The NOTCH signaling pathway: role in the pathogenesis of T-cell acute lymphoblastic leukemia and implication for therapy

Abstract

T-cell acute lymphoblastic leukemia/lymphoma (T-ALL) is characterized by aberrant activation of NOTCH1 in over 60% of T-ALL cases. The high prevalence of activating NOTCH1 mutations highlights the critical role of NOTCH signaling in the pathogenesis of this disease and has prompted the development of therapeutic approaches targeting the NOTCH signaling pathway. Small molecule gamma secretase inhibitors (GSIs) can effectively inhibit oncogenic NOTCH1 and are in clinical testing for the treatment of T-ALL. Treatment with GSIs and glucocorticoids are strongly synergistic and may overcome the gastrointestinal toxicity associated with systemic inhibition of the NOTCH pathway. In addition, emerging new anti-NOTCH1 therapies include selective inhibition of NOTCH1 with anti-NOTCH1 antibodies and stapled peptides targeting the NOTCH transcriptional complex in the nucleus.

Clinical challenges and molecular features of T-ALL

T-cell acute lymphoblastic leukemia/lymphoma (T-ALL) is an aggressive hematologic tumor that generally affects children and adolescents but also arises in adults. T-ALL accounts for 10–15% of pediatric and 25% of adult ALL cases [Ferrando et al. 2002] and is characteristically more prevalent in males than in females (male-to-female ratio 3:1). Patients presenting with diffuse (>25%) bone marrow infiltration are diagnosed with T-cell acute lymphoblastic leukemia, while those with mediastinal masses and limited or no bone marrow involvement are diagnosed as T-cell acute lymphoblastic lymphoma. T-ALL was originally associated with very poor prognosis. However, the introduction of intensified chemotherapy protocols have resulted in marked improvements in the clinical outcome of this disease over the last decades. Current long-term survival rates in T-ALL are about 75–85% in children and adolescents and about 40–50% in adults [Pui et al. 2008]. Despite this progress, the outcome of T-ALL patients with primary resistant or relapsed disease remains very poor [Oudot et al. 2008].

