Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Translocations involving anaplastic lymphoma kinase (ALK)

Abstract

Anaplastic large-cell lymphoma (ALCL) comprises a group of non-Hodgkin's lymphomas (NHLs) that were first described in 1985 by Stein and co-workers and are characterized by the expression of the CD30/Ki-1 antigen (Stein et al., 1985). Approximately half of these lymphomas are associated with a typical chromosomal translocation, t(2;5)(p23;q35). Much confusion about the exact classification and clinicopathological features of this subgroup of NHL was clarified with the identification of NPM–ALK (nucleophosmin-anaplastic lymphoma kinase) as the oncogene created by the t(2;5) (Morris et al., 1994). With the discovery of NPM–ALK as the specific lymphoma gene mutation, this NHL subtype could be redefined on the molecular level. This achievement was enhanced by the availability of specific antibodies that recognize ALK fusion proteins in paraffin-embedded lymphoma tissues. Several excellent recent reviews have summarized the histopathological and molecular findings of ALCL and their use in the classification of this lymphoma entity (Anagnostopoulos and Stein, 2000; Benharroch et al., 1998; Drexler et al., 2000; Foss et al., 2000; Gogusev and Nezelof, 1998; Kadin and Morris, 1998; Ladanyi, 1997; Morris et al., 2001; Shiota and Mori, 1996; Skinnider et al., 1999; Stein et al., 2000). This review will focus on the molecular function and signal transduction pathways activated by ALK fusion oncogenes, with recent advances and possible clinical implications to be discussed.

ALCL

ALCL belongs to the group of high-grade NHLs and typically presents as an aggressive systemic disease, with or without extranodal involvement (for reviews see Kadin, 1997; Kinney and Kadin, 1999; Morris et al., 2001; Skinnider et al., 1999; Stein et al., 2000). ALCL tumor cells are characterized by the expression of the CD30/Ki-1 antigen. These lymphomas normally arise from T-cells; in some cases, however, neither T-cell markers nor T-cell receptor rearrangements can be detected, leading to a null-cell categorization. Rare cases of ALK-positive ALCL have been reported to be of B-cell phenotype (Gascoyne et al., 1999); however, most investigators do not consider these B-cell lymphomas to be classical ALCLs. For example, the REAL and the new WHO classifications exclude B-cell cases from the entity of ALCL and place these cases into a different category (diffuse large B-cell lymphoma) (Harris et al., 1994). ALCL arises most frequently in children and young adults, but also occurs in older (60 years) patients. These lymphomas have a broad morphological spectrum, ranging from cases dominated by small tumor cells to cases marked by very large tumor cells, as well as combinations thereof (Benharroch et al., 1998; Falini et al., 1998). A histological hallmark found in most cases is the common classical ALCL cell, which is characterized by an eccentric localized nucleus with a kidney-like shape (Benharroch et al., 1998). The histological variability of ALCL led to considerable difficulties in the unequivocal diagnosis of this entity in the past.

The genes altered in the classical t(2;5) translocation were identified and cloned in 1994 (Morris et al., 1994). Morris et al. (1994) demonstrated that nucleophosmin (NPM) was fused to ALK (anaplastic lymphoma kinase), a previously unknown tyrosine kinase. With the development of sensitive diagnostic tools to detect ALK, the percentage of CD30+ ALCL cases carrying ALK rearrangements has been rising in recent years (for a table summarizing the frequency of ALK positivity in ALCL see Morris et al., 2001). Depending on the histological subtype (ranging from large cell- to small cell-predominant), ALK rearrangements can be demonstrated in about 60% of all CD30-positive ALCL (Morris et al., 2001). Within specific ALCL subtypes, 30–50% of pleomorphic ALCL (Kinney et al., 1996; Kinney and Kadin, 1999), more than 80% of monomorphic ALCL (Kinney et al., 1996), 75–100% of small-cell cases (Falini et al., 1998; Kinney et al., 1996), and 60–100% of lymphohistiocytic ALCL (Falini et al., 1998; Pileri et al., 1997) have been shown to be ALK-positive. A higher incidence of ALK-positive cases is consistently found in children and young adults than older patients (Falini et al., 1999a; Nakamura et al., 1997; Pulford et al., 1997). In most clinical trials published, ALK-positive lymphomas (so-called ‘ALKomas') have had a better prognosis than ALK-negative tumors (Nakamura et al., 1997; Sandlund et al., 1994a,b; Shiota et al., 1995). This difference in prognosis may be due in part, however, to the younger age of the ALK-positive ALCL population.

Nucleophosmin (NPM)/B23

NPM/B23 is a 38-kD nucleolar protein (Figure 1) encoded on chromosome 5 that contains bipartite nuclear localization signals (NLS) (Michaud and Goldfarb, 1991), binds nuclear proteins, and engages in cytoplasm/nuclear trafficking (Borer et al., 1989; Szebeni et al., 1995; Valdez et al., 1994). The protein has been suggested to be involved in the assembly and intranuclear transport of preribosomal particles (Borer et al., 1989; Yung et al., 1985). Because it is phosphorylated by casein kinase II in interphase (Chan et al., 1990) and by p34 cdc2 in mitosis (Peter et al., 1990), an active role for NPM/B23 in the cell cycle has been proposed. A decisive involvement of NPM/B23 in controlling centrosome duplication initiated by CDK2/cyclin E-mediated phosphorylation in the events of mitosis was revealed recently. NPM/B23 associates with unduplicated centrosomes and dissociates upon its phosphorylation by CDK2/cyclin E. Thus, NPM/B23 functions as a target of CDK2/cyclin E in the initiation of centrosome duplication (Okuda et al., 2000). In cells entering mitosis (while the nucleoli disassemble), NPM/B23 relocates to the cytoplasm at the periphery of the chromosomes and migrates to the poles of the mitotic spindle (Ochs et al., 1983; Zatsepina et al., 1999). NPM/B23 normally undergoes self-oligomerization (Chan and Chan, 1995), as well as hetero-oligomerization with NPM–ALK, which leads to the nuclear localization of the NPM–ALK protein (Bischof et al., 1997; Liu and Chan, 1991), the NPM portion of which does not contain any NLS (see section Subcellular localizations of ALK fusion proteins). Interestingly, several cases of the t(5;17) variant translocation of acute promyelocytic leukemia (APL) have been found to fuse the NPM gene to the retinoic acid receptor (RAR) gene (Brunel et al., 1995; Redner et al., 1996). Within this entity, the NPM portion is implicated to be serving in oligomerization with wild-type NPM (Liu and Chan, 1991) and in mediating the homo-oligomerization of NPM–RAR1, therefore contributing to the aberrant transcriptional activity of this fusion protein (Redner et al., 1996). NPM is involved in yet another oncogenic fusion, the NPM–MLF1 chimera formed by the t(3;5) in myelodysplastic syndrome and acute nonlymphocytic leukemia (Yoneda-Kato et al., 1996), in which the NPM segment mediates NPM–MLF1 self-association, hetero-association with normal NPM, and aberrant nucleoplasmic/nucleolar localization of the predominantly cytoplasmic normal MLF1 protein. In the case of NPM–ALK, the strong ubiquitous NPM promoter drives the high-level expression of NPM–ALK in lymphoma cells, and oligomerization mediated by the NPM segment leads to the activation of the NPM–ALK fusion protein.

Figure 1
figure1

Schematic illustration of the NPM–ALK fusion protein resulting from the t(2;5) chromosomal translocation. Fusion of the chromosome 5 gene encoding nucleophosmin (NPM) to the chromosome 2 gene encoding anaplastic lymphoma kinase (ALK) results in the expression of a constitutively activated chimeric tyrosine kinase, NPM–ALK. NPM contains an oligomerization domain (OD; residues 1–117), a metal binding domain (MB; residues 104–115), two acidic amino acid clusters (AD: Asp/Glu-rich acidic domain; residues 120–132 and 161–188) that function as acceptor regions for nucleolar targeting signals, and two nuclear localization signals (NLS; residues 152–157 and 191–197). ALK contains a MAM (Meprin/A5/protein tyrosine phosphatase Mu) domain, a region of about 170 amino acids present in the extracellular portions of a number of functionally diverse proteins that may have an adhesive function (PROSITE database: PDOC 00604, Jiang et al., 1993), in its extracellular segment (ALK residues 480–635). The putative ligand binding site (LBS) for pleiotrophin (ALK residues 391–401) is indicated. TM: transmembrane domain; TK: tyrosine kinase catalytic domain

ALK (anaplastic lymphoma kinase)

The ALK gene on chromosome 2p23 codes for a receptor tyrosine kinase (RTK). Two groups independently cloned the human and mouse ALK receptor (Iwahara et al., 1997; Morris et al., 1997), which was shown to code for a protein with a predicted molecular mass of 180 kD. N-linked glycosylation of ALK leads to a glycoprotein that migrates in SDS gels at about 210 kD. ALK has the typical structure of a transmembrane RTK, with a large extracellular domain, a lipophilic transmembrane segment, and a cytoplasmic tyrosine kinase domain (Figure 1). Only the cytoplasmic tyrosine kinase domain is fused to NPM in NPM–ALK.

