Oncogene (2001) 20, 5644 ± 5659 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc

Modeling Philadelphia chromosome positive leukemias

Stephane Wong3 and Owen N Witte*1,2,3

1Howard Hughes Medical Institute, University of California, Los Angeles, California, CA 90095-1662, USA; 2Department of Microbiology, Immunology, and Molecular Genetics, University of California, Los Angeles, California, CA 90095-1662, USA; 3Molecular Biology Institute, University of California, Los Angeles, California, CA 90095-1662, USA

The Ph chromosome has been genetically linked to CML common. In the blast crisis phase, blood and marrow and ALL. Its chimeric fusion gene product, BCR ± ABL, mature cells are displaced by immature blasts. More can generate leukemia in mice. This review will discuss than 50% of patients enter a myeloid blast stage selected model systems developed to study BCR ± ABL resembling acute myeloblastic leukemia (AML). A pre- induced leukemia and focuses on what we have learned B blast stage similar to acute lymphoid leukemia (B- about the human disease from these models. Five main ALL) accounts for 30% of patients, and erythroid experimental approaches will be discussed including: (i) blasts develop in 10% of patients. Rarely do T cell Reconstitution of mice with bone marrow cells retro- blasts (T-ALL) evolve (Colleoni et al., 1996; Fabbiano virally transduced with BCR ± ABL; (ii) Transgenic mice et al., 1998). During progression to blast phase, over expressing BCR ± ABL; (iii) Knock-in mice with BCR ± 80% of patients develop additional chromosome ABL expression driven from the endogenous bcr locus; abnormalities (reviewed in Mitelman, 1993). This (iv) Development of CML-like disease in mice with loss suggests that genetic aberrations in addition to the of function mutations in heterologous genes; and (v) ES Ph chromosome are necessary for chronic phase CML in vitro hematopoietic di€erentiation coupled with to develop to blast crisis. regulated BCR ± ABL expression. Oncogene (2001) 20, The Ph chromosome results from a reciprocal 5644 ± 5659. translocation, between chromosomes 9 and 22 t(9;22)- (q34;q11) (Nowell and Hungerford, 1960; Rowley, Keywords: BCR ± ABL; mouse model; leukemia 1973). The telomeric segment of chromosome 9q34 encoding the ABL tyrosine kinase gene is fused to the centromeric segment of chromosome 22q11 encoding the BCR gene (see Figure 1). This forms a chimeric BCR ± ABL mRNA and protein product. The alter- Introduction native reciprocal translocation product, ABL ± BCR, is thought not to play a role in leukemogenesis and will The Philadelphia chromosome was the ®rst consistent not be discussed further (Lazaridou et al., 1994; Melo chromosomal abnormality linked to a speci®c human et al., 1993, 1996). Depending on the breakpoint cancer, chronic myeloid leukemia (CML) (Nowell and positions within the BCR and ABL genes, di€erent Hungerford, 1960). CML accounts for 15 ± 20% of forms of BCR ± ABL are generated. Most breakpoints human leukemia (Gha€ari et al., 1999). CML is a within ABL occur between exons 1a and 1b to generate clonal disorder originating in the hematopoietic stem BCR ± ABL mRNA with BCR sequences fused to ABL cell (HSC) (Delforge et al., 1999; Fialkow et al., 1967, exon a2. Breakpoints within the BCR locus occur over 1977; Holyoake et al., 1999; and reviewed in Kabar- a large area spanning over 20 exons. Three breakpoint owski and Witte, 2000; Raskind and Fialkow, 1987; cluster regions (bcr) within BCR are most frequently Takahashi et al., 1998; Yo€e et al., 1987). It can be involved. The major breakpoint cluster region (M-bcr) regarded as a biphasic disease (reviewed in Clarkson et generates an mRNA product with either a b2a2 or al., 1997). The ®rst stage, chronic phase (cpCML), is b3a2 junction due to alternative splicing which encodes characteristically indolent with clinical features includ- a 210 Kd chimeric protein p210BCR ± ABL (Ben-Neriah et ing a large increase in the number of myeloid al., 1986; Heisterkamp et al., 1985; Konopka et al., precursors and their progeny. Numbers of mature 1984; Shtivelman et al., 1985). A minor breakpoint neutrophils as well as basophils and eosinophils are cluster region (m-bcr) junction in BCR produces a elevated in the blood. In the bone marrow there is also smaller 185/190 Kd chimeric protein (Chan et al., 1987; an increase in the ratio of myeloid to erythroid cells. Clark et al., 1987; Fainstein et al., 1987; Walker et al., Splenomegaly, hepatomegaly and mild anemia are 1987). Recently, a larger 230-kd chimeric protein P230BCR ± ABL has been identi®ed generated from a g- bcr breakpoint (Saglio et al., 1990). Other fusion events leading to expression of *Correspondence: ON Witte, HHMI/UCLA, 675 Charles E Young Drive South, Room 5-748 MRL Bldg, Los Angeles, CA 90095-1662, activated forms of ABL occur in leukemia. Rare cases USA; E-mail: [email protected] of human ALL express a TEL ± ABL fusion (Golub et Models of Ph+ leukemias S Wong and ON Witte 5645

Figure 1 Locations of the breakpoints in the BCR and ABL genes and structure of the chimeric proteins derived. Three main breakpoint regions within the BCR genome are responsible for generating the three predominant forms of BCR ± ABL protein. The minor breakpoint cluster (m-BCR) spans 54-kb and results in an e1a2 7.0 kb mRNA that generates P185. The major breakpoint cluster (M-BCR) spans 5.8 kb and results in either a b2a2 or b3a2 8.5 kb mRNA producing P210. A third breakpoint cluster located at the 3' end of the gene (g-BCR) generates an e19a2 9.0 kb mRNA forming P230. Regardless of the exact breakpoint in ABL, BCR sequences are most often fused to ABL exon a2 in the hybrid transcript. BCR domains include: OLIGO=oligomerization domain; A and B=SH2 domain binding regions; S/TKINASE=serine/threonine kinase domain; DBL/ CDC24=Dbl homology domain; PH=Pleckstrin homology domain; RACGAP=Rac-GTPase domain. ABL domains include: MYR=myristoylation site; SH3 and SH2 domains; SH1KINASE=tyrosine kinase domain; NLS=nuclear localization domains; DNA=DNA binding site; ACTIN=F and G actin binding sites al., 1996; Papadopoulos et al., 1995). In mice, the its ecacy, with 98 ± 100% complete hematologic original viral form of ABL (GAG ± ABL) is a chimeric remission (Druker et al., 2001). protein in which sequences encoding the ®rst exon and In addition to the tyrosine kinase domain, BCR ± part of the second exon (a2) of ABL are replaced by ABL contains oligomerization (McWhirter et al., those encoding the viral GAG protein. GAG ± ABL is 1993; Muller et al., 1991) SH2 (Afar et al., 1994; associated with B-ALL as well as other leukemias Goga et al., 1995), autophosphorylation (Pendergast (reviewed in Rosenberg and Witte, 1988). et al., 1993a), actin binding (McWhirter and Wang, Oncogenic forms of ABL have highly elevated levels 1991, 1993; Van Etten et al., 1994), and DNA of tyrosine kinase activity compared to c-ABL. The binding (Kipreos and Wang, 1992) motifs important tyrosine kinase activities of GAG ± ABL, P185BCR ± ABL for transformation (see Figure 1). The oligomeriza- and P210BCR ± ABL correlate with their transformation tion domain of BCR ± ABL is essential for transfor- potencies in several systems (Lugo et al., 1990). mation (Maru et al., 1996a; McWhirter et al., 1993). Tyrosine kinase inactive mutants of BCR ± ABL and Other domains such as the Grb2 binding site are GAG ± ABL lose their transforming potential (Engel- required for transformation of ®broblasts to ancho- man and Rosenberg, 1987; Pendergast et al., 1993a; rage independence (Pendergast et al., 1993b) but not Ponticelli et al., 1982; Witte et al., 1980, 1981). These of hematopoietic cell lines to growth factor indepen- ®ndings have supported the essential role of deregu- dence (Cortez et al., 1995). In vivo leukemogenesis lated tyrosine kinase activity in transformation by ABL studies are important to determine the signi®cance of oncogenes, a factor exploited by the ABL kinase these BCR ± ABL domains. inhibitor STI571, a 2-phenylaminopyrimidine that There is strong association of the three main forms speci®cally inhibits the tyrosine kinase activity of of BCR ± ABL with speci®c types of human leukemias. ABL, PDGF, and C-KIT (Buchdunger et al., 1995). P210BCR ± ABL is generally considered as the pathogno- Phase 1 clinical trial studies of Ph+ chronic phase monic marker of CML and is also associated with CML patients treated with STI571 have demonstrated approximately one third of adult Ph+ ALLs (Deininger

