Report

A TEL-JAK2 Fusion Protein with Constitutive Kinase Activity in Human Leukemia

See allHide authors and affiliations

Science  14 Nov 1997:
Vol. 278, Issue 5341, pp. 1309-1312
DOI: 10.1126/science.278.5341.1309

Abstract

The Janus family of tyrosine kinases (JAK) plays an essential role in development and in coupling cytokine receptors to downstream intracellular signaling events. A t(9;12)(p24;p13) chromosomal translocation in a T cell childhood acute lymphoblastic leukemia patient was characterized and shown to fuse the 3′ portion ofJAK2 to the 5′ region of TEL, a gene encoding a member of the ETS transcription factor family. The TEL-JAK2 fusion protein includes the catalytic domain of JAK2 and the TEL-specific oligomerization domain. TEL-induced oligomerization of TEL-JAK2 resulted in the constitutive activation of its tyrosine kinase activity and conferred cytokine-independent proliferation to the interleukin-3–dependent Ba/F3 hematopoietic cell line.

Chromosomal translocations in human tumors frequently produce fusion genes whose chimeric protein products play an essential role in oncogenesis (1).TEL is located on chromosome 12p13 and is frequently involved in chromosomal translocations seen in a variety of human leukemias (2-4). We observed a t(9;12)(p24;p13) translocation in a 4-year-old male with T cell acute lymphoblastic leukemia (ALL) in relapse (Fig. 1A). The involvement of TEL was investigated by Southern (DNA) blot analysis with a genomic probe corresponding to intron 5 ofTEL (5, 6). In addition to the expected germline bands, rearranged fragments were detected in the Eco RI and Bam HI digests of the patient's DNA, suggesting that the chromosome 12 breakpoint localized to TEL intron 5 (Fig. 1B). To identify the fusion partner on chromosome 9, we used anchored polymerase chain reaction (PCR) to characterize the fusion transcript in leukemic cells. Sequence analysis of the amplified cDNA clones identified a non-TEL sequence fused in frame upstream of TEL exon 6. This sequence showed extensive similarity to the genes encoding the JAK family of protein tyrosine kinases, with the greatest similarity to murine Jak2 (7).

Figure 1

Identification and molecular analysis of t(9;12)(p24;p13). (A) Partial karyotype showing the t(9;12) (RHG bands) of a T cell ALL patient in relapse with 46,XY,del(6) (q14q27),t(9;12)(p24;p13)[12]/46,XY[3]. Arrows show the breakpoints on the rearranged chromosomes. Blast cells were CD2+, CD5+, and CD7+. (B) Southern blot analysis of the TEL locus in t(9;12)(p24;p13). Genomic DNA of the patient's t(9;12)-positive cells and of the control (C) HL-60 cell line was analyzed with aTEL genomic probe [probe B in (5)]. Arrows indicate the rearranged fragments in the Eco RI and Bam HI digests. The Eco RI fragments detected in control DNA are polymorphic. No TEL rearrangement was detected at diagnosis (8). (C) RT-PCR analysis of TEL,JAK2, TEL-JAK2, and JAK2-TELtranscripts in the patient [t(9;12)] and the control (C) sample. BothTEL-JAK2 [298 nucleotides (nt)] and JAK2-TEL(355 nt) junction DNA fragments were amplified from the patient's cDNA but not from the cDNA of the control K562 cell line. TEL(313 nt) and JAK2 (340 nt) fragments were amplified from both samples. Numbers at right are molecular size markers (in nucleotides).

Using these putative JAK2 cDNA sequences, we isolated lambda phage clones from a normal human genomic library and showed by fluorescent in situ hybridization (FISH) that they hybridized to chromosome 9p24 (8), the chromosomal location of humanJAK2 (9). Partial sequence analysis of these genomic clones allowed us to design JAK2 oligonucleotides corresponding to sequences on either side of the chromosome 9 breakpoint. These oligonucleotides were used together withTEL-specific primers to analyze RNA from leukemic or control cells by reverse transcriptase (RT)–PCR (10). BothTEL-JAK2 and JAK2-TEL cDNAs were specifically amplified from patient cells but not from control cells (Fig. 1C). The sequence of the amplified human JAK2 was 90% identical to that of murine Jak2 at the nucleotide level and 96% identical at the protein level over a stretch of 114 amino acids (Fig.2A). These results establish that humanJAK2 is the gene fused to TEL as a result of the t(9;12) translocation.