The extracellular domain of ALK is highly similar to the extracellular domain of leukocyte tyrosine kinase (LTK), placing ALK within the family of insulin RTKs. Other RTKs with significant homology to ALK include the IGF-1R and cROS (Figure 2). Computer analysis has also identified a Drosophila protein, suggesting that ALK is an evolutionarily conserved tyrosine kinase (Figure 2). Despite their high homology, the ALK receptor extracellular domain is much larger than the extracellular segment of LTK (1033 amino acids (aa) versus 421 aa). Both extracellular domains contain a conserved cysteine-rich EGF-like motif and a glycine-rich region (Iwahara et al., 1997; Morris et al., 1997). In the part of the extracellular domain present only in ALK, no common motif is present except for an amino acid sequence also found in the LDL receptor (ACDFXXDCAXGED) (Iwahara et al., 1997); the function of this motif is unclear. By Northern blotting, expression of Alk was detected in the murine brain and spinal cord (Iwahara et al., 1997; Morris et al., 1997). Immunoblotting showed highest expression in the murine neonatal brain and to a lesser extent in the adult brain. RNA in situ hybridization revealed expression in specific regions of the developing brain, with highest expression detectable in thalamus, mid-brain, olfactory bulb and selected cranial, as well as dorsal root, ganglia of mice (Iwahara et al., 1997; Morris et al., 1997). Expression of Alk in hematopoietic tissues could not be demonstrated. This restricted expression pattern of ALK has been confirmed in human tissues (Pulford et al., 1997; Shiota et al., 1994b). The fact that the most abundant expression of ALK occurs in the neonatal brain suggests a role for the receptor in brain development. For example, ALK may serve as a receptor for a yet unidentified neurotrophic factor. This hypothesis is supported by the observation that ALK expression partially overlaps with the expression of the TRK family of neurotrophin receptors (Barbacid, 1995).

Figure 2
figure2

Amino acid sequence alignments of NPM–ALK, hLTK, dALK, hIGF-1R and c-ROS. The tyrosine kinase catalytic domain of NPM–ALK is shown from aa 178 to 440. hLTK: Human leukocyte tyrosine kinase; dALK: Drosophila homologue of anaplastic lymphoma kinase; hIGF-1R: Human insulin-like growth factor 1 receptor; c-ROS: proto-oncogene c-ROS, a receptor tyrosine kinase related to the viral oncoprotein v-Ros (Neckameyer et al., 1986). : ATP-binding site; ▪: Active site; •: NPM–ALK tyrosine residues Y338, Y342 and Y343

A putative ALK ligand, pleiotrophin

A recent paper describes the identification of pleiotrophin as a possible ALK ligand (Stoica et al., 2001). Pleiotrophin is a polypeptide growth factor that has been shown to induce proliferation in a wide range of cells including epithelial, endothelial and mesenchymal cell lineages (Chauhan et al., 1993; Courty et al., 1991; Fang et al., 1992; Li et al., 1990; Wellstein et al., 1992). By screening a phage display library with radiolabeled pleiotrophin, a peptide spanning a 10-aa stretch of the extracellular domain of ALK could be identified. This putative pleiotrophin recognition sequence lies within the extracellular part of ALK not shared with LTK (Figure 1). Thus, pleiotrophin could be a ligand for ALK but not for LTK. Interestingly, this binding region is also conserved in the potential homologue of ALK in Drosophila, suggesting a comparable pleiotrophin-ALK pathway in the fly. In vitro studies demonstrated high affinity binding of pleiotrophin to ALK in Farwestern blots (Stoica et al., 2001). By Scatchard analysis, a Kd value of approximately 35 pM was calculated. A cell line overexpressing ALK showed enhanced growth in soft agar upon pleiotrophin stimulation. In addition, pleiotrophin induced autophosphorylation of ALK following stimulation, and phosphorylation of several potential ALK substrates including IRS-1, SHC, PLC-gamma and p85 could be demonstrated. While these data convincingly argue for a functional link between pleiotrophin and ALK, several questions remain including whether ALK is indeed the sole receptor for pleiotrophin. The overlapping expression patterns of ALK and pleiotrophin in the nervous system, including the thalamus and midbrain in the adult and developing brain, suggest that the molecules indeed could represent a growth factor/receptor pair (Schulte and Wellstein, 1997). However, pleiotrophin effects have been demonstrated also in endothelial cells and fibroblasts (Chauhan et al., 1993; Courty et al., 1991; Fang et al., 1992; Li et al., 1990; Wellstein et al., 1992). Elevated pleiotrophin levels in the serum of patients suffering from a variety of solid tumors have been demonstrated (Fang et al., 1992; Souttou et al., 1998) and animal studies have suggested that pleiotrophin may contribute to tumor growth, invasion and metastasis (Chauhan et al., 1993; Choudhuri et al., 1997; Czubayko et al., 1994, 1996; Schulte et al., 1996). However, ALK expression has not been detected in any of these different cell types in mice or humans. The paper by Stoica et al. (2001) suggests for the first time that ALK expression may occur in some solid cancer cell lines and NIH3T3 cells. The expression of ALK was correlated with the responsiveness to pleiotrophin in these cells. However, ALK expression could not be demonstrated by immunoblotting with anti-ALK antibodies, the detection requiring an RT–PCR procedure with 30 cycles of amplification followed by hybridization with a radiolabeled internal oligonucleotide. Thus, ALK receptor density seems to be extremely low. It seems surprising that the profound and multiple effects of pleiotrophin might be mediated solely by the ALK receptor. Indeed, other cellular receptors for pleiotrophin have been identified, protein tyrosine phosphatase zeta/RPTP beta and syndecan-3, that may mediate some of these effects (Maeda et al., 1996; Raulo et al., 1994). In addition, further studies are required to establish pleiotrophin as the sole ligand for the ALK receptor, and the physiological relevance of this receptor/ligand pair.

Alk knock-out mice

Alk knock-out mice have been established in the laboratory of SW Morris (L Xue and SW Morris, manuscript submitted). These animals have a normal lifespan and possess no grossly evident abnormalities. In addition, no neural defects were observed that would argue for an essential function of Alk in the development of the nervous system. Thus, at present, the normal function of ALK and its ligand remains an open question.

NPM–ALK

The t(2;5) chromosomal rearrangement leads to the expression of an 80-kD fusion protein containing the first 117 aa of NPM fused to the C-terminal residues 1058–1620 of ALK (Fujimoto et al., 1996; Morris et al., 1994). The reciprocal ALK–NPM fusion gene is not transcribed at significant levels and thus does not contribute to the molecular pathogenesis of ALCL (Beylot-Barry et al., 1998; Cordell et al., 1999; Delsol et al., 1997; Morris et al., 1994). The chromosomal break within the ALK genomic sequence occurs in a 1935-bp intron that is located between the exons encoding the transmembrane and juxtamembrane domains of ALK. The chromosomal break within the NPM genomic sequence occurs in intron 4 of NPM (Ladanyi and Cavalchire, 1996a; Chan et al., 1997; Luthra et al., 1998). Interestingly, all fusion proteins containing rearranged ALK identified so far contain exactly the same 563 aa comprising the cytoplasmic tail of ALK. Only a single ALCL carrying a variant breakpoint has been described to date (with the break located in the ALK exon coding for the juxtamembrane region), leading however to an essentially identical fusion protein product (Ladanyi and Cavalchire, 1996b). These observations suggest that the ALK sequences present in NPM–ALK are the minimum residues of the protein required to lead to ALCL.

NPM–ALK is a potent oncogene capable of transforming a wide array of different cell types in vitro, including mouse and rat fibroblasts like NIH3T3, Fr3T3 and Rat-1 cells, and hematopoietic cell lines such as the myeloid line 32Dcl3 and the lymphocytic cell line Ba/F3 (Bai et al., 1998b; Bischof et al., 1997; Fujimoto et al., 1996; Kuefer et al., 1997; Mason et al., 1998). In addition, primary mouse bone marrow cells can be efficiently transformed by NPM–ALK in vitro (Bai et al., 2000). In comparison to other oncogenic fusion proteins such as BCR–ABL, NPM–ALK possesses strong transforming potential in both hematopoietic and fibroblast cell lines. For example, soft-agar colonies obtained by transfection with NPM–ALK grow faster and are larger than colonies transformed by BCR–ABL (unpublished observation). Given the demonstration that NPM–ALK-transduced bone marrow transplanted into irradiated recipient mice can induce a lymphoma-like disease, it is reasonable to assume that NPM–ALK is the causative oncogene in ALK-positive ALCL (Kuefer et al., 1997).