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5646 et al., 2000) and on occasion, acute myeloid leukemia had a 16% incidence of macrophage tumors, and a (AML) (Kantarjian et al., 1991) and myeloma (Martiat much higher incidence of erythroid leukemia. Neither et al., 1990). P185BCR ± ABL is linked to the remaining strain developed a CML-like disease (see Table 1). two-thirds of Ph+ ALL cases not associated with In 1990, several groups demonstrated that P210 can P210BCR ± ABL (Kantarjian et al., 1991; Melo, 1996; induce a CML-like disease in mice (Daley et al., 1990; Secker-Walker et al., 1988). In addition, P185BCR ± ABL Kelliher et al., 1990). The experimental strategy was to is linked to 3% of atypical CML cases with enrich for HSC expression of P210. Mice were treated monocytosis (Melo et al., 1994), and in some AML with 5-FU to kill cycling cells and enrich for HSC cases (Kurzrock et al., 1987) and lymphoma (Mitani et populations. Quiescent HSCs are induced to cycle with al., 1990). The recently identi®ed P230BCR ± ABL is mainly cytokine mixtures prior to infection with retroviral associated with neutropenic CML (Pane et al., 1996) as vectors expressing P210. Reconstitution of lethally well as some cases of CML (Wilson et al., 1997). It is irradiated BALB/c mice with P210 infected cells not understood why di€erent forms of BCR ± ABL are generated a CML-like disease. preferentially associated with distinct leukemias. Dif- This retroviral transduction system generated a ferences in target cell as well as BCR ± ABL form are CML-like disease in 20 ± 30% of reconstituted possible explanations. BALB/c mice with a mean disease latency of 63 days A correlation between the type of leukemia formed (Daley et al., 1990; Kelliher et al., 1990) (see Table 1). and the speci®c target cell containing the Ph chromo- P210 also induced lymphoid leukemias/lymphomas and some is evident in human patients. In CML patients monocytic/macrophage tumors in recipient mice. Clin- the Ph chromosome is present in a quiescent ical features of CML mice included high white blood hematopoietic stem cell (HSC) population (Takahashi cell counts (WBC) with elevated levels of granulocytes, et al., 1998; Delforge et al., 1999; Holyoake et al., hypogranulated basophils and metamyelocytes in the 1999; reviewed in Kabarowski and Witte, 2000). The blood. The bone marrow was hypercellular with a Ph+ quiescent HSC is often maintained in CML dominance of myeloid cells at all stages of develop- patients treated with irradiation and bone marrow ment. The spleen and liver were enlarged with leukemic transplantation (Deisseroth et al., 1994; Diekmann et in®ltrates. This CML-like disease was transplantable, al., 1994) and is likely the cause of disease recurrence. suggesting the initial leukemic clone was of stem cell In most cases of Ph+ ALL, on the other hand, the origin (Daley et al., 1990). presence of the Ph chromosome is restricted to the Could this CML model system be used to study the lymphoid lineage (Haferlach et al., 1997; Kasprzyk et progression of CML disease from chronic phase to al., 1999). Although some cases of Ph+ ALL are blast crisis? Gishizky et al. (1993) studied the associated with the presence of the Ph chromosome in development of P210 induced leukemia in primary, multiple lineages, suggesting HSC origin, many of these secondary, tertiary and quaternary transplant cases may represent CML lymphoid blast crises recipients. In primary recipients, three types of following a clinically silent/undiagnosed chronic phase leukemia, B lymphoid, myelo/monocytic, and granulo- (Pajor et al., 2000). cytic leukemia resembling chronic phase CML, devel- oped with a wide range in latency. The leukemias were categorized into short (5140 days) and long (4140 Retroviral transduction of bone marrow cells with days) latency (see Table 1). ABL oncogene The question of whether the short and long latency leukemias originate from di€erent target cells was In 1970, Abelson and Rabstein showed that mice tested. `Early' and `late' leukemic mouse bone marrow infected with Abelson murine leukemia virus reprodu- samples were tested by secondary transplantation. cibly generated B cell lymphomas (Abelson and Myelo/monocytic and B lymphoid leukemias were Rabstein, 1970). It was later demonstrated that equally transplantable from `early' and `late' leukemic GAG ± ABL was the essential etiological component. mice, suggesting that P210 induced B-ALL and myelo/ GAG ± ABL can also transform mast cells and generate monocytic leukemia may arise from both transduced plasmacytomas in vivo (reviewed by Rosenberg and HSCs and committed progenitors. Witte, 1988). Abelson and Rabstein (1970) tested if However, `early' and `late' CML disease were not alternative strains of mice were equally susceptible to equally transplantable. Only one out of 30 secondary virus induced lymphoma. Although all strains of mice recipients transplanted with `early' CML donor cells developed the same type of B cell lymphoma, the developed leukemia (see Table 1). In sharp contrast, disease latency varied between strains, demonstrating `late' CML disease was readily transplantable in 75% that genetic context can vary the susceptibility to of secondary recipients (see Table 1). The transplanted disease establishment. chronic phase CML disease progressed to blast crisis in Do di€erent strains of mice vary in their penetrance the majority of secondary recipients. Acute myelo/ to P210 induced leukemia? Elefanty and Cory (Elefanty monocytic blast stage (AMML) was the predominant et al., 1990) tested two mouse strains, DBA/2 and disease, but chronic phase CML, B-ALL and T-ALL C57BL/6. DBA/2 mice reconstituted with P210 also occurred (see Table 1). Clonal analyses revealed transduced bone marrow developed macrophage that leukemic cells from di€erent blast crises had the tumors 80% of the time. However, C57BL/6 mice only same P210 viral integration site as the initial chronic

Oncogene Table 1 (a) MLV/MPSV retroviral transduction ABL mouse models Infection Oncogenic Animal Disease Latency Penetrance Disease route Promoter ABL form strain phenotype (Average) (%) transplantation Reference i.p. MLV helper + MLV GAG ± ABL BALB/C Pre-B lymphoma 25 days 100 100% Pre-B lymph. Abelson and Rabstein, 1970 GAG-ABL NIH swiss Pre-B lymphoma 28 days 100 ND virus DBA/2 Pre-B lymphoma 39 days 100 ND C3H Pre-B lymphoma 48 days 100 ND C57BL/6 Pre-B lymphoma 64 days 100 ND

i.v. infected MPSV P210 BALB/C CML 63 days 27 CML Daley et al., 1990 5-FU+IL-3, IL-6 b3a2 ALL 98 days 10 ND treated BM MAC tumor 116 days 7ND 44

i.v. infected MPSV P210 DBA/2 MAC tumor 115 days 80 58% MAC Elefanty et al., 1990 5-FU+SCM b3a2 Pre-B lymphoma 61 days 12 Pre-B lymph. treated BM MPSV Erythroid leuk. 21 days 4 rarely Mast tumor 112 days 4 mast tumor 100 C57BL/6 MAC tumor 243 days 16 58% MAC T lymphoma 147 days 13 T lymph. erythroid leuk. 50 ± 88 days 37 rarely Mast tumor 344 days 3 mast tumor 69

i.v. infected Abelson GAG ± ABL BALB/C MYEL+MONO leuk. 33 days 50 ND Kelliher et al., 1990 5-FU+IL-3 MLV Pre-B lymphoma 45 days 50 ND oeso Ph of Models treated BM 100 Witte ON and Wong S JW-RX P210 MYEL+MONO leuk. 70 days 22 ND b3a2 Gran. leuk 63 days 22 ND Pre-B lymphoma 78 days 44 ND

88 + leukemias i.v. infected Abelson GAG ± ABL BALB/C T lymphoma 28 days 17 ND Chung et al., 1991 DAY 12 MLV MONO+lymphoma 35 days 25 ND FL cells MYEL+blast leuk. 35 days 50 ND Mast cell leuk. 115 days 8 100

i.v. infected+ MPSV+ GAG ± ABL BALB/C Lymphoid+MONO 28 ± 70 days 100 ND Scott et al., 1991 MLV helper MLV leuk. 5-FU+WEHI BM

i.v. infected JW-RX+ P185 BALB/C Myeloid leuk. 41 days 31 ND Kelliher et al., 1991 MLV helper+ MLV e1a2 Pre-B lymphoma 46 days 42 ND 5-FU+IL-3 B-ALL 30 ± 100 days 15 ND Treated BM MAC tumor 30 ± 100 days 8ND 96 JW-RX+ P210 Myeloid leuk. 61 days 26 ND MLV b3a2 Pre-B lymphoma 70 days 56 ND B-ALL 41 ± 100 days 13 ND 95 Continued Oncogene 5647 Oncogene 5648

Table 1 (a) (Continued ) Infection Oncogenic Animal Disease Latency Penetrance Disease route Promoter ABL form strain phenotype (Average) (%) transplantation Reference i.v. infected MPSV P210 BALB/C Early myel. leuk. 84 days 5 50% Elefanty and Cory, 1992 5-FU b3a2 Late myel. leuk. 140 days 27 100% +SCM B lymphoma 49 days 5 ND 5 days T lymphoma 140 days 21 ND treated BM MAC tumor 126 days 42 ND 100

i.v. infected Early myel. leuk. 77 days 5 50% 5-FU Late myel. leuk. 196 days 85 100% +SCM B lymphoma 42 days 10 ND 2 days T lymphoma ± 0 ± treated BM MAC tumor ± 0±

100 Ph of Models

i.v. infected Early myel. leuk. 91 days 5 50% Witte ON and Wong S 5-FU Late myel. leuk. 196 days 28 100% IL-3+IL-6 B lymphoma 42 days 39 ND 2 days T lymphoma 182 days 11 ND + leukemias treated BM MAC tumor 252 days 17 ND 100

i.v. infected Early myel. leuk. ± ± ± 5-FU Late myel. leuk. 196 days 73 100% No added cyto. B lymphoma 112 days 27 ND 2 days T lymphoma ± 0 ± treated BM MAC tumor ± 0± 100

i.v. infected MLV P210 BALB/C CML 5140 days 11 3.3% AMML Gishizky et al., 1993 5-FU+IL-3, IL-6 (pMV) b3a2 B lymphoma 5140 days 3 B leukemia treated BM MYEL+MONO 5140 days 8 MAC+MONO CML 4140 days 11 48%AMML: 10% B-ALL: 5% T-ALL: 10% CML B lymphoma 4140 days 20 B leukemia MYEL+MONO 4140 days 22 MAC+MONO 75

i.t. MLV helper+ MLV GAG ± ABL BALB/C T lymphoma 32 days 62 ND Clark et al., 1993 virus (pMV) p185e1a2 T lymphoma 70 days 35 ND to thymus P210b3a T lymphoma 80 days 32 ND

i.v. infected MSV P210 C57BL ± SCID CML 182 days 70 ND Skorski et al., 1996 5-FU+IL-3, IL-6 (pSRa) b3a2 SzJ +KL treated BM Continued Table 1 (a) (Continued ) Infection Oncogenic Animal Disease Latency Penetrance Disease route Promoter ABL form strain phenotype (Average) (%) transplantation Reference i.v. infected MSV P185 CB.17 SCID B-ALL 45 days 25 ND Afar et al., 1997 total BM (pSRa) e1a2 no cytokines

i.v. infected MSC P210 BALB/C CML 20 ± 30 days 100 28=100% CML Pear et al., 1998 5-FU+IL-3, IL-6 (MSCV) b3a2 38=80% CML:10% T-ALL Zhang and Ren, 1998 +SCF treated BM 48=100% T-ALL

i.v. infected MSC P190 BALB/C CML 21 days 100 28=100% CML Li et al., 1999 5-FU+IL-3, IL-6 (MSCV) e1a2 +SCF treated BM P230 e19a2 BALB/C CML 25 days 100 28=100% CML

i.v. infected MSC P210 BALB/C CML 43 days 60 ND IL-3+IL-6+SCF (MSCV) b3a2 B-ALL 43 days 30 28=B-ALL Total BM MAC+MONO 43 days 10 ND