T-cell transformation is a multistep process resulting from the accumulation of genetic alterations in oncogenes and tumor suppressor genes, which together disrupt key pathways responsible for the control of cell growth, proliferation survival in normal T-cell progenitors. The most frequent of these genetic lesions in pediatric T-ALL is the deletion of the CDKN2A locus in chromosome 9p22 present in up to 70% of cases. CDKN2A comprises the p14/INK4A and 16/ARF tumor suppressor genes which control cell cycle regulation and P53-mediated apoptosis, respectively [Ferrando et al. 2002; Hebert et al. 1994]. Moreover, over 60% of T-ALL cases show aberrant activation of the NOTCH1 signaling pathway making NOTCH1 the most prevalent oncogene in this disease [Weng et al. 2004]. In addition, about 40% of T-ALLs harbor chromosomal translocations juxtaposing developmentally important transcription factor oncogenes next to strong regulatory elements located in the vicinity of the T-cell receptor β (TCRB) gene in chromosome 7q34 or the T-cell receptor α-δ (TCRAD) locus in chromosome 14q11 [Ferrando and Look, 2000]. These T-ALL-specific transcription factor oncogenes include basic helix–loop–helix (bHLH) family members such as TAL1 [Begley et al. 1989; Bernard et al. 1990; Chen et al. 1990; Finger et al. 1989], TAL2 [Xia et al. 1991], LYL1 [Mellentin et al. 1989] and BHLHB1 [Wang et al. 2000]; LIM-only domain (LMO) proteins such as LMO1 and LMO2 [Boehm et al.; Greenberg et al. 1990; McGuire et al. 1989; Royer-Pokora et al. 1991; Warren et al. 1994]; the TLX1/HOX11 [Dube et al. 1991; Hatano et al. 1991; Kennedy et al. 1991] TLX3/HOX11L2, [Hansen-Hagge et al. 2002]; NKX2.5 [Nagel et al. 2003; Przybylski et al. 2006]; and HOXA homeobox genes [Hansen-Hagge et al. 2002; Soulier et al. 2005]; the MYC [Erikson et al. 1986; Inaba et al. 1990; Shima-Rich et al. 1997; Shima et al. 1986; Urashima et al. 1992] and MYB [Clappier et al. 2007] oncogenes; and TAN1, a truncated and constitutively activated form of the NOTCH1 receptor [Ellisen et al. 1991; Palomero et al. 2006a]. In some cases, these factors can also be activated in the context of other non-TCR-associated chromosomal abnormalities. This is the case for small deletions activating TAL1 [Aplan et al. 1990] and LMO2 [Van Vlierberghe et al. 2006], duplications of the MYB oncogene [Clappier et al. 2007; Lahortiga et al. 2007; O’Neil et al. 2007b] and the t(5;14)(q32;q11) translocation which activates the TLX3 oncogene in chromosome 5 by relocating it to the vicinity of the BCL11B locus in chromosome 14 [Nagel et al. 2007; Su et al. 2006]. Finally, about 10% of T-ALL cases harbor translocations resulting in the expression of fusion transcripts encoding chimerical transcription factor oncogenes. These include PICALMMLLT10/CALM-AF10 [Asnafi et al. 2003; Carlson et al. 2000; Dreyling et al. 1996], MLL-MLLT1/MLL-ENL [Chervinsky et al. 1995; Rubnitz et al. 1996] SET-NUP214 [Van Vlierberghe et al. 2008], NUP98-RAP1GSD1 [Hussey et al. 1999; Mecucci et al. 2000]. More recently, a series of loss of function mutations and deletions have been described for transcription factor tumor suppressor genes, such as WT1 [Tosello et al. 2009] [Renneville et al. 2010], LEF1 [Gutierrez et al. 2010], PHF6 [Van Vlierberghe et al., 2010], BCL11B [De Keersmaecker et al. 2010]; RUNX1 [Della Gatta et al. 2012; Grossmann et al. 2011; Zhang et al. 2012] and ETV6 [Gutierrez et al. 2010; Van Vlierberghe et al., 2011]. In addition, genetic alterations in signal transduction pathways can also contribute to the pathogenesis of T-ALL. These include deletions and mutations in PTEN [Palomero et al. 2007], ABL1 fusion oncogenes (NUP214-ABL1, EML1-ABL1 and ETV6-ABL1) [De Keersmaecker et al. 2005; Graux et al. 2004]; mutations activating the RAS signaling pathway [Balgobind et al. 2008; Bar-Eli et al. 1989; Campbell and Der, 2004], activating mutations in JAK1 and JAK3 [Flex et al. 2008; Zhang et al. 2012], activating mutations in IL7R [Zenatti et al. 2011; Zhang et al. 2012], FLT3 mutations [Paietta et al. 2004], PTPN2 deletions [Kleppe et al. 2011] and IRS4 translocations [Karrman et al. 2009]. Finally, deletions and loss of function mutations in histone-modifying genes such as SUZ12, EED, EZH2, and SETD2 have been found in T-ALL, which highlight the role of altered epigenetic regulation in T-cell transformation [Ntziachristos et al. 2012; Zhang et al. 2012].