The t(2;5) brings the truncated ALK gene under the control of the strong NPM promoter, leading to the high-level expression of the NPM–ALK fusion protein in cells. NPM–ALK-positive cells show abundant tyrosine phosphorylation of NPM–ALK itself as well as a number of other proteins, even when the cells are cultured in the absence of growth factors (Bai et al., 1998b; Bischof et al., 1997; Fujimoto et al., 1996; Shiota et al., 1994a). The fusion of NPM to the cytoplasmic tyrosine kinase domain of ALK leads to constitutive activation of the ALK tyrosine kinase function. It is known that the normal NPM protein forms oligomers, although a classical coiled-coil domain could not be identified. The fusion of the N-terminal oligomerization motif in NPM, or functionally similar motifs in the alternative ALK fusion partners, to the cytoplasmic tyrosine kinase domain of ALK leads to activation of the kinase catalytic domain, mimicking oligomerization of the receptor by a natural ligand. While activation of a given receptor by a ligand occurs only transiently (being terminated rapidly by cellular internalization and receptor degradation), oligomerization of oncogenic fusion kinases leads to unregulated, constitutive activation. The commonly accepted paradigm is that oncogenic tyrosine kinases recruit, and in turn constitutively activate, promitogenic signaling cascades, leading to the transformation of cells. By recruiting SRC homology 2 (SH2) or phosphotyrosine binding (PTB) domain-containing molecules, specific signaling pathways are activated. NPM–ALK is a highly autophosphorylated molecule and contains 21 putative autophosphorylation sites that could serve as docking regions for SH2- or PTB-containing molecules.

Role of NPM

The normal NPM protein is able to form hexamers, and this oligomerization ability leads to the constitutive activation of NPM–ALK However, no classical oligomerization motifs are present in NPM–ALK. To further study the domains in NPM–ALK required for self-association, several NPM deletion mutants have been constructed (Bischof et al., 1997). Deletion of the first 64 aa of NPM (Δ64NPM–ALK), which have been reported to mediate hexamer formation of the normal protein (Liu and Chan, 1991), still retains autokinase ability in vitro but lacks kinase activity in vivo. Consequently, this mutant does not possess transforming abilities in cell lines, as judged by focus formation or soft-agar growth (Bischof et al., 1997). Loss of transforming ability was also observed with deletion of NPM residues 65–103 (Δ65–103NPM–ALK) and upon deletion of NPM residues 104–114, reported to comprise a putative metal binding motif (ΔMBNPM–ALK) (Chan et al., 1989). These data suggested that the complete 117-aa portion of NPM must be present in NPM–ALK for its transforming function. Earlier studies had indicated that methionine residues at positions 5, 7 and 9 of normal NPM are essential for oligomer formation (Liu and Chan, 1991). An NPM–ALK mutant in which these residues were mutated to leucines was still capable of forming homo-oligomers, as shown in both sucrose gradient and SDS–PAGE analysis (Bischof et al., 1997). This mutant also retained the ability to transform fibroblasts. Thus, the NPM sequence requirements for homo-oligomerization may differ somewhat between NPM–ALK and NPM. The exact motif and the mechanism of homo-oligomerization of NPM–ALK are still incompletely understood and require further investigation.

NPM sequences could serve solely as an oligomerization motif, with no further essential functions for the transforming activity of NPM–ALK, or they could potentially serve additional required roles. To address this issue, ALK chimeras have been constructed in which the NPM part of NPM–ALK was replaced with unrelated domains known to mediate oligomerization (Bischof et al., 1997). The TPR (translocated promoter region) protein is a nuclear pore protein that contains two leucine zipper motifs, and is rearranged in the TPR–MET oncogene (Rodrigues and Park, 1993). A TPR–ALK chimera displayed full transforming function in fibroblasts. The oncogenic activation of ALK by a fusion partner with no homology to NPM, such as TPR, supports a singular function for the NPM residues in NPM–ALK of mediating oligomerization.

Subcellular localizations of ALK fusion proteins

The localization of a portion of the total cellular NPM–ALK protein within the nuclear compartment is unexpected, given that the NPM nuclear localization motifs are not retained in the fusion. As mentioned above, NPM is a protein involved in transport processes between the nucleolus and the cytoplasm and is able to oligomerize, being present in cells primarily as hexamers. Thus, it is likely that the partial nuclear localization of NPM–ALK is due to its association with endogenous nuclear-localized NPM. Indeed, such an association has been demonstrated in the NPM–ALK-positive cell line SUP-M2 by both coimmunoprecipitation and sedimentation gradient experiments (Bischof et al., 1997). Δ64NPM–ALK, Δ65–103NPM–ALK and ΔMBNPM–ALK are all unable to oligomerize with endogenous NPM and show the anticipated cytoplasmic localization.

The nuclear localization of NPM–ALK does not seem to be required for lymphomagenesis. The chimeric TPR–ALK protein, although displaying full transforming capabilities in fibroblasts, shows an exclusive cytoplasmic location (Bischof et al., 1997). This assumption is also supported by the fact that the majority of the recently identified alternative ALK fusion proteins are localized solely to the cytosol (see Figure 5). ALCLs bearing alternative ALK gene rearrangements are indistinguishable from ALCLs with the classical t(2;5). The fact that a variety of oligomerization domain-containing proteins can replace NPM and the exclusively cytoplasmic location of these ALK fusion variants together support the notion that only the cytoplasmic pool of NPM–ALK contributes to oncogenesis.

Figure 5
figure5

ALK fusion proteins, the chromosomal rearrangements that generate them, and their frequency and subcellular localization. C: cytosolic; N: nuclear; NM: nuclear membrane. The numbers above the fusion proteins indicate the length (aa) of the fusion partners incorporated into each chimeric protein, and the total number of residues in each ALK fusion

Signaling pathways activated by NPM–ALK

Structural homology studies place NPM–ALK into the family of insulin RTKs, with the highest homology to LTK. The two proteins share 64% aa identity in their kinase catalytic domains (Figure 2) (Fujimoto et al., 1996; Iwahara et al., 1997; Morris et al., 1997). LTK has been demonstrated to activate the RAS–MAP kinase pathway by recruiting the adaptor molecule SHC (Ueno et al., 1996). In addition, IRS-1 has been shown to be recruited by LTK and seems to be of importance in delivering an antiapoptotic signal by this tyrosine kinase (Ueno et al., 1997). These findings prompted investigators to elucidate the role of these and other adaptor and signaling proteins for the mitogenicity induced by NPM–ALK.

RAS pathway

Analysis of the ALK sequence revealed two putative binding sites for PTB domains, NPNY156 and NPTY567; subsequently, it was shown that the adaptor molecules IRS-1 and SHC bind to these sites, respectively, probably via their PTB domains (Bischof et al., 1997; Fujimoto et al., 1996). SHC is an important signal transducer known to couple growth factor receptors such as the EGF or insulin receptors to the mitogenic RAS pathway. SHC binds directly to the activated EGF receptor and then forms a complex with GRB2 and SOS (mammalian homologue of Drosophila son of sevenless), leading to the activation of RAS. IRS-1 is involved in signal transduction from the insulin receptor. Site-directed mutagenesis has confirmed that Tyr156 and Tyr567 in NPM–ALK are indeed responsible for the complex formation of NPM–ALK with IRS-1 and SHC, respectively. However, NPM–ALK mutants defective in recruiting these molecules still show transforming potential in NIH3T3 cells and in the pro-B cell line Ba/F3, suggesting that these signaling cascades are not essential for the transforming function of the chimeric protein in these cells (Figure 3) (Bai et al., 1998b; Fujimoto et al., 1996). The importance of these signaling pathways for the oncogenicity of NPM–ALK, however, has not been evaluated in more definitive in vivo models of ALCL (see section Animal models of ALCL). Thus, whether these signaling pathways are required for the transforming ability of NPM–ALK in vivo remains unanswered.

Figure 3
figure3

Tyrosine-to-phenylalanine mutations of the putative autophosphorylation sites of NPM–ALK and their biological impact. Wild-type NPM–ALK or mutants were transfected and stably selected in IL3-dependent Ba/F3 cells. Dependency on IL-3 was tested by culturing the cells at 105/ml in IL3-free RPMI 1640 medium supplemented with 5% FCS for 3 weeks (Bai et al., 1998b). The identified binding sites of IRS-1, SHC and PLC-gamma are indicated (Bai et al., 1998b; Fujimoto et al., 1996)

Besides SHC and IRS-1, GRB2 has also been demonstrated to coprecipitate with NPM–ALK in ALCL cells (Fujimoto et al., 1996). GRB2 binds NPM–ALK in a GST pull-down assay even under denaturing conditions, suggesting a direct interaction between the GRB2 SH2 domain and a phosphotyrosine residue on NPM–ALK. However, site-directed mutagenesis failed to idenfity a single tyrosine residue in NPM–ALK responsible for the recruitment of GRB2 (Bai et al., 1998b; Fujimoto et al., 1996). Complex formation between GRB2 and NPM–ALK therefore may involve multiple tyrosine residues or perhaps intermediate proteins. It is possible that NPM–ALK(Y567F), which is unable to recruit SHC, is still capable of activating the promitogenic RAS pathway by recruiting GRB2. Thus, the importance of the RAS pathway for NPM–ALK-mediated oncogenicity has to be further evaluated.

