia virus D N A depends on mitosis. E M B O 12:2099. 194. Purcel l , D .F .J . , C M . Brosc ius, E .F . Van in , C . E . Buckler, A . W . Nienhuis, and M.A. Mart in. 1996. A n array of murine leukemia virus-related e lements is transmitted and expressed in a primate recipient of retroviral gene transfer. J . Virol . 70:887. 195. M a n n , R., R . C . Mul l igan, and D. Balt imore. 1983. Construct ion of a retrovirus packaging mutant and its use to produce helper-free defect ive retrovirus. Ce l l 33 :153. 196. C o n e , R.D. , and R . C . Mul l igan. 1984. High-eff iciency gene transfer into mammal ian cel ls: Generat ion of helper-free recombinant retrovirus with broad mammal ian host range. Proc . Natl. A c a d . S c i . U S A 81:6349. 126 197. Miller, A . D . , and C . Buttimore. 1986. Redes ign of retrovirus packag ing cel l l ines to avoid recombinat ion leading to helper virus product ion. Mo lec . C e l l . B io l . 6 :2895. 198. Markowitz, D., S . Goff, and A . Bank. 1988. Construct ion and use of a safe and efficient amphotropic packaging cell line. Vi ro l . 167:400 199. Markowitz, D., S . Goff, and A . Bank. 1988. A safe packaging line for gene transfer: Separart ing viral genes on two different p lasmids. J . Virol . 62 :1120 . 200 . Danos , O. , and R . C . Mul l igan. 1988. Safe and Efficient generat ion of recombinant retroviruses with amphotropic and ecotropic host ranges. P roc . Natl . A c a d . S c i . U S A 85:6460. 2 0 1 . Pear , W . S . , G . P . No lan, M.L. Scott, and D. Balt imore. 1993. Product ion o f ' high-titer helper-free retroviruses by transient transfection. Proc . Natl . A c a d . S c i . U S A 90:8392. 202 . Bod ine , D.M. , K.T. M c D o n a g h , N.E. Se ide l , and A . W . Nienhuis. 1991. Surv ival of retroviral infection of murine hematopoiet ic s tem cel ls in vitro: Effects of 5 - F U and method of infection. Exp . Hemato l . 19:206. 2 0 3 . Weider , R., K. Cornetta, S .W. Kess ler , and W . F . Anderson . 1991. Increased eff iciency of retroviral-mediated gene transafer and express ion in primate bone marrow progenitors after 5-f luorouraci l- induced hematopoiet ic suppress ion and recovery. B lood 77:448. 204 . Wi l l iams, D.A., I.R. Lemischka , D .G. Nathan, and R . C . Mul l igan. 1984. Introduction of new genet ic material into pluripotent haematopoiet ic . Nature 310:476. 2 0 5 . Eglit is, M.A., P. Kantoff, E. G i lboa, and W . F . Anderson . 1985. G e n e express ion in mice after high eff iciency retroviral mediated gene transfer. S c i e n c e 230 :1395 . 206 . Luskey , B.D. , M. Rosenblatt , K. Zsebo , and D.A. Wi l l iams. 1992. S tem cel l factor, lnterleukin-3, and interleukin-6 promote retroviral-mediated gene transfer into murine hematopiet ic stem cel ls. B lood 80:396. 207 . E inerhand, M.P. , T.A. Bakx, A . Kukler, and D. V a l e r i c 1993. Factors affecting the transduction of pluripotent hematopoiet ic s tem cel ls : long-term express ion of a human adenos ine deaminase gene in mice. B lood 81:254. 208 . Fletcher, F.A., K.A. Moore, M. Ashkenaz i , V . P . De, P.A. Overbeek, D .E . Wi l l iams, and J . W . Belmont. 1991. Leukemia inhibitory factor improves survival of retroviral vector-infected hematopoiet ic s tem cel ls in vitro, al lowing efficient long-term express ion of vec tor -encoded human adenos ine deaminase in vivo. J . Exp . M e d . 174:837. 127 209 . Fletcher, F.A., D.E. Wi l l iams, C . Mal iszewsk i , D. Anderson , M. R ives , and J . W . Belmont. 1990. Murine leukemia inhibitory factor enhances retroviral-vector infection eff iciency of hematopoiet ic progenitors. B lood 76:1098. 210 . Nolta, J .A . , G . M . Crooks , R.W. Overal l , D.E. Wi l l iams, and D.B. Kohn . 1992. Retroviral vector-mediated gene transfer into primitive human hematopoiet ic progenitor cel ls: effects of mast cell growth factor ( M F G ) combined with other cytokines. Exp . Hemato l . 20 :1065. 2 1 1 . Dilber, M .S . , B. Bjorkstrand, K . J . Li , C L E . Smith, K . G . Xanthopoulos, and G . G a h r o n . 1994. Bas i c fibroblast growth factor increases retroviral-mediated gene transfer into human hematopoiet ic peripheral blood progenitor ce l ls . Exp . Hemato l . 22:1129. 212 . van B e u s e c h e m , V .W. , J . A . K . Bart-Baumeister, P . M . Hoogerbrugge, and D. Valer io . 1995. Influence of interleukin-3, interleukin-6, and s tem cel l factor on retroviral transduction of R h e s u s monkey C D 3 4 + hematopoiet ic progenitor cel ls measured in vitro and in vivo. G e n e Therapy 2:245. 2 1 3 . C rooks , G . M . , and D.B. Kohn . 