P190 CML 28 days 40 ND e1a2 B-ALL 28 days 60 28=B-ALL MAC+MONO ± ± ±

P230 e19a2 CML 50 days 40 ND B-ALL 50 days 50 28=B-ALL MAC+MONO 50 days 10 ND oeso Ph of Models ogadO Witte ON and Wong S

i.v. infected MSC P210 CB.17 SCID B-ALL 22 days 100 ND Lim et al., 2000 total BM (MSCV) b3a2 no cytokines + i.v. infected MSC P210 C57BL/6 CML 39 days 50 ND Li et al., 2001 leukemias IL-3+IL-6+SCF (MSCV) b3a2 B-ALL 39 days 5 ND 5-FU treated BM B-ALL+MAC 49 days 5 ND CML+B-ALL 50 days 10 ND CML+MAC 45 days 15 ND CML+B+MAC 43 days 5 ND

Table 1 (b) Transgenic mouse models Oncogenic Animal Founder (F) Disease Latency Disease Promoter ABL form strain TX line (T) phenotype (Average) Penetrance % transplantation Reference EuVH BCR-V-ABL C57BL/6JWEHI X F=12 Lymphoma 14 ± 77 days 25 ± Hariharan et al., 1989 SJL/JWEHI T=1 T-lymphoma 52 ± 244 days 75 45% T lymphoma B-lymphoma 17 ± 54 days 25 doesn't

MPSV LTR BCR-V-ABL C57BL/6JWEHI X F=3 Lymphoma 55 days 33 ± SJL/JWEHI T=1 T-lymphoma 55 ± 165 days 36 66% T lymphoma

Continued Oncogene 5649 Oncogene 5650

Table 1 (b) (Continued ) Oncogenic Animal Founder (F) Disease Latency Disease Promoter ABL form strain TX line (T) phenotype (Average) Penetrance % transplantation Reference MT P190e1a2 C57BL X CBA F=10 Myeloblast leuk. 10 ± 13 days 20 ± Heisterkamp et al., 1990 ALL 12 ± 58 days 60 ± T=0

BCR P210b3a2 F=0 Embryonic ± ± ± Heisterkamp et al., 1991 T=0 Lethal MT P190e1a2 C57BL X CBA F=14 B-blast leuk/lymp. 28 ± 198 days 43 ± Voncken et al., 1992a

B-ALL 70 days 14 ± Ph of Models T=7 B-Blast leuk/lymp. 31 ± 199 days 67 100% B-Blast B-ALL 31 ± 199 days 67 Up to 100% B-ALL Witte ON and Wong S

MT P210b3a2 C57BL/6 X F=6 T leukemia 90 days 33 ± Honda et al., 1995 +

DBA/2F2 T=1 T-leukemia 120 days 33 ND leukemias

MT P210b3a2 C57BL X CBA F+T=36 B-ALL 98 ± 196 days 30 ND Voncken et al., 1995a T-ALL 126 ± 308 days 60 ND Myeloblast leuk. 378 days 7 ND P190e1a2 F+T=42 B-ALL 56 days 100 ND

TEC P210b3a2 C57BL/6 X F=5 T-ALL 90 ± 120 days 40 ± Honda et al., 1998

DBA/2F2 T=1 CML 365 days 100 ND MMTV-tTA P210b3a2 BL/6 X SJL F=? ± ± ± ± Huettner et al., 2000 +TET-O-P210 X FVB/N T=4 B-ALL 24 ± 80 days 100 B-ALL

Table 1 (c) Knock-in ABL mouse model Oncogenic Animal Disease Latency Disease Promoter ABL form strain Chimeras phenotype (Average) Penetrance % transplantation Reference BCR Locus P190e1a2 129Sv X C57BL/6 40 Chimeras B-ALL 120 days 95 94% B-ALL Castellanos et al., 1997

leuk.=leukemia; lymph.=lymphoma; CML=chronic myeloid leukemia; ALL=acute lymphoid leukemia; AMML=acute myeloid/monocytic blast stage leukemia; MYEL=myeloid; MAC=macrophage; MONO=monocyte Models of Ph+ leukemias S Wong and ON Witte 5651 phase CML cells. This demonstrates that the P210 mas. These ®ndings further support the hypothesis transduced HSC that initially generated `late' chronic that target cell type may primarily determine the phase CML in primary recipients was also the origin of leukemic phenotype. blast crisis AMML, B-ALL, and T-ALL in secondary A disadvantage of the CML animal model recipients. `Late' chronic phase CML had a long described above is the long disease latency and low disease latency and clonal progression to blast crisis. penetrance. Advances in retroviral transduction and These observations suggest that, as in the human hematopoietic expression have led to the development disease, additional genetic aberrations are required for of new models which develop leukemia with 100% CML progression in this mouse model. penetrance and a short latency period of 18 ± 30 days. Using the same retroviral transduction system, Critical improvements in retroviral transduction Chung et al. (1996) noticed some mice did not develop protocols include more e€ective infection protocols leukemia after transplantation with GAG ± ABL in- and use of the MSCV LTR promoter engineered for fected bone marrow cells. This group questioned high expression in immature hematopoietic cells whether GAG ± ABL expression in HSCs was silenced (Hawley et al., 1994). The MSCV BCR ± ABL in these mice. Reconstituted mice that did not develop leukemia model retains the use of 5-FU treatment leukemia were treated with 5-FU to activate quiescent to enrich for HSC populations, and BALB/c mice as HSC carrying a GAG ± ABL integrant. The bone a highly penetrant host environment. A comparison marrow was transplanted into secondary recipients, of BALB/c and C57BL/6 strains for susceptibility to which developed a CML-like leukemia. The leukemia P210 induced CML-like disease showed 100% arose from a di€erent infected GAG ± ABL clone than penetrance with BALB/c and 50% penetrance with that from which the primary leukemia developed. This C57BL/6 (Li et al., 2001). demonstrated that activating a quiescent HSC which Several groups have demonstrated the generation harbored an oncogenic form of ABL could generate a of CML-like disease in P210 MSCV retroviral mouse CML-like disease. CML patients treated with irradia- models (Li et al., 1999; Pear et al., 1998; Zhang and tion have most of their Ph+ cells destroyed. The cause Ren, 1998). The clinical features of the disease in this of relapse in these patients is most likely the presence animal model include elevated white blood cell of radiation resistant Ph+ quiescent HSCs which counts, with a dominance of mature granulocytes become re-activated to generate CML (Deisseroth et and the presence of metamyelocyes and nucleated al., 1994; Diekmann et al., 1994; Holyoake et al., erythroid cells. Splenomegaly and hepatomegaly occur 1999). with an in®ltration of immature and mature As mentioned earlier, there is a strong correlation granulocytes. The bone marrow is hypercellular with between BCR ± ABL form and disease phenotype in a predominance of granulocytes. The disease is Ph+ human leukemias. However, a third of Ph+ B- polyclonal with an average of seven clones dominat- ALL involve the P210 form of BCR ± ABL. Although ing leukemic progression. The CML-like disease is these cases may not represent true de novo B-ALLs transplantable to secondary and tertiary recipients. but rather B-lymphoblastic blast crisis following an The transplanted disease is monoclonal and can undiagnosed chronic phase, it poses the question of develop to a T lymphoblastic blast crisis (T-ALL) whether the structure of BCR ± ABL or the target cell in quaternary recipients. However, T-ALL blast crisis acquiring the Ph translocation and BCR ± ABL is rare in human CML. Another unusual character- expression are most critical in determining the type istic of this model is death from pulmonary of leukemia formed. Kelliher et al. (1990, 1991) hemorrhage, not common in human CML, which addressed this issue by infecting the same 5-FU the authors suggest could be due to a rapid treated target cell population with GAG ± ABL, P185 development of leukostasis, blast/basophilic degranu- or P210. They observed that all forms of oncogenic lation, or platelet dysfunction. ABL were capable of generating a similar spectrum of The high reproducibility of a chronic phase CML disease: Pre-B lymphomas, myeloid leukemias and B blood picture and short latency period in this MSCV- leukemias. Scott et al. (1991) disagrees with this based system has enabled researchers to address conclusion. This group observe that GAG ± ABL many questions regarding BCR ± ABL induced leuke- induced leukemias generated by a similar method are mia. Li and Van Etten used this system to test associated with less severe splenomegaly and archi- whether the three forms of BCR ± ABL (P185, P210 tectural disruption compared to that observed in P210 and P230) generate identical or di€erent leukemias. induced leukemias. GAG ± ABL leukemic mice devel- All three forms of BCR ± ABL induced a CML-like op paraspinous masses which do not occur in P210 disease with 100% penetrance (Li et al., 1999). This generated leukemias. In addition, GAG ± ABL expres- again supports the hypothesis that the target cell sion is only detected in leukemic monocytes and expressing BCR ± ABL, rather than the BCR ± ABL lymphoid cells, but not granulocytes. form, may be most critical in determining the type of Clark et al. (1993) tested whether GAG ± ABL, leukemia. However, the short disease latencies (21 P185, and P210 are equally capable of generating days for P185, 23 days for P210 and 25 days for thymomas. This group injected GAG ± ABL, P185, or P230) precluded the establishment of whether these P210 virus directly into thymic lobes. All three forms di€erent BCR ± ABL forms induce more or less of oncogenic ABL induced indistinguishable thymo- aggressive leukemias.

