The Effect of Chromosomal Translocations in Acute Leukemias: the LMO2 Paradigm in Transcription and Development I

[CANCER RESEARCH (SUPPL.) 59, 1794s-1798s, April 1, 1999] The Effect of Chromosomal Translocations in Acute Leukemias: The LMO2 Paradigm in Transcription and Development I

Terence H. Rabbitts, 2 Katharina Bucher, Grace Chung, Gerald Grutz, Alan Warren, and Yoshi Yamada

Medical Research Council Laboratory of Molecular Biology, Division of Protein and Nucleic Acid Chemistry, Hills Road, Cambridge, CB2 2QH United Kingdom

Abstract proto-oncogene from a Burkitt's lymphoma translocation breakpoint t(8;14)(q24;q32) revealed that, while the translocated gene was intact, Two general features have emerged about genes that are activated after there were many mutations in both the noncoding first exon and chromosomal translocations in acute forms of cancer. The protein prod- within the coding region itself (4-11). In T-cell acute leukemias, there ucts of these genes are transcription regulators and are involved in are several different chromosomal translocations found in individual developmental processes, and it seems that the subversion of these normal functions accounts for their role in tumorigenesis. The features of the tumors, and the analysis of these has led to the discovery of many LMO family of genes, which encode LIM-domain proteins involved in different, novel genes that can contribute to T-cell tumorigenesis (1). T-cell acute leukemia through chromosomal translocations, typify these These and many other studies have led to two conclusions: abnormal functions in tumorigenesis. For example, the LMO2 protein is (a) many of the chromosomal translocation-activated oncogenes are involved in the formation of multimeric DNA-binding complexes, which transcription regulators in their normal sites of expression, and it is may vary in composition at different stages of hematopoiesis and function this property that is instrumental in their involvement in tumor etiol- to control differentiation of specific lineages. In T cells, enforced expres- ogy after the chromosomal translocation. This was suggested previ- sion of Lmo2 causes aberrant protein complex formation that primarily ously from the early studies on hematopoietic tumor translocations seems to hinder the T-ceU differentiation program. These observations and later, in addition, from the very large range of changes found in underscore the conclusion that protein-protein interaction (in this case, hematopoietic and mesenchymal tumors (reviewed in Ref. 1); and through the LIM domain) is a key determinant in tumorigenesis. Further- (b) the biological role of many chromosomal translocation-acti- more, the study of chromosomal translocations as naturally occurring mutations has been informative about mechanisms in hematopoiesis as vated genes is normally in developmental processes leading to the well as in tumor etiology. notion that subversion of development may be their crucial biological role in tumorigenesis, perhaps explaining why so few of the chromo- Introduction somal translocation-activated genes were previously identified as oncogenes from other experimental approaches. The LMO 1 and LMO2 The cytogenetic analysis of tumors, particularly those of hemato- LIM-only proteins are involved in T-cell leukemia because of their poietic origin, have revealed that reciprocal chromosomal transloca- ability to interact with other transcription factors. This enhances the tions are recurring features of these tumors (reviewed in Ref. (1). conclusion that protein-protein interaction is a common component of Furthermore, it has become clear that particular chromosomal trans- oncogene function. locations are consistently found in specific tumor subtypes. Through the cloning of the chromosomal breakpoints and identification of Results and Discussion oncogenes at many different breakpoints, followed by transgenic (2) and homologous recombination knock-in analysis (3), it has become The LMO Family Genes. The LMO family of genes was uncov- clear that these abnormal tumor-associated chromosomes are impor- ered (Fig. 1) by the association of LM01 (previously called RBTN1 or tant in the etiology of tumors. The scientific challenge of the last TTG1) with the chromosomal translocation t(11;14)(p15;ql 1). The decade has been to define the contribution of the genes activated by transcription unit was first observed in a T-cell line (12), and the translocations to the course of tumor development and to ascertain mRNA sequence was obtained from its cDNA sequence (13, 14) and whether any general principles can be discerned about these "trans- shown to encode a protein essentially consisting of two zinc-binding location" genes. LIM domains (15). Using an LMO1 probe, the two related genes There are a variety of chromosomal translocations in leukemias and LM02 and LM03 were isolated (previously called RBTN2 or TTG2 in solid tumors of mesenchymal origin (sarcomas), and two main and RBTN3, respectively; Refs. (16, and 17), of which LM02 was outcomes are apparent (1). One of these is confined to lymphoid found located at the junction of the chromosomal translocation t(11; tumors in which the process of antigen receptor rearrangement (im- 14)(p13;ql 1) in T-ALL 3 (16, 18). The LMO-associated chromosomal munoglobulin and T-cell receptor) occurs and which occasionally translocations seem to have occurred by an error of the usual RAG- aberrantly mediates chromosomal translocation. This type of chromo- mediated variable diversity joining recombinase process inasmuch as somal translocation causes oncogene activation resulting from the sequence analysis of the breakpoints on chromosome 1 lp13 detected new chromosomal environment of the rearranged gene, and, in gen- recombinase signal sequences at the junctions and because the joins eral, this means inappropriate gene expression. In this first category of on chromosome 14 in the TCR~ locus or on chromosome 7 in the chromosomal translocation, the B- and T-cell tumors exemplify many TCR~ locus occur precisely at the end of D-segments (19). consistently occurring features. For instance, the cloning of the CMYC The unique feature of the LMO-derived protein sequences is that they are small proteins comprising two tandem LIM domains. These