Figure 2

(A) Nucleotide and deduced amino acid sequences of fused and normal TEL andJAK2. The fusion occurs within codon 337 of TEL(4) and codon 811 of JAK2(7). Intronic sequences are shown in lowercase letters. (B) Nucleotide sequence of the RT-PCR–amplified junctions of the TEL-JAK2 fusion in different leukemic patients (26). Patient 1 is the patient analyzed in this study. The karyotype and the details of the structural analysis of theTEL-JAK2 fusions in patients 2 and 3 are reported elsewhere (18). No reciprocal JAK2-TEL fusion transcript was detected in patients 2 and 3 (18). (C) Schematic structure of TEL and JAK2 proteins, together with the TEL-JAK2 fusion protein and derivatives used in this study. Arrows indicate the locations of the t(9;12) fusion points in the different patients. Fusion points in patient 1 are shown by black arrows. Fusion points in patients 2 and 3 are shown by open arrows. TheTEL-JAK2 cDNA as characterized in patient 1 was reconstructed to encode the 336 NH2-terminal residues of human TEL and the 318 COOH-terminal residues of murine JAK2 by PCR-mediated amplification of the appropriate regions ofTEL and JAK2 cDNAs and subsequent subcloning of the products. The nucleotide sequence of the amplified cDNA was determined and found to be devoid of mutations. Arrowheads indicate the two alternative TEL initiation codons identified previously (27). Black ovals indicate the hemagglutinin (HA) epitope tags. K → R indicates the mutation of lysine to arginine. Single-letter abbreviations for the amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.

JAK2 is a widely expressed protein tyrosine kinase that associates with the intracellular domains of a number of cytokine receptors and is essential to receptor function (11, 12). JAK2 shares with other members of the JAK family (such as JAK1, JAK3, and TYK2) seven regions of homology, referred to as JH1 to JH7 (Fig. 2C) (11, 13). JH1 is the catalytic domain, whereas JH2 to JH7 appear to be involved in protein-protein interactions and in the specificity of the different JAK family members (14,15). In the TEL-JAK2 fusion, the NH2-terminal TEL sequences are fused to the kinase JH1 domain (Fig. 2C). TEL is a member of the ETS family of transcription factors. In addition to a COOH-terminal ETS domain that is conserved in all ETS proteins, TEL specifically contains a 60–amino acid homotypic oligomerization domain at its NH2-terminus (16, 17). Fusion of TEL and JAK2 has been observed in two other leukemic patients, which also resulted in a TEL-JAK2 fusion transcript but not in the expression of the reciprocalJAK2-TEL fusion (Fig. 2B) (18). The conserved feature of the fusion proteins encoded by the TEL-JAK2transcripts characterized in all three patients is the presence of both the TEL oligomerization domain and the JAK2 catalytic domain (Fig. 2C).

To investigate the oligomerization properties and protein kinase activity of TEL-JAK2, we compared the properties of intact TEL-JAK2, in which the 336 NH2-terminal residues of TEL are fused to 318 COOH-terminal residues of JAK2, with those of two mutants: TELΔ-JAK2, in which the TEL oligomerization domain (amino acids 40 to 117) was deleted, and TEL-JAK2 K, a kinase-defective mutant in which a lysine residue essential to the catalytic function of JAK2 (K882) (19) was changed to arginine (Fig. 2C).

Previous experiments have shown that TEL-induced oligomerization as it occurs in vivo can be analyzed with in vitro–translated proteins (16, 17). As expected, in vitro–translated TEL and TEL-JAK2 were specifically immunoprecipitated by antibodies directed against the COOH-terminal domains of TEL [antibody to C-TEL (anti–C-TEL)] and JAK2 (anti-JAK2), respectively (Fig.3A, top). When coexpressed, TEL and TEL-JAK2 formed a complex because they were both immunoprecipitated by either of the antibodies (Fig. 3A, bottom). In vitro association of TEL and TEL-JAK2 required the presence of the TEL oligomerization domain, because neither the TELΔ-JAK2 mutant nor the full-length JAK2 coimmunoprecipitated with TEL (Fig. 3A, bottom). These experiments show that the TEL oligomerization domain is functional in the TEL-JAK2 fusion.

Figure 3

Oligomerization and tyrosine kinase activity of TEL-JAK2. (A) In vitro oligomerization of TEL-JAK2. pBS expression plasmids encoding JAK2, TEL, TEL-JAK2, or TELΔ-JAK2 were in vitro translated alone or in combination, as indicated at the top of each panel, with the TNT kit (Pharmacia) in the presence of l-(35S)methionine. Translated proteins were analyzed directly (left lanes) or with either anti-JAK2 or anti–C-TEL, as indicated at the bottom of each panel (middle and right lanes). The antibodies used were rabbit polyclonal antibodies raised against amino acids 1110 to 1129 of JAK2 (anti-JAK2; Santa Cruz Biotechnology, Santa Cruz, California) and against the COOH-terminal end of TEL (anti–C-TEL) (27). Radiolabeled proteins were visualized on denaturing polyacrylamide gels by fluorography with Amplifier (Amersham). The TEL and TEL-JAK2 proteins migrate as doublets, because of alternative translational initiation at two in-frame ATGs (27). (B) Requirement of oligomerization for activation of TEL-JAK2 tyrosine kinase activity. In vitro–translated TEL-JAK2 and TEL-JAK2 mutants were analyzed by immunoblotting with the phosphotyrosine-specific 4G10 monoclonal antibody (anti-PTyr). After stripping of the membrane, expression levels of the proteins were analyzed with a rabbit antiserum directed to the NH2-terminal domain of TEL (anti–N-TEL) (27).