PLC-gamma

Additional studies have investigated the importance of autophosphorylation of NPM–ALK by site-directed mutagenesis of each of the possible autophosphorylation sites in the protein (Figure 3) (Bai et al., 1998b). Interestingly, individual mutation of each of three tyrosine residues, Tyr338, Tyr342 and Tyr343, led to a severe impairment of NPM–ALK tyrosine kinase activity. These tyrosine residues are located within the kinase catalytic domain of NPM–ALK close to the catalytic core (Figure 2) and it is likely that mutation of these sites interferes with the ability of NPM–ALK to catalyze phosphorylation. All other individual tyrosine-to-phenylalanine mutations retained kinase activity (Figure 3). To evaluate the importance of different tyrosine residues for the mitogenicity of NPM–ALK, all mutants were tested for their ability to induce IL-3-independent growth in Ba/F3 cells, a pro-B cell line that requires IL-3 for survival and growth. Induction of IL-3 independence in this cell line is an indication of the transforming/mitogenic potency of a particular NPM–ALK mutant. Interestingly, only one mutant that retained kinase activity failed to induce stable IL-3-independent growth, NPM–ALK(Y664F). Subsequently, Tyr664 was shown to recruit PLC-gamma, which leads to enhanced IP3 production in NPM–ALK-expressing cells (Bai et al., 1998b). Mutation of Tyr664 abolishes binding and activation of PLC-gamma by NPM–ALK. Overexpression of wild-type PLC-gamma could partially restore the transforming function of the NPM–ALK(Y664F) mutant. The PLC-gamma pathway seems to be of general importance for NPM–ALK-mediated mitogenicity since NPM–ALK(Y664F) also showed impaired transforming capabilities in NIH3T3 cells. The fact that all mutants retaining kinase activity except NPM–ALK(Y664F) still rendered Ba/F3 cells growth factor-independent emphasizes the importance of PLC-gamma signaling for the transforming function of the fusion protein (Figure 3). However, as is the case for the other signaling molecules mentioned here, verification of the significance of the PLC-gamma pathway in an in vivo model of ALCL still remains to be demonstrated.

PI-3 kinase

NPM–ALK(Y664F)-expressing Ba/F3 cells do not stably grow without IL-3. However, in contrast to parental Ba/F3 cells or Ba/F3 cells expressing a kinase-defective mutant of NPM–ALK, NPM–ALK(Y664F) cells do not undergo complete apoptosis after IL-3 withdrawal and re-addition of IL-3 within a week can rescue these cells (Bai et al., 1998b). This observation suggested that PLC-gamma is predominantly required for the promitogenic properties of NPM–ALK and for its ability to replace IL-3 for mitogenicity but not for the protection against apoptosis.

It is well documented that apoptosis protection in Ba/F3 cells by IL-3 is mediated in part by activation of PI-3 kinase (phosphatidylinositol-3 kinase)/AKT and subsequent phosphorylation and inactivation of the proapoptotic molecule BAD. A widely distributed isoform of PI-3 kinase is a heterodimer of a 110-kd catalytic subunit and an 85-kd (p85) regulatory subunit, which is found in cellular complexes with ligand-activated growth factor receptors and non-receptor tyrosine kinases (Cantley et al., 1991; Leevers et al., 1999). Specific phosphorylated tyrosine residues present on the tyrosine kinase and the two SH2 domains of p85 mediate formation of these complexes. The association of p85 with activated tyrosine kinases is believed to be sufficient for the activation of the catalytic function of the p110 subunit. This activation leads to phosphorylation of the D-3 position of phosphatidylinositol (PI), phosphatidylinositol 4-phosphate (PI4P) and PI4,5P (Leevers et al., 1999). The 3-phosphoinositide that is generated is able to bind to pleckstrin homology (PH) domains of signaling molecules and to activate various downstream molecules, including the serine/threonine kinase AKT/PKB. AKT in turn is capable of preventing caspase-9 activation by maintaining mitochondrial integrity, phosphorylating and inactivating the proapoptotic BCL-2 family member BAD, activation of NF-κB, and reduction of the transcription of the Fas ligand, all of which contribute to its antiapoptotic function (Khwaja, 1999).

Two groups have reported the activation of PI-3 kinase by NPM–ALK (Bai et al., 2000; Slupianek et al., 2001). Blocking the activation of PI-3 kinase with specific inhibitors like wortmannin or LY294002 was demonstrated to prevent NPM–ALK-mediated mitogenicity to a much greater extent than IL-3-mediated mitogenicity in Ba/F3 cells. PI-3 kinase inhibitors were also able to induce apoptosis in the ALCL cell line JB6 (Bai et al., 2000). Expression of dominant-negative p85 or AKT mutants induced apoptosis in NPM–ALK-expressing cell lines and prevented tumor growth in syngeneic mice (Slupianek et al., 2001). Furthermore, BAD-induced apoptosis in COS cells could be partially blocked by the overexpression of NPM–ALK (Bai et al., 1998a). It is important to note that AKT activation was also demonstrated in ALCL patient samples (Slupianek et al., 2001). The complex of NPM–ALK and p85 is mediated mainly by the C-terminal SH2 domain of p85, with the N-terminal SH2 domain not apparently important for the interaction (Bai et al., 1998a). In an effort to identify the binding site of the p85-CSH2 domain on NPM–ALK, Tyr418 was shown to be a potential interaction site in vitro; however, mutation of this site failed to abolish binding in coimmunoprecipitation experiments using whole cell extracts. The most likely explanation for these findings is that an indirect association occurs between NPM–ALK and p85 through an additional adaptor molecule(s). For example, IRS-1 could recruit p85 to NPM–ALK. However, mutation of the IRS-1 binding site (NPM–ALK Y156F) failed to reduce the binding affinity of NPM–ALK for p85. Given the numerous adaptor molecules that can function as linkers between tyrosine kinases and SH2 domain-containing effector molecules, a complex involving p85, NPM–ALK, and an adaptor molecule is likely (Pawson and Gish, 1992). Indeed, a simultaneous complex among NPM–ALK, p85 and Gab2, SHC or CrkL has been demonstrated (Bai et al., 2000). Thus, these adaptor molecules are candidates linking NPM–ALK via the regulatory subunit p85 to the antiapoptotic PI-3 kinase pathway. LTK, in contrast to NPM–ALK, has been shown to bind directly to the p85 subunit of PI-3 kinase. Mutation of the p85-binding phosphotyrosine residue within LTK results in loss of the antiapoptotic function of an EGF–LTK chimear in Ba/F3 cells (Ueno et al., 1997). Thus, PI-3 kinase seems to be another important player in the signaling events that lead to NPM–ALK-mediated transformation (summarized in Figure 4).

Figure 4
figure4

Schematic illustration of signaling cascades activated by NPM–ALK

STAT (signal transducer and activator of transcription) proteins

Additional signaling molecules that have recently been reported to be of relevance for the genesis of NPM–ALK-positive lymphomas include STAT3 and STAT5. For example, NPM–ALK has been shown to activate STAT5 and this activation may be essential for lymphomagenesis (Nieborowska-Skorska et al., 2000). Retroviral infection of NPM–ALK-positive cells with a dominant-negative STAT5B mutant inhibited the anti-apoptotic activity of NPM–ALK in these studies.

Role of CD30

Most, if not all, NPM–ALK-positive lymphomas express the CD30 transmembrane receptor protein. Therefore, it is intriguing to speculate that CD30 and NPM–ALK are functionally connected, and that both molecules may be required for the development of ALCL. In support of this notion, it was first shown in 1994 that CD30 can physically interact with NPM–ALK (Shiota et al., 1994a).

CD30 belongs to the TNF receptor superfamily and it is expressed on activated T-cells and on tumor cells of hematopoietic origin, including the neoplastic cells of Hodgkin's disease and ALCL (Ellis et al., 1993; Schwab et al., 1982; Stein et al., 1985). Stimulation of CD30 by its ligand, CD30L, has been shown to elicit a plethora of cellular effects, including both proliferative and apoptotic responses in different cell types (Lee et al., 1996; Smith et al., 1993). It has been proposed that the TNF receptor superfamily can either induce apoptosis by recruitment of molecules like FADD and TRADD, leading to a caspase activation and apoptosis, or promote survival through the recruitment of TRAF proteins and activation of NF-κB (Beg and Baltimore, 1996; Hsu et al., 1996; Liu et al., 1996; Van Antwerp et al., 1996). However, in contrast to other TNF receptor family members, CD30 lacks the so-called death domain that is required for the recruitment of molecules such as FADD and TRADD that are responsible for caspase activation (Durkop et al., 1992). Nevertheless, like other TNF receptor family members, CD30 induces pleiotropic effects in cells upon its activation. Recently, it was observed that CD30L induces cell death in NPM–ALK-positive Karpas299 cells but not in an NPM–ALK-negative Hodgkin's disease-derived cell line, HDLM-2 (Hubinger et al., 1999a). These studies also demonstrated that CD30 forms a complex with NPM–ALK in Karpas299 cells (Hubinger et al., 1999a). However, no definite functional relationship between the two proteins could be demonstrated, although it was speculated that CD30 may target NPM–ALK to the inner face of the cell membrane, allowing the phosphorylation of proteins in this region and perhaps contributing to the apoptotic response observed after CD30 activation in NPM–ALK-positive cells. Another paper recently confirmed the proapoptotic effect of CD30L on ALCL cells (Mir et al., 2000). It was demonstrated in this study that CD30 activation leads to the induction of apoptotic death of ALCL cells, with the selective reduction of TNF receptor-associated factor 2 (TRAF2) function and impairment in the ability of the cells to activate the pro-survival transcription factor NF-κB (Mir et al., 2000). Again, however, no clearcut connection of these effects to the expression of NPM–ALK was demonstrated.