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5652 Testing potential BCR ± ABL signaling pathways using atypical CML picture similar to human monocytic retroviral transduction CML induced by P185. Nevertheless, the issue of whether development of BCR ± ABL induced CML-like Loss of P53 function is a common event in the disease in this system involves, and perhaps requires, development of multiple human malignancies (reviewed the IL-3RBc itself remains to be addressed. by Hupp et al., 2000; Soussi, 2000) including 30% of Many groups have shown that STAT5 activation human CML blast crises (Prokocimer and Rotter, correlates with BCR ± ABL transformation (reviewed 1994; Rovira et al., 1995). Several groups have studied by Sattler and Salgia, 1997). Expression of a dominant the role of P53 in BCR ± ABL and GAG ± ABL negative form of STAT5 inhibited BCR ± ABL induced induced leukemia. A CML model using 5-FU treated bone marrow transformation in vitro (Nieborowska- donor bone marrow cells retrovirally transduced with Skorska et al., 1999). Moreover, a dominant negative P210 to reconstitute lethally irradiated mice was form of STAT5 inhibited both BCR ± ABL induced IL- utilized. Mice reconstituted with P53 transduced 3 independent growth and leukemogenesis of 32D cells knock-out cells developed a more aggressive myelo- (Nieborowska-Skorska et al., 1999). However, studies blastic leukemia with shorter latency compared to with STAT5A/B knock-out cells clearly demonstrate those reconstituted with transduced wild-type cells that P210 and P185 do not require STAT5A/B for (Skorski et al., 1996). In human CML, loss of P53 leukemogenesis (Sexl et al., 2000). This result exem- function is preferentially associated with progression to pli®es the importance of using alternative genetic myeloid rather than lymphoid blast crisis leukemia approaches to de®nitively address the requirement of (Prokocimer and Rotter, 1994). Over 40% of Pre-B speci®c pathways in BCR ± ABL oncogenesis. cells transformed by GAG ± ABL harbor a P53 BCR ± ABL expression is linked in cis to GFP mutation (Thome et al., 1997). GAG ± ABL lymphoma expression via an internal ribosomal entry site (IRES) studies also demonstrate a faster rate of disease element in the MSCV BCR ± ABL retrovirus. Several progression in the absence of P53 (Unnikrishnan et groups have demonstrated that expression of GFP in al., 1999; Zou et al., 2000). In addition, 75% of this system is an accurate indicator of BCR ± ABL GAG ± ABL lymphomas in P53 heterozygous (P53+/7) expression (Li et al., 1999; Pear et al., 1998; Zhang and mice had lost function of the remaining P53 allele. It is Ren, 1998). The e€ect of enforced expression of a clear that multiple genetic alterations may determine candidate gene, or its dominant negative counterpart, the severity of disease phenotype. upon BCR ± ABL induced leukemogenesis can be Many pathways have been implicated as being tested by appropriate replacement of GFP sequences. important for BCR ± ABL induced transformation. Other animal model systems with a longer disease One is the IL-3 signaling pathway. BCR ± ABL latency and lower penetrance may be better suited to expression in IL-3 dependent cell lines confers IL-3 identify genes that enhance BCR ± ABL induced independence (Anderson and Mladenovic, 1996; Hari- leukemia. Afar et al. (1997) generated leukemia in haran et al., 1988). Human Ph+ CML cells have been 25% of animals reconstituted with P185 transduced shown to have a reduced requirement of IL-3 for in bone marrow between 30 and 90 days. Rin-1 has been vitro (Jiang et al., 1999). The MSCV P210 driven CML shown to bind BCR ± ABL and synergize with BCR ± model has demonstrated an elevation in IL-3 levels in ABL in in vitro transformation studies (Afar et al., CML mice (Zhang and Ren, 1998). BCR ± ABL binds 1997). The complementing activity of Rin-1 was tested to the common subunit of the IL-3 receptor (IL-3RBc) in this mouse model. Co-expression of Rin-1 with P185 and lead to its phosphorylation (Wilson-Rawls et al., decreased the latency of Pre-B leukemia and increased 1996). Mice reconstituted with constitutively active its penetrance from 25 to 60% (Afar et al., 1997). forms of IL-3RBc transduced bone marrow develop a In Ph+ human cell lines, BCR ± ABL associates with myeloproliferative disorder and leukemia (McCormack Rin-1 and leads to Rin-1 phosphorylation (Afar et al., and Gonda, 1999). 1997). 5-FU treated IL-3 de®cient murine bone marrow cells transduced with P210 retroviruses have been used to reconstitute mice. P210 transduced wild-type and BCR ± ABL domain requirement studies IL-37/7 bone marrow cells were equally e€ective at generating CML disease in reconstituted mice (Li et al., There are examples where domains of BCR ± ABL 2001). To address whether IL-3 is required in the bone shown to be critical for transformation in vitro have marrow environment, as opposed to the P210 target also been demonstrated as essential for induction of cell, wild-type or IL-37/7 mice were used as recipients. leukemia. An oligomerization mutant of BCR ± ABL P210 induced leukemia with similar latencies irrespec- (BCR ± ABLD61) lacking the ®rst 61 amino acids of tive of recipient genotype. However, an interesting BCR cannot induce CML-like disease in the MSCV observation was noted. When both donor cells and P210 transduction system (Zhang et al., 2001a). recipient were IL-37/7, the leukemic mice had an Surprisingly, this P210 mutant generates a T cell elevated level of monocytes and macrophages in the leukemia/lymphoma with a long latency of 115 days. blood. Interestingly, in rare cases of human CML These results support in vitro studies (Maru et al., associated with P185, monocytosis is frequently 1996b; McWhirter et al., 1993; Muller et al., 1991; observed. In the absence of IL-3, P210 generates an Tauchi et al., 1998) demonstrating that the oligomer-

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5653 ization domain is critical for BCR ± ABL transform- tional regulation in the human disease. Bcr P210 ation and suggest that the BCR oligomerization transgenic embryos failed to develop (Heisterkamp et domain may be a potential therapeutic target. al., 1991). However, using the Methallothienin (MT) The role of the SH3 domain in BCR ± ABL promoter, both P210 and P185 founder animals were oncogenesis is controversial. A deletion of the SH3 obtained. MT P185 transgenic lines develop B-ALL domain in P210 BCR ± ABL (P210DSH3) delays the (Heisterkamp et al., 1990; Voncken et al., 1992a, latency of leukemias developing in mice inoculated 1995a) with an expansion of Pre-B lymphoblasts in with expressing 32D cells from 6 ± 9 weeks to 17 ± 45 hematopoietic organs such as the bone marrow and weeks (Skorski et al., 1998). However, deletion of the spleen. Non-hematopoietic organs including brain, SH3 domain potentiates BCR ± ABL transformation of heart, and muscle also expressed P185 but appeared Rat1 ®broblasts to anchorage independence (Maru et una€ected. The Pre-B leukemias were oligoclonal and al., 1996b). Does the SH3 domain potentiate or inhibit transplantable. Two groups generated MT promoter BCR ± ABL induced leukemogenesis? To address this driven P210 transgenic lines with di€erent outcomes. question, P210DSH3 was compared to wild-type P210 MT P210 transgenics generated by Voncken et al. for leukemogenic potential in the 5-FU, MSCV (1995a) developed B-ALL in 30% of cases (latency retroviral transduction system. Deletion of the SH3 100 ± 200 days), 60% developed T-ALL (latency 130 ± domain did not alter the leukemogenic potential of 300 days) and 7% developed myeloblastic leukemia by P210 with respect to both latency and disease 380 days. MT P210 mice generated by Honda et al. all phenotype (Gross et al., 1999). Although only two developed T cell leukemia within 90 days (Honda et al., Ph+ patients harboring an SH3 domain deleted form of 1995). One notable di€erence between the two MT P210 have been reported, neither showed distinguishing P210 transgenic lines is the mouse strain background clinical features (Tiribelli et al., 2000). These results (see Table 1), again demonstrating the importance of suggest that the SH3 domain may not signi®cantly genetic context. in¯uence BCR ± ABL induced leukemogenesis. Limitations of the MSCV retroviral CML model Hematopoietic lineage restricted promoters driving have been revealed in studies of SH2 domain point/ BCR ± ABL expression deletion mutants of BCR ± ABL (P210DSH2). Two groups have observed di€erent disease phenotypes in To restrict BCR ± ABL expression to the hematopoietic mice reconstituted with bone marrow transduced with system the tec promoter has been employed to drive retroviruses encoding P210DSH2. One group observed BCR ± ABL expression preferentially in hematopoietic that P210DSH2 mice develop a CML-like myelo- precursors (Mano et al., 1990). Tec promoter driven proliferative disease with a longer latency compared P210 founders develop clonal T cell thymomas with to wild-type P210 mice (Zhang et al., 2001b). The 100% penetrance (Honda et al., 1998). It is interesting CML-like disease was also distinct from that induced to note that the strain of these founders is C57BL/ by P210 WT; a pre-leukemic stage with an expansion 66DBA/2F2, which is also associated with the of mature and immature B-cells occurred prior to onset preferential development of T cell leukemias in MT of a CML-like disease. Another group showed that promoter driven P210 founders (see Table 1). P210DSH2 mice develop mainly B-ALL with a long A transgenic line derived from a tec promoter driven disease latency compared to wild-type P210 P210 founder, however, developed a CML-like disease (Roumiantsev et al., 2001). There is no obvious at 1 year of age (Honda et al., 1998). These mice explanation for the di€erent phenotypes observed from develop an indolent disease with elevations in platelet both groups. count and mature granulocytes at 5 ± 8 months of age. At one year of age, these mice succumb to a clonal chronic phase CML-like disease. The tec promoter Generation of transgenic mice expressing BCR ± ABL driven P210 transgenic is the only transgenic line that Three main approaches have been taken to generate routinely succumbs to a CML-like disease. The disease transgenic mice expressing BCR ± ABL. latency in these mice is extremely long; during the ®rst year of life (pre-leukemic phase), P210 expression (1) Use of global promoters driving BCR ± ABL ex- cannot be detected at the protein level in these mice. pression. However, a low level of P210 expression is detectable (2) Hematopoietic restricted expression of BCR ± by RT ± PCR, suggesting that low levels of P210 ABL. expression may underlie the long latency of disease in (3) Regulatable expression of BCR ± ABL. this model.