The NOTCH1 signaling pathway

NOTCH signaling is a highly evolutionary conserved pathway responsible for the direct transduction of developmental signals at the cell surface into changes in gene expression in the nucleus. The NOTCH family of receptors is composed of four highly related proteins (NOTCH1–4). NOTCH receptors are heterodimeric transmembrane proteins composed of an extracellular subunit and a transmembrane and intracellular subunit, which interact via a specialized heterodimerization domain (HD). The extracellular subunit of NOTCH1 contains several EGF-like repeats involved in ligand-receptor interaction and three LIN-12/NOTCH repeats (LNRs), which stabilize the dimerization domain by holding the two NOTCH subunits together. The transmembrane-intracellular subunit of the receptor is composed of a short extracellular juxtamembrane peptide followed by a transmembrane sequence and a series of cytoplasmatic domain including RAM domain, nuclear localization signals (NLS), a series of ankyrin repeats, a region rich in glutamine (OPA), and a C-terminal PEST domain, which together function as a ligand-activated transcription factor [Aster et al. 2008]. Upon interaction with its ligands (Delta-like 1, 3 and 4; and Jagged 1 and 2), NOTCH1 undergoes a conformational change in the LNR–HD complex, which results in the proteolytic cleavage of the transmembrane-intracellular domain of the receptor, first by an ADAM metalloprotease, and subsequently by the γ-secretase complex [Brou et al. 2000; Mumm and Kopan, 2000]. This final cleavage releases the intracellular domains of NOTCH1 (ICN1) from the membrane, allowing its translocation to the nucleus where it activates gene expression via association with the RBPJ DNA binding protein and members of the mastermind-like family of coactivators [Fryer et al. 2002; Lubman et al. 2007]. Finally, activated NOTCH1 is quickly targeted for proteasomal degradation by FBXW7, an E3 ubiquitin ligase that recognize the PEST domain of ICN1 and mediates the termination of NOTCH1 signaling in the nucleus [Hubbard et al. 1997; Sundaram and Greenwald, 1993].

NOTCH1 signaling in normal T-cell development

In the hematopoietic system, NOTCH1 is strictly required for the commitment of multipotent hematopoietic progenitors to the T-cell lineage and to support cell growth, proliferation and survival at multiple stages of thymocyte development [Hozumi et al. 2008; Tanigaki and Honjo, 2007]. Indeed, conditional inactivation of RBPJ in T-cells results in the blockage of T development and in ectopic B-cell development in the thymus [Han et al. 2002; Radtke et al. 1999]. These results suggest that Notch provides a key regulatory signal in determining T lymphoid versus B lymphoid cell fate, In addition, NOTCH1 is involved in the progression through the early DN1, DN2 and DN3 stages of thymocyte development [Schmitt et al. 2004] and in the regulation of TCRB rearrangement [Wolfer et al. 2002].

NOTCH1 signaling in T-ALL

The first indication that Notch signaling could be an important element in leukemogenesis came in the early 1990s when Ellisen and colleagues identified a rare chromosomal translocation t(7;9)(q34;q34.3) involving the human NOTCH1 gene in T-ALL [Ellisen et al. 1991]. However, the prominent role of NOTCH1 signaling in the pathogenesis of T-ALL was only realized in 2004 with the identification of activating NOTCH1 mutations in over 60% of T-ALLs [Weng et al. 2004]. NOTCH1 mutations localized in the HD domain found in 20% of T-ALLs result in ligand-independent activation of the receptor, while mutations of the PEST domain present in 15% of T-ALLs cause increased ICN1 stability and aberrantly prolonged NOTCH1 activation [Weng et al. 2004]. In addition, juxtamembrane expansion (JME) alleles [Sulis et al. 2008] and intragenic deletions encompassing the 5’ region of the NOTCH1 locus (NOTCH1 del-N) [Haydu et al. 2012] result in very high levels of NOTCH1 signaling in rare cases of T-ALL. Finally, 20% of T-ALL cases show activation of NOTCH1 via mutations in the FBXW7 gene [Malyukova et al. 2007; O’Neil et al. 2007a; Thompson et al. 2007]. Mechanistically, FBXW7 mutations are related to NOTCH1 PEST mutations as they result in increased ICN1 protein stability. However, FBXW7 mutations may also be associated with additional oncogenic functions as this F-box protein is also involved in the degradation of other important oncoproteins such as MYC, JUN, MCL1, and Cyclin E [Clurman et al. 1996; Inuzuka et al. 2011; Sugimoto and Himeno, 1992; Tan et al. 2008; Wei et al. 2005]. Notably, in about 25% of T-ALL cases HD mutations are associated with PEST or FBXW7 mutations which results in a dual NOTCH1 activation that combines ligand-independent activation and prolonged ICN1 stability [Weng et al. 2004].