1993. Growth factors increase amphotropic retrovirus binding to human CD34+ bone marrow progenitor ce l ls . B lood 82 :3290 . 214 . Peters, S . O . , E .L .W. Kittler, H.S. Ramshaw, and P . J . Quesenberry . 1996. Ex vivo expans ion of murine marrow cel ls with interleukin-3 (IL-3), IL-6, IL-11, and s tem cel ls factors leads to imparied engraftment in irradiated hosts. B lood 87:30. 2 1 5 . Apper ley , J . F . , B.D. Luskey, and D.A. Wi l l iams. 1991. Retroviral gene transfer of human adenos ine deaminase in murine hematopoiet ic ce l ls : effect of selectable marker sequences on long-term express ion . B lood 7 8:310. 216 . Wi l l iams, D.A., S . H . Ork in, and R . C . Mul l igan. 1986. Retrovirus-mediated transfer of human adenos ine deaminase gene s e q u e n c e s into cel ls in culture and into murine hematopoiet ic cel ls in vivo. P roc . Nat l . A c a d . S c i . U S A 83:2566. 217 . L im, B., J . F . Apperly, S . H . Orkin, and D.A. Wi l l iams. 1989. Long term express ion of human adenos ine deaminase in mice t ransplanted with retrovirus-infected hematopoiet ic stem cel ls. Proc . Natl . A c a d . S c i . U S A 86 :8892 . 218 . Kohn , D.B., P .W. Kantoff, M.A. Eglitis, J .R . McLachl in , R . C . Moen , E. Ka rson , J .A . Zwiebe l , A . Nienhius, S . Kar lsson, R. O'Rei l ly , A . Gil l io, C . Bord ignon, E. G i lboa , E.D. Zanjani , R . M . B laese , and W . F . Ande rson . 1987. Retroviral-mediated gene transfer into mammal ian cel ls . B lood Ce l l s 13:285. 219 . Kantoff, P.W., A . P . Gil l io, J . R . McLach l in , C . Bordignon, M.A. Eglit is, N.A. Kernan , R . C . Moen , D.B. Kohn , S . F . Y u , and E.e .a . Ka rson . 1987. 128 Express ion of human adenos ine deaminase in nonhuman pr imates after retrovirus-mediated gene transfer. J . Exp . M e d . 166:219. 220 . Anderson , W .F . , P. Kantoff, M. Eglit is, J . McLach l in , E. Ka rson , J . Zwiebe l , A . Nienhuis, S . Kar lsson, R . M . B laese , D. Kohn , E. G i lboa , D. Armentano, E .D . Zanjani , A . Dlake, M.R. Harr ison, A . Gil l io, C . Bord ignon, and R. O'Rei l ly . 1986. G e n e transfer and express ion in nonhuman pr imates using retroviral vectors. C o l s Spr ing Harbor S y m p o s i a on Quanti tat ive Bio logy L l :1073 . 2 2 1 . X u , L .C . , S . Kar lsson, E .R. Byrne, S . Kluepfel-Stahl , S . W . Kess le r , B.A. Agr ico la , S . Se l l , M. Kirby, C . E . Dunbar, R .O . Brady, A . W . Nienhuis, and R . E . Donahue. 1995. Long-term in vivo express ion of the human g lucocerebros idase gene in nonhuman primates after C D 3 4 + hematopoiet ic cel l transduction with cell-free retroviral vector preparat ions. Proc . Natl . A c a d . S c i . U S A 92:4372. 222 . Nolta, J .A . , E . M . Smogo rzewska , and D.B. Kohn . 1995. Ana lys is of optimal condi t ions for retroviral-mediated transduction of primitive human hematopoiet ic cel ls . B lood 86:101. 2 2 3 . X u , L . - C , S . Kluepfel-Stahl , M. B lanco, R. Schi f fmann, C . Dunbar, and S . Kar l sson . 1995. Growth factors and stromal support generate very efficient retroviral t ransduct ion of peripheral blood C D 3 4 + cel ls from G a u c h e r patients. B lood 86:141. 224 . Moore , K.A., A . B . Deisseroth, C .L . Read ing, D.E. Wi l l iams, and J . W . Belmont. 1992. Stromal support enhances cell-free retroviral vector transduct ion of human bone marrow long-term culture-initiating ce l ls . B lood 79 :1393. 2 2 5 . Moritz, T., V . P . Pate l , and D.A. Wi l l iams. 1994. Bone marrow extracel lular matrix molecu les improve gene transfer into human hematopoiet ic cel ls v ia retroviral vectors. J . C l in . Invest. 93:1451. 226 . Chuck , A . S . , and B.O. Pa l sson . 1996. Consis tent and high rates of gene transfer can be obtained using flow-through transduct ion over a wide range of retroviral titres. Human G e n e Therapy 7:743. 227 . Kar l sson , S . , D .M. Bodine, L. Perry, T. Papayannopou lou , and A . Nienhuis . 1988. Express ion of the human B-globin gene fol lowing retroviral-mediated transfer into multipotent hematopoiet ic progenitors in mice. P roc . Nat l . A c a d . S c i . U S A 85:6062. 228 . Schi f fmann, R., J .A . Medin , J . M . Ward , S . Stahl , M. Cott ier-Fox, and S . Kar l sson . 1995. Transfer of the human g lucocerebros idase gene into hematopoiet ic stem cel ls of nonablated recipients: success fu l engraftment and long-term express ion of the t ransgene. B lood 86:1218. 229 . Fink, J .K . , P . H . Correl l , L K . Perry, R .O. Brady, and S . Kar lsson . 1990. Correct ion of g lucocerebros idase def ic iency after retroviral mediated gene 129 transfer into hematopoiet ic progenitor cel ls from patients with G a u c h e r d i sease . Proc . Natl . A c a d . S c i . U S A 87:2334. 