Global promoter driven BCR ± ABL expression Regulatable promoter driving BCR ± ABL expression The ®rst generation of transgenic models tested global promoters to drive P210 and P185 BCR ± ABL To control the level and timing of P210 BCR ± ABL expression in mice. One obvious choice of promoter expression in an adult mouse, a tetracycline regulatable to test was the mouse bcr promoter. This would system has been employed. Crossing mouse mammary conceivably mimic aspects of BCR ± ABL transcrip- tumor virus (MMTV) promoter driven Tet-transact-

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5654 ivator transgenic mice with Tet-responsive promoter between normal mice, pre-leukemic mice, or Pre-B- driven P210 transgenic mice generated doubly trans- ALL MT P185 transgenic mice (Salloukh et al., 2000). genic mice. These compound transgenic mice were In addition, Ph+ human leukemic Pre-B-ALL cells highly responsive to Tet-regulated expression of P210 exhibit normal apoptotic responses to ionizing radia- by administration and removal of tetracyline from tion (Li et al., 1994). These observations suggest that a drinking water (Huettner et al., 2000). In response to major inhibition of apoptosis may not be instrumental P210 expression these mice develop B-ALL within 24 ± in P185 induced B-ALL. 80 days. Sustained P210 expression was required for Expression of dominant negative RAS or other the maintenance of P210 induced B-ALL, as shutting dominant negative proteins in the RAS pathway such down BCR ± ABL expression resulted in disease as c-JUN inhibit BCR ± ABL transformation of remission. Previously, a requirement for sustained ®broblasts in vitro (Pendergast et al., 1993b) and BCR ± ABL expression in the maintenance of the (Raitano et al., 1995; Sawyers et al., 1995; Skorski et transformed state had only been demonstrated in cell al., 1994). Active RAS requires farnesylation of the lines expressing temperature sensitive forms of BCR ± RAS CAAX box and localization to the plasma ABL or GAG ± ABL proteins (Cleveland et al., 1989; membrane. Inhibition of the RAS pathway using the Engelman and Rosenberg, 1990; Kabarowski et al., farnesyl transferase inhibitor SCH66336 has been 1994). This suggests that despite the association of Ph+ tested as a potential therapeutic approach in B-ALL. disease progression with the acquisition of additional Administration of SCH66366 at a dose of 40 mg/kg genetic mutations, a requirement for BCR ± ABL prevented development of leukemias in 80% of P185 tyrosine kinase is retained. This is also re¯ected in transgenic mice (Reichert et al., 2001). Disease free MT human chronic phase CML, where blocking the P185 mice appeared to maintain normal hematopoiesis tyrosine kinase activity of P210 by STI571 treatment with no signs of leukemic blasts during the 200 day leads to disease remission (Druker et al., 2001). observation period. In primary human cells, the The Tet regulatable P210 transgenic mouse model farnesyl transferase inhibitor SCH66336 selectively does not develop a CML-like disease. P210 expression inhibits hematopoietic colony formation of Ph+ cells is only detected in Pre-B cells in these animals, possibly (Peters et al., 2001). Although inhibition of RAS accounting for the sole development of B-ALL. The function is likely to have deleterious e€ects, low doses authors suggest that the MMTV promoter controlling of SCH66336 may selectively target BCR ± ABL expression of the Tet-transactivator may not be active expressing cells and suppress leukemic cell expansion. in HSCs. Use of alternative HSC speci®c promoters Further studies in animal model systems such as the tec and/or enhancers in combination with this Tet- promoter driven P210 transgenic line which develop regulated system may generate a regulatable CML exclusively CML-like disease are perhaps warranted. mouse model.

Generation of knock-in mice expressing BCR ± ABL Studies using the MT P185 transgenic line from the mouse bcr locus

In human CML, progression from chronic phase to To more accurately mimic P185 expression in human blast crisis involves a non-random selection of Ph+ B-ALL, P185 knock-in mice expressing P185 from cytogenetic alterations (reviewed by Mitelman, 1993). the endogenous murine bcr locus were generated. By 4 Karyotype analyses in preleukemic MT P185 mice months of age, 95% of chimeric mice expressing one showed no detectable chromosomal aberration (Vonck- P185 allele and one normal bcr allele developed Pre-B- en et al., 1992b). Leukemic mice, however, had a high ALL (Castellanos et al., 1997). Pre-B cells dominated frequency (88%) of aneuploidy (Voncken et al., 1992b). the majority of WBCs in the blood and in®ltrated the Trisomy of chromosomes 10, 12, 14 and 17 were most spleen and liver in leukemic mice. BCR ± ABL common in this mouse model (Voncken et al., 1992b). expression was detected in the spleen, bone marrow A comparison of chromosomal abnormalities in human and thymus, but no leukemic involvement was detected Ph+ leukemia with these using the complete human in the thymus. The 94% transplantability of this genome sequence and soon complete mouse genome disease suggests the leukemic clone(s) is of stem cell sequence may expand the list of candidate gene clusters origin. It will be of interest to compare leukemias or even validate further studies of genes already induced by other forms of BCR ± ABL using this bcr identi®ed in BCR ± ABL transformation/leukemo- knock-in strategy. genesis systems. The BCR protein is phosphorylated by BCR ± ABL BCR ± ABL expression has been shown to inhibit and is present within BCR ± ABL oligomers (Liu et al., apoptosis in growth factor dependent cell lines 1993; Lu et al., 1993; Voncken et al., 1995b). To test (Amarante-Mendes et al., 1998; Bedi et al., 1995; whether BCR is required for P185 induced Pre-B-ALL, Cambier et al., 1999; Cortez et al., 1995; Kabarowski P185 knock-in ES cells were ablated of their second bcr et al., 1994; Zhao et al., 2001). However, an anti- allele and used to generate chimeric mice. A compar- apoptotic function of BCR ± ABL in cpCML is ison of chimeric mice expressing P185 and BCR or controversial (Kabarowski and Witte, 2000). No P185 alone, did not show any di€erence in leukemic signi®cant di€erence in apoptosis was observed penetrance, latency, or phenotype. MT P185 bcr null

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5655 transgenic mice also produce leukemia with indis- U937, in which transfection of all three components tinguishable latency and clinical pattern compared to (ICSBP, IRF-1 and PU.1) was required to produce a genetically matched MT P185 transgenic counterparts highly active transcriptional complex (Eklund et al., (Voncken et al., 1998). Together, these observations 1998). Another plausible explanation is that ISCBP suggest that a functional bcr gene product may not downregulation is a consequence of P210 induced play a role in the development of P185 associated myeloid cell expansion rather than a direct e€ect of leukemias. BCR ± ABL expression instrumental in the leukemo- genic process.

Knock-outs with CML phenotype Lack of JunB expression in the myeloid lineage generates a CML-like disease in mice Interferon consensus sequence binding protein (ICSBP) JunB expression increases during myeloid di€erenti- knock-out mice develop a CML-like disease ation (Lord et al., 1993; Mollinedo et al., 1991), ICSBP is a transcription factor of the interferon suggesting it may play an important role in regulatory factor (IRF) family, members of which myelopoiesis. JunB inactivation leads to early leth- regulate interferon (IFN) stimulated gene expression ality in mice (Schorpp-Kistner et al., 1999). To rescue (Driggers et al., 1990). ICSBP is poorly active at the early lethality of JunB knock-out mice, viable regulating gene expression by itself (Eklund et al., JunB knock-out mice were generated by expressing 1998; Eklund and Kakar, 1999; Sharf et al., 1995) but JunB via the human ubiquitin promoter transgene is highly active when bound to IRF-1 or IRF-2 and (JunB7/7Ubi-junB) (Passegue et al., 2001). Surpris- PU.1 (Eklund et al., 1998; Eklund and Kakar, 1999). ingly, these mice gradually develop elevated levels of IFNa treatment leads to remission in chronic phase mature myeloid cells in the blood, spleen, bone CML patients (Gutterman, 1994). It is tempting to marrow, and lymph nodes beginning at 4 months of speculate therefore that lack of ICSBP predisposes age. Sixteen per cent of the mice progress from this animals to the development of a CML-like disease by chronic CML-like disease to a syndrome resembling altering the balance between active and inactive IRF blast crisis characterized by an expansion of complexes. immature myeloid cells. Clonal analysis of these The CML-like disease which develops in ICSBP leukemic cells will determine if JunB down-regulation knock-out mice evolves to a blast stage leukemia. The directly leads to leukemia, or confers a predisposition chronic phase disease is 100% penetrant by 10 ± 16 to the development of leukemia. Symptom-free 1 ± 3- weeks of age, and by 50 weeks of age, 33% of the mice month-old mice express JunB in all hematopoietic develop blast stage disease (Holtschke et al., 1996). The lineages analysed. As the myeloproliferative disease chronic stage disease in these mice is characterized by progresses, JunB expression is down regulated or lost an elevation of neutrophils in hematopoietic tissues. in myeloid cells. This suggests that down-regulation Development of blast crisis is associated with a huge or loss of JunB expression contributes to the myeloid expansion of myeloblasts in the bone marrow and cell expansion. To test this hypothesis, JunB de®cient spleen. B lymphoblasts are dominant in the blood and fetal liver cells were used to reconstitute lethally lymph nodes. irradiated mice. The reconstituted mice develop a ICSBP+/7 mice have a lower penetrance of CML- myeloproliferative disorder similar to that in JunB7/7 like disease compared to ICSBP7/7 mice, suggesting Ubi-junB transgenic mice (Passegue et al., 2001). that development of CML by loss of ICSBP These results demonstrate that loss of JunB expres- expression is dose-dependent. This would be expected sion is a pathogenetic factor in myeloid leukemogen- if the critical role of ICSBP is in conjunction with esis. It remains to be determined if downregulation of other transcriptional regulators (IRF-1, IRF-2, PU.1) JunB transcription or activity is an etiological factor within multi-protein complexes. The ability of ISCBP in human CML. overexpression to suppress P210 induced CML was JunB expression inhibits cyclin D1 function by examined by expressing ICSBP and P210 in the same activating P16INK4a (Bakiri et al., 2000; Passegue and MSCV driven bicistronic retroviral vector (Hao and Wagner, 2000) Cyclin D1 over-expression complements Ren, 2000). Although leukemic progression in mice BCR ± ABL induced transformation (Afar et al., 1995). reconstituted with P210 transduced bone marrow is Therefore, cyclin D1 up-regulation may be a common associated with a reduction in ICSBP expression, signaling event in myeloid leukemogenesis induced by enforced ICSBP expression did not block P210 both BCR ± ABL expression and loss of JunB activity. induced leukemia. However, the latency to leukemic Although JunB7/7Ubi-junB transgenic mice lose development was increased from 3 to 5 weeks. JunB expression in the myeloid compartment during Several possibilities can explain these apparently disease progression, JunB expression is maintained in T contradictory ®ndings. The most obvious is that cells, B cells and erythrocytes. This suggests that loss of ICSBP functions as part of a multiprotein complex JunB expression in HSCs is not required for the and overexpression of ISCBP as well as its partner(s) induction of the myeloproliferative disorder. Further, it may be necessary to inhibit P210 leukemogenesis. suggests that a genetic lesion restricted to myeloid This has been shown in the myelomonocytic cell line progenitors and their progeny can induce a CML-like

