Concise Review International Journal of Cell Cloning 10:269-276 (1992)

SCL and Related Hemopoietic Helix-Loop-Helix Transcription Factors A. R. Green’, C. G . Begley“.b ’The Walter and Eliza Hall Institute of Medical Research and bDepartrnent of Diagnostic Haernatology, PO Royal

Melbourne Hospital, Victoria, Australia Key Words. MEL cells

K562 cells Leukemia Translocation

Abstract. The helix-loop-helix (HLH) proteins are a family of transcription factors that include proteins critical to differentiationand development in species ranging from plants to mammals. Five members of thii family (MYC, SCL, TAL-2, LYL-1 and E2A) are implicated in oncogenic events in human lymphoid tumors because of their consistent involvement in chromosomal translocations. Although activated in T cell leukemias, expression of SCL and LYL-1 is low or undetectable in normal T cell populations. SCL is expressed in erythroid, megakaryocyte and mast cell populations (the same cell lineages as GATA-1, a zinc-finger transcription factor). In addition, both SCL and GATA-1 undergo coordinate modulation during chemically induced erythroid differentiation of mouse erythroleukemia cells and are downmodulated during myeloid differentiation of human K562 ceUs, thus implying a role for SCL in erythrold differentiation events. However, in contrast to GATA-1, SCL is expressed in the developing brain. Studies of the function of SCL suggest it is also important in prolieration and self-renewal events in erythroid cells. Introduction

One of the central issues of biology concerns the mechanisms whereby differentiated cell types arise from a single multipotential cell. Although most of our current understanding stems from invertebrate studies, several recent advances have been made in mammalian systems, especially in the areas of myogenesis and hemopoiesis. Hemopoiesis offers a uniquely tractable experimental system for the study of mammalian lineage commitment. At least nine well-characterized differentiated cell types arise from a hierarchy of oligoCorrespondence: Dr. A. R. Green, Department of Haernatology, MRC Building, Hills Road, Cambridge CB22QH, United Kingdom. Received May 28, 1992; accepted for publication May 28, 1992. 0737-1454/92/$2.00/0 QAlphaMed Press

or multipotential progenitors which in turn derive from a pluripotential stem cell [l]. Clonal assays are available for many of the progenitor cell types. Over the past two decades much has been learned about the extracellular signals that regulate hemopoiesis-many of the growth factors and the cellular receptors involved in hemopoiesis have been characterized, purified and molecularly cloned. In contrast, the intracellular mechanisms that are responsible for developmental or differentiative “decisions” remain largely obscure. However, transcription factors are likely to play a prominent role in these “decisions” in view of their central role in invertebrate development and differentiation, together with the evidence that tissue-specific gene regulation occurs mainly at the level of transcription in mammalian cells [2]. The helix-loop-helix (HLH) motif unites a family of transcription factors with roles in the regulation of tissue-specific gene expression and developmental processes such as neurogenesis [3], myogenesis 141, germ layer formation [5] and sex determination [6].The HLH motif consists of two amphipathic helices joined by an intervening loop of varying length. Immediately upstream, many members have a basic region. Site-directed mutagenesis studies have shown that the basic region is important for DNA binding, whereas the HLH domain is required for both DNA binding and for protein-protein dimerization 171. A minimal consensus DNA binding site for HLH proteins has been defined as consisting of CA-TG (the so-called “E-box” element) [8]. Protein dimerization patterns have been used to classify HLH proteins into three categories [9]. HLH proteins in one class are ubiquitously expressed and will form heterodimers with a second class of tissuerestricted HLH proteins. A third class of HLH proteins does not form heterodimers with either of the other classes and presumably these proteins interact with other as yet undefined HLH proteins. This prediction has been recently confirmed for the MYC proteins by the isolation of a novel HLH protein,

Green/Begley

270

Table I. Helix-loop-helix genes activated in human lymphoid tumors

Consequence of Translocation

Normal Expression

t(8; 14)(q24;q32) t(8;22)(q24;qll) t(2;8)(pl1;q24)

Overexpression

Ubiquitous

T-ALL

t(1; 14)(p32;q11)

