Toward The Rational Design of Cell Fate Modifiers: Notch Signaling as a Target for Novel Biopharmaceuticals

A. Zlobin, M. Jang and L. Miele*

Cardinal Bernardin Cancer Center, Loyola University Medical Center, 2160 South First Avenue, Maywood, IL 60153, USA

*Address correspondence to this author at the Cardinal Bernardin Cancer Center, Loyola University Medical Center, 2160 South First Avenue, Maywood, IL 60153, USA; Tel: 708-327-3362; Fax: (708)-327-3238; Email: lmiele@luc.edu

Abstract: Recent advances in our understanding of highly conserved mechanisms that control cell fate determination are paving the way towards rationally designed biologics that modulate specific cell fate decisions. Cell fate decisions leading to proliferation, differentiation or apoptosis are crucial elements in the pathogenesis of countless human diseases. Biopharmaceuticals designed to regulate such processes in specific cell types in vivo or ex vivo have vast potential applications in oncology, stem cell technology, immunomodulation and neuropathology. One of the most conserved mechanisms controlling cell fate determination is based upon Notch-ligand interactions and subsequent signaling events. Recent studies have shown that this mechanism regulates cell differentiation, proliferation and apoptosis in a wide variety of cell maturation processes and in neoplastic cells. These observations identify the Notch signaling network as a promising drug target for numerous indications. In this review, we describe: 1) potential drug targets in the Notch signaling network; 2) the Notch agonists and antagonists developed so far, including recombinant proteins, antibody-based agents, synthetic peptides, antisense oligonucleotides and gene therapy approaches, as well as possible strategies to design novel Notch-targeting biopharmaceuticals; 3) the possible clinical applications of such biopharmaceuticals and 4) a model strategy for the selection and developement of a Notch-targeting biopharmaceutical.

INTRODUCTION

     Notch genes encode evolutionarily conserved transmembrane receptors that regulate cell fate determination [1-7]. In recent years, Notch signaling has been identified as a crucial mechanism controlling numerous cell fate decisions during development and postnatal life in organisms from Drosophila to humans [1; 5-7]. Strong experimental evidence indicates that Notch signaling regulates all three branches of the cell fate decision tree: differentiation, cell cycle progression and apoptotic cell death. Notch activation promotes proliferation and inhibits differentiation of bone marrow stem cells and it prevents differentiation of neural and glial precursors. Recent data suggest a possible connection of Notch-1 with the pathogenesis of Alzheimer’s disease (see below). Notch signaling plays multiple, crucial roles in T cell development, and most likely in hematopoiesis and the immune response. Transformed cells of virtually all embryonic lineages and various human malignancies overexpress Notch receptors and ligands. Chronic Notch activation is likely to be a survival advantage for neoplastic cells, in light of the recently discovered anti-apoptotic activity of Notch signaling.

     In summary, numerous observations indicate that the Notch signaling network, including Notch receptors, ligands, mediators, modulators and target genes, is one of the most attractive targets for biopharmaceutical development identified in recent years. Promising areas of clinical development for Notch-targeting biologics include cancer treatment, stem cell technology, immunomodulation and the treatment of neurodegenerative disorders. In this review, we shall describe in separate sections: 1) the drug targets, including Notch receptors, ligands, mediators and modulators, and the current knowledge about their biological roles; 2) the agents, including Notch agonists and antagonists that have been recently developed, and possible alternative strategies for the development of novel candidates; 3) the potential clinical applications of Notch-targeting biopharmaceuticals and 4) a model strategy for the selection and development of a putative Notch-targeting cell fate modifier. For a comprehensive discussion of Notch biology, the reader is referred to excellent recent reviews [1-7].