Received 11/11/98; accepted 2/4/99. zinc-binding finger-like structures have structural similarities to the 1 Presented at the "General Motors Cancer Research Foundation Twentieth Annual DNA-binding GATA fingers (20, 21) but as yet no case of a specific Scientific Conference: Developmental Biology and Cancer," June 9-10, 1998, Bethesda, LIM-DNA interaction has been reported; rather the function of this MD. This work was supported by the Medical Research Council and by Grants from the Leukaemia Research Fund (United Kingdom) and the National Foundation for Cancer domain seems to be restricted to protein-protein interaction (see Research (United States). K.B. is supported by the Roche Foundation; G.G. was "LMO2 Functions by Protein Interaction"). supported by an EMBO Fellowship, and Y. Y. by Kyoto University. 2 To whom requests for reprints should be addressed, at MRC Laboratory of Molecular Biology, Division of Protein and Nucleic Acid Chemistry, Hills Road, Cambridge, CB2 3 The abbreviations used are: T-ALL, T-cell acute lymphocytic leukemia; bHLH, 2QH United Kingdom. basic-helix-loop-helix; DN, double negative; Tall, Tall/Scl. 1794s

LMO2 Is a Regulator of Mouse Hematopoiesis. To gain insights into the function of the LMO genes in tumorigenesis, an integrated approach has been adopted to attempt to understand the normal Pluri-potent cell Myeloid function at the biological level as well as the function of LMO proteins and to use this information to explore the function in tumors. Mes od~erm-'~~ lineage As a first step, gene targeting was used to introduce null mutations in Lymphoid mouse Lmo2 gene, which showed that Lmo2 is necessary for yolk sac Self-renewa/j lineage erythropoiesis in mouse embryogenesis (22). Furthermore, the use of embryonic stem cells with null mutations of both alleles of Lmo2 in Lmo2 null defect chimeric mice has shown that adult hematopoiesis, including lympho- poiesis and myelopoiesis, fails completely in the absence of Lmo2 Fig. 2. Lmo2 is necessary for adult hematopoiesis in mice. The tissue contribution of embryonic stem cells with homozygous null mutation of Lmo2 showed that hematopoiesis (23). These data show that Lmo2 must function at early stages of in adult mice is dependent on this gene (23). Thus, the gene product is required for all of hematopoiesis (Fig. 2) either at the level of the pluripotent stem cell the stages of adult hematopoiesis, functioning at least before the bifurcation of myeloid and lymphoid lineages. The function may be restricted to the bone marrow stem cell or even perhaps before this when ventral mesoderm gives rise to these (either at the self-renewal stage or at the proliferative stages that produce committed precursor cells. Remarkably similar results have been shown for progenitors) or in the ventral mesoderm bone marrow precursor cells. another T-cell oncogene, Tall/Scl (herein called Tall; Refs. (24-27). The notion of genes that function in developmental processes is clearly delineated by the Lmo2 gene. LDB1 interact in vivo in T-ALL and in neuroblastoma cell lines (41). LMO2 Functions by Protein Interaction: The Role of the LIM This array of interactions led to the observation that Lmo2 is found in Domain. The concept that oncogenes can interfere with developmen- an oligomeric complex in erythroid cells that involves--in addition to tal processes because they are themselves normally performing this Lmo2--Tall, E47, Ldbl and Gata-I (40). This complex is able to function has been suggested to explain their role in tumor etiology. bind DNA at least in vitro and in reporter assays, with the Lmo2 and The finding that genes that encode proteins such as Lmo2 control Ldbl components seeming to bridge a bipartite DNA-binding com- development of hematopoietic lineages led to clear tests of molecular plex (Fig. 3A). These findings suggest that Lmo2 is part of a tran- function in both normal and tumor contexts. A particularly relevant scription complex in hematopoiesis. The involvement of Lmo2, Tall, observation was that the T-cell oncogenes Lmo2 and Tall were and Gatal in a common DNA-binding complex suggests that this coexpressed in erythrocytes (22), and this suggested a functional complex regulates downstream target genes, perhaps explaining why synergy between the two proteins. This turned out to be even more the null mutation of these genes leads to the lack of primitive eryth- germane because it was shown that the Lmo2 and Tal 1 proteins--the ropoiesis (22, 25, 42, 43). Conversely, there are differences in some latter is a bHLH protein that is activated in T-cell leukemias by specific aspects of hematopoiesis related to the individual null muta- chromosomal translocations or promoter deletions (28-33) and which tions, which suggests that each of them can act in separate complexes can also interact with the bHLH E47 protein in a DNA-binding at different stages of hematopoiesis. complex (34, 35)--could interact directly with each other (36, 37), The distinct roles of Lmo2 in primitive and definitive hematopoi- and this was through the LIM domains. This observation and those esis, uncovered by the analysis of null mutations of Lmo2 (22, 23), made with isolated LIM fragments (38) establish a role for the LIM suggested that the molecular complexes in which Lmo2 is involved domain in protein-protein interaction. The LIM domain does not seem may also differ at different stages (40). In erythroid cells, we observed to directly bind DNA despite the similarity to the GATA DNA- a complex of Lmo2 with GATA-1, E47, Tall and Ldbl (40) that binding zinc finger. could bind to a unique bipartite E box-GATA motif (Fig. 4B). This The ability of the LIM protein interaction domain to bind various suggested that earlier stages of hematopoiesis may have different proteins was also shown in the ability of Lmo2 to bind to Gata-1 (39) complexes; for instance, GATA-1 is absent from early hematopoietic and to bind to the Ldbl/NLI protein (40). In addition, LMO1 and progenitors; therefore, it may be replaced by GATA-2 (39). Indeed, there may also be a variation in the bHLH pair present, although phenotypes of Tal 1/Scl null mutations (26, 27) parallel those of Lmo2 (22, 23). In our model, variations in the composition of Lmo2- LMO family containing complexes may influence lineage differentiation (40, 23), Human presumably by controlling distinct sets of target genes (Fig. 4). Gene Chromosome Translocation The Role of LMO2 in T-ALL. The existence of an oligomeric Man Mouse complex in which Lmo2 protein is apparently a linking molecule in RBCs suggests that the function of the Lmo2 protein in T-cell acute LMO1 11p15 7 t(11;14)(p15;q11) leukemias may be to participate in an analogous but aberrant complex after the chromosomal translocation has enforced expression of the Lmo2 gene in T cells. The emulation of the human T-cell LMO2- enforced expression after chromosomal translocation has been LMO2 11p13 2 t(11;14)(p13;q11) achieved using transgenic expression of Lmo2 in the T-cell lineage (44-47). This results in clonal T-cell leukemia arising in the transgenic mice with a long latency, on average about 9 months. This latency period indicates that the transgene is necessary but not suffi- LMO3 12 p12-13 6 ND cient to cause tumors in this model, as is the case for many transgenic Fig. 1. LMO family of genes and oncogenes. The LMO genes (so designated as LIM-Only genes and previously known as RBTN or TTG genes) have three known oncogene models, and that mutations in other oncogenes must be members. LM01 (previously RBTN1 or TTG1) was identified first and then, LM02 occurring to allow development of overt disease. (previously RBTN2 or TTG2) and LM03 (previously RBTN3). LM01 and LM02 are both The long latency period facilitated the detailed study of possible located on the short arm of chromosome 11 and are both involved in independent chromosomal translocations in T-ALL. As yet, LMO3 has not been found in association effects in the asymptomatic thymuses of transgenic mice (45, 47). An with any chromosomal translocations. outline of normal T-cell differentiation is shown in Fig. 4, which 1795s