Thus, the molecular and functional relationship of the coexpression of CD30 and NPM–ALK in ALCL cells remains enigmatic. At present, it is unclear whether the expression of CD30 contributes to the development of ALCL at the molecular level. Rather, it may be that the chromosomal translocations leading to the expression of ALK fusion proteins occur at the stage of a specifically differentiated T-cell that happens to express CD30. Thus, CD30 expression would be an epiphenomenon, and not functionally related to the development of ALCL. The fact that NPM–ALK is able to transform CD30-negative hematopoietic cell lines and that lymphomas induced by NPM–ALK in mice are CD30 negative also argue against a functional role for CD30 in the development of ALK-positive ALCL. Nonetheless, whether CD30 functionally contributes to the molecular pathogenesis of ALCL is still an open issue and requires further investigation. The fact, however, that stimulation of CD30 leads to apoptotic death of ALCL cells provides a rationale for the use of CD30-activating antibodies in therapeutic approaches. Indeed, in animal models, this approach has shown potentially promising results (Pfeifer et al., 1999; Tian et al., 1995).

Other oncogenic fusions involving ALK

Immunohistochemical staining has revealed that about 15–25% of ALK-positive ALCLs do not exhibit the typical staining pattern of NPM–ALK in both the nucleus and cytoplasm, but rather possess exclusively cytoplasmic staining (Benharroch et al., 1998). Variant chromosomal translocations involving the ALK locus at chromosome 2p23 have also been observed, suggesting the existence of ALK fusion partners other than NPM. In addition, immunoblotting with ALK- and NPM-specific antibodies has revealed variant ALK proteins of 85-, 97-, 104- and 113-kD in ALCL cases containing t(2;3)(p23;q21) and t(1;2)(q21;p23), and these proteins did not contain NPM (Pulford et al., 1999).

Over the past 3 years, a number of variant ALK rearrangements have been cloned, and the ALK fusion partners identified (Figure 5). Some of these rearrangements have been observed only in ALCL cases, such as NPM–ALK itself, ATIC–ALK, TFG–ALKL and TFG–ALKS, while others have been found in both ALCL and inflammatory myofibroblastic tumors (IMTs), like TPM3–ALK and CLTC–ALK. Finally, two rearrangements have been seen to date in IMT only, RanBP2–ALK and TPM4–ALK. In all of these variant fusions, NPM is replaced by a protein that contains some form of oligomerization motif, suggesting that the mechanism underlying ALK tyrosine activation is similar in all of these chimeric proteins and that the signal transduction pathways activated are identical. This hypothesis is supported by the fact that the clinico-pathological features of ALCL containing alternative ALK chromosomal rearrangements seem to be indistinguishable from those with the classical t(2;5) (Falini et al., 1999b). The prognosis of these variant ALK-positive lymphomas seems comparable to NPM–ALK-positive ALCL and is significantly better than ALK-negative cases. In addition, like NPM–ALK-positive tumors, all variant ALK-positive cases display a T- or null-cell phenotype, show comparable nodal and extranodal involvement, and occur predominantly in children and young adults (Morris et al., 2001).

TPM3–ALK is expressed from a t(1;2)(q25;p23) translocation (Lamant et al., 1999). This translocation leads to the fusion of the N-terminal 221 residues of TPM3 to the cytoplasmic portion of ALK. TPM3 is a nonmuscle tropomyosin that contains an N-terminal coiled-coil structure, allowing its self-association and leading to the activation of the TPM3–ALK fusion protein. TPM3 has been shown to be involved in other chromosomal translocations as well; for example, in colon and papillary thyroid cancers, a fusion of TPM3 with the truncated NTRK1 receptor tyrosine kinase has been demonstrated. As for TPM3–ALK, fusion with TPM3 leads to the constitutive activation of the truncated NTRK1 (Butti et al., 1995). Like most variant ALK fusions, TPM3–ALK is located exclusively in the cytoplasm of lymphoma cells.

The t(2;3)(p23;q21) translocation leads to the formation of two different fusion proteins, TFG–ALKL and TFG–ALKS, depending upon the chromosomal breakpoint in the TFG gene (Hernandez et al., 1999). TFG stands for TRK-fused gene and was originally cloned as a fusion partner for NTRK1 in papillary thyroid carcinomas (Greco et al., 1995). As for TPM3, TFG contains a coiled-coil domain that can mediate the oligomerization and constitutive activation of the fused tyrosine kinase.

Three groups independently identified the ALK fusion partner involved in the inv(2)(p23q35), ATIC, which encodes 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase, the enzyme responsible for the final steps of de novo purine nucleotide biosynthesis (Colleoni et al., 2000; Ma et al., 2000; Trinei et al., 2000). The N-terminal 229 residues of ATIC are fused to ALK and are capable of leading to homo-oligomerization and constitutive activation of the ATIC–ALK chimeric kinase.

The largest variant ALK fusion protein identified thus far, CLTC–ALK, is expressed from the t(2;17)(p23;q23) (Touriol et al., 2000). CLTC is the heavy chain of the clathrin molecule, which is composed of both light and heavy chains and is found in the coated vesicles responsible for the intracellular transport of molecules. Clathrin proteins normally form trimolecular complexes, indicating that they are able to oligomerize. In contrast to other variant ALK fusions, CLTC–ALK does not show a diffuse cytoplasmic localization but rather exhibits a fine granular staining pattern that is presumed to be due to the binding of CLTC–ALK to clathrin-containing vesicles. CLTC–ALK was initially misidentified as involving the closely related CLTCL clathrin heavy chain gene located at chromosome 22q11.2 (Touriol et al., 2000).

Two additional ALK rearrangements have been cloned in cases of ALK-positive inflammatory myofibroblastic tumors (IMTs). The t(2;19)(p23;p13.1) initiates the expression of a TPM4–ALK fusion protein, TPM4 being a nonmuscle tropomyosin highly related to the TPM3 protein described above (Lawrence et al., 2000). Another novel ALK fusion results from a t(2;2)(p23;q11-13) or inv(2)(p23q11-13) (SW Morris, manuscript submitted). Ran binding protein 2 (RanBP2) (Mahajan et al., 1997), also known as Nup358, is a large nucleopore protein encoded at chromosome 2q11-13 that is localized at the cytoplasmic side of the nuclear pore complex. The N-terminal 867 residues of this protein are fused to ALK in RanBP2–ALK. In contrast to the other variant ALK fusion proteins, RanBP2–ALK displays a nuclear membrane staining pattern that is thought to be due to association of the fusion with normal RanBP2 at the nuclear pores.

Interestingly, all of the variant rearrangements cloned thus far contain the same breakpoint within the ALK gene, suggesting that this chromosomal region is very susceptible to breakage. Alternatively, shorter fragments of ALK may not be sufficient to induce lymphoma in vivo. This would suggest that the N-terminal sequences of the ALK cytoplasmic segment outside of the catalytic domain must be present and are of importance for lymphomagenesis. Yet additional variant ALK fusions may be identified in the future. For example, a number of rearrangements in ALCL involving 2p23 with loci other than 5q35 remain in which fusion partners have not yet been cloned, including t(1;2)(q21;p23), t(2;2)(p23;q23), t(2;13)(p23;q34) and t(2;22)(p23;q11.2) (Ladanyi, 1997; Mitev et al., 1998; Park et al., 1997; Rosenwald et al., 1999).

Other diseases associated with ALK expression

Large B-cell lymphomas with full-length ALK expression

Rare cases of large B-cell lymphoma have been described that express the full-length ALK receptor (Delsol et al., 1997). PCR and FISH failed to identify ALK rearrangements in these cases and the mechanism of full-length ALK expression in these lymphomas is unknown. The tumor cells in these lymphomas lack CD30 expression but do express the epithelial membrane antigen (EMA). In contrast to NPM–ALK-positive lymphomas, these B-cell tumors occur predominantly in adults and have a poor outcome. Whether the full-length ALK receptor is constitutively activated or functionally contributes to this type of lymphoma is at present unclear.

ALK expression in inflammatory myofibroblastic tumors (IMTs)

Inflammatory myofibroblastic tumors (IMTs) possess a pseudosarcomatous inflammatory appearance and can occur in nearly every soft tissue and viscera of the body (Morris et al., 2001). These tumors mainly affect children and young adults and histopathologically are composed of myofibroblasts together with infiltrating lymphocytes, eosinophils and plasma cells. The diagnosis is often difficult depending on the relative contribution of the myofibroblastic versus inflammatory components. Most IMTs can be cured by surgical resection of the tumor; however, cases of malignant, invasive and metastatic IMTs have been described. Recently, chromosomal rearrangements involving band 2p23 were identified in IMT and it was demonstrated that these rearrangements lead to the generation of ALK fusion proteins such as TPM3–ALK and TPM4–ALK (Griffin et al., 1999; Lawrence et al., 2000). IMT cases that express CLTC–ALK and RanBP2–ALK have also been demonstrated, and immunostaining studies of 68 IMTs showed ALK expression in 42 cases (61.8%) (SW Morris, manuscript submitted). This high frequency of involvement suggests that ALK fusions are of pathogenic importance on the molecular level for this neoplasm.