230 . Nolata, J .A . , L .S . Sender , J . A . Barranger, and D.B. Kohn . 1990. Express ion of human g lucocerebros idase in murine long-term bone marrow cultures after retroviral vector-mediated transfer. B lood 75:787. 2 3 1 . Bodine, D.M. , T. Moritz, R . E . Donahue, B.D. Luskey, S . W . Kess ler , D.I. Mart in, S . H . Orkin, A . W . Nienhuis, and D.A. Wi l l iams. 1993. Long-term in vivo express ion of a murine adenos ine deaminase gene in rhesus monkey hematopoiet ic cel ls of multiple l ineages after retroviral mediated gene transfer into C D 3 4 + bone marrow cel ls. B lood 82:1975. 232 . Lothrop, C . J . , L.Z. A l , G . P . Niemeyer, J . B . Jones , M . G . Peterson, J . R . Smith, H.J . Baker , R.A. Morgan, M.A. Eglit is, and W . F . Anderson . 1991. Express ion of a foreign gene in cats reconstituted with retroviral vector infected auto logous bone marrow. B lood 78:237. 2 3 3 . K iem, H.P. , B. Darovsky, C . Von Kal le, S . Goeh le , D. Stewart, T. G r a h a m , R. Hackman , F.R. Appe lbaum, H.J . Deeg , A . D . Miller, R. Storb, and F . G . Schuen ing . 1994. Retrovirus-mediated gene transduct ion into can ine per ipheral blood repopulat ing cel ls . B lood 83:1467. 234 . S tead , R.B. , W . W . Kwok, R. Storb, and A . D . Miller. 1988. C a n i n e model for gene therapy: inefficient gene express ion in dogs reconstituted with auto logous marrow infected with retroviral vectors. B lood 71:742. 2 3 5 . B ienz le , D., A . Ab rams-Ogg , S .A . Kruth, J . Ack land-Snow, R .F . Carter, J . E . Dick, R . M . J a c o b s , S . Kame l -Re id , and I.D. Dube. 1994. G e n e transfer into hematopoiet ic stem cel ls: Long-term maintenance of in vitro act ivated progenitors without marrow ablat ion. Proc . Natl . A c a d . S c i . U S A 91:350. 236 . Bord ignon, C , L.D. Notarangelo, N. Nobil i , G . Ferrari , G . Casora t i , P. Pan ina , E. Mazzo lar i , D. Maggioni , C . R o s s i , P. Serv ida , A . G . Ugaz io , and F. Mavi l io. 1995. G e n e therapy in peripheral b lood lymphocytes and bone marrow for A D A - immunodeficient patients. Sc i ence 270:470. 237 . Hoogerbrugge, P . M . , V . W . van B e u s e c h e m , A . F ischer , M. Debree, F. Ie Deist, J .L . Per ignon, G . Morgan, B. Gaspar , L.D. Fa i rbanks, C . H . S k e o c h , A . Mose ley , M. Harvey, R . J . Levinsky, and D. Valer io. 1996. B o n e marrow gene transfer in three patients with adenos ine deaminase def ic iency. G e n e Therapy 3:179. 238 . V o n Kal le, C , H.P. K iem, S . Goeh le , B. Darovsky, S . Heimfeld, B. Torok-Storb, R. Storb, and F . G . Schuen ing . 1994. Increased gene transfer into human hematopoiet ic progenitor cel ls by extended in vitro exposure to a pseudotyped retroviral vector. B lood 84:2890. 239 . Eglit is, M.A., R.D. Schne iderman, P . M . R ice , and M.V. E iden . 1995. Evaluat ion of retroviral vectors based on the gibbon ape leukemia virus. G e n e Therapy 2:486. 130 240 . Bauer , T .R. , A . D . Miller, and D.D. Hickstein. 1995. Improved transfer of the leukocyte integrin C D 1 8 subunit into hematopoiet ic cel l l ines by using retroviral vectors having a gibbon ape leukemia virus enve lope. B lood 86 :2379 . 2 4 1 . Burns, J . C , T. Fr iedmann, W. Driever, M. Bur rascano, and J . -K . Y e e . 1993. Ves icu la r stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentrat ion to very high titer and efficient gene transfer into mammal ian and nonmammal ian cel ls . Proc . Natl. A c a d . S c i . U S A 90:8033. 242 . Y e e , J . -K . , A . Miyanohara, P. LaPor te, K. Boutic, J . C . Burns, and T. Fr iedman. 1994. A general method for the generat ion of high-titre, pantropic retroviral vectors: Highly efficient infection of primary hepatocytes. Proc . Natl. A c a d . S c i . U S A 91:9564. 2 4 3 . K a s a h a r a , N., A . M . Dozy, and Y . W . Kan . 1994. T issue-spec i f ic targeting of retroviral vectors through l igand-receptor interactions. S c i e n c e 266 :1373 . 244 . S o m i a , N.V., M. Zoppe , and I.M. Ve rma . 1995. Geberat ion of targeted retroviral vectors by using s ingle-chain var iable fragment: an approach to in vivo gene delevery. Proc . Natl . A c a d . S c i . U S A 92:7570. 2 4 5 . Cosse t , F . - L , F . J . Morl ing, Y . Takeuch i , R.A. We iss , M.K.L. Col l ins, and S . J . Russe l l . 1995. Retroviral retargeting by enve lopes express ing N-terminal binding domains . J . Virol . 69:6314. 246 . Roux, P. , P. Jeanteur, and M. P iechaczyk . 