Fig. 3. Lmo2 lbrms part of a complex that can bind a bipartite DNA site. A, the ability of the Lmo2 protein to interact with Tall and Molecular consequences of enforced LMO2 expression I with GATA1 appears to facilitate the formation of a complex in which two DNA contacting regions comprise a Tal 1-E47 dimer [binding an E box (CANNTG)] and a GATA1 molecule [binding a GATA site] as part of an erythroid complex. In model experiments (40), the two parts AI Sm of the bipartite recognition sequence are separated by approximately one turn of the DNA helix. This erythroid complex can specifically Erythroid progenitor complex Aberrant T cell complex bind a unique bipartite DNA site. Thus, the probable function of the complex is to bind to the DNA of chromosomal target genes and to regulate their expression (positively or negatively; Ref. 40). B, an analogous DNA-binding complex has been identified in T-cell of Lmo2-expressing transgenic mice (48). In this case, however, a novel complex is formed which can recognize a dual E-box motif, apparently via two bHLH dimers linked by Lmo2 and Ldbl proteins. Therefore, enforced Lmo2 expression by transgenesis or by chromosomal translocations may cause formation of an aberrant T-cell complex that binds to and controls the expression of target genes (48). Q~A:NNTG ...... G'A~T~A ...... CAt',~'TC, CAt'~'TC These may well be different from those genes normally the focus of Lmo2 function. serves to illustrate the individual points in T-cell development. There bridging molecule between the two DNA-binding arms of a complex. was a marked accumulation of immature CD4-, CD8-, CD25 +, The ability of Lmo2 to interact with other proteins to perhaps mediate CD44 + T cells (herein referred to as DN T cells) in transgenic the formation of a multiprotein DNA binding complex suggested a thymuses compared with nontransgenic litter mates, an effect that was similar molecular mechanism for LMO2 protein in T-cell tumors. exacerbated in mice transgenic for both Lmo2 and Tall (47). Thus, the After enforced expression of LM02, either by chromosomal translo- role of the transgene products is to cause an inhibition of T-cell cation or by transgenesis, the presence of the protein could, therefore, differentiation that appears reversible, presumably by antigenic stim- mediate the formation of an aberrant protein complex. This aberrant ulation occurring after birth because different transgenic mice exhibit complex may contain DNA-binding factors that can recognize target different levels of DN cell accumulation. genes (perhaps with altered specificity) either activating or repressing It is of note that the DN T-cell population is RAG variable diversity gene expression, or the aberrant complex could simply inhibit normal joining recombinase-positive, and thus is a population where, in binding of components of the complex to their cognate DNA sites, humans, the recombinase-associated translocations may occur. This thereby effectively repressing gene activity. suggests that the T-cell acute leukemia precursors in humans acquire LMO2 expression in T cells may result in a complex with other the chromosomal translocations within the DN T-cell population, and proteins that would not normally form. In addition, it might be that, that this produces a cell with inhibited differentiation analogous to although a complex of proteins does form, the crucial interaction is that of the transgenic mouse model (Fig. 5). It is proposed that the with only one of the interacting partners, sequestering it from per- overt tumor eventually arises in the targeted, predisposed population forming its normal function, which might not involve DNA binding. of T cells that emerges from the event of chromosomal translocation For instance, the binding of LDB 1 in the T cell by enforced LMO2 by the accumulation of secondary mutations in other genes. Therefore, expression might simply remove LDB 1 from its normal function. these tumors can display any other stage of T-cell differentiation. In a search for evidence of oligomeric Lmo2-complexes, T-cell Protein Complexes Involving LMO2 in T-ALL. The work on lines were derived from CD2-Lmo2 transgenic mice (48) and used as Lmo2 gene targeting clearly showed that the gene is essential for both a source of Lmo2 protein complexes for in vitro random site selection yolk sac erythropoiesis and for adult definitive hematopoiesis. This CASTing experiments (48). This work resulted in the detection of a information indicated that erythroid cells might possess a functional Lmo2-containing complex, which, like its analogue in erythroid cells, Lmo2 protein and made possible the discovery that the Lmo2 protein binds to a bipartite recognition site (Fig. 3B) but, in the T-cell context, is part of a DNA-binding oligomeric protein complex in which Lmo2, recognizes a dual E-box motif, in which the two E-box sequences are rather than participating in direct DNA contacts, is a connecting or separated by about one DNA helical turn. Analysis of the components

Fig. 4. Variation of LMO2 complexes may occur at different hematopoietic stages We have found that the LMO2 protein is part of a multimeric complex in erythroid cells (40), which suggests that it is involved in erythropoiesis. Our data on Lmo2 .... ~NLI null mutations in mice suggest possible distinct roles in primitive and definitive hematopoiesis (22, Ao 23). Therefore, distinct Lmo2-multimeric complexes may have distinct roles in different hemato- Early haematopoiesis poietic lineages (40), regulating different sets of target genes. The finding of a novel LMO2- containing multimeric complex in T-cells after enforced Lmo2 expression (48) supports the notion of the flexibility of Lmo2 to be involved in different complexes in different environments. Thus, in erythroid cells (B), Lmo2 is involved in a complex B. .... 'NLI with a number of proteins, and it was suggested that T in early hematopoiesis (A) it may be in complexes with a different composition (40). Variations in these complexes may function in lineage differen- Erythropoiesis tiation programs as suggested by mutational studies of Lmo2 (22, 23).

1796s

topoietic lineages (40). The nature of Lmo2 protein-protein interactions in early hematopoiesis will guide additional investigations. It is also very evident from the study of LM02 and of other translocation- Differentiation block . ITransge~ activated oncogenes, that transcription regulation is a key target within the tumor precursor. Finally, as the LIM domain interactions demonstrate, protein-protein associations are crucial molecular events Pre-T CD4- CO8- in tumor formation. cell CD25+ CD44+