Neuroblastoma

ALK expression detectable by immunoreactive methods is normally restricted to neural tissues. A recent paper showed that in cell lines derived from neuroblastomas and neuroectodermal tumors approximately 45% express full-length ALK (Lamant et al., 2000). In primary neuroblastomas, ALK expression was demonstrated in 92% of the cases examined. However, no correlation between ALK expression and known prognostic factors for neuroblastoma could be identified. There was also no evidence of constitutive activation of the full-length ALK receptor in these tumors (although in vivo exposure to ligand would be expected to produce regulated, but contextually aberrant, receptor activation). Therefore, confirmation of a pathophysiological role for ALK in the development of neuroblastoma currently requires additional study.

NPM–ALK expression in normal cells

Using a highly sensitive RT–PCR procedure, NPM–ALK expression has been demonstrated in peripheral blood cells of healthy individuals (Trumper et al., 1998). As has been previously shown for the t(9;22) and BCR–ABL, this observation suggests that t(2;5) chromosomal rearrangements do occur in cells other than the lymphoma-inducing target cell of NPM–ALK. It is likely that the detection of NPM–ALK in cases of CD30-positive Hodgkin's disease, which has been very controversially discussed in the past, may be due to the detection of NPM–ALK in otherwise normal bystander cells.

Animal models for ALCL

To study the in vivo oncogenicity of NPM–ALK, transplantations of retrovirally infected bone marrow into mice have been performed. For example, Kuefer et al. (1997) demonstrated the development of B-lineage lymphomas with a 4–6 month latency in BALB/cByJ mice, using the pSRαMSVtkneo construct to direct NPM–ALK expression. Tumors arose in the mesenteric lymph nodes, with metastases to the lungs, kidneys, liver, spleen, and paraspinal area. More recently, lymphomas have been successfully induced in BALB/c mice after a latency of approximately 6 weeks using an MSCV-based EGFP bicistronic vector that programs high-level expression in hematopoietic stem cells (Miething et al., 2000). A massive infiltration of NPM–ALK-positive cells in the spleen, lymph nodes, and bone marrow could be observed in these animals. More interestingly, cytologic examination revealed relatively large and immature cells of lymphoid appearance, which by surface marker analysis showed a null-cell phenotype.

These approaches open the possibility to investigate the molecular pathogenesis of ALCL caused by NPM–ALK in an in vivo model. The contribution of different signaling pathways, which have been characterized only in cell lines thus far, to the oncogenicity of NPM–ALK in vivo could be investigated by employing different NPM–ALK mutants in these mouse models. In addition, the availability of knock-out mice for certain signal transduction pathways now allows evaluation of the oncogenicity of NPM–ALK within a specific mouse knock-out background. This powerful approach has been used very successfully for studies of transformation mediated by the BCR–ABL oncogene (Sexl et al., 2000; Li et al., 2001). Finally, these mice models are useful tools to evaluate potential NPM–ALK inhibitors in animals.

Clinical implications

ALK-positive lymphomas (‘ALKomas’) define a distinct class of non-Hodgkin's lymphomas. The molecular characterization of this entity now allows the development of molecularly targeted therapies. Since the malignant clone is distinguishable based on the expression of oncogenic ALK, the design of specific drugs targeted only against the malignant clone seems achievable. Several strategies to specifically target ALK-positive tumor cells can be envisioned. For example, NPM–ALK-specific ribozymes could prevent lymphoma cells from translating the NPM–ALK fusion protein. Such an approach has shown promising results in vitro (Hübinger et al., 1999b). However, this technique is hampered by the difficulties of introducing ribozymes into cells, and cell-permeable ribozymes against NPM–ALK have not yet been employed successfully. In addition, NPM–ALK-specific ribozymes only showed antiproliferative effects when highly overexpressed in NPM–ALK-positive cells (Hübinger et al., 1999b). Thus, while certainly promising, this approach has significant limitations that must be overcome to be feasible.

Another therapeutic approach involves the inhibition of signal transduction pathways known to be crucial for NPM–ALK-mediated oncogenicity. Along this line, inhibition of PLC-gamma has been shown to be quite effective in preventing the proliferation of cell lines engineered to express NPM–ALK, as well as cell lines derived from NPM–ALK-positive ALCL patients (Bai et al., 1998a). However, since NPM–ALK is capable of activating several promitogenic and antiapoptotic pathways, inhibition of only one of these pathways may not be sufficient to completely eradicate the malignant clone in ALCL. Thus, inhibition of NPM–ALK at the very beginning of the signaling cascade – i.e., of the constitutively activated tyrosine kinase itself – may be more efficient. Indeed, tyrosine kinase inhibitors have gained much attention upon the introduction of STI571, an ABL tyrosine kinase inhibitor that shows very promising results in clinical trials with BCR–ABL-positive leukemias. STI571 is considered by many clinicians to become a standard treatment of these leukemias in the near future. Therefore, much effort in the next few years is focusing on the development of analogous compounds for inhibition of ALK. Whether such inhibitors will be able to induce apoptosis of ALK-positive lymphoma cells and complete remissions in ALCL patients, as appears to be the case for STI571 and BCR–ABL-positive leukemias, remains to be seen.

Finally, immune-based therapeutic strategies could potentially be employed for ALK-positive ALCL. It has been demonstrated that many patients with NPM–ALK-positive lymphomas have significant levels of circulating antibodies directed against the ALK portion of NPM–ALK, suggesting that ALK fusions are quite immunogenic (Pulford et al., 2000). Thus, several strategies for anti-ALK immune-based treatments of chemotherapy-resistant ALCL can be imagined.

Conclusions

The identification of activated ALK fusion proteins in ALCL has revolutionized understanding of the genesis, diagnosis, and classification of these non-Hodgkin's lymphomas. In addition, recent, ongoing studies are examining the role of ALK fusions in the genesis of inflammatory myofibroblastic tumors. In both of these neoplasms, ALK promises to be an excellent target for directed therapeutic approaches for the future.

References

  1. Anagnostopoulos I, Stein H . 2000 Pathology 21: 178–189

  2. Bai RY, Coutinho S, Morris SW, Peschel C, Duyster J . 1998a Blood 92: 2110a

  3. Bai RY, Dieter P, Peschel C, Morris SW, Duyster J . 1998b Mol. Cell. Biol. 18: 6951–6961

  4. Bai RY, Ouyang T, Miething C, Morris SW, Peschel C, Duyster J . 2000 Blood 96: 4319–4327

  5. Barbacid M . 1995 Ann. NY Acad. Sci. 766: 442–458

  6. Beg AA, Baltimore D . 1996 Science 274: 782–784

  7. Benharroch D, Meguerian-Bedoyan Z, Lamant L, Amin C, Brugieres L, Terrier-Lacombe MJ, Haralambieva E, Pulford K, Pileri S, Morris SW, Mason DY, Delsol G . 1998 Blood 91: 2076–2084

  8. Beylot-Barry M, Groppi A, Vergier B, Pulford K, Merlio JP . 1998 Blood 91: 4668–4676

  9. Bischof D, Pulford K, Mason DY, Morris SW . 1997 Mol. Cell. Biol. 17: 2312–2325

  10. Borer RA, Lehner CF, Eppenberger HM, Nigg EA . 1989 Cell 56: 379–390

  11. Brunel V, Sainty D, Carbuccia N, Arnoulet C, Costello R, Mozziconacci MJ, Simonetti J, Coignet L, Gabert J, Stoppa AM, Birg F, Lafagepochitaloff M . 1995 Genes Chrom. Cancer 14: 307–312

  12. Butti MG, Bongarzone I, Ferraresi G, Mondellini P, Borrello MG, Pierotti MA . 1995 Genomics 28: 15–24

  13. Cantley LC, Auger KR, Carpenter C, Duckworth B, Graziani A, Kapeller R, Soltoff S . 1991 Cell 64: 281–302

  14. Chan PK, Chan FY . 1995 Biochim. Biophys. Acta 1262: 37–42

  15. Chan PK Chan FY, Morris SW, Xie Z . 1997 Nucleic Acids Res. 25: 1225–1232

  16. Chan PK, Liu QR, Durban E . 1990 Biochem. J. 270: 549–552

  17. Chan WY, Liu QR, Borjigin J, Busch H, Rennert OM, Tease LA, Chan PK . 1989 Biochemistry 28: 1033–1039

  18. Chauhan AK, Li YS, Deuel TF . 1993 Proc. Natl. Acad. Sci. USA 90: 679–682

  19. Choudhuri R, Zhang HT, Donnini S, Ziche M, Bicknell R . 1997 Cancer Res. 57: 1814–1819

  20. Colleoni GW, Bridge JA, Garicochea B, Liu J, Filippa DA, Ladanyi M . 2000 Am. J. Pathol. 156: 781–789

  21. Cordell JL, Pulford KA, Bigerna B, Roncador G, Banham A, Colombo E, Pelicci PG, Mason DY, Falini B . 1999 Blood 93: 632–642