Overexpression

Hernopoietic restricted

TAL-2

T-ALL

t(7;9)(q34;q32)

Overexpression

Unknown

LYL- 1

T-ALL

t(7;19)(q35;pl3)

Overexpression

Hemopoietic restricted

E2A

Pre-B-ALL

t(1; 19)(q23;pl3)

Fusion protein

Ubiquitous

Gene

Tumor

Translocation

MYC

Burkitt’s Lymphoma

SCL

HLH genes SCL, LYL-1 and TAL-2 [48] are translocated in T-ALL, and E2A forms a fusion gene with the homeobox gene PBX 1 in pre-B-ALL. MYC is involved in chromosomal translocation in Burkitt’s lymphomas; the t(8;14) is observed in 75% of cases, t(8;22) in 16% of cases and t(2;8) in 9% of cases.

HLH Genes and Hematological Malignancies

Fig. 1. The helix-loop-helix (HLH) class of DNA-binding proteins undergo dimerization to interact with specific DNA target sequences. Dimerization is mediated via the HLH motif. DNA binding occurs via a basic domain that is absent in some HLH proteins. These negative regulators are thereby able to sequester other HLH proteins in inactive complexes.

MAX, which forms heterodimers with C-, N- and L-MYC [ 10, 111. Although this classification is useful as a working model, it is already clear that further levels of complexity exist. First, several HLH proteins contain additional protein or DNA binding motifs; thus MYC and TFE3 contain HLH and leucine zipper domains [12, 131. Interestingly, the leucine zipper and not the HLH domain of TFE3 direct the specificity of protein dimerization [ 131. Second, a number of negative regulatory HLH proteins have been described which lack a basic DNA binding domain and which dimerize with and inhibit other HLH proteins (Fig. 1) [ 14-16].

Five members of the HLH family have clearly been implicated in oncogenic events in lymphoid cells: MYC, SCL, TAL-2, LYL-1 and E2A. In each case it is because of their involvement in chromosomal translocations (Table I). The most studied translocation event is that of Burkin’s lymphomas where the t(8;14) translocation moves the MYC gene into the immunoglobulin heavy chain locus to come under the influence of the immunoglobulin enhancer. There is now considerable evidence that deregulation of MYC is a critical component in the pathogenesis of several lymphoid tumors [17]. SCL and the t(1;14) Translocation in T-ALL SCL was first identified because of its involvement in a t( 134) translocation in a patient with T cell acute lymphoblastic leukemia (T-ALL). Intriguingly, the leukemic T cells differentiated into myeloid cells when the patient was treated with 2’-deoxycofomycin. The karyotype and multipotential phenotype were recapitulated by a cell line established from the patient, and this suggested that the leukemia probably arose in a multipotential hemopoietic precursor cell [18]. Analogous to the situation for MYC in B cell tumors, molecular studies demonstrated that SCL was translocated into the T cell receptor (TCR) for antigen TCR 6 locus. This event also involved a rearrangement of the diverse D62 and D63 genes in a manner reminiscent of the normal physiological rearrangement observed in developing T lymphocytes. Thus, DNA recombinases were strongly implicated in the