BM ~ | --~ | "~ | ~ CD3+ Note Added in Proof CD8+ OR CD4+ TCR~- RAG+ A fourth member of the LMO family (LMO4) has recently been identified (Gmtz, G., Forster, A., and Rabbitts, T. H., Identification of the LM04 gene encoding an interaction partner of the LIM-binding protein LDB1/NLII: a Onset of TCR Chromosomal ] :Chiiaho0d [ candidate for displacement by LMO proteins in T cell acute leukaemia. -"- translocation --~ i T!ALL,,,,,, ' I rearrangement during TCR joining Oncogene, 17: 2799-2803, 1998). Fig. 5. Model of Lmo2 function in T-ALL by inhibition of T-cell differentiation. Bone marrow (BM) produces pre-T cells (designated as TN because these cells do not yet References express typical T-cell markers such as CD3, CD4 and CD8). These pre-T cells become DN cells (having a CD4-, CD8 , CD25 +, CD44 + marker phenotype) which do not yet 1. Rabbitts, T. H. Chromosomal translocations in human cancer. Natm-e (Lond.), 372: express T-cell receptor (TCR) but begin to express the RAG recombinase proteins. Further 143-149, 1994. differentiation of this immature DN T-cells subset results after TCR rearrangement, 2. Adams, J. M., and Cory, S. Transgenic models of tumor development. Science resulting in mature functional TCR-bearing T cells. The Lmo2 transgenic mice (45, 46) (Washington DC), 254:1161-1167, 1991. and double Lmo2-Tall transgenics (45, 47) accumulate the DN cells, apparently as a 3. CoiTal, J., Lavenir, I., lmpey, H., Warren, A. J., Forster, A., Larson, T. A., Bell, S., result of differentiation inhibition caused by the transgene. Because the LMOI- and McKenzie, A. N. J., King, G., and Rabbitts, T. H. An Mll-Af9 fusion gene made by LMO2-associated chromosomal translocations seem to result from RAG-mediated recom- homologous recombination causes acute leukemia in chimeric mice: a method to binase errors, it is proposed that this same target population of DN T cells is affected in create fusion oncogenes. Cell, 85: 853-861, 1996. humans with the chromosomal translocation. The enforced LMO2 expression affects 4. Rabbitts, T. H., Hamlyn, P. H., and Baer, R. Altered nucleotide sequences of a multimeric protein complex formation, which seems to function by alteration of the T-cell translocated c-myc gene in Burkitt lymphoma. Nature (Lond.), 306: 760-765, 1983. differentiation program. Thus, altered T-cell clones provide the precursors from which the 5. Rabbitts, T. H., Forster, A., Hamlyn, P. H., and Baer, R. Effect of somatic mutation overt tumor arises after secondary mutations. within translocated c-myc genes in Burkitt's lymphoma. Nature (Lond.), 309: 592- 597, 1984. 6. Showe, L. C., Ballantine, M., Nishikura, K., Erikson, J., Kaji, H., and Croce, C. M. Cloning and sequencing of a c-myc oncogene in a Burkitt's lymphoma cell line that of this complex showed that E47-Tall bHLH heterodimeric elements is translocated to a germ line a switch region. Mol. Cell. Biol., 5: 501-509, 1985. were present as well as Lmo2 and the Ldbl protein (illustrated in Fig. 7. Murphy, W., Sarid, J., Taub, R., Vasicek, T., Battey, J., Lenoir, G., and Leder, P. Translocated human c-myc oncogene is altered in a conserved coding sequence. Proc. 3B). The function of this complex may, therefore, be to bind to unique Natl. Acad. Sci. USA, 83: 2939-2943, 1986. sites in chromosomal target genes to control their expression (posi- 8. Cesarman, E., Dalla-Favera, R., Bentley, D., and Groudine, M. Mutations in the first tively or negatively). The formation of an aberrant Lmo2-containing exon are associated with altered transcription of c-myc in Burkitt lymphoma. Science (Washington DC), 238:1272-1275, 1987. complex suggests that the target genes are affecting T-lineage devel- 9. Morse, B., South, V. J., Rothberg, P. G., and Astrin, S. M. Somatic mutation and opment, providing a cell population in which secondary mutations transcriptional deregulation of myc in endemic Burkitt's lymphoma disease: hep- occur preceding overt tumors. tamer-nonamer recognition mistakes'? Mol. Cell. Biol., 9: 74-82, 1989. 10. Bhatia, K., Huppi, K., Spangler, G., Siwarski, D., Iyer, R., and Magrath, I. Point Conclusions. The extensive work that has been carried out on the mutations in the c-Myc transactivation domain are common in Burkitt's lymphoma cloning and characterization of chromosomal translocation break- and mouse plasmacytomas. Nat. Genct., 5: 56-61, 1993. 