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5656 disease. As BCR ± ABL is expressed in all hematpoietic change in the rate of apoptosis. Discordant hemato- lineages (except NK cells) of chronic phase CML poietic maturation (Clarkson and Strife, 1993) is a patients, this raises the question of whether targeting feature in human cpCML which may underlie the its expression exclusively to the common myeloid frequent incidence of anemia resulting from a severe progenitor (CMP) (Akashi et al., 2000) and its progeny reduction in erythropoiesis, as well as the myeloid in animal model systems leads to the development of a expansion. The discordant maturation hypothesis CML-like disease. This will also provide information as proposes that BCR ± ABL may a€ect growth re- to the involvement of the HSC in CML beyond that of sponses by deregulating di€erentiation programs of the cellular origin/target. primitive hematopoietic cells. It remains to be seen if our observations in the ES-di€erentiation system are associated with delayed maturation within critical ES in vitro di€erentiation system hematopoietic populations.

One major obstacle in studying BCR ± ABL e€ects on the HSC is the rarity of this population in vivo. Conclusion BCR ± ABL studies in vivo are complicated by the occurrence of cell autonomous as well as indirect Various animal model systems have demonstrated that e€ects of BCR ± ABL expression, often in multiple BCR ± ABL expression is sucient to initiate the cell-types depending on the nature of transcriptional/ development of a variety of di€erent leukemias. regulatory elements employed. We have modi®ed an Targeting speci®c cell-types has been shown to be ES in vitro di€erentiation system by adding tetracy- critical in determining the type of leukemia formed. cline-regulation to study the direct impact of BCR ± The use of global promoters such as MT and murine ABL on hematopoietic development (Era and Witte, bcr to drive P185 expression have generated leukemia 2000). models mimicking many features of human P185 B- We adapted an ES cell di€erentiation system that ALL. Enriching for stem and progenitor cell popula- speci®cally favors hematopoietic development (Nakano tions expressing P210 by 5-FU bone marrow treatment et al., 1994). ES cells are induced to di€erentiate on an or Tec promoter driven P210 transgenic expression M-CSF7/7 stromal cell line (OP9) which preferentially have generated CML models. These models have supports di€erentiation of myeloid and erythroid cells utilized knock-out cells to genetically determine the without an overproduction of macrophages. Up to 5 requirement of pathways implicated in BCR ± ABL days of di€erentiation, ES cells develop into hemangio- transformation. Thus far, many signaling pathways blasts capable of forming hematopoietic and endothe- thought to be critical in BCR ± ABL oncogenesis (such lial cells. At this stage, early hematopoietic cells begin as BCR, STAT5, IL-3, and GM ± CSF) have been to form and by day 7 ± 8 many hematopoietic shown to be dispensable for BCR ± ABL induced progenitors are present. From day 8 ± 14, di€erentia- leukemogenesis. ICSBP and JunB knock-out mice tion proceeds generating mature myeloid and erythroid develop a CML-like disease. Are there common cells. signaling pathways operating in these and P210 We investigated the e€ect of BCR ± ABL expression leukemic mice? on immature and mature hematopoietic cell popula- In human Ph+ leukemias, CML is overwhelmingly tions in this system by controlling the dose and associated with P210 BCR ± ABL, while P185 is timing of P210 expression via a Tet-regulatable typically associated with B-ALL. However, BCR ± promoter. P210 expression led to an expansion of ABL mouse models have demonstrated that di€erent early progenitors and mature myeloid cells with a forms of BCR ± ABL can generate a similar spectrum concomitant reduction in the erythroid compartment of leukemias. Despite this, these mouse models do (Era and Witte, 2000). An acute 48 h exposure to show that P185 generates leukemia in mice at a faster P210 BCR ± ABL was sucient for the ampli®cation rate compared to P210, consistent with the chronic of progenitors, demonstrating that BCR ± ABL ex- indolent nature of CML. The Tet-regulatable P210 pression is necessary and sucient to expand mouse model and ES in vitro di€erentiation systems hematopoietic progenitors. The cell expansion in demonstrate that BCR ± ABL expression is necessary response to BCR ± ABL expression was directly and sucient to deregulate hematopoietic develop- proportional to the dose of BCR ± ABL, with ment. Combining these systems will allow the maximal P210 expression leading to a threefold identi®cation of critical hematopoietic populations increase in cell number over 2 days. The expansion directly a€ected by BCR ± ABL expression in vivo,as of myeloid cells is reversible, demonstrating that well as the e€ect, if any, of BCR ± ABL expression in sustained BCR ± ABL expression is required for its the HSC. It is important to consider that CML maintenance. Possible explanations for this BCR ± originates in a single HSC. Although a single HSC ABL induced cell expansion include an increase in can repopulate the hematopoietic system of a mouse cell cycling, resistance to apoptosis, or a delay in (Osawa et al., 1996), no one has yet reported the di€erentiation. An increase in cell proliferation induction of a CML-like disease in mice inoculated measured by BrdU uptake was observed during with one or a few pure BCR ± ABL transduced or BCR ± ABL induction, but there was no dramatic BCR ± ABL transgenic HSCs.

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5657 Acknowledgments from this laboratory described in the manuscript was We thank Dr Janusz HS Kabarowski and Lu Quang Le for supported by funds from the National Cancer Institute and helpful discussions and critical reading of this manuscript the Howard Hughes Medical Institute. and JC White for the preparation of this manuscript. Work