27 1 mechanism of this chromosomal translocation. The translocation interrupted the 3' untranslated region (UTR) of the SCL gene and therefore did not alter the coding region. As a result, SCL was highly expressed in the leukemic cells, although it is not normally detected in T cell populations [18, 191. These findings were confirmed by others [20], and, despite initial conclusions to the contrary [21], the SCL (also called TAL-1) protein coding region was preserved intact. In these later studies the translocation occurred in the 5' UTR of the SCL gene [22]. In addition to the t( 1;14) translocation event that occurs in 3% of patients with T-ALL, SCL is activated in 12-25% of T-ALL by a mechanism that involves an interstitial deletion on chromosome 1 [23, 241. Again, DNA recombinases are implicated in the mechanism of this event. The deletional event interrupts in the 5' UTR and is functionally analogous to translocations involving the 5' UTR of SCL [22]. Thus, SCL appears to be a common target for rearrangement/translocationevents in T-ALL. These events have in common the likely involvement of DNA recombinases in their generation and the preservation of the SCL coding region. This implies that aberrant regulation and subsequent overexpression of a normal SCL gene product may play an important role in the generation of T-ALL. Other HLH Genes and Chromosome Translocations LYL-1 has a number of striking similarities to SCL. Like SCL, LYL-1 was identified via translocation of t(7;19) in a case of T-ALL [25]. This translocation occurred between the first intron of LYL-1 and the TCR locus. As a consequence, the LYL-1 transcript was truncated and lacked a potential CTG initiation codon in exon 1. Although this was interpreted as resulting in an altered LYL-1 protein product, this seems unlikely given the absence of this CTG codon in the murine LYL-1 sequence and the otherwise high homology between the mouse and human genes [26]. Thus it seems that overexpression of a normal LYL-1 protein may have contributed to the genesis of the T-ALL. Both LYL-1 and SCL are closely related in the HLH domain and upstream basic region with the majority of differences representing conservative substitutions (84% identity overall). However, despite their involvement in T-ALL, both genes display low level or undetectable expression in most T lymphoid cell sources. LYL- 1, for example, is expressed in myeloid, erythroid and B lymphoid cells [26, 271. It seems likely that ectopic expression of SCL and LYL-1 proteins in a T cell environment allows their interaction with other proteins to which they are not normally exposed. Such pathological

SCL and Hemopoietic Differentiation

interactions presumably produce inappropriate gene regulation and thus contribute to tumorigenesis. The product of the E2A gene was originally identified because of its ability to bind to a site in the enhancer for the immunoglobulin K chain gene. The E2A gene encodes the related HLH transcription factors El2 and E47 [3] that are ubiquitously expressed. This gene has been identified as one partner in the t( 1;19) translocation found in 30% of pre-B-ALL [28, 291. The translocation disrupts the coding region of the E2A gene and generates a fusion gene. Although the breakpoints on chromosome 1 are distributed over a large segment of DNA, identical chromosome 1 sequences are joined to E2A to generate the fusion transcripts. The resultant chimeric protein contains the transcriptional activating motif of E2A but the DNA-binding motif of a homeobox gene on chromosome 1 called PBX 1 [30, 3 11. This fusion protein is therefore likely to bind to and aberrantly activate a completely different set of target genes to those normally regulated by the E2A protein and is likely to be very important in the pathogenesis of pre-B-ALL.

The Regulation of Normal Hemopoiesis Invertebrate evidence suggests that mammalian differentiation and development are likely to be regulated by transcription factors in a combinatorial manner [32]. Hence the precise pattern and concentration of lineage-restricted and ubiquitous transcription factors within a particular hemopoietic progenitor cell are likely to play a prominent role in dictating developmental decisions. Within this scenario, lineage-restricted transcription factors such as SCL and LYG 1 are likely to play a particularly crucial role since it is their presence or absence which determines distinct lineages. The SCL Gene The SCL gene is located on human chromosome 1~32-33[ 181 and on mouse chromosome 4 [33]. The predicted protein product is 34 kDa with striking homology among the mouse, human and chicken sequences; the HLH and upstream basic regions are entirely conserved among the three species with 94% amino acid identity overall between mouse and human and 68% identity between chicken and human proteins [33, 341. The human gene consists of 8 exons that span 15 kb of genomic sequence, and it contains multiple 5'untranslated exons and a large 3' exon with a 3.4 kb UTR[22]. A complex pattern of alternate exon usage in the 5' UTR has been demonstrated in human cells,

272

Green/Begle y

A.