11. Yano, T., Sander, C. A., Clark, H. M., Dolezal, M. V., Jaffe, E. S., and Raffeld, M. points, coupled with functional studies of the oncogenes found to be Clustered mutations in the second exon of the MYC gene in sporadic Burkitt's associated with these breakpoints, has allowed some general conclu- lymphoma. Oncogene, 8: 2741-2748, 1993. sions to be made, which are typified by the studies discussed in this 12. Boehm, T., Baer, R., Lavenir, I., Forster, A., Waters, J. J., Nacheva, E., and Rabbitts, T. H. The mechanism of chromosomal translocation t(1 t ;14) involving the T-cell paper on the LM02 gene. receptor C6 locus on human chromosome 14ql I and a transcribed region of chro- Clearly, chromosomal translocations are important for the etiology mosome 1 lp15. EMBO J., 7." 385-394, 1988. of the tumors that contain them. It seems likely that in many cases, the 13. McGuire, E. A., Hockett, R. D., Pollock, K. M., Bartholdi, M. F., O'Brien, S. J., and Korsmeyer, S. J. The t(11; 14)(p 15 ;q 11 ) in a T-cell acute lymphoblastic leukcmia cell translocation event occurs early in tumor formation and involves the line activates multiple transcripts, including Ttg-l, a gene encoding a potential zinc alteration of developmental processes in some way. In terms of the finger protein. Mol. Cell. Biol., 9: 2124-2132, 1989. 14. Boehm, T., Greenberg, J. M., Buluwela, L., Lavenir, I., Forster, A., and Rabbitts, formation of the chromosomal translocation, there are two crucial T. H. An unusual structure of a putative T cell oncogene which allows production of points: similar proteins from distinct mRNAs. EMBO J., 9: 857-868, 1990. (a) the mechanism of the formation of the aberrant chromosome 15. Boehm, T., Foroni, L., Kennedy, M., and Rabbitts, T. H. The rhombotin gene belongs to a class of transcriptional regulators with a potential novel protein dimerisation itself. In the lymphoid tumors, this can be through mistakes of T-cell motif. Oncogene, 5:1103-1105, 1990. receptor rearrangement creating an interchromosomal event rather 16. Boehm, T., Foroni, L., Kaneko, Y., Perutz, M. P., and Rabbitts, T. H. The rhombotin than the intrachromosomal VDJ join, whereas in cases where the family of cysteine-rich LlM-domain oncogenes: Distinct members are involved in T-cell translocations to human chromosomes 1 lpl 5 and 1 l p13. Proc. Natl. Acad. Sci. fusion of genes occurs, it is far from clear what the mechanism is; and USA, 88: 4367-4371, 1991. (b) in either situation, the outcome is a cell with some growth 17. Foroni, L., Boehm, T., White, L., Forster, A., Sherrington, P., Liao, X. B., Brannan, C. I., Jenkins, N. A., Copeland, N. G., and Rabbitts, T. H. The rhombotin gene family advantage over its partners, and this cell may eventually appear as an encode related LIM-domain proteins whose differing expression suggests multiple overt tumor after accumulation of other mutations. roles in mouse development. J. Mol. Biol., 226: 747-761, 1992. The discovery of multimeric complexes involving Lmo2 in normal, 18. Royer-Pokora, B., Loos, U., and Ludwig, W.-D. TTG-2, a new gene encoding a cysteine-rich protein with the LIM motif, is overexpressed in acute T-cell leukaemia erythroid cells (40) and in T-cell tumors (48) suggests that different with the t(11 ;14)(p13;ql 1). Oncogene, 6: 1887-1893, 1991. chromosomal target genes are controlled in the different settings. The t9. Sanchez-Garcia, I., Kaneko, Y., Gonzalez-garmiento, R., Campbell. K., White, L., study of these genes will prove interesting in the future. The combined Boehm, T., and Rabbitts, T. H. A study of chromosome llp13 translocations involving TCR~9 and TCR6 in human T cell leukaemia. Oncogene, 6: 577-582, 1991. study of the etiology and function of the naturally occurring chromo- 20. Sanchez-Garcia, I., and Rabbitts, T. H. The LIM domain: a new structural motif in somal mutations (i.e., translocations) and the function of genes found zinc-finger-like proteins. Trends Genet., IO: 315-320, 1994. 21. Perez-Alvarado, G. C., Miles, C., Michelsen, J. W., Louis, H. A., Winge, D. R., at their junctions has been informative about general mechanisms. The Beckerle, M. C., and Summers, M. F. Structure of the carboxy-terminal LIM domain data suggest that variation of Lmo2-complexes may influence hema- from the cysteine rich protein CRP. Nat. Stnlct. Biol., 1: 388-398, 1994. 1797s