  22. Courty J, Dauchel MC, Caruelle D, Perderiset M, Barritault D . 1991 Biochem. Biophys. Res. Commun. 180: 145–151

  23. Czubayko F, Riegel AT, Wellstein A . 1994 J. Biol. Chem. 269: 21358–21363

  24. Czubayko F, Schulte AM, Berchem GJ, Wellstein A . 1996 Proc. Natl. Acad. Sci. USA 93: 14753–14758

  25. Delsol G, Lamant L, Mariame B, Pulford K, Dastugue N, Brousset P, Rigal-Huguet F, al Saati T, Cerretti DP, Morris SW, Mason DY . 1997 Blood 89: 1483–1490

  26. Drexler HG, Gignac SM, von Wasielewski R, Werner M, Dirks WG . 2000 Leukemia 14: 1533–1559

  27. Durkop H, Latza U, Hummel M, Eitelbach F, Seed B, Stein H . 1992 Cell 68: 421–427

  28. Ellis TM, Simms PE, Slivnick DJ, Jack HM, Fisher RI . 1993 J. Immunol. 151: 2380–2389

  29. Falini B, Bigerna B, Fizzotti M, Pulford K, Pileri SA, Delsol G, Carbone A, Paulli M, Magrini U, Menestrina F, Giardini R, Pilotti S, Mezzelani A, Ugolini B, Billi M, Pucciarini A, Pacini R, Pelicci PG, Flenghi L . 1998 Am. J. Pathol. 153: 875–886

  30. Falini B, Pileri S, Zinzani PL, Carbone A, Zagonel V, Wolf-Peeters C, Verhoef G, Menestrina F, Todeschini G, Paulli M, Lazzarino M, Giardini R, Aiello A, Foss HD, Araujo I, Fizzotti M, Pelicci PG, Flenghi L, Martelli MF, Santucci A . 1999a Blood 93: 2697–2706

  31. Falini B, Pulford K, Pucciarini A, Carbone A, De Wolf-Peeters C, Cordell J, Fizzotti M, Santucci A, Pelicci PG, Pileri S, Campo E, Ott G, Delsol G, Mason DY . 1999b Blood 94: 3509–3515

  32. Fang W, Hartmann N, Chow DT, Riegel AT, Wellstein A . 1992 J. Biol. Chem. 267: 25889–25897

  33. Foss HD, Marafioti T, Stein H . 2000 Pathologe 21: 124–136

  34. Fujimoto J, Shiota M, Iwahara T, Seki N, Satoh H, Mori S, Yamamoto T . 1996 Proc. Natl. Acad. Sci. USA 93: 4181–4186

  35. Gascoyne RD, Aoun P, Wu D, Chhanabhai M, Skinnider BF, Greiner TC, Morris SW, Connors JM, Vose JM, Viswanatha DS, Coldman A, Weisenburger DD . 1999 Blood 93: 3913–3921

  36. Gogusev J, Nezelof C . 1998 Hematol. Oncol. Clin. North. Am. 12: 445–463

  37. Greco A, Mariani C, Miranda C, Lupas A, Pagliardini S, Pomati M, Pierotti MA . 1995 Mol. Cell. Biol. 15: 6118–6127

  38. Griffin CA, Hawkins AL, Dvorak C, Henkle C, Ellingham T, Perlman EJ . 1999 Cancer Res. 59: 2776–2780

  39. Harris NL, Jaffe ES, Stein H, Banks PM, Chan JK, Cleary ML, Delsol G, De Wolf-Peeters C, Falini B, Gatter KC . 1994 Blood 84: 1361–1392

  40. Hernandez L, Pinyol M, Hernandez S, Bea S, Pulford K, Rosenwald A, Lamant L, Falini B, Ott G, Mason DY, Delsol G, Campo E . 1999 Blood 94: 3265–3268

  41. Hsu H, Shu HB, Pan MG, Goeddel DV . 1996 Cell 84: 299–308

  42. Hübinger G, Scheffrahn I, Muller E, Bai R, Duyster J, Morris SW, Schrezenmeier H, Bergmann L . 1999a Exp. Hematol. 27: 1796–1805

  43. Hübinger G, Wehnis E, Maurer U, Morris SW, Bergmann L . 1999b Blood 94: 598a

  44. Iwahara T, Fujimoto J, Wen D, Cupples R, Bucay N, Arakawa T, Mori S, Ratzkin B, Yamamoto T . 1997 Oncogene 14: 439–449

  45. Jiang YP, Wang H, D'Eustachio P, Musacchio JM, Schlessinger J, Sap J . 1993 Mol. Cell. Biol. 13: 2942–2951

  46. Kadin ME . 1997 Cancer Surv. 30: 77–86

  47. Kadin ME, Morris SW . 1998 Leuk. Lymphoma 29: 249–256

  48. Khwaja A . 1999 Nature 401: 33–34

  49. Kinney MC, Kadin ME . 1999 Am. J. Clin. Pathol. 111: S56–S67

  50. Kinney MC, Greer JP, Kadin ME, DeCouteau JF, Collins RD . 1996 Lab. Invest. 74: 114A

  51. Kuefer MU, Look AT, Pulford K, Behm FG, Pattengale PK, Mason DY, Morris SW . 1997 Blood 90: 2901–2910

  52. Ladanyi M . 1997 Cancer Surv. 30: 59–75

  53. Ladanyi M, Cavalchire G . 1996a Diagn. Mol. Pathol. 5: 154–158

  54. Ladanyi M, Cavalchire G . 1996b Genes Chromos. Cancer 15: 173–177

  55. Lamant L, Dastugue N, Pulford K, Delsol G, Mariame B . 1999 Blood 93: 3088–3095

  56. Lamant L, Pulford K, Bischof D, Morris SW, Mason DY, Delsol G, Mariame B . 2000 Am. J. Pathol. 156: 1711–1721

  57. Lawrence B, Perez-Atayde A, Hibbard MK, Rubin BP, Dal Cin P, Pinkus JL, Pinkus GS, Xiao S, Yi ES, Fletcher CD, Fletcher JA . 2000 Am. J. Pathol. 157: 377–384

  58. Lee SY, Park CG, Choi Y . 1996 J. Exp. Med. 183: 669–674

  59. Leevers SJ, Vanhaesebroeck B, Waterfield MD . 1999 Curr. Opin. Cell. Biol. 11: 219–225

  60. Li S, Gillessen S, Tomasson MH, Dranoff G, Gilliland DG, van Etten RA . 2001 Blood 97: 1442–1450

  61. Li YS, Milner PG, Chauhan AK, Watson MA, Hoffman RM, Kodner CM, Milbrandt J, Deuel TF . 1990 Science 250: 1690–1694

  62. Liu QR, Chan PK . 1991 Eur. J. Biochem. 200: 715–721

  63. Liu ZG, Hsu H, Goeddel DV, Karin M . 1996 Cell 87: 565–576

  64. Luthra R, Pugh WC, Waasdorp M, Morris W, Cabanillas F, Chan PK, Sarris AH . 1998 Hematopathol. Mol. Hematol. 11: 173–183

  65. Ma Z, Cools J, Marynen P, Cui X, Siebert R, Gesk S, Schlegelberger B, Peeters B, De Wolf-Peeters C, Wlodarska I, Morris SW . 2000 Blood 95: 2144–2149

  66. Maeda N, Nishiwaki T, Shintani T, Hamanaka H, Noda A . 1996 J. Biol. Chem. 271: 21446–21452

  67. Mahajan R, Delphin C, Guan T, Gerace L, Melchior F . 1997 Cell 88: 97–107

  68. Mason DY, Pulford KA, Bischof D, Kuefer MU, Butler LH, Lamant L, Delsol G, Morris SW . 1998 Cancer Res. 58: 1057–1062

  69. Michaud N, Goldfarb DS . 1991 J. Cell. Biol. 112: 215–223

  70. Miething C, Bai R, Morris SW, von Schilling C, Schmidt B, Peschel C, Duyster J . 2000 Blood 96: 93a

  71. Mir SS, Richter BW, Duckett CS . 2000 Blood 96: 4307–4312

  72. Mitev L, Christova S, Hadjiev E, Guenova M, Oucheva R, Valkov I, Manolova Y . 1998 Leuk. Lymphoma 28: 613–616

  73. Morris SW, Kirstein MN, Valentine MB, Dittmer KG, Shapiro DN, Saltman DL, Look AT . 1994 Science 263: 1281–1284

  74. Morris SW, Naeve C, Mathew P, James PL, Kirstein MN, Cui X, Witte DP . 1997 Oncogene 14: 2175–2188

  75. Morris SW, Xue L, Ma Z, Kinney MC . 2001 Br. J. Haematol. 113: 275–279

  76. Nakamura S, Shiota M, Nakagawa A, Yatabe Y, Kojima M, Motoori T, Suzuki R, Kagami Y, Ogura M, Morishima Y, Mizoguchi Y, Okamoto M, Seto M, Koshikawa T, Mori S, Suchi T . 1997 Am. J. Surg. Pathol. 21: 1420–1432