DRUG TARGETS

a)    Notch Receptors: Biological Functions and Expression Patterns

     Notch signaling is thought to mediate primarily interactions between contiguous cells during cell-cell contact. This is because Notch receptors are activated by ligands that are predominantly cell membrane-associated. However, a soluble form of the Drosophila Notch ligand Delta has been recently identified, suggesting that Notch may also mediate interactions between non-contiguous cells [8]. In general terms, Notch receptors regulate cell fate determination in three different situations [1; 7]: 1) lateral specification/inhibition, in which initially identical cells with multiple possible differentiation fates regulate each other’s fates; 2) inductive signaling in which one cell type regulates another cell type’s differentiation choices and 3) cell-autonomous effects, in which a cell regulates it own fate through Notch signaling. The latter may be due to expression of ligand and receptor by the same cell, in a fashion similar to autocrine production of growth factors, or to ligand-independent activation of Notch. A classical paradigm of lateral specification/ inhibition is Drosophila neurogenesis [9-11]. In this model, putative stochastic oscillations in the levels of expression of Notch and its ligand Delta in neuroectodermal cells identify cells that will commit to the neuronal phenotype. Recently, such oscillations in the levels of Notch and Delta have been shown to be controlled by Wingless signaling [12; 13]. Neuronal precursor cells express higher levels of Delta and lower levels of Notch than surrounding cells. Each neuronal precursor, through Delta, activates Notch in the cells surrounding it, and causes them to further upregulate Notch expression. The activation of Notch in the cells surrounding neuronal precursors prevents them from differentiating towards the neuronal lineage. Subsequently, these cells switch to an epidermal differentiation program. Thus, notch deletions or loss of function mutations result in excessive numbers of neuroectodermal cells differentiating towards a neuronal fate. This is defined a “neurogenic” phenotype [9-11]. The C. elegans notch homologues LIN-12 and GLP-1 play analogous lateral specification roles during the development of uterine and vulvar precursors and of germ cells and pharyngeal epithelium respectively [5; 14-16]. Examples of inductive signaling can be found in mammalian odontogenesis [17; 18] and hematopoiesis [for review, see 6]. In these cases, Notch signaling is used to mediate communications between different cell types (ameloblasts and dental mesenchyme, bone marrow stroma and hematopoietic precursors). Cell-autonomous effects of Notch have been observed in Drosophila [19-21]. The anti-apoptotic effect of Notch-1 in murine erythroleukemia (MEL) cells [22] may fall within this category.

     The effects of Notch signaling on cell fate decisions in vertebrates have been extensively studied in tissue culture, ex vivo systems and transgenic animals. Notch signaling appears to be a highly conserved mechanism, which is used in cell fate determination control in many different tissues. In general, Notch activation modulates the response of a cell to a differentiation stimulus, rather than directly specifying a cell fate. Notch and its ligands have been shown to be essential for Xenopus neurogenesis [23-25] and mesoderm segmentation [26]. In mice, Notch-1 is necessary for embryonic development and targeted disruption of the NOTCH-1 gene results in disorganized somitogenesis and embryonic lethality [27; 28]. Similarly, targeted disruption of Notch ligand genes JAGGED-1 or -2 and the NOTCH-2 gene result in severe developmental defects or embryonic lethality [29-31]. Expression of constitutively active forms of Notch receptors inhibits terminal differentiation in vitro in murine models of myogenesis and granulocytopoiesis [32-35]. In chick retina explants, expression of constitutively active Notch-1 inhibits differentiation of retinal progenitors to ganglion cells, while NOTCH-1 antisense oligonucleotides increase differentiation towards a neuronal phenotype [36]. Similarly, ligand-induced activation of Notch-1 inhibits oligodendrocyte maturation in vitro [37]. Overall, these data suggest that in many contexts Notch activation inhibits or delays differentiation toward a specific fate until the cell is able to respond to signals which specify an alternate fate. However, in some experimental models, such as CD4 versus CD8 and a/b versus g/d T-cell receptor (TCR) lineage decisions in thymocytes [38; 39] and in vitro adipocyte differentiation [40], Notch signaling appears to be required for correct processing of differentiation stimuli.