22. Warren, A. J., Colledge, W. H., Carlton, M. B. L., Evans, M. J., Smith, A. J. H., and 35. Hsu, H.-L., Huang, L., Tsan, J. T., Funk, W., Wright, W. E., Hu, J-S., Kingston, R. E, Rabbitts, T. H. The oncogenic cysteine-rich LIM domain protein rbtn2 is essential for and Baer, R. Preferred sequences for DNA recognition by the TALl helix-loop-helix erythroid development. Cell, 78: 45-58, 1994. proteins. Mol. Cell. Biol., 14: 1256-1265, 1994. 23. Yamada, Y., Warren, A. W., Dobson, C., Forster, A., Pannell, R., and Rabbitts, T. H. 36. Valge-Archer, V. E., Osada, H., Warren, A. J., Forster, A., Li, J., Baer, R., and The T cell leukaemia LIM protein Lmo2 is necessary for adult mouse haematopoiesis. Rabbitts, T. H. The LIM protein RBTN2 and the bHLH protein TALl are present in Proc. Natl. Acad. Sci. USA, 95: 3890-3895, 1998. a complex in erythroid cells. Proc. Natl. Acad. Sci. USA, 91:8617 8621, 1994. 24. Robb, L., Lyons, 1., Li, R., Hartley, L., Kontgen, F., Harvey, R. P., Metcalf, D., and 37. Wadman, I., Li, J., Bash, R. O., Forster, A., Osada, H., Rabbitts, T. H., and Baer, R. Begley, C. G. Absence of yolk sac hematopoiesis from mice with a targeted disrup- Specific in vivo association between the bHLH and LIM proteins implicated in human tion of the scl gene. Proc. Nail. Acad. Sci. USA, 92: 7075-7079, 1995. T cell leukaemia. EMBO J., 13: 4831-4839, 1994. 25. Shivdasani, R. A., Mayer, E., and Orkin, S. H. Absence of blood formation in mice 38. Schmeichel, K. L., and Beckerle, M. C. The LIM domain is a modular protein-binding lacking the T-cell leukaemia oncoprotein tal-I/SCL. Nature (Lond.), 373: 432-434, interface. Cell, 79: 211-219, 1994. 1995. 39. Osada, H., Grutz, G., Axelson, H., Forster, A., and Rabbitts, T. H. Association of 26. Robb, L., Elwood, N. J., Elefanty, A. G., Kontgen, F., Li, R., Bamet, L. D., and erythroid transcription factors: complexes involving the LIM protein RBTN2 and the Begley, C. G. The scl gene product is required for the generation of all hematopoietic zinc-finger protein GATA1. Proc. Natl. Acad. Sci. USA, 92: 9585-9589, 1995. lineages in the adult mouse. EMBO J., 15: 4123-4129, 1996. 40. Wadman, 1. A., Osada, H., Grutz, G. G., Agulnick, A. D., Westphal, H., Forster, A., 27. Porcher, C., Swat, W., Rockwell, K., Fujiwara, Y., Alt, F. W., and Orkin, S. H. The and Rabbitts, T. H. The LIM-only protein Lmo2 is a bridging molecule assembling an T cell leukemia oncoprotein SC1/tal-1 is essential for development of all hematopoi- erythroid, DNA-binding complex which includes the TALl, E47, GATA-1 and etic lineages. Cell, 86: 47-57, 1996. Ldbl/NLI proteins. EMBO J., 16: 3145-3157, 1997. 28. Begley, C. G., Aplan, P. D., Davey, M. P., Nakahara, K., Tchorz, K., Kurtzberg, J., 41. Valge-Archer, V., Forster, A., and Rabbitts, T. H. The LMO1 and LDBI proteins Hershfield, M. S., Haynes, B. F., Cohen, D. 1., Waldmann, T. A., and Kirsch, 1. R. interact in human T-cell acute leukemia with the chromosomal translocation Chromosomal translocation in a human leukemic stem-cell line disrupts the T-cell t(11;14)(p15;ql 1). Oncogene, 17: 3199-3202, 1998. antigen receptor &chain diversity region and results in a previously unreported fusion 