0 b -0

s

I

I

xo

xo

c

n I

B

lkb

8. N -I-

w m X

Y

N

a w m

I Y

3 %

2

28S-

18S-

probe: a

b

c

Fig. 2. Heterogeneity in the 3' UTR of human SCL transcripts. A) Map of human SCL cDNA showing the coding region (boxed) and long 3' UTR.Probes a, b, c are shown above. S=Sstl; Xa=Xbal; Xo=Xhol; B=Bam. B) Northem blot analysis of polyA+mRNA ( 3 ~from ) two human erythroid cell lines (HEL and K562). The predominant SCL transcript migrates with the 28s ribosomal band. The lower transcript differs in the 3' UTR.

although this does not appear to affect the coding region [22,35]. Both exon la and l b have upstream TATA and CCAAT boxes, and, in view of the transcription pattern.of SCL (see below), the presence of a potential binding site for the transcription factor GATAL1 upstream of exon la suggests a possible regulatory role for this protein. There is remarkable similarity in the structure of the murine SCL gene (unpublished data) and complexity in the 3' UTR in both species. As shown in Figure 2, two major transcripts of 5 kb and 3 kb are seen in human hemopoietic cells, although their relative intensity varies in different cell types. The 3 kb band is faint and lacks sequences from the 3' UTR. Similar heterogeneity exists in murine cells [33]. The functional significance of this complexity is presently unclear, but it is intriguing that the relative abundance of 5 kb and .3 kb transcripts may differ in distinct erythroid subpopulations [361. The presence of an HLH motif in the SCL protein strongly suggests that it functions as a transcription factor and probably as a heterodimer. Indeed, heterodimers of SCL with the protein products

of the E2A HLH gene (El2 and E47) have recently been shown to bind specifically to two E-box elements [37] that are present in the regulatory regions of several erythroid-specific genes. Several clues to the biological function of SCL within erythroid cells have also emerged recently. SCL Is Coexpressed and Comodulated with GATA-1 SCL expression is almost exclusively hemopoietic in both human and murine tissues. Although SCL is frequently involved in T-ALL, there is no evidence that SCL is normally expressed in the T lymphocyte lineage. Transcripts are undetectable in thymus and peripheral blood T lymphocytes [19, 26, 361 and, although mRNA is found in a proportion of leukemic T cell lines, this reflects aberrant rearrangement events with consequent transcriptional activation of the SCL locus. In contrast, SCL mRNA is expressed at high levels in erythroid, mast and megakaryocytic cell lines. Moreover, similar high levels are seen in tissues rich in erythroid cells, in response to phenylhydrazine-induced hemolysis and in normal mast cells [j6,38]. SCL is thereforeexpressed in the three hemcpoietic lineages that also express GATA-1. Intriguingly, SCL and GATA- 1 are also coexpressed in IL-3dependent primitive myeloid cell lines [38], suggesting that both transcription factors may play a role early in the process of hemopoietic commitment. Induced terminal erythroid differentiation of murine erythroleukemia (MEL) cells is accompanied by a striking coordinate biphasic modulation of both SCL and GATA-1 mRNA expression [38]. These changes consist of an early fall followed by a delayed, sustained rise. The late rise precedes upregulation of multiple erythroid-specific genes and is likely to be an important part of the erythroid differentiation program. This conclusion is further supported by the observation that chemically induced myeloid differentiation of human leukemic cells is accompanied by a late, coordinate and sustained fall in GATA-1 and SCL mRNA levels [38, 391. Previous studies utilizing MEL cells have demonstrated that chemically induced differentiation produces altered expression of several transcription factors before irreversible commitment to differentiation is detected. The functional significance of the early transcription factor changes is unclear, but continued exposure to differentiation-inducerhexamethylene bisutamide (HMBA) results in an increasing proportion of MEL cells becoming committed to terminal differentiation(reviewed in [a]). The significance of the early fall in SCL and GATA-1 mRNA levels is also unclear, and these early changes are not restricted to transcription factors (transient modulation of erythropoietin receptor mRNA levels also