     Notch receptors and ligands are widely expressed in postnatal animals. Notch-1 and -2 mRNA can be detected in various organs of humans and mice, particularly thymus, spleen, lung, heart, testis, ovary and central nervous system, with partially overlapping organ distributions [41; 42]. The same is true of Notch ligands Jagged-1 and Delta-1 in human organs [43]. Notch-1 and-2 are expressed in CD34⁺ human bone marrow stem cells and other hematopoietic precursors and are thought to participate in the control of cell fate decisions during hematopoiesis [6; 34; 44-46]. Bone marrow stromal cells express the Notch ligand Jagged-1 [46; 47], as do thymic stromal cells [48]. There is strong evidence that Notch-1 participates in the regulation of T cell development [38; 39; 49-52]. Notch-3 is expressed in the developing central and peripheral nervous system [53] and in the thymus [48] and developing pancreas [54]. Mutations of the NOTCH-3 gene in humans are associated with CADASIL, a neurological disorder characterized by multiple subcortical strokes and dementia [55]. Notch-4 is thought to be expressed primarily in endothelial cells during development and adult life [56].

     Recent data indicate that, in addition to its well-characterized effects on cell differentiation, Notch signaling affects cell cycle progression and apoptotic cell death. In HL-60 promyelocytic leukemia cells and CD34⁺ bone marrow stem cells, constitutively active Notch-1 and Notch-1 stimulation by Notch ligand Jagged-2 accelerated progression through G₁ [57]. Transfected constitutively active Notch-1 inhibits apoptosis induced by glucocorticoids [50] and by TCR engagement [51] in T cell hybridomas. Moreover, spontaneously expressed Notch-1 inhibits pharmacologically induced apoptosis in MEL cells [22]

b)    Structure and Processing of Notch Receptors and Ligands

     Notch receptors have an evolutionarily conserved structural organization. Vertebrate NOTCH genes are strongly related to each other and to Drosophila notch [58; 59]. Humans and mice have 4 NOTCH genes, denominated NOTCH-1 through 4. Similarly, multiple Notch ligands have been identified in vertebrates [60-66]. These are homologous to the Drosophila ligands Delta and Serrate and the C. elegans ligand Lag-2, and thus, are commonly identified as DSL from the initials of Delta, Serrate and Lag-2 [3; 67]. Mammalian Delta homologues are denominated “Delta-like” and mammalian Serrate homologues are denominated “Jagged”. [60; 62]. Notch proteins are synthesized as single polypeptide precursors, which are proteolytically processed to a heterodimeric, mature form [see Fig. (1)]. During receptor maturation, Notch pre-proteins are cleaved into an extracellular subunit (N^EC) containing multiple EGF-like repeats and a transmembrane subunit including the intracellular region (N^TM) [68], a single pass transmembrane domain and a short extracellular tail. This cleavage is catalyzed by a furin-like convertase [69]. It is still unclear whether the N^EC subunit is further processed by an ADAM family protease [70-72]. Such proteases have been shown to facilitate Notch signaling [70-72]. However, more recently the Drosophila ADAM family protease, Kuzbanian, has been shown to process the Notch ligand Delta [8]. Therefore, the effect of ADAM proteases on Notch signaling may be indirect. The number of EGF repeats in the Notch pre-proteins varies from 36 repeats in Drosophila Notch and mammalian Notch-1 and-2 to the 13 and 10 repeats of C. elegans Lin-12 and Glp-1 respectively. In Drosophila Notch, EGF repeats 11 and 12 are responsible for binding both ligands Delta and Serrate [73]. These repeats are highly conserved in mammalian Notch receptors, particularly Notch-1 and -2, and are thought to be the main ligand -binding site of these receptors. The N^TM subunit contains 6 ankyrin-like repeats, a polyglutamine region (OPA) and a proline/glutamic acid/serine/threonine rich (PEST) sequence [1]. A sequence denominated RAM23, immediately distal to the transmembrane region and proximal to the ankyrin repeats is thought to be a high affinity binding site for transcription factors of the CSL group (see below) [74]. The N^TM subunit is further cleaved to release an intracellular fragment (N^IC) during Notch signaling in a presenilin-dependent step (see below). Notch ligands also contain multiple EGF-like domains and an additional, N-terminal cysteine-rich domain (the DSL domain) that appears to be responsible for Notch binding [67], as well as short cytoplasmic tails.