References

Abelson HT and Rabstein LS. (1970). Cancer Res., 30, Deisseroth AB, Zu Z, Claxton D, Hanania EG, Fu S, 2213 ± 2222. Ellerson D, Goldberg L, Thomas M, Janicek K and Afar DEH, Goga A, McLaughlin J, Witte O and Sawyers Anderson WF. (1994). Blood, 83, 3068 ± 3076. CL. (1994). Science, 264, 424 ± 426. Delforge M, Boogaerts MA, McGlave PB and Verfaillie CM. Afar DEH, Han L, McLaughlin J, Wong S, Dhaka A, (1999). Blood, 93, 284 ± 292. Parmar K, Rosenberg N, Witte ON and Colicelli J. (1997). Diekmann L, Beelen DW, Quabeck K, Becher R, Schulte Immunity, 6, 773 ± 782. Holthausen H, Butzler R, Schaefer UW and Opalka B. Afar DEH, McLaughlin J, Sherr CJ, Witte ON and Roussel (1994). Acta Haematol., 92, 169 ± 175. MF. (1995). Proc. Natl. Acad. Sci. USA, 92, 9540 ± 9544. Driggers PH, Ennist DL, Gleason SL, Mak WH, Marks MS, Akashi K, Traver D, Miyamoto T and Weissman IL. (2000). Levi BZ, Flanagan JR, Appella E and Ozato K. (1990). Nature, 404, 193 ± 197. Proc. Natl. Acad. Sci. USA, 87, 3743 ± 37477. Amarante-Mendes GP, McGahon AJ, Nishioka WK, Afar Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, DE, Witte ON and Green DR. (1998). Oncogene, 16, Ford JM, Lydon NB, Kantarjian H, Capdeville R, Ohno- 1383 ± 1390. Jones S and Sawyers CL. (2001). N.Engl.J.Med.,344, Anderson SM and Mladenovic J. (1996). Blood, 87, 238 ± 1084 ± 1086. 244. Eklund EA, Jalava A and Kakar R. (1998). J. Biol. Chem., Bakiri L, Lallemand D, Bossy-Wetzel E and Yaniv M. 273, 13957 ± 13965. (2000). EMBO J., 19, 2056 ± 2068. Eklund EA and Kakar R. (1999). J. Immunol., 163, 6095 ± Bedi A, Barber JP, Bedi GC, el-Deiry WS, Sidransky D, Vala 6105. MS, Akhtar AJ, Hilton J and Jones RJ. (1995). Blood, 86, Elefanty AG and Cory S. (1992). Mol. Cell. Biol., 12, 1755 ± 1148 ± 1158. 1763. Ben-Neriah Y, Daley GQ, Mes-Masson A-M, Witte ON and Elefanty AG, Hariharan IK and Cory S. (1990). EMBO J., 9, Baltimore D. (1986). Science, 233, 212 ± 214. 1069 ± 1078. Buchdunger E, Zimmermann J, Mett H, Meyer T, Muller M, Engelman A and Rosenberg N. (1987). Proc. Natl. Acad. Sci. Regenass U and Lydon NB. (1995). Proc. Natl. Acad. Sci. USA, 84, 8021 ± 8025. USA, 92, 2558 ± 2562. Engelman A and Rosenberg N. (1990). J. Virol., 64, 4242 ± Cambier N, Zhang Y, Vairo G, Kosmopoulos K, Metcalf D, 4251. Nicola NA and Elefanty AG. (1999). Oncogene, 18, 343 ± Era T and Witte ON. (2000). Proc. Natl. Acad. Sci. USA, 97, 352. 1737 ± 1742. Castellanos A, Pintado B, Weruaga E, Are valo R, Lo pez A, Fabbiano F, Santoro A, Felice R, Catania P, Cannella S and Orfao A and Sa nchez-Garcõ a I. (1997). Blood, 90, 2168 ± Majolino I. (1998). Haematologica, 83, 856 ± 857. 2174. Fainstein E, Marcelle C, Rosner A, Canaani E, Gale RP, Chan LC, Karhi KK, Rayter SI, Heisterkamp N, Eridani S, Dreazen O, Smith SD and Croce CM. (1987). Nature, 330, Powles R, Lawler SD, Gro€en J, Foulkes JG, Greaves MF 386 ± 388. and Wiedemann LM. (1987). Nature, 325, 635 ± 637. Fialkow PJ, Gartler SM and Yoshida A. (1967). Proc. Natl. Chung S-W, Chen H and Wong PMC. (1996). Oncogene, 13, Acad.Sci.USA,58, 1468 ± 1471. 2397 ± 2405. Fialkow PJ, Jacobson RJ and Papayannopoulou T. (1977). Chung S-W, Wong PMC, Durkin H, Wu YS and Petersen J. Am. J. Med., 63, 125 ± 130. (1991). Proc. Natl. Acad. Sci. USA, 88, 1585 ± 1589. Gha€ari S, Daley GQ and Lodish HF. (1999). Leukemia, 13, Clark SS, Chen E, Fizzotti M, Witte ON and Malkovska V. 1200 ± 1206. (1993). J. Virol., 67, 6033 ± 6046. Gishizky ML, Johnson-White J and Witte ON. (1993). Proc. Clark SS, McLaughlin J, Crist WM, Champlin R and Witte Natl. Acad. Sci. USA, 90, 3755 ± 3759. ON. (1987). Science, 235, 85 ± 88. Goga A, McLaughlin J, Afar DEH, Sa€ran DC and Witte Clarkson B and Strife A. (1993). Leukemia, 7, 1683 ± 1721. ON. (1995). Cell, 82, 981 ± 988. ClarksonBD,StrifeA,WisniewskiD,LambekCand Golub TR, Goga A, Barker GF, Afar DE, McLaughlin J, Carpino N. (1997). Leukemia, 11, 1404 ± 1428. Bohlander SK, Rowley JD, Witte ON and Gilliland DG. Cleveland JL, Dean M, Rosenberg N, Wang JYJ and Rapp (1996). Mol. Cell. Biol., 16, 4107 ± 4116. UR. (1989). Mol. Cell. Biol., 9, 5685 ± 5695. Gross AW, Zhang X and Ren R. (1999). Mol. Cell. Biochem., Colleoni GW, Yamamoto M, Kerbauy J, Sera®m RC, 19, 6918 ± 6928. Lindsey CJ, Costa FF, Grignolli CR and Saad ST. Gutterman JU. (1994). Proc. Natl. Acad. Sci. USA, 91, (1996). Am.J.Hematol.,53, 277 ± 278. 1198 ± 1205. Cortez D, Kadlec L and Pendergast AM. (1995). Mol. Cell. Haferlach T, Winkemann M, Ramm-Petersen L, Meeder M, Biol., 15, 5531 ± 5541. Schoch R, Weber-Matthiesen K, Schlegelberger B, Schoch Daley GQ, Van Etten RA and Baltimore D. (1990). Science, C, Ludwig WD, Hiddemann W and Lo‚er H. (1997). Br. 247, 824 ± 830. J. Haematol., 99, 452 ± 459. Deininger MW, Goldman JM and Melo JV. (2000). Blood, Hao SX and Ren R. (2000). Mol. Cell. Biol., 20, 1149 ± 1161. 96, 3343 ± 3356.

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5658 Hariharan IK, Adams JM and Cory S. (1988). Onc. Res., 3, Lugo TG, Pendergast A, Muller AJ and Witte ON. (1990). 387 ± 399. Science, 247, 1079 ± 1082. HariharanIK,HarrisAW,CrawfordM,AbudH,WebbE, Mano H, Ishikawa F, Nishida J, Hirai H and Takaku F. Cory S and Adams JM. (1989). Mol. Cell. Biol., 9, 2798 ± (1990). Oncogene, 5, 1781 ± 1786. 2805. Martiat P, Mecucci C, Nizet Y, Stul M, Phillippe M, Hawley RG, Lieu FHL, Fong ZC and Hawley TS. (1994). Cassiman JJ, Michaux JL, Van den Berghe H and Sokal Gene Ther., 1, 136 ± 138. G. (1990). Leukemia, 4, 751 ± 754. Heisterkamp N, Jenster G, Kioussis D, Pattengale PK and Maru Y, Afar DE, Witte ON and Shibuya M. (1996a). J. Gro€en J. (1991). Transgenic Res., 1, 45 ± 53. Biol. Chem., 271, 15353 ± 15357. Heisterkamp N, Jenster G, ten Hoeve J, Zovich D, Maru Y, Witte ON and Shibuya M. (1996b). FEBS Lett., Pattengale PK and Gro€en J. (1990). Nature, 344, 251 ± 379, 244 ± 246. 253. McCormack MP and Gonda TJ. (1999). Oncogene, 18, Heisterkamp N, Stam K, Gro€en J, de Klein A and Grosveld 7109 ± 7199. G. (1985). Nature, 315, 758 ± 761. McWhirter JR, Galasso DL and Wang JYJ. (1993). Mol. Holtschke T, Lohler J, Kanno Y, Fehr T, Giese N, Cell. Biol., 13, 7587 ± 7595. Rosenbauer F, Lou J, Knobeloch KP, Gabriele L, Waring McWhirter JR and Wang JYJ. (1991). Mol. Cell. Biol., 11, JF, Bachmann MF, Zinkernagel RM, Morse III HC, 1553 ± 1565. Ozato K and Horak I. (1996). Cell, 87, 307 ± 317. McWhirter JR and Wang JYJ. (1993). EMBO J., 12, 1533 ± Holyoake T, Jiang X, Eaves C and Eaves A. (1999). Blood, 1546. 94, 2056 ± 2064. Melo JV. (1996). Blood, 88, 2375 ± 2384. Honda H, Fujii T, Takatoku M, Mano H, Witte ON, Yazaki MeloJV,GordonDE,TuszynskiA,DhutS,YoungBDand Y and Hirai H. (1995). Blood, 85, 2853 ± 2861. Goldman JM. (1993). Blood, 81, 2488 ± 2491. Honda H, Oda H, Suzuki T, Takahashi T, Witte ON, Ozawa Melo JV, Hochhaus A, Yan XH and Goldman JM. (1996). K, Ishikawa T, Yazaki Y and Hirai H. (1998). Blood, 91, Br. J. Haematol., 92, 684 ± 686. 2067 ± 2075. Melo JV, Myint H, Galton DA and Goldman JM. (1994). Huettner CS, Zhang P, Van Etten RA and Tenen DG. Leukemia, 8, 208 ± 211. (2000). Nat. Genet., 24, 57 ± 60. Mitani K, Sato Y, Tojo A, Ishikawa F, Kobayashi Y, Miura Hupp TR, Lane DP and Ball KL. (2000). Biochem. J., 352, Y, Miyazono K, Urabe A and Takaku F. (1990). Br. J. 1±17. Haematol., 76, 221 ± 225. Jiang X, Lopez A, Holyoake T, Eaves A and Eaves C. (1999). Mitelman F. (1993). Leuk. Lymphoma, 11, 11 ± 15. Proc. Natl. Acad. Sci. USA, 96, 12804 ± 12809. Mollinedo F, Vaquerizo MJ and Naranjo JR. (1991). Kabarowski JHS, Allen PB and Wiedemann L. (1994). Biochem. J., 273, 477 ± 479. EMBO J., 13, 5887 ± 5895. Muller AJ, Young JC, Pendergast A-M, Pondel M, Landau Kabarowski JHS and Witte ON. (2000). Stem Cells, 18, NR, Littman DR and Witte ON. (1991). Mol. Cell. Biol., 399 ± 408. 11, 1785 ± 1792. Kantarjian HM, Talpaz M, Dhingra K, Estey E, Keating Nakano T, Kodama H and Honjo T. (1994). Science, 265, MJ, Ku S, Trujillo J, Huh Y, Stass S and Kurzrock R. 1098 ± 1101. (1991). Blood, 78, 2411 ± 2418. Nieborowska-Skorska M, Wasik MA, Slupianek A, Salo- Kasprzyk A, Harrison CJ and Secker-Walker LM. (1999). moni P, Kitamura T, Calabretta B and Skorski T. (1999). Leukemia, 13, 2000 ± 2006. J. Exp. Med., 189, 1229 ± 1242. Kelliher M, Knott A, McLaughlin J, Witte ON and Nowell PC and Hungerford DA. (1960). Science, 132, 1497. Rosenberg N. (1991). Mol. Cell. Biol., 11, 4710 ± 4716. Osawa M, Hanada K-i, Hamada H and Nakauchi H. (1996). Kelliher MA, McLaughlin J, Witte ON and Rosenberg N. Science, 273, 242 ± 245. (1990). Proc. Natl. Acad. Sci. USA, 87, 6649 ± 6653. Pajor L, Vass JA, Kereskai L, Kajtar P, Szomor A, Egyed M, Kipreos ET and Wang JYJ. (1992). Science, 256, 382 ± 385. Ivanyi J and Jakso P. (2000). Leukemia, 14, 1122 ± 1126. Konopka JB, Watanabe SM and Witte ON. (1984). Cell, 37, Pane F, Frigeri F, Sindona M, Luciano L, Ferrara F, Cimino 1035 ± 1042. R, Meloni G, Saglio G, Salvatore F and Rotoli B. (1996). Kurzrock R, Shtalrid M, Talpaz M, Kloetzer WS and Blood, 88, 2410 ± 2414. Gutterman JU. (1987). Blood, 70, 1584 ± 1588. Papadopoulos P, Ridge SA, Boucher CA and Stocking C. Lazaridou A, Chase A, Melo J, Garicochea B, Diamond J (1995). Cancer Res., 55, 34 ± 38. and Goldman J. (1994). Leukemia, 8, 454 ± 457. Passegue E, Jochum W, Schorpp-Kistner M, Mohle- Li S, Gillessen S, Tomasson MH, Drano€ G, Gilliland DG Steinlein U and Wagner EF. (2001). Cell, 104, 21 ± 32. and Etten RAV. (2001). Blood, 97, 1442 ± 1450. Passegue E and Wagner EF. (2000). EMBO J., 19, 2969 ± Li S, Ilaria RL, Million RP, Daley GQ and Van Etten RA. 2979. (1999). J. Exp. Med., 189, 1399 ± 1412. Pear WS, Miller JP, Xu L, Pui JC, So€er B, Quackenbush Li X, Gong J, Feldman E, Seiter K, Traganos F and RC, Pendergast AM, Bronson R, Aster JC, Scott ML and Darzynkiewicz Z. (1994). Leuk. Lymph., 13, 65 ± 70. Baltimore D. (1998). Blood, 92, 3780 ± 3792. Lim Y-M, Wong S, Lau G, Witte ON and Colicelli J. (2000). Pendergast AM, Gishizky ML, Havlik MH and Witte ON. Proc. Natl. Acad. Sci. USA, 97, 12233 ± 12238. (1993a). Mol. Cell. Biol., 13, 1728 ± 1736. LiuJ,CampbellM,GuoJQ,LuD,XianYM,AnderssonBS Pendergast AM, Quilliam LA, Cripe LD, Bassing CH, Dai Z, and Arlinghaus RB. (1993). Oncogene, 8, 101 ± 109. Li N, Batzer A, Rabun KM, Der CJ, Schlessinger J and Lord KA, Abdollahi A, Ho€man-Liebermann B and Gishizky ML. (1993b). Cell, 75, 175 ± 185. Liebermann DA. (1993). Mol. Cell. Biol., 13, 841 ± 851. Peters DG, Hoover RR, Gerlach MJ, Koh EY, Zhang H, Lu D, Liu J, Campbell M, Guo JQ, Heisterkamp N, Gro€en Choe K, Kirschmeier P, Bishop WR and Daley GQ. J, Canaani E and Arlinghaus R. (1993). Blood, 82, 1257 ± (2001). Blood, 97, 1404 ± 1412. 1263.