273 occurs) [38]. These observations suggest that HMBA induces rapid alterations in mRNA levels in a plethora of genes. However, while the significance of the early decrease in SCL and GATA- 1 mRNA is unclear, the later changes occur in a lineage-specific manner and have a likely role in determining the ultimate cellular phenotype [38, 391. One early alteration of particular interest was the transient upregulation of ID-1 mRNA levels in HMBA-induced MEL cells [38]. ID-1 is a member of the HLH family, but the protein lacks an upstream basic domain and is thus able to bind other HLH proteins and sequester them in inactive complexes (Fig. 1). In myogenesis the ID-1 protein inhibits differentiation by forming a complex with the protein products of the ubiquitously expressed E2A gene. This prevents the formation of heterodimers between the E2A product and the product of the muscle determination gene myoD and hence inhibits transcriptional activation of various muscle-specific genes [ 141. In erythroid differentiation of MEL cells, ID-I mRNA levels increase just before SCL and GATA-1 levels decrease, suggesting that ID-1 may regulate SCL and GATA-1 transcription by an analogous mechanism. In contrast to GATA-1 [41], SCL mRNA has also been observed in neural tissues and in the developing brain where it displays domain-like expression previously reported for homeobox genes 1381. The almost identical expression patterns seen for SCL and GATA-1 in hemopoietic tissues is therefore not reproduced in the brain. In developing neural tissues, SCL expression is confined to differentiating neural cells with predominant expression in the dorsal part of the metencephalon (comprising the floor of the fourth ventricle) and the roof of the mesencephalon. Intriguingly, we have also recently identified a novel HLH gene, called NSCL, that is closely related to SCL (61% amino acid identity in the HLH domain) but is predominantly expressed in neurological tissue [42]. The close relationship among SCL, LYL- 1, TAL-2 and NSCL in the HLH domain raises the possibility that these genes may belong to a subfamily of HLH proteins that are expressed in a tissue-specific manner.

SCL Is Implicated in Proliferation and SelfRenewal in Erythroid Cells In additional studies examining the role of SCL in erythroid cells, an antisense SCL construct significantly inhibited proliferation of the human erythroleukemia (HEL) cell line K562 [43]. Clonal cell lines electroporated with the antisense construct grew much more slowly than those that had incorporated the vector alone. K562 cells expressing antisense

SCL and Hemopoietic Differentiation

SCL also exhibited fewer and smaller colonies when grown in agar cultures, and recloning experiments demonstrated a striking suppression of self-renewal. Concomitant with these effects, an increase in spontaneous erythroid differentiation was noted in K562 cells expressing antisense SCL. Nonspecific toxicity was excluded by showing that the antisense vector did not influence the growth of Raji cells, a B cell line that does not express endogenous SCL mRNA. These effects of antisense SCL are remarkably similar to the action of certain hemopoietic growth factors in suppressing the growth of leukemic cell lines [44]. These experiments have two important implications. First, they strengthen the analogy between SCL and MYC. Both are implicated in the pathogenesis of lymphoid malignancies by virtue of their involvement in chromosomal rearrangements mediated by sequence-specific recombination mechanisms. Moreover, both encode HLH proteins and are therefore likely to function as sequence-specific transcription factors. These recent observations provide a third parallel between SCL and MYC, since there exists considerable evidence that implicates MYC in the regulation of cell proliferation and DNA replication [12]. In particular, antisense MYC constructs have been shown to inhibit proliferation and promote differentiation in a human leukemic cell line [45,46], observations which are entirely analogous to the situation with SCL. However, there exists at least one major difference between SCL and MYC: whereas MYC is ubiquitously expressed in all proliferating cells, SCL expression is restricted to a subset of hemopoietic lineages. This suggests that the nuclear mechanisms responsible for the regulation of proliferation may entail lineage-restricted as well as ubiquitous components. Secondly, these results provide the first evidence that SCL can influence self-renewal and proliferation in erythroid cells, a lineage in which SCL is normally expressed. Furthermore. they suggest one possible direct role SCL may play in providing a growth advantage when aberrantly expressed in T cells. The contrasting results obtained in the K562 and MEL experiments (SCL influenced proliferation of K562 cells but was implicated in differentiation of MEL cells) is analogous to the action of colony-stimulating factors (CSFs) that stimulate both proliferation and differentiation [ 11 and may also reflect intrinsic differences between the two cell lines. Although both exhibit erythroid features, they represent radically different hemopoietic progenitors. MEL cells are derived by infection of post-natal mice with the Friend retroviral complex, express adult globins and seem to represent an erythroprotein-responsivestage of