1989. A verditile and potentially general approach to the targeting of speci f ic cel l types by retroviruses: appl icat ion to the infection of human cel ls by means of major histocompatibil i ty complex c lass I and c lass II ant igens by mouse ecotropic murine leukemia virus-derived v i ruses. Proc . Natl . A c a d . S c i . U S A 86 :9079 . 247 . Mal ik, P. , W . J . Kral l , X . Y u , C . Zhou , and D.B. Kohn . 1995. Retroviral-mediated gene express ion in human myelomonocyt ic cel ls : A compar ison of hematopoiet ic cell promoters to viral promoters. B lood 86 :2993. 248. Dziennis , S . , A . V a n Etton, H.L. Pah l , D.L. Morris, T.L. Rothstein, C M . B losch , R . M . Perlmutter, and D .G . Tenen . 1995. The C D 1 1 b promoter directs high-level express ion of reporter genes in mac rophages in t ransgenic mice. B lood 85:319. 249 . Corre l l , P . H . , S . Col i l la , and S . Kar lsson. 1994. Retroviral vector des ign for long-term express ion in murine hematopoiet ic cel ls in vivo. B lood 84:1812. 250 . Riviere, I., K. Brose, and R . C . Mul l igan. 1995. Effects of retroviral vector des ign on express ion of human adenos ine deaminase in murine bone marrow transplant recipients engrafted with genetical ly modif ied ce l ls . P roc . Natl . A c a d . S c i . U S A 92:6733. 131 2 5 1 . Ka leko , M., J .V . Garc ia , W . R . Osborne, and A . D . Miller. 1990. Express ion of human adenos ine deaminase in mice after transplantation of genet ical ly modif ied bone marrow. B lood 75:1733. 252 . Kempler , G . , B. Freitag, B. Berwin, O. Nanassy , and E. Barkl is. 1993. Character izat ion of the Moloney murine leukemia virus s tem cel l -speci f ic repressor binding site. Virol . 193:690. 2 5 3 . Pe terson , R., G . Kempler , and E. Barkl is. 1991. A stem cel l -speci f ic s i lencer in the primer-binding site of a retrovirus. Molec . Ce l l . B io l . 11:1214. 254 . Taketo, M., and M. Tanaka . 1987. A cellular enhancer of retrovirus gene express ion in embryonal carc inoma cel ls. Proc . Natl . A c a d . S c i . U S A 84 :3748 . 2 5 5 . Sel iger , B., R. Kol lek, C . Stocking, T. Franz, and W . Oster tag. 1986. Viral transfer, transcription, and rescue of a selectable myeloproli ferative s a r c o m a virus in embryonal cel l l ines: express ion of the mos oncogene . Mo lec . Ce l l . B io l . 6:286. 256 . B a u m , C , S . Hegewisch-Becker , H . -G . Eckert, C . Stock ing, and W. Oster tag. 1995. Novel retroviral vectors for efficient express ion of the multidrug resistance (mdr-1) gene in early hematopoiet ic cel ls . Journa l of Vi ro l . 69 :7541. 257 . Chal l i ta , P . M . , and D.B. Kohn . 1994. Lack of express ion from a retroviral vector after transduction of murine hematopoiet ic s tem cel ls is assoc ia ted with methylation in vivo. Proc . Natl. A c a d . S c i . U S A 91:2567. 258 . Hi lberg, F., C . Stock ing, W. Ostertag, and M. G rez . 1987. Funct ional ana lys is of a retroviral host-range mutant: altered long terminal repeat s e q u e n c e s al low express ion in embryonal ca rc inoma cel ls . P roc . Nat l . A c a d . S c i . U S A 84:5232. 259 . Col icel l i , J . , and P . W . J . Rigby. 1987. Isolation of a recombinant murine leukemia virus utilizing a new primer t R N A . J . Virol . 57:37. 260 . Bernad , A. , F. Va ras , J . M . Gal lego, J . M . A lmendra l , and J . A . Bueren . 1994. Ex vivo expans ion and select ion of retrovirally t ransduced bone marrow: s tem cel ls. Br. J . Haemato l . 87:6. 2 6 1 . Ko rman , A . J . , J . D . Frantz, J .L . Strominger, and R . C . Mul l igan. 1987. Express ion of human c lass II major histocompatibil ity complex ant igens using retrovirus vectors. Proc . Natl. A c a d . S c i . U S A 84:2150. 262 . Y a n g , Z . , A . J . Korman, J . Cooper , D. P ious, R .S . Acco l la , R . C . Mul l igan, and J . L . Strominger. 1987. Express ion of H L A - D R antigen in human c lass II mutant B-cell l ines by double infection with retrovirus vectors. Mo lec . C e l l . B io l . 7 :3923. 132 2 6 3 . Corey , C .A . , A . D . DeS i l va , C . A . Hol land, and D.A. Wi l l iams. 1990. Ser ia l transplantation of methotrexate-resistant bone marrow: protection of murine recipients from drug toxicity by progeny of t ransduced s tem cel ls . B lood 75:337. 264 . F lasshove , M., D. Banerjee, S . Mineichi , M.-X. Li , J . R . Bert ino, and M . A . S . Moore . 1995. Ex vivo expans ion and select ion of human C D 3 4 + peripheral b lood progenitor cel ls after introduction of a mutated dihydrofolate reductase c D N A v ia retroviral gene transfer. B lood 85:566. 2 6 5 . Mul len, C .A . , M. Kilstrup, and R . M . B laese . 