Genes and pathways deregulated by activated NOTCH1 in T-ALL

Recent progress in the identification of the transcriptional regulatory networks that control T-cell transformation downstream of NOTCH1 has shown a close relationship between oncogenic NOTCH1 signaling and the transcriptional control of cell growth and metabolism. Thus, NOTCH1 directly controls multiple genes involved in anabolic pathways and further promotes cell growth via direct transcriptional upregulation of MYC [Palomero et al. 2006c; Sharma et al. 2006; Weng et al. 2006]. The positive effects of activated NOTCH1 in cell growth are further sustained by interaction of NOTCH1 signaling with the PI3K–AKT–mTOR pathway [Chan et al. 2007; Ciofani and Zuniga-Pflucker, 2005; Palomero et al. 2007]. In addition, NOTCH1 transcriptionally upregulates molecules upstream of the PI3K such as interleukin 7 receptor alpha chain (IL7RA), the pre-T-cell receptor alpha (PTCRA) and the insulin-like growth factor 1 receptor (IGF1R) [Gonzalez-Garcia et al. 2009; Medyouf et al. 2011; Reizis and Leder, 2002].

Another important mechanism in T-cell transformation downstream of oncogenic NOTCH is the promotion of the G1/S cell cycle progression mediated by direct and indirect regulation of cell cycle regulator genes. NOTCH1 signaling promotes G1/S cell cycle progression via upregulation of CDK4 and CDK6 [Joshi et al. 2009] and downregulation of p27/KIP1 and p18/INK4C cell cycle inhibitors [Dohda et al. 2007; Palomero et al. 2006c].

Although the role and mechanisms of NOTCH1 signaling in promoting increased survival in T-ALL are less well characterized, constitutively active NOTCH1 can activate the NF-kappaB signaling supporting a major role for NF-kappaB in NOTCH1 induced T-cell transformation [Vilimas et al. 2007]. Consistently, HES1, a transcriptional repressor downstream of NOTCH1, sustains IKK activation in T-ALL by repressing the CYLD deubiquitinase [Espinosa et al. 2010].

NOTCH1 contributes to the control of epigenetic modulators of gene expression as demonstrated by the loss of the repressive mark Lys27 trimethylation of histone 3 in NOTCH1 target genes [Ntziachristos et al. 2012]. Thus, NOTCH1 antagonizes the function of the Polycomb repressive complex 2 (PRC2), responsible for writing the H3K27 mark [Ntziachristos et al. 2012]. Moreover, mutational loss of key components of the pRC2 complex such as EZH2, SUZ12, and EED are frequently found in T-ALL [Ntziachristos et al. 2012; Zhang et al. 2012].

Finally, an emerging and still developing concept is the role of microRNA regulation in NOTCH1 induced T-ALL. In this regard, expression of miR19 can accelerate the development of NOTCH1-induced T-ALL in mice [Mavrakis et al. 2010]. Moreover, the identification of a T-ALL patient with dual translocations activating the 17-92 cluster, where miR-19 is located, and NOTCH1 indicate the relevance of this interaction in T-cell transformation [Mavrakis et al. 2010]. In addition, NOTCH1 signaling downregulates miR-451 and miR-709 in mouse NOTCH1-induced leukemias by inducing degradation of the E2A tumor suppressor [Li et al. 2011].