42. Pevny, L., Simon, M. C., Robertson, E., Klein, W. H., Tsai, S-F., D'Agati, V., Orkin, transcript. Proc. Natl. Acad. Sci. USA, 86: 2031-2035, 1989. S. H., and Costantini, F. Erythroid differentiation in chimaeric mice blocked by a 29. Begley, C. G., Aplan, P. D., Denning, S. M., Haynes, B. F., Waldmann, T. A., and targeted mutation in the gene for transcription factor GATA-1. Nature (Lond.), 349: Kirsch, I. R. The gene SCL is expressed during early hematopoiesis and encodes a 257-260, 1991. differentiation-related DNA-binding motif. Proc. Natl. Acad. Sci. USA, 86:10128- 43. Robb, L., Rasko, J. E. J., Bath, M. L., Strasser, A., and Begley, C. G. scl, a gene 10132, 1989. frequently activated in human T cell leukaemia, does not induce tymphomas in 30. Bernard, O., Gugliehni, P., Jonveaux, P., Cherif, D., Gisselbrecht, S., Mauchauffe, transgenic mice. Oncogene, 10: 205-209, 1995. M., Berger, R., Larson, C-J,, and Mathieu-Mahul, D. Two distinct mechanisms for the 44. Fisch, P., Boehm, T., Lavenir, I., Larson, T., Amo, J,, Forster, A., and Rabbitts, T. H. SCL gene activation in the t(1;14) translocation of T-cell leukemias. Genes Chromo- T-cell acute lymphoblastic lymphonra induced in transgenic mice by the RBTN1 and somes Cancer, l: 194-210, 1989. RBTN2 LIM-domain genes. Oncogene, 7: 2389-2397, 1992. 31. Finger, k. R., Kagan, J., Christopher, G., Kurtzberg, J., Hershfield, M. S., Nowell, 45. Larson, R. C., Osada, H., Larson, T. A., Lavenir, I., and Rabbitts, T. H. The oncogenic P. C., and Croce, C. M. involvement of the TCL5 gene on human chromosome 1 in LIM protein Rbtn2 causes thymic developmental abelrations that precede malignancy T-cell leukemia and melanoma. Proc. Natl. Acad. Sci. USA, 86: 5039-5043, 1989. in transgenic mice. Oncogene, 11: 853-862~ 1995. 32. Chen, Q., Cheng, J.-T., Tsai, L-H., Schneider, N., Buchanan, G., Carroll, A., Crist, 46. Neale, G. A., Rehg, J. E., and Goorha, R. M, Ectopic expression of rhombotin-2 W., Ozanne, B., Siciliano, M. J., and Baer, R. The tal gene undergoes chromosome causes selective expansion of the thymus and T-cell tumours in transgenic mice. translocation in T cell leukaemia and potentially encodes a helix-loop-helix protein. Blood, 86: 3060-3071, 1995. EMBO J., 9: 415-424, 1990. 47. Larson, R. C., Lavenir, I., Larson, T. A., Baer, R., Wan'en, A. J., Wadman, I., Nottage, 33, Chen, Q., Yang, C. Y-C., Tsan, J., Xia, Y., Ragab, A. H., Peiper, S. C., Carroll, A., K., and Rabbitts, T, H. Protein dimerisation between Lmo2 (Rbm2) and Tall alters and Baer, R. Coding sequences of the tat-I gene are disrupted by the chromosome thymocyte development and potentiates T cell tumorigenesis in transgenic mice. translocation in human T cell leukaemia. J. Exp. Med., 172: 1403-1408, 1990. EMBO J., 15: 1021-1027, 1996. 34. Hsu, H-L., Cheng, J-T., Chen, Q., and Baer, R. Enhancer-binding activity of the tall 48. Grutz, G., Bucher, K., Lavenir, l., Larson, R., Larson, T., and Rabbitts, T. H. The oncoprotein in association with the E47/EI2 helix-loop-helix proteins. Mol. Cell. oncogenic T cell LIM-protein Lmo2 forms part of a DNA-binding complex specifi- Biol., 11: 3037-3042, 1991. cally in immature T cells. EMBO J., 17: 4594-4605, 1998.