1992. Transfer of the bacterial gene for cytosine deaminase to mammal ian cel ls confers lethal sensit ivity to 5-f luorocytosine: A negative select ion sys tem. Proc . Natl . A c a d . S c i . U S A 89:33. 266 . Sorrentino, B .P . , S . J . Brandt, D. Bodine, M. Got tesman, I. Pas tan , A . C l ine , and A . W . Nienhuis. 1992. Select ion of drug-resistant bone marrow cel ls in vivo after retroviral transfer of human M D R 1 . Sc i ence 257:99. 267 . P lautz , G . , E . G . Nabe l , and G . J . Nabe l . 1991. Select ive el imination of recombinant genes in vivo with a suic ide retroviral vector. The N e w Biologist 3:709. 268 . Strair, R.K. , M. Towle, and B.R. Smith. 1990. Retroviral mediated gene transfer into bone marrow progenitor cel ls: use of beta-ga lactos idase a s a se lectab le marker. Nucle ic Ac ids R e s . 18:4759. 269 . No lan , G . P . , S . Fier ing, J . F . Nico las, and L.A. Herzenberg . 1988. F luorescence-act iva ted cel l ana lys is and sorting of viable mammal ian cel ls based on B-D-ga lac tos idase activity after transduction of Escher i ch ia coli lacZ. Proc . Natl . A c a d . S c i . U S A 85:2603. 270 . Henthorn, P., P. Zervos, M. Raducha , H. Harris, and T. K a d e s c h . 1988. Express ion of a human placental alkal ine phosphatase gene in t ransfected cel ls : U s e as a reporter for studies of gene express ion. Proc . Natl . A c a d . S c i . U S A 85:6342. 2 7 1 . Strair, R.K., M .J . Towle, and B.R. Smith. 1988. Recombinant retroviruses encod ing cell sur face ant igens as selectable markers. Virol . 62 :4756. 272 . R ichardson , C , and B. A . 1995. Preselect ion of t ransduced murine hematopoiet ic s tem cel l populat ions leads to increased long-term stability and express ion of the human multiple drug resistance gene. B lood 86 :2579 . 2 7 3 . Valt ieri , M. , R. Schi ro, C . Che lucc i , B. Mase l la , U. Tes ta , I. C a s e l l a , E. Montesoro, G . Mariani , H .J . H a s s a n , and C . Pesch le . 1994. Efficient transfer of selectable and membrane reporter genes in hematopoiet ic progenitor and s tem cel ls purified from human peripheral b lood. C a n c e r R e s e a r c h 54:4398. 133 274 . O l s e n , J . C . , C . G . Johnson , M.L. W o n g - S u n , K . L Moore, R. Swans t rom, and R . C . Boucher . 1993. Retrovirus-mediated gene transfer to cyct ic f ibrosis airway epithelial cel ls : effect of selectable marker s e q u e n c e s on long-term express ion . Nucle ic Ac ids R e s . 21:663. 275 . Connea l ly , E. , P. Bardy, C . J . E a v e s , T. Thomas , S . C h a p p e l , E . J . Shpa l l , and R.K. Humphr ies. 1996. Rapid and efficient select ion of human hematopoiet ic cel ls express ing murine heat-stable ant igen as an indicator of retroviral-mediated gene transfer. B lood 87:456. 276 . Tumas , D.B., G . J . Spangrude, D.M. Brooks, C D . Wi l l iams, and B. C h e s e b r o . 1996. High-frequency cel l surface express ion of a foreign protein in murine hematopoiet ic stem cel ls using a new retroviral vector. B lood 87:509. 277 . Chal f ie , M. , Y . Tu , G . Eusk i rchen, W . W . Ward , and D . C Prasher . 1994. G r e e n f luorescent protein as a marker for gene express ion . S c i e n c e 263 :802 . 278 . Pawl iuk, R., R. Kay , P. Lansdorp, and R.K. Humphr ies. 1994. Select ion of retrovirally t ransduced hematopoiet ic cel ls using C D 2 4 as a marker of gene transfer. B lood 84:2868. 279 . P lane l les , V . , A . Haisl ip, E . S . Wi thers-Wad, S .A . Stewart, Y . X ie , N .P . S h a h , and I.S.Y. C H e n . 1995. A new reporter sys tem for detection of retroviral infection. G e n e Therapy 2:369. 280 . G r e z , M. , E. Akgun , F. Hilberg, and W. Ostertag. 1990. Embryonic stem cel ls virus, a recombinant murine retrovirus with express ion in embryonic s tem cel ls . Proc . Natl. A c a d . S c i . U S A 87:9202. 2 8 1 . G r e z , M. , M. Zornig, J . Nowock, and M. Ziegler. 1991. A single point mutation act ivates the Moloney murine leukemia virus long terminal repeat in embryonal stem cel ls. J . Virol. 65 :4691. 282 . Pr ince, V . E . , and P . W . J . Rigby. 1991. Derivat ives of Moloney murine s a r c o m a virus capab le of being transcr ibed in embryonal ca rc inoma cel ls have gained a functional Sp1 binding site. J . Virol . 65 :1803. 2 8 3 . Hawley, R . G . , F.H.L. L ieu, A . Z . C . Fong , and T . S . Hawley. 1994. Versat i le retroviral vectors for potential use in gene therapy. G e n e Therapy 1:136. 284 . Garver , R.I., A . Chyti l , S . Kar lsson, G .A . Fel ls, M.L. Brantly, M . Courtney, P .W. Kantoff, A . W . Nienhuis, W . F . Anderson, and R . G . Crysta l . 