Clinical impact of aberrant NOTCH1 signaling in T-ALL

Given the high prevalence of activating NOTCH1 mutations in T-ALL, a number of groups have analyzed the possible association of NOTCH1 activation with clinical outcome in this disease. By analyzing the effects of activating NOTCH1 mutations on early treatment response and long-term outcome in 157 pediatric patients with T-ALL enrolled in the ALL-Berlin–Frankfurt–Munster (BFM) 2000 study, Breit and coworkers originally showed that the presence of NOTCH1 mutations was significantly correlated with a good prednisone response, favorable minimal residual disease kinetics and improved long-term outcome in children with T-ALL [Breit et al. 2006]. Similarly, analysis of the prognostic impact of both NOTCH1 and FBXW7 mutations in 55 pediatric T-ALL patients treated on the Japan Association of Childhood Leukaemia Study (JACLS) protocols ALL-97 show a favorable prognosis for pediatric T-ALL patients carrying NOTCH1 or a FBXW7 mutation [Park et al. 2009]. A similar result was obtained in the analysis of 141 adult T-ALL patients treated in either the Lymphoblastic Acute Leukemia in Adults 94 (LALA-94) (n = 87) and GRAALL-2003 (n = 54) trials, in which NOTCH1 or FBXW7 mutations correlated with favorable outcome [Asnafi et al. 2009]. However, analysis of a series of T-ALL patients treated on the United Kingdom Acute Lymphoblastic Leukaemia XII (UKALLXII)/Eastern Cooperative Oncology Group (ECOG) E2993 protocol found no significant association between NOTCH1 and FBXW7 mutations and prognosis [Mansour et al. 2009]. Similarly, while NOTCH1 and FBXW7 mutations are associated with favorable early response to treatment in pediatric T-ALL, this early advantage did not ultimately translate to long-term outcome in DCOG or COALL protocols and in the EORTC 58881 and 58951 clinical trials [Clappier et al. 2010; Zuurbier et al. 2010]. Interestingly, the favorable prognosis of NOTCH1/FBXW7 mutations in adult T-ALL is found in more intense, pediatric-inspired GRAALL chemotherapy protocols but not in the less-intense LALA-94 study [Ben Abdelali et al. 2011], suggesting that differences in therapy intensification may influence the prognostic impact of NOTCH activating mutations in T-ALL.

Gamma secretase inhibitors, an emerging targeted therapy for T-ALL

The high prevalence of NOTCH1 mutations and the central role of NOTCH signaling in T-ALL make of NOTCH1 a potentially important therapeutic target in this disease. Much of the current effort targeting NOTCH1 in T-ALL aims to block NOTCH signaling with small molecule inhibitors of the γ-secretase complex. Indeed a number of γ-secretase inhibitors (GSI) originally developed for the treatment of Alzheimer’s disease is already in clinical development. GSIs reduce levels of intracellular NOTCH1 and transcriptional down regulation of NOTCH1-target genes in T-ALL [Palomero et al. 2006c]. Proof of principle for the antileukemic effects of GSIs was shown by the ability of compound E, a highly specific GSI to block NOTCH1 signaling and induce cell cycle arrest and decreased proliferation in T-ALL cell lines [Palomero et al. 2006b; Weng et al. 2004]. In addition, GSI treatment of mouse models of T-ALL showed marked antileukemic effects in vivo [Cullion et al. 2009; Tatarek et al. 2011]. Still, to date, only one clinical trial using GSIs in T-ALL has been reported [Deangelo et al. 2006]. This phase I study tested MK-0752, an oral GSI developed by Merck in seven patients with T-ALL. One patient with a NOTCH1 mutated T-ALL achieved a 45% reduction in a mediastinal mass after 28 days of treatment. However, this patient subsequently progressed and overall no patient achieved an objective durable response to treatment. In addition, most patients in the study showed dose-limiting gastrointestinal toxicity. The toxic effects of GSIs are an on target side effect resulting from the inhibition of NOTCH1 and NOTCH2 in intestinal progenitor cells [Riccio et al. 2008]. NOTCH1 and NOTCH2 play an important role in the specification of absorptive cell fate in intestinal stem cells. As a result, genetic and pharmacologic inhibition of NOTCH1 and NOTCH2 signals results in intestinal secretory cell metaplasia with accumulation of mucus-secreting goblet cells and malabsorption [Riccio et al. 2008; van Es et al. 2005]. Thus, overcoming GSI-induced gut toxicity has become a major imperative for the clinical development of GSIs. In this regard, the use of parenteral formulations and intermittent dosing schedules has been proposed as possible approaches to ameliorate the toxic effects of GSIs. Notably, inhibition of NOTCH signaling with GSIs can reverse glucocorticoid resistance in T-ALL and abrogate the gastrointestinal toxicity induced by GSIs in animal models [Real and Ferrando, 2009; Real et al. 2009; Samon et al. 2012; Wei et al. 2010]. The synergistic effects of GSI and glucocorticoids and the enteroprotective effects of dexamethosone against GSI-induced gut toxicity warrant the design of clinical trials testing the safety and efficacy of this drug combination in T-ALL [Cullion et al. 2009; Tammam et al. 2009; Wei et al. 2010]. Alternatively, combination therapies of GSIs with chemotherapy or other molecularly target agents could increase the antileukemic effects of these drugs facilitating the use of decreased GSI dosing with reduced GSI-induced gut toxicity. Combination treatment of GSIs with drugs targeting NFkB signaling [Vilimas et al. 2007], CDK inhibitors [Rao et al. 2009], PI3K–AKT–mTOR inhibitors [Chan et al. 2007; Cullion et al. 2009; Palomero et al. 2007; Sanda et al. 2010], and HDAC inhibitors [Sanda et al. 2010] have been proposed.