Greeaegley adult erythroid differentiation somewhere between the erythroid burst-forming unit (BFU-E) and erythroid colony-forming unit (CFU-E) [40]. In contrast, K562 cells more closely resemble fetal or embryonic erythroblasts since they express fetal and embryonic globin, a fetaVembryonic LDH isoenzyme pattern and the i antigen [47].Together with our previous suggestion that SCL transcript patterns may differ in distinct erythroid cell populations, these results raise the possibility that both the function and regulation of SCL may vary in distinct populations of erythroid precursors. Based on these results, it seems likely that SCL plays an important role in proliferation and differentiation events in erythroid/megakaryocyte/mast cell lineages. In contrast, in developing neural cells SCL is implicated in differentiation events alone. Furthermore, aberrant expression of SCL (or LYL- l ) in T lymphocytes is associated with leukemogenesis. Thus, studies over the last few years have provided intriguing insights into this fascinating new class of transcriptional regulators.

Acknowledgments This work was supported by grants from the Victorian Health Promotion Foundation, The AntiCancer Council of Victoria and the National Health and Medical Research Council, Canberra, Australia. A. R. Green is a Wellcome Senior Fellow in Clinical Science.

References 1 Metcalf D. The hemopoietic colony stimulating

factors. Elsevier, Amsterdam, 1984.

2 Darnell JE. Variety in the level of gene control in eukaryotic cells. Nature 1982;297:365-371. 3 Murre C, McCaw PS, Baltimore D. A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, myoD and myc proteins. Cell 1989;56:777-783. 4 Weintraub J, Davis R, Tapscott S, Thayer M, Krause

M. Benzra R, Blackwell TK, Turner D, Rupp R, Hullengerg S, Zhuang Y, Lassar A. The myoD gene family: nodal point during specification of the muscle cell lineage. Science 1991;251:761-766. 5 Thisse B, Stoetzel C, Gorostiza-Thisse C, PemnSchmitt F. Sequence of the twist gene and nuclear

localisation of its protein in endomesodermal cells of early Drosophila embryos. EMBO J 1988;7: 2175-2183.

214

6 Parkhust SM. Bopp D, Ish-Horowicz D. X: a ratio, the primary sex-determining signal in Drosophila, is transduced by helix-loop-helix proteins. Cell 1990; 63:1179-1191. 7 Voronova A. Baltimore D. Mutations that disrupt DNA binding and dimer formation in the E47 helixloop-helix protein map to distinct domains. Proc Natl Acad Sci USA 1990;87:4722-4726. 8 Villares R, Cabrera CV. The achaete-scute gene complex of D melanogasrer: conserved domains in a subset of genes required for neurogenesis and their homology to myc. Cell 1987;50:415-424. 9 Murre C, McCaw PS, Vaessin H, Caudy M, Jan LY, Jan YN, Cabrera CV, Buskin JN, Hauschka SD, Lassar AB,Weintraub H, Baltimore D. Interactions between heterologous helix-loop-helix proteins generate complexes that bind specifically to a common DNA sequence. Cell 1989;58:537-544. 10 Blackwood EM, Eisenmann RN. Max: a helix-loophelix zipper protein that forms a sequence-specific DNA-binding complex with myc. Science 1991;251: 1211-1217. 11 Prendergast GC, Lawe D, Ziff ER. Association of myn, the murine homolog of m a with c-myc stimulates methylation-sensitive DNA binding and ras cotransformation. Cell 1991;65:395-408. 12 Liischer B, Eisenmann RN. New light on myc and myb. Part 1. Myc. Genes Dev 1990;4:2025-2035.

13 Beckman H, Kadesch T. The leucine zipper of TFE3 dictates helix-loop-helix dimerization specificity. Genes Dev 1991;5:1057-1066.