1987. Product ion of g lycosylated physiological ly "normal" human a lpha 1-antitrypsin by mouse fibroblasts modif ied by insertion of a human a lpha 1-antitrypsin c D N A using a retroviral vector. Proc . Natl. A c a d . S c i . U S A 84 :1050 . 2 8 5 . Mclvor , R .S . , J . M . Johnson , D.A. Miller, S . Pitts, S . R . Wi l l iams, D. Valer io, D.W. Mart in, and I.M. Ve rma . 1987. Human purine nuc leos ide 134 phosphory lase and adenos ine dema inase gene transfer into cul tured cel ls and murine hematopoiet ic stem cel ls by using recombinant amphotropic retroviruses. Molec . Ce l l . Bio l . 7:838. 286 . Bowtel l , D.D.L., S . Cory , G . R . Johnson , and G . J . Thomas . 1988. Compar i son of express ion in hemopoiet ic cel ls by retroviral vectors carrying two genes . J . Virol . 62:2464. 287 . Wi l l iams, D.A., B. L im, E. Spooncer , J . Longtine, and T . M . Dexter. 1988. Restr ict ion of express ion of an integrated recombinant retrovirus in primary but not immortal ized murine hematopoiet ic s tem cel ls . B lood 71:1738. 2 8 8 . Y e e , J . -K . , J . C . Moores, D.J . Jolly, J .A . Wolff, J . G . R e s p e s s , and T. F r iedmann. 1987. G e n e express ion from transcriptionally d isab led retroviral vectors. Proc . Natl. A c a d . S c i . U S A 84:5197. 289 . Hawley, R . G . , L. Covarrub ias, T. Hawley, and B. Mintz. 1987. Hand icapped retroviral vectors efficiently t ransduce foreign genes into hematopoiet ic s tem cel ls . Proc . Natl. A c a d . S c i . U S A 84:2406. 290 . Y u , S . - F . , T. von Ruden , P .W. Kantoff, C . Garber , M. Se iberg , U. Ruther, W . F . Anderson , E .F . Wagner , and E. G i lboa . 1986. Self- inactivating retroviral vectors des igned for transfer of whole genes into mammal ian cel ls . P roc . Natl. A c a d . S c i . U S A 83:3194. 2 9 1 . Hol lander, G .A . , B.D. Luskey , D.A. Wi l l iams, and S . J . Burakoff. 1992. Funct ional express ion of human C D 8 in fully reconstituted mice after retroviral-mediated gene transfer of hematopoiet ic s tem cel ls . J Immunol. 42 :321 . 292 . Correl l , P .H . , J .K . Fink, R .O. Brady, L.K. Perry, and S . Kar lsson. 1989. Product ion of human g lucocerebros idase in mice after retroviral gene transfer into multipotential hematopoiet ic progenitor cel ls. P roc . Nat l . A c a d . S c i . U S A 86:8912. 2 9 3 . Y a o , S . - N . , J . M . Wi lson, E . G . Nabe l , S . Kurachi , H.L. Hach iya , and K. Kurach i . 1991. Express ion of human factor IX in rat capil lary endothel ial ce l ls : toward somat ic gene therapy for hemophi la B. Proc . Natl . A c a d . S c i . U S A 88 :8101 . 294 . E m e r m a n , M., and H.M. Temin . 1984. G e n e s with promoters in retrovirus vectors can be independently suppressed by an epigenet ic mechan i sm . Ce l l 39 :459. 2 9 5 . Evstaf ieva, A . G . , T .Y . Ugarova, B.K. Chernov, and I.N. Shatsky. 1990. A complex R N A sequence determines the internal initiation of encephalomyocard i t is virus R N A translation. Nucle ic A c i d s R e s . 16:665. 2 9 6 . Skinner, M.A., V . R . Racanie l lo , G . Dunn, J . Cooper , P .D. Minor, and J . W . A lmond . 1989. New model for the secondary structure of the 5' non-coding R N A of poliovirus is supported by b iochemical and genet ic data that a lso 135 show that R N A secondary structure is important is neurovirulence. J . Mo lec . Bio l . 207:379. 297 . K im, D .G. , H .M. Kang , S .K . J a n g , and H. -S. Sh in . 1992. Construct ion of a bifunctional m R N A in the mouse by using the internal r ibosomal entry site of the encephalomyocard i t is virus. Mo lec . Ce l l . Bio l . 12:3636. 298 . Morgan, R.A., L. Couture, S . O . Elroy, J . Ragheb , B. M o s s , and W . F . Ande rson . 1992. Retroviral vectors containing putative internal r ibosome entry sites: to human gene therapy. Nucle ic Ac ids R e s . 20 :1293. 299 . Kay , R., P . M . Rosten, and R.K. Humphries. 1991. C D 2 4 , a s ignal t ransducer modulat ing B cel l activation responses , is a very short peptide with a glycocyl phosphatidyl inositol membrane anchor. J . Immunol. 147 :1412 . 300 . Kay , R., and R.K. Humphr ies. 1991. New vectors and procedures for isolating c D N A s encoding cell sur face proteins by express ion c loning in C O S cel ls . Methods in Molecular and Cel lu lar Biology 2:254. 3 0 1 . T h o m a s , K.R. , and M.R. Capecch i . 1987. Site-directed mutagenes is by gene targetting in mouse embryo-der ived stem cel ls . Ce l l 51 :503. 302 . G r e z , M. , E. Akgun , F. Hi lberg, and W. Ostertag. 