Finally, early observations of in vitro resistance to GSI treatment suggest that T-ALL cells can become resistance to NOTCH inhibition. Notably, PTEN loss and consequent constitutive activation of the PI3K–AKT pathway is present in 17% of primary T-ALL samples at diagnosis and can drive resistance to GSI treatment in T-ALL cell lines [Palomero et al. 2007]. In addition, FBXW7 mutations are also more prevalent in GSI-resistant T-ALL cell lines, compared with GSI-sensitive tumors [O’Neil et al. 2007a; Thompson et al. 2007]. However, additional T-ALL cooperating mutations may be needed to confer full resistance to NOTCH inhibition [Medyouf et al. 2010].

Novel approaches for NOTCH1 inhibition

Specific antibodies directed against individual NOTCH proteins can selectively block the activities of different NOTCH receptors [Aste-Amezaga et al. 2010; Li et al. 2008; Wu et al. 2010]. Anti-NOTCH1 antibodies recognizing the HD LNR repeat region of the receptor can effectively inhibit the activity of wild-type NOTCH1 and leukemia associated NOTCH1 mutants and effectively block tumor growth in vitro and in vivo [Aste-Amezaga et al. 2010; Wuet al. 2010]. Inhibition of ADAM10 may also facilitate effective inhibition of wild-type and mutant NOTCH receptors [Sulis et al. 2011]. An alternative strategy is to block NOTCH transcriptional complexes in the nucleus using chemically modified peptides. In this regard, the resolution of the NOTCH-RBPJ-MAML1 transcriptional complex [Nam et al. 2006] was instrumental in developing SAHM1, a stapled peptide designed to displace MAML1 and block the transcriptional activity of NOTCH1 RBPJ complex [Moellering et al. 2009].

Closing remarks

Clinical challenges remain ahead in the development of anti-NOTCH1 therapies in T-ALL. Logistically, relapsed T-ALL is a rare disease and patients at relapse typically show rapid disease progression making accruals in clinical trials slow. In addition, anti-NOTCH1 therapies may be most effective in the context of combination therapies, particularly with glucocorticoids, but possibly also with drugs targeting additionally important signaling pathways involved in the pathogenesis of T-ALL. Still, all GSI clinical trials designed to date have tested these drugs as single agents. However, it is worth noting the growing relevance of NOTCH1 as a therapeutic target outside T-ALL. The identification of prototypical activating mutations in NOTCH1 in chronic lymphocytic leukemia [Di Ianni et al. 2009; Fabbri et al. 2011; Puente et al. 2011] and the association of NOTCH1 mutations with advanced disease and chemotherapy resistance strongly suggest a potentially major role for anti NOTCH1 therapies in this disease. Similarly, rare but recurrent activating mutations in NOTCH1 have been found in lung cancer [Westhoff et al. 2009] and a pathogenic role for NOTCH signaling has been proposed other solid tumors [Roy et al. 2007; Shih Ie and Wang, 2007], with recent phase I clinical trials showing relevant results on the safety and potential therapeutic activity of GSIs in nonhematologic malignancies [Krop et al. 2012; Tolcher et al. 2012].