14 Benezra R, Davis RL, Lockshon D, Turner DL, Weintraub H. The protein ID: a negative regulator of helix-loophelix DNA binding proteins. Cell 1990; 61~49-59. 15 Christy BA, Saunders LK, Lau LF, Copeland NG. Jenkins NA, Nathans D. An Id-related helix-loophelix protein encoded by a growth factor-inducible gene. Proc Natl Acad Sci USA 1991;88:1815-1819. 16 Sun X-H.Copeland NG, Jenkins NA, Baltimore D. Id proteins Id1 and Id2 selectively inhibit DNA binding by one class of helix-loop-helix proteins. Mol Cell Biol 1991;11:5603-5611. 17 Spencer CA. Groudine M.Control of c-myc regulation in normal and neoplastic cells. Adv Cancer Res 199136:1-48. 18 Begley CG,Aplan PD, Davey MP. Nakahara K, Tchon K, Kurtzberg J, Herschfield MS, Haynes BE Cohen DI, Waldmann TA. Kirsch IR. Chromosomal translocation in a human stem-cell line disrupts the T-cell antigen receptor &chain diversity region and results in a previously unreported fusion transcript. Proc Natl Acad Sci USA 1989;86:2031-2035. 19 Begley CG, Aplan PD. Haynes B, Waldmann TA, Kirsch IR.The gene SCL is expressed during early

SCL and Hemopoietic Differentiation

275 hematopoiesis and encodes a differentiation-related DNA-binding motif. Proc Natl Acad Sci USA 1989; 86: 10128-10132.

sistent fusion of E2A and PBXl coding sequences in acute lymphoblastic leukemia. Blood 1991;77: 681-693.

20 Bernard 0, Guglielmi P, Jonveaux P, Cherif D, Gisselbrecht S, Mauchauffe M, Berger R, Larsen CJ, Mathien-Mahul D. Two distinct mechanisms for the SCL gene activation in the t(1;14) translocation of T-cell leukaemias. Genes, Chrom and Cancer 1990; 1:194-208. 21 Chen Q,Cheng J-T, Tsai L-H, Schneider N, Buchanan G, Carroll A, Crist W, Ozanne B, Siciliano W, Baer R. The ral gene undergoes chromosome translocation in T cell leukemia and potentially encodes a helix-loop-helix protein. J Exp Med 1990;172: 1403-1408.

32 Reid L. From gradients to axes, from morphogenesis to differentiation. Cell 1990;63:875-882.

22 Aplan PD, Begley CG. Bertness V, Nussmeier MA, Ezguerra A, Coligan J, Kirsch IR. The SCL gene is formed from a transcriptionally complex locus. Mol Cell Biol 1990;10:6426-6435. 23 Brown L, Cheng J-T, Chen Q, Siciliano MJ, Crist W, Buchanan G, Baer R. Site-specific recombination of the tul-1 gene is a common occurrence in human T cell leukemia. EMBO J 1990;9:3343-3351. 24 Aplan PD, Lombardi DP, Reaman GH, Sather HN, Hammond GD, Kirsch IR.Involvement of the putative hematopoietic transcription factor SCL in Tcell acute lymphoblastic leukemia. Blood 1992;79: 1327-1333. 25 Mellentin JD, Smith SD. Cleary ML. Lyl-1, a novel gene altered by chromosomal translocation in T cell leukemia, codes for a protein with a helix-loophelix DNA binding motif. Cell 1989;58:77-83. 26 Visvader J. Begley CG, Adams JM. Differential expression of the L n , SCL and E2A helix-loophelix genes within the hemopoietic system. Oncogene 1991;6: 195-204. 27 Kuo SS, Mellentin JD, Copeland NG, Gilbert DJ, Jenkins NA, Cleary ML. Structure, chromosome mapping and expression of the mouse lyl-1 gene. Oncogene 1991;6:961-968. 28 Nourse J, Mellentin JD, Galili N, Wilkinson J, Stanbridge E. Smith S , Cleary ML. Chromosomal translocation t( 1;19) results in synthesis of a homeobox fusion mRNA that codes for a potential chimeric transcription factor. Cell 1990;60535-545.