1990. Embryonic s tem cell virus, a recombinant murine retrovirus with express ion in embryonic s tem cel ls . Proc . Natl . A c a d . S c i . U S A 87:9202. 3 0 3 . Miller, A . D . , and G . J . R o s m a n . 1989. Improved retroviral vectors for gene transfer and express ion . B ioTechn iques 7:980. 304 . Migita, M. , J .A . Medin , R. Pawliuk, S . J a c o b s o n , J . W . Nagle, S . Ande rson , M. Amir i , R.K. Humphr ies, and S . Kar lsson. 1995. Select ion of t ransduced C D 3 4 + progenitors and enzymat ic correction of cel ls from G a u c h e r patients, with bicistronic vectors. Proc . Natl . A c a d . S c i . U S A 92:12075. 3 0 5 . Pa lac ios , R., and M. Steinmetz. 1985. IL-3 dependent mouse c lones that express B-220 surface ant igen, contain Ig genes in germ-l ine configuration, and generate B lymphocytes in vivo. Ce l l 41 :727. 306 . Porter, E .H . , and R . J . Berry. 1963. The efficient design of transplantable tumour a s s a y s . Br. J . C a n c e r 17:583. 307 . Taswe l l , C . 1981. Limiting dilution a s s a y s for the determination of immunocompetent cell f requencies. J . Immunol. 126:1614. 308 . Sambrook , J . , E .F . Fr i tsch, and T. Maniat is. 1989. Molecular c loning: a laboratory manua l . S e c o n d edition. Co ld Spr ing Harbour P r e s s , C o l d Spr ing Harbour, N Y . E.3 pp. 136 309 . Unke less , J . C . 1979. Character izat ion of a monoclonal ant ibody directed against mouse macrophage and lymphocyte F c receptors. Journa l of Exp . M e d . 150:580. 310 . Lansdorp, P . M . , R . C . Aa lberse, R. Bos , W . G . Schulter, and E . F . G . V a n Bruggen . 1986. Cyc l i c tetramolecular comp lexes of monoc lonal ant ibodies: a new type of c ross linking reagent. European Journa l of Immunol. 16:679. 3 1 1 . W o g n u m , A . W . , T . E . Thomas , and P . M . Lansdorp. 1987. U s e of tetrameric antibody complexes to stain cel ls for flow cytometry. Cytometry 8:366. 312 . Miller, A .D. , T .D. Palmer, and R.A. Hock. 1986. Transfer of genes into human somat ic cel ls using retrovirus vectors. C o l d Spr ing Harbour S y m p o s i a on Quantum Biology 51:1013. 3 1 3 . Palmer , T.D., R.A. Hock, W . R . A . Osborne, and A .D. Miller. 1987. Efficient retrovirus mediated transfer and express ion of a human adenos ine deaminase gene in diploid skin f ibroblasts from an adenos ine deaminase deficient human. Proc . Natl . A c a d . S c i . U S A 84:1055. 314 . W o o d , P.A. , G . E . Herman, C . V . C h a o , W . E . O'Br ien, and A . L . Beaudet . 1986. Retrovirus mediated gene transfer of arg inosuccinate synthetase into cultured rodent cel ls and human citrulinemic f ibroblasts. C o l d Spr ing Harbour S y m p o s i a on Quantum Biology 51:1027. 315 . Miller, A . D . , M.F. Law, and I.M. Ve rma . 1985. Generat ion of helper-free amphotropic retroviruses that t ransduce a dominant-act ing methotrexate-resistant dihydrofolate reductase gene. Molec . Ce l l . B io l . 5 :431. 316 . Stuh lmann, H., R. C o n e , R . C . Mul l igan, and R. Jaen i sch . 1984. Introduction of a selectable gene into different animal t issue by a retrovirus recombinant vector. Proc . Natl . A c a d . S c i . U S A 81 :7151. 317 . Gu i ld , B .C . , R . C . Mul l igan, P. Gros , and D.E. Housman . 1988. Retroviral transfer of a murine c D N A for multidrug resistance confers pleiotropic drug resistance to cel ls without prior drug select ion. Proc . Natl . A c a d . S c i . U S A 85 :1595 . 318 . Kane , S . E . , D.H. Reinhard, C M . Fordis, I. Pas tan , and M . M . Got tesman. 1989. A new vector using the human multidrug resistance gene as a se lectab le marker enab les overexpress ion of foreign genes in eukaryot ic cel ls . G e n e 84:439. 319 . Cho i , K., T .O . Frommel , R.K. Stern, C . F . Perez , M. Kriegler, T. Tsusuo , and I.B. Ron inson . 1991. Multidrug resistance after retroviral transfer of the human M D R 1 gene correlates with P-glycoprotein density in the p l asma membrane and is not affected by cytotoxic select ion. Proc . Natl . A c a d . S c i . U S A 88:7386. 137 320 . Wi lson, J . M . , D .M. Jef ferson, J . R . Chowdhury, P . M . Novikoff, D .E . Johns ton , and R . C . Mul l igan. 1988. Retrovirus mediated transduct ion of adult hepatocytes. Proc . Natl. A c a d . S c i . U S A 85:3014. 3 2 1 . F ie lds-Berry , S . C , A . L . Holl iday, and C L . C e p k o . 1992. A recombinant retrovirus encoding alkal ine phosphatase conf i rms c lonal boundary ass ignment in l ineage analys is of murine retina. Proc . Natl . A c a d . S c i . U S A 89 :693 . 322 . Nabe l , E. , G . Plautz, and G . J . Nabe l . 