  77. Neckameyer WS, Shibuya M, Hsu MT, Wang LH . 1986 Mol. Cell. Biol. 6: 1478–1486

  78. Nieborowska-Skorska M, Slupianek A, Zhang Q, Raghunath PN, Xue L, Morris SW, Wasik M, Skorski T . 2000 Blood 96: 2017a

  79. Ochs R, Lischwe M, O'Leary P, Busch H . 1983 Exp. Cell. Res. 146: 139–149

  80. Okuda M, Horn HF, Tarapore P, Tokuyama Y, Smulian AG, Chan PK, Knudsen ES, Hofmann IA, Snyder JD, Bove KE, Sukasawa K . 2000 Cell 103: 127–140

  81. Park JP, Curran MJ, Levy NB, Davis TH, Elliott JH, Mohandas TK . 1997 Cancer Genet. Cytogenet. 96: 118–122

  82. Pawson T, Gish D . 1992 Cell 71: 359–362

  83. Peter M, Nakagawa J, Doree M, Labbe JC, Nigg EA . 1990 Cell 60: 791–801

  84. Pfeifer W, Levi E, Petrogiannis-Haliotis T, Lehmann L, Wang Z, Kadin ME . 1999 Am. J. Pathol. 155: 1353–1359

  85. Pileri SA, Pulford K, Mori S, Mason DY, Sabattini E, Roncador G, Piccioli M, Ceccarelli C, Piccaluga PP, Santini D, Leone O, Stein H, Falini B . 1997 Am. J. Pathol. 150: 1207–1211

  86. Pulford K, Falini B, Banham AH, Codrington D, Roberton H, Hatton C, Mason DY . 2000 Blood 96: 1605–1607

  87. Pulford K, Falini B, Cordell J, Rosenwald A, Ott G, Muller-Hermelink HK, MacLennan KA, Lamant L, Carbone A, Campo E, Mason DY . 1999 Am. J. Pathol. 154: 1657–1663

  88. Pulford K, Lamant L, Morris SW, Butler LH, Wood KM, Stroud D, Delsol G, Mason DY . 1997 Blood 89: 1394–1404

  89. Raulo E, Chernousov MA, Carey DJ, Nolo R, Rauvala H . 1994 J. Biol. Chem. 269: 12999–13004

  90. Redner RL, Rush EA, Faas S, Rudert WA, Corey SJ . 1996 Blood 87: 882–886

  91. Rodrigues GA, Park M . 1993 Mol. Cell. Biol. 13: 6711–6722

  92. Rosenwald A, Ott G, Pulford K, Katzenberger T, Kuhl J, Kalla J, Ott MM, Mason DY, Muller-Hermelink HK . 1999 Blood 94: 362–364

  93. Sandlund JT, Pui CH, Roberts WM, Santana VM, Morris SW, Berard CW, Hutchison RE, Ribeiro RC, Mahmoud H, Crist WM . 1994a Blood 84: 2467–2471

  94. Sandlund JT, Pui CH, Santana VM, Mahmoud H, Roberts WM, Morris S, Raimondi S, Ribeiro R, Crist WM, Lin JS . 1994b J. Clin. Oncol. 12: 895–898

  95. Schulte AM, Lai S, Kurtz A, Czubayko F, Riegel AT, Wellstein A . 1996 Proc. Natl. Acad. Sci. USA 93: 14759–14764

  96. Schulte AM, Wellstein A . 1997 Tumor Angiogenesis Bicknell R, Lewis CM and Ferrara N. (eds). Oxford University Press: Oxford, New York, Tokyo pp. 273–289

    Google Scholar 

  97. Schwab U, Stein H, Gerdes J, Lemke H, Kirchner H, Schaadt M, Diehl V . 1982 Nature 299: 65–67

  98. Sexl V, Piekorz R, Moriggl R, Rohrer J, Brown MP, Bunting KD, Rothammer K, Roussel MF, Ihle JN . 2000 Blood 96: 2277–2283

  99. Shiota M, Mori S . 1996 Leuk. Lymphoma 23: 25–32

  100. Shiota M, Fujimoto J, Semba T, Satoh H, Yamamoto T, Mori S . 1994a Oncogene 9: 1567–1574

  101. Shiota M, Fujimoto J, Takenaga M, Satoh H, Ichinohasama R, Abe M, Nakano M, Yamamoto T, Mori S . 1994b Blood 84: 3648–3652

  102. Shiota M, Nakamura S, Ichinohasama R, Abe M, Akagi T, Takeshita M, Mori N, Fujimoto J, Miyauchi J, Mikata A, Nanba K, Takami T, Yamabe H, Takano Y, Izumo T, Nagatani T, Mohri N, Nasu K, Satoh H, Katano H, Fujimoto J, Yamamoto T, Mori S . 1995 Blood 86: 1954–1960

  103. Skinnider BF, Connors JM, Sutcliffe SB, Gascoyne RD . 1999 Hematol. Oncol. 17: 137–148

  104. Slupianek A, Nieborowska-Skorska M, Hoser G, Morrione A, Majewski M, Xue L, Morris SW, Wasik MA, Skorski T . 2001 Cancer Res. 61: 2194–2199

  105. Smith CA, Gruss HJ, Davis T, Anderson D, Farrah T, Baker E, Sutherland GR, Brannan CI, Copeland NG, Jenkins NA . 1993 Cell 73: 1349–1360

  106. Souttou B, Juhl H, Hackenbruck J, Rockseisen M, Klomp HJ, Raulais D, Vigny M, Wellstein A . 1998 J. Natl. Cancer Inst. 90: 1468–1473

  107. Stein H, Foss HD, Durkop H, Marafioti T, Delsol G, Pulford K, Pileri S, Falini B . 2000 Blood 96: 3681–3695

  108. Stein H, Mason DY, Gerdes J, O'Connor N, Wainscoat J, Pallesen G, Gatter K, Falini B, Delsol G, Lemke H, Schwarting R, Lennert K . 1985 Blood 66: 848–858

  109. Stoica GE, Kuo A, Aigner A, Sunitha I, Souttou B, Malercyk C, Caughey DJ, Went D, Karavanov A, Riegel AT, Wellstein A . 2001 J. Biol. Chem. 18: 16772–16779

  110. Szebeni A, Herrera JE, Olson MO . 1995 Biochemistry 34: 8037–8042

  111. Tian ZG, Longo DL, Funakoshi S, Asai O, Ferris DK, Widmer M, Murphy WJ . 1995 Cancer Res. 55: 5335–5341

  112. Touriol C, Greenland C, Lamant L, Pulford K, Bernard F, Rousset T, Mason DY, Delsol G . 2000 Blood 95: 3204–3207

  113. Trinei M, Lanfrancone L, Campo E, Pulford K, Mason DY, Pelicci PG, Falini B . 2000 Cancer Res. 60: 793–798

  114. Trumper L, Pfreundschuh M, Bonin FV, Daus H . 1998 Br. J. Haematol. 103: 1138–1144

  115. Ueno H, Honda H, Nakamoto T, Yamagata T, Sasaki K, Miyagawa K, Mitani K, Yazaki Y, Hirai H . 1997 Oncogene 14: 3067–3072

  116. Ueno H, Sasaki K, Kozutsumi H, Miyagawa K, Mitani K, Yazaki Y, Hirai H . 1996 J. Biol. Chem. 271: 27707–27714

  117. Valdez BC, Perlaky L, Henning D, Saijo Y, Chan PK, Busch H . 1994 J. Biol. Chem. 269: 23776–23783

  118. Van Antwerp DJ, Martin SJ, Kafri T, Green DR, Verma IM . 1996 Science 274: 787–789

  119. Wellstein A, Fang WJ, Khatri A, Lu Y, Swain SS, Dickson RB, Sasse J, Riegel AT, Lippman ME . 1992 J. Biol. Chem. 267: 2582–2587

  120. Yoneda-Kato N, Look AT, Kirstein MN, Valentine MB, Raimondi SC, Cohen KJ, Carroll AJ, Morris SW . 1996 Oncogene 12: 265–275

  121. Yung BY, Busch H, Chan PK . 1985 Biochim. Biophys. Acta 826: 167–173

  122. Zatsepina OV, Rousselet A, Chan PK, Olson MO, Jordan EG, Bornens M . 1999 J. Cell. Sci. 112: 455–466

Download references

Acknowledgements

This work was supported by a grant from the Wilhelm-Sander Stiftung and Mildred-Scheel Stiftung to J Duyster, by SFB grant No 456 to J Duyster, by National Cancer Institute (NCI) grants CA69129 and CA76301, and CORE grant CA21765 to SW Morris, and by the American-Lebanese Syrian Associated Charities (ALSAC), St Jude Children's Research Hospital.

Author information

Affiliations

Authors

Corresponding authors

Correspondence to Justus Duyster or Stephan W Morris.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Duyster, J., Bai, RY. & Morris, S. Translocations involving anaplastic lymphoma kinase (ALK). Oncogene 20, 5623–5637 (2001). https://doi.org/10.1038/sj.onc.1204594

Download citation

Keywords

  • ALK
  • NPM–ALK
  • variant ALK fusions
  • ALCL
  • IMT
  • pleiotrophin

Further reading

Search

Quick links