Discussion

Speaker: Can you tell us... you said for lack of time you couldn't DNA binding site for the LMO 1-associated complex. And that is why tell us the other components of the complex in the LMO2 transgenic I am beginning now to think in terms of the sequestration LDB protein T cells. What other factors are present in that E-box binding complex? function as being the mechanism. Dr. Rabbitts: The only factors that we know about at the present Dr. Phillip Sharp: If I understood it correctly, you believe you form time are LDB, E47, and TALl. We are talking about complexes an aberrant complex when LMO2 is overexpressed in the T cell that within the transgenic mouse T-cell tumors. So, of course, that situa- might activate genes that would normally not be activated under those tion is slightly artificial, because we already know that the TALl conditions. The direct prediction of that would be that you could do protein is present there because of the transgene. plus/minus screening in those cell populations and see novel RNAs Dr. Stanley Korsmeyer: I am wondering, do you think LMO1, that would appear only in that population compared with normal or LMO3 would all be frilly substitutable in terms of the mechanism? other tumor types of the same type but not expressing that complex. Have you had a chance to compare them with LMO2? Has that ever been looked at? Dr. Rabbitts: This is a complex question actually. With LMO3, we Dr. Rabbitts: Those are ongoing experiments, and we haven't haven't really done anything at all. That protein, although extremely actually been able to identify any differences. We have taken a related to the LMO1 protein (the LIM domains themselves are iden- slightly complicated route by trying to fractionate the double-negative tical) has not yet been identified as far as I am aware in any chromo- asymptomatic T cells away from the double-positive T cells and somal translocations, so we have really ignored it. comparing those two populations. We think the double-negative cells With LMO1, some recent data on T-cell tumors which express are the ones where the pathogenic effect is occurring. LMO1 have shown two things. One is that LMO1 binds to the So, although those experiments are important, we just haven't got endogenous LDB protein. And, two, we have been unable to find any data yet.

1798s

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1999 American Association for Cancer Research. The Effect of Chromosomal Translocations in Acute Leukemias: The LMO2 Paradigm in Transcription and Development

Terence H. Rabbitts, Katharina Bucher, Grace Chung, et al.

Cancer Res 1999;59:1794s-1798s.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/59/7_Supplement/1794s

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/59/7_Supplement/1794s. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.