1990. Site specif ic gene express ion in vivo by direct gene transfer into the arterial wal l . Sc i ence 249 :1285 . 323 . Nabe l , E . G . , G . Plautz, F .M . Boyce , C . Stanley, and G . J . Nabe l . 1989. Recombinant gene express ion in vivo within endothel ial ce l ls of the arterial wal l . S c i e n c e 244:1342. 324 . Ferry, N., O . Dupless is , D. Houss in , O . Danos , and J . M . Heard . 1991. Retroviral-mediated gene transfer into hepatocytes in vivo. Proc . Nat l . A c a d . S c i . U S A 88:8377. 3 2 5 . Pr ice, J . , D. Turner, and C . C epko . 1987. L ineage analys is in the vertebrate nervous sys tem by retrovirus-mediated gene transfer. Proc . Natl . A c a d . S c i . U S A 84:156. 326 . Mavi l io, F., G . Ferrari, S . Ross in i , N. Nobil i , C . Bonini , G . Casora t i , C . Traversar i , and C . Bordignon. 1994. Per ipheral blood lymphocytes as target cel ls of retroviral vector-mediated gene transfer. B lood 83:1988. 327 . Kay , R., F. Take i , and R.K. Humphr ies. 1990. Express ion cloning of a c D N A encod ing M1 /69 -J11d heat-stable ant igens. J . Immunol. 145:1952. 328 . Hough, M.R., F. Take i , R.K. Humphries, and R. Kay . 1994. Defect ive development of thymocytes overexpress ing the costimulatory molecule, heat-stable ant igen. J . Exp . M e d . 179:177. 329 . L iu, Y . , B. Jones , A . Aruffo, K . M . Sul l ivan, P . S . Linsley, and C A . J a n e w a y J r. 1992. Heat-stable antigen is a costimulatory molecule for C D 4 T cel l growth. J . Exp . M e d . 175:437. 330 . Lamb, B.T., K. Satyamoorthy, L. L i , D. Solter, and C . C . Howe. 1991. C p G methylation of an endogenous retroviral enhancer inhibits transcript ion factor binding and activity. G e n e Express ion 1:185. 3 3 1 . Hoeben , R . C , A . A . J . Migchie lsen, R . C . M . van der Jagt, H. van Ormondt, and A . J . van der E b . 1991. Inactivation of the Moloney Murine leukemia virus long terminal repeat in murine fibroblast cel l l ines is assoc ia ted with methylation and dependent on its chromosomal locus. J . Vi ro l . 65:904. 332 . Corre l l , P . H . , S . Col i l la , H . P . G . Dave, and S . Kar lsson . 1992. High levels of human g lucocerebros idase activity in macrophages of long-term 138 reconstituted mice after retroviral infection of hematopoiet ic s tem cel ls . B lood 80 :331 . 3 3 3 . R ichardson, C , M. Ward , and A . Bank. 1995. M D R gene transfer into live mice. C l in . Investig. 73:189. 334 . Tsuk i yama, T., H. U e d a , S . Hirose, and O. N iwa. 1992. Embryona l long terminal repeat-binding protein is a murine homolog of F T Z - F 1 , a member of the steroid receptor superfamily. Molec. Ce l l . B io l .12:1286. 3 3 5 . Rebe l , V.I., C .L . Miller, G . R . Thornbury, W . H . Dragowska, C . J . E a v e s , and P . M . Lansdorp . In P ress . A compar ison of long-term repopulat ing hematopoiet ic stem cel ls in fetal liver and adult bone marrow from the mouse . Exp . Hemato l . 336 . M a u c h , P. , J . S . Greenberger , L. Botnick, E. Hannon, and S . He l lman. 1980. Ev idence for the structured variation in sel f -renewal capaci ty within long-term bone marrow cultures. Proc . Natl . A c a d . S c i . U S A 77:2927. 337 . E a v e s , C . J . , J . D . C a s h m a n , S .D . Wo lpe , and A . C . E a v e s . 1993. Unrespons iveness of primitive chronic myeloid leukemis cel ls to macrophage inflammatory protein-a , an inhibitor of primitive normal hematopoiet ic cel ls . Proc . Natl. A c a d . S c i . U S A 90:12015. 338 . Schof ie ld , R. 1970. A comparat ive study of the repopulating potential of grafts from var ious haematopoiet ic sources : C F U repopulat ion. Cel lu lar T i ssue Kinet ics 3:119. 339 . Suther land, H.J . , P . M . Lansdorp, D.H. Henke lman, A . C . E a v e s , and C . J . E a v e s . 1990. Funct ional character izat ion of individual human hematopoiet ic s tem cel ls cultured at limiting dilution on support ive marrow stromal layers. Proc . Natl . A c a d . S c i . U S A 87:3584. 340 . Kral l , W . J . , D .C. Skel ton, X . J . Y u , I. Riviere, P. Lehn , R . C . Mul l igan, and D.B. Kohn . 1996. Increased levels of spl iced R N A account for augmented express ion from the M F G retroviral vector in hematopoiet ic cel ls . G e n e Therapy 3:37. 3 4 1 . Med in , J .A . , M. Migita, R. Pawliuk, S . J a c o b s o n , M. Amir i , S . Kluepfe l -Stahl , R .O . Brady, R.K. Humphr ies, and S . Kar lsson. 1996. A bicistronic t herapeut ic retroviral vector enab les sorting of t ransduced C D 3 4 + cel ls and corrects the enzyme def iciency in cel ls from G a u c h e r patients. B lood 87 :1754 . 342 . Che rkezov , Z . H . 1996. B lood in press. 139