To investigate whether TEL-induced oligomerization results in activation of TEL-JAK2 tyrosine kinase activity, we compared the autophosphorylation of JAK2, TEL-JAK2, and TEL-JAK2 mutants. In vitro–translated proteins were analyzed by immunoblotting with a phosphotyrosine-specific antibody (anti-PTyr). For comparison of expression levels, the membranes were stripped and reprobed with antibodies specific to the NH2-terminal domain of TEL (anti–N-TEL) (Fig. 3B). A high level of tyrosine phosphorylation occurred only with TEL-JAK2. The phosphorylation of TEL-JAK2 reflected its intrinsic tyrosine kinase activity because the TEL-JAK2 K mutant was not phosphorylated. Constitutive tyrosine kinase activity of TEL-JAK2 required TEL-induced oligomerization because the oligomerization-defective TELΔ-JAK2 mutant was inactive.

To assay the transforming properties of TEL-JAK2, we stably expressed the cDNAs encoding JAK2, TEL-JAK2, and the TEL-JAK2 mutants in the murine Ba/F3 hematopoietic cell line, which is strictly dependent on interleukin-3 (IL-3) for survival and proliferation. Mock-transfected cells and transfectants harboring the pBabeNeo vector alone were used as controls. Analysis of transfectants obtained after G418 selection in the presence of IL-3 showed that expression of exogenous proteins was comparable with that of endogenous JAK2 (Fig.4A). Consistent with our in vitro data, only the oligomerization-competent TEL-JAK2 was tyrosine phosphorylated in vivo (Fig. 4A). To assay the ability of TEL-JAK2 to confer IL-3–independent proliferation, we deprived Ba/F3 transfectants of IL-3 and seeded the cells in 96-well trays (20). In contrast to control cells, which died by apoptosis under these conditions (0 of 96 wells displaying proliferating cells), the TEL-JAK2 transfectants showed sustained proliferation (96 of 96 wells with proliferating cells). Similar results were obtained when cell proliferation was monitored by daily cell counting, although the rate of proliferation in the absence of IL-3 was slightly less than that in its presence. TEL-JAK2–induced proliferation required the constitutive activation of its protein tyrosine kinase activity, because Ba/F3 cells transfected with TELΔ-JAK2, TEL-JAK2 K, and wild-type JAK2 did not proliferate in the absence of IL-3 (0 of 96 wells showing proliferating cells).

Figure 4

Induction of IL-3–independent proliferation of Ba/F3 cells by TEL-JAK2. (A) The cDNAs encoding JAK2, TEL-JAK2, or the TEL-JAK2 mutants were subcloned into the pBabeNeo retroviral expression vector (28). These constructs and the pBabeNeo control were introduced by electroporation into the murine IL-3–dependent lymphoid Ba/F3 cells, and stably transfected cells were selected in the presence of G418 and IL-3. For evaluation of protein expression and phosphorylation, lysates of transfected cells were immunoprecipitated with either anti–N-TEL or anti-JAK2 (IP). After electrophoresis in denaturing polyacrylamide gels, proteins were blotted to nitrocellulose, and the blot was probed with anti-HA, anti-JAK2, or anti-PTyr (blot). TEL-JAK2 and TEL-JAK2 mutants are indicated by solid arrowheads. JAK2 is indicated by an open arrowhead. (B) Constitutive STAT5 activation in Ba/F3 cells expressing TEL-JAK2. Electrophoretic mobility–shift assay with a β-casein probe showed a specific STAT5 complex in lane 1 (arrow) (29). Specificity was established by the disappearance of the complex on competition with a 100-fold molar excess of the unlabeled β-casein probe (lane 3), but not after competition with the same molar excess of a nonspecific unlabeled probe (lane 4). Furthermore, addition of anti-STAT5 supershifted the complex (lane 2; the supershift is indicated by an arrowhead).

JAK2 couples cytokine receptors to downstream signaling events that control cell survival, proliferation, and differentiation by transiently activating several pathways, including RAS, phosphatidylinositol 3-kinase, and STAT5 (signal transducer and activator of transcription 5). However, the respective role and potential redundancy of these pathways in these cellular responses are presently unclear (11, 13).

An inhibitor study has shown that JAK2 is constitutively activated in human ALL cells (21), but no causal link to leukemogenesis has been established. A JAK pathway has been implicated in a leukemia-like defect in the fruit fly Drosophila(22-24). Point mutations in the hopscotch locus, which encodes a JAK homolog, cause the hyperactivation of JAK activity and, in turn, phosphorylation and activation ofDrosophila STAT. The role of STAT proteins in theDrosophila leukemia-like defect has also been genetically established (23). Ba/F3 cells expressing TEL-JAK2 show constitutive activation of STAT5 (Fig. 4B), suggesting that alteration of the normal JAK2 signaling pathway participates in the deregulation of cell proliferation. Alternatively, TEL-JAK2 could phosphorylate protein substrates that are not normally involved in JAK2 signaling but are critical to its oncogenic properties.

REFERENCES AND NOTES

View Abstract

Navigate This Article