Oncogene Models of Ph+ leukemias S Wong and ON Witte 5659 Ponticelli AS, Whitlock CA, Rosenberg N and Witte ON. Takahashi N, Miura I, Saitoh K and Miura AB. (1998). (1982). Cell, 29, 953 ± 960. Blood, 92, 4758 ± 4763. Prokocimer M and Rotter V. (1994). Blood, 84, 2391 ± 2411. Tauchi T, Okabe S, Miyazawa K and Ohyashiki K. (1998). Raitano AB, Halpern JR, Hambuch TM and Sawyers CL. Int. J. Oncol., 12, 1269 ± 1276. (1995). Proc. Natl. Acad. Sci. USA, 92, 11746 ± 11750. Thome KC, Radfar A and Rosenberg N. (1997). J. Virol., 71, Raskind WH and Fialkow PJ. (1987). Adv. Cancer Res., 49, 8149 ± 8156. 127 ± 167. Tiribelli M, Tonso A, Ferrro D, Parzilae A, Cambrin GR, Reichert A, Heisterkamp N, Daley GQ and Gro€en J. Scaravaglio P, Cilloni D, Gottardi E and Saglio G. (2000). (2001). Blood, 97, 1399 ± 1403. Blood, 95, 4019 ± 4020. Rosenberg N and Witte ON. (1988). Adv. Virus Res., 39, 39 ± Unnikrishnan I, Radfar A, Jenab-Wolcott J and Rosenberg 81. N. (1999). Mol. Cell. Biol., 19, 4825 ± 4831. Roumiantsev S, De Aos IEA, Varticovski L, Ilaria RL and VanEttenRA,JacksonPK,BaltimoreD,SandersMC, Etten RAV. (2001). Blood, 97, 4 ± 13. Matsudaira PT and Janmey PA. (1994). J. Cell. Biol., 124, Rovira A, Urbano-Ispizua A, Cervantes F, Rozman M, 325 ± 340. Vives-Corrons JL, Montserrat E and Rozman C. (1995). Voncken JW, Griths S, Greaves MF, Pattengale PK, Ann. Hem., 70, 129 ± 133. Heisterkamp N and Gro€en J. (1992a). Cancer Res., 52, Rowley JD. (1973). Nature, 243, 290 ± 293. 4534 ± 4539. Saglio G, Guerrasio A, Rosso C, Zaccaria A, Tassinari A, VonckenJW,KaartinenV,Gro€enJandHeisterkampN. Serra A, Rege-Cambrin G, Mazza U and Gavosto F. (1998). Oncogene, 16, 2029 ± 2032. (1990). Blood, 76, 1819 ± 1824. Voncken JW, Karrtinen V, Pattengale PK, Germeraad Salloukh HF, Vowles I, Heisterkamp N, Gro€en J and WTV, Gro€en J and Heisterkamp N. (1995a). Blood, 86, Laneuville P. (2000). Oncogene, 19, 4362 ± 4374. 4603 ± 4611. Sattler M and Salgia R. (1997). Cytokine Growth Factor Rev., Voncken JW, Morris C, Pattengale P, Dennert G, Kikly C, 8, 63 ± 79. Gro€en J and Heisterkamp N. (1992b). Blood, 79, 1029 ± Sawyers CL, McLaughlin J and Witte ON. (1995). J. Exp. 1036. Med., 181, 307 ± 313. Voncken JW, Van Schaick H, Kaartinen V, Deemer K, Schorpp-Kistner M, Wang ZQ, Angel P and Wagner EF. Coates T, Landing B, Pattengale P, Dorseuil O, Bokoch (1999). EMBO J., 18, 934 ± 948. GM, Gro€en J and Heisterkamp N. (1995b). Cell, 80, Scott ML, Van Etten RA, Daley GQ and Baltimore D. 719 ± 728. (1991). Proc. Natl. Acad. Sci. USA, 88, 6506 ± 6510. Walker LC, Ganesan TS, Dhut S, Gibbsons B, Lister TA, Secker-Walker LM, Cooke HM, Browett PJ, Shippey CA, Rothbard J and Young BD. (1987). Nature, 329, 851 ± 853. Norton JD, Coustan-Smith E and Ho€brand AV. (1988). Wilson G, Frost L, Goodeve A, Vandenberghe E, Peake I Blood, 72, 784 ± 791. and Reilly J. (1997). Blood, 88, 2410 ± 2414. Sexl V, Piekorz R, Moriggl R, Rohrer J, Brown MP, Bunting Wilson-Rawls J, Xie S, Liu J, Laneuville P and Arlinghaus KD,RothammerK,RousselMFandIhleJN.(2000). RB. (1996). Cancer Res., 56, 3426 ± 3430. Blood, 96, 2277 ± 2283. Witte ON, Go€ S, Rosenberg N and Baltimore D. (1980). Sharf R, Azriel A, Lejbkowicz F, Winograd SS, Ehrlich R Proc. Natl. Acad. Sci. USA, 8, 4993 ± 4997. and Levi BZ. (1995). J. Biol. Chem., 270, 13063 ± 13069. Witte ON, Ponticelli A, Gi€ord A, Baltimore D, Rosenberg Shtivelman E, Lifshitz B, Gale RP, Roe BA and Canaani E. N and Elder J. (1981). J. Virol., 39, 870 ± 878. (1985). Nature, 315, 550 ± 554. Yo€e G, Chinault AC, Talpaz M, Blick MB, Kantarjian Skorski T, Kanakaraj P, Ku D-H, Nieborowska-Skorska M, HM, Taylor K and Spitzer G. (1987). Exp. Hematol., 15, Canaani E, Zon G, Perussia B and Calabretta B. (1994). J. 725 ± 728. Exp. Med., 179, 1855 ± 1865. Zhang X and Ren R. (1998). Blood, 92, 3829 ± 3840. Skorski T, Nieborowska-Skorska M, Wlodarski P, Perrotti Zhang X, Subrahmanyam R, Wong R, Gross AW and Ren D, Martinez R, Wasik MA and Calabretta B. (1996). Proc. R. (2001a). Mol. Cell. Biol., 21, 840 ± 853. Natl. Acad. Sci. USA, 93, 13137 ± 13142. Zhang X, Wong R, Hao SX, Pear WS and Ren R. (2001b). Skorski T, Nieborowska-Skorska M, Wlodarski P, Wasik M, Blood, 97, 277 ± 287. TrottaR,KanakarajP,SalomoniP,AntonyakM, Zhao RC, Jiang Y and Verfaillie CM. (2001). Blood, 97, Martinez R, Majewski M, Wong A, Perussia B and 2406 ± 2412. Calabretta B. (1998). Blood, 91, 406 ± 418. Zou X, Cong F, Coutts M, Cattoretti G, Go€ SP and Calame Soussi T. (2000). Ann. NY Acad. Sci., 910, 121 ± 139. K. (2000). Mol. Cell. Biol., 20, 628 ± 633.

Oncogene