33 Begley CG, Visvader J, Green AR, Aplan PD, Metcalf D. Kirsch IR, Gough NM. Molecular cloning and chromosomal localization of the murine homolog of the human helix-loop-helix gene SCL. Proc Natl Acad Sci USA 1991;88:869-873. 34 Goodwin G. MacGregor A, Zhu J, Crompton MR. Molecular cloning of the chicken SCL cDNA. Nucleic Acids Res 1992;20:368. 35 Bernard 0, Lecointe N, Jonveaux P, Souyri M, Mauchauffe M, Berger R, Larsen CJ, Mathieu-Mahul D. Two site-specific deletions and t( 1;14) translocation restricted to human T-cell acute leukemias disrupt the 5’ part of the ral-1 gene. Oncogene 1991;6: 1477- 1488. 36 Green AR, Salvaris E, Begley CG. Erythroid expression of the “helix-loop-helix” gene, SCL. Oncogene 1991;6:475-379. 37 Hsu HL, Cheng JT, Chen Q, Baer R. Enhancerbinding activity of the ral-1 oncoprotein in association with the E47E12 helix-loop-helix proteins. Mol Cell Biol 1991;11:1912-1920. 38 Green AR, Lints T. Visvader J, Harvey R. Begley CG. SCL is coexpressed with GATA-1 in hemopoietic cells but is also expressed in developing brain. Oncogene 1992;7:653-660. 39 Green AR, Rockman S. DeLuca E, Begley CG. Induced myeloid differentiation of K562 cells with downregulation of erythroid and megakaryotic transcription factors: a novel experimental model for hemopoietic lineage restriction. 1992. Submitted. 40 Marks PA, Rifkind RA. Induced differentiation of erythroleukaemia cells by hexamethylene bisacetamide: a model for cytodifferentiation of transformed cells. Env Health Pers 1989;80:181-188. 41 Whitelaw E, Tsai S-F, Hogben P, Orkin SH. Regulated expression of globin chains and the erythroid transcription factor GATA-1 during erythroporesis in the developing mouse. Mol Cell Biol 1990;lO: 6596-6606.

29 Kamps MP. Murre C, Sun X-H, Baltimore D. A new homeobox gene contributes the DNA binding domain of the t(1;19) translocation protein in preB-ALL. Cell 1990;60:547-555.

42 Begley CG, Lipkowitz S, Gobel V, Mahon KA, Bertness V, Green AR, Gough NM,Kirsch IR. Molecular characterization of a helix-loop-helix gene NSCL that is expressed in the developing nervous system. Proc Natl Acad Sci USA 1991;89:38-42.

30 Mellentin JD. Nourse J, Hunger SP, Smith SD, Cleary ML. Molecular analysis of the t( 1;19) breakpoint cluster region in pre-B cell acute lymphoblastic leukemia. Genes, Chrom and Cancer 1990,2:239-247.

43 Green AR, DeLuca E. Begley CG. SCL suppresses self-renewal and enhances spontaneous erythroid differentiation of the human leukemic cell line K562. EMBO J 1991;10:4153-4158.

31 Hunger SP. Galili N, Carroll AJ, Crist WM,Link MP, Cleary ML. The t(1;19)(q23p13) results in con-

44 Begley CG, Nicola NA, Metcalf D. Purified colony stimulating factors (G-CSF and GM-CSF) induce

GreenBegley differentiation in human HL60 leukemic cells with suppression of clonogenicity. Int J Cancer 1987;39: 99-105. 45 Holt JT, Rodner RL, Nienhuis AW. An oligomer complementary to c-myc mRNA inhibits proliferation of HL-60 promyelocytic cells and induces differentiation. Mol Cell Biol 1988;8:963-973. 46 Wickstrom EL, Bacon TA, Gonzalez A, Lyman GH, Wickstrom E. Anti c-myc DNA increases differentiation and decreases colony formation by HL-60 cells. In Vitro Cell & Devel Biol 1989;24:297-302.

276 47 Benz El, Mumane MJ, Tonkonow BL, Berman BW, Mazur EM, Cavallesco C. Jenko T, Snyder EL, Forget BG, Hoffman R. Embryonic-fetal erythroid characteristics of a human leukemic cell line. Proc Natl Acad Sci USA 1980;88:869-873. 48 Xia Y, Brown L, Yang CY-C. Tsan JT, Siciliano MJ, Espinosa R, LeBeau MM, Baer RJ. Tal-2, a helix-loop-helix gene activated by the (7;9)(q34;q32) translocation in human T-cell leukemia. Proc Natl Acad Sci USA 1991;88:11416-11420.