Vertebrate Twist homologs: Protein Interaction with Daughterless homologs

Paradoxically, M-twist inhibits myogenesis by blocking DNA binding by MyoD, by titrating E proteins (the vertebrate homologs of Daughterless), and by inhibiting trans-activation by MEF2 (the vertebrate homolog of Drosophila MEF2). For inhibition of MEF2, M-twist requires heterodimerization with E proteins and an intact basic domain and carboxyl-terminus. The inhibitory role of M-twist is consistent with exclusion of M-twist from the myotome (Spicer, 1996).

In vertebrates, the basic helix-loop-helix (bHLH) protein Twist may be involved in the negative regulation of cellular determination and in the differentiation of several lineages, including myogenesis, osteogenesis, and neurogenesis. Although it has been shown that mouse twist (M-Twist) (1) sequesters E proteins, thus preventing formation of myogenic E protein-MyoD complexes and (2) inhibits the MEF2 transcription factor, a cofactor of myogenic bHLH proteins, overexpression of E proteins and MEF2 fails to rescue the inhibitory effects of M-Twist on MyoD. M-Twist physically interacts with the myogenic bHLH proteins in vitro and in vivo; this interaction is required for the inhibition of MyoD by M-Twist. In contrast to the conventional HLH-HLH domain interaction formed in the MyoD/E12 heterodimer, this novel type of interaction uses the basic domains of the two proteins. While the MyoD HLH domain without the basic domain fails to interact with M-Twist, a MyoD peptide containing only the basic and helix 1 regions is sufficient to interact with M-Twist, suggesting that the basic domain contacts M-Twist. The replacement of three arginine residues by alanines in the M-Twist basic domain is sufficient to abolish both the binding and inhibition of MyoD by M-Twist, while the domain retains other M-Twist functions, such as heterodimerization with an E protein and inhibition of MEF2 transactivation. These findings demonstrate that M-Twist interacts with MyoD through the basic domains, thereby inhibiting MyoD (Hamamori, 1997).

The murine bHLH protein Twist has been shown to inhibit muscle differentiation in mammalian cells. This inhibition is cell autonomous and does not alter cell proliferation. By overexpression of E12, the inhibitory mechanisms of Twist and the dominant negative HLH factor Id can be distinguished. A difference is seen both for the native muscle-specific enhancers of myogenin and myosin light chain 1/3 and for an enhancer consisting of only four E-boxes. Mutagenesis experiments reveal that both the basic region and an evolutionarily conserved carboxy-terminal domain are required for the Twist-specific type of inhibition. Loss of either of these regions renders Twist less efficient and more similar to Id. Twist can bind to the muscle creatine-kinase E-box and inhibit DNA binding of E12 heterodimers with myogenic bHLH transcription factors like MyoD. However, a fourfold excess of Twist compared to MyoD is required for both effects. These results suggest that Twist inhibits muscle-specific gene activation by formation of actively inhibitory complexes rather than by sequestering E-proteins (Hebrok, 1997).

Vertebrate Twist homologs: Effects of Mutation

Saethre-Chotzen syndrome is one of the most common autosomal dominant disorders of craniosynostosis in humans and is characterized by craniofacial and limb anomalies. The locus for Saethre-Chotzen syndrome maps to chromosome 7p21-p22. TWIST has been evaluated as a candidate gene for this condition because its expression pattern and mutant phenotypes in Drosophila and mouse are consistent with the Saethre-Chotzen phenotype. TWIST maps to human chromosome 7p21-p22 and mutational analysis reveals nonsense, missense, insertion and deletion mutations in patients. These mutations occur within the basic DNA binding, helix I and loop domains, or result in premature termination of the protein. Studies in Drosophila indicate that twist may affect the transcription of fibroblast growth factor receptors (FGFRs), another gene family implicated in human craniosynostosis. The emerging cascade of molecular components involved in craniofacial and limb development now includes TWIST, which may function as an upstream regulator of FGFRs (Howard, 1997).

Saethre-Chotzen syndrome (acrocephalo-syndactyly type III [ACS III]) is an autosomal dominant craniosynostosis with brachydactyly, soft tissue syndactyly and facial dysmorphism including ptosis, facial asymmetry and prominent ear crura. ACS III has been mapped to chromosome 7p21-22. Of interest, TWIST, the human counterpart of the murine Twist gene, has been localized on chromosome 7p21 as well. The Twist gene product is required in head mesenchyme for cranial neural tube morphogenesis in mice. The co-localisation of ACS III and TWIST prompted a screen ACS III patients for TWIST gene mutations especially as mice heterozygous for Twist null mutations display skull defects and duplication of hind leg digits. 21-bp insertions and nonsense mutations were found for the TWIST gene (S127X, E130X) in seven ACS III probands. Impairment of head mesenchyme induction by TWIST is a novel pathophysiological mechanism in human craniosynostoses (el Ghouzzi, 1997).

The TWIST gene maps to 7p21; mutations in the gene have been reported in the Saethre-Chotzen form of craniosynostosis. The position of the Saethre-Chotzen gene has previously been refined by FISH analysis of four patients carrying balanced translocations involving 7p21, which suggests that it is located between D7S488 and D7S503. The breakpoints in four translocation patients do not interrupt the coding sequence of the TWIST gene and thus most likely act through a positional effect. Twelve Saethre-Chotzen cases have been found to have TWIST mutations. Four of these families had been used as part of the linkage study of the Saethre-Chotzen locus. The mutations detected included missense and nonsense mutations and three cases of a 21 bp duplication. Although phenotypically diagnosed as having Saethre-Chotzen syndrome, three families were found to have a pro250arg mutation of FGFR3 (Rose, 1997).

Saethre-Chotzen syndrome, a common autosomal dominant craniosynostosis in humans, is characterized by brachydactyly, soft tissue syndactyly and facial dysmorphism including ptosis, facial asymmetry, and prominent ear crura. A yeast artificial chromosome has been identified that encompasses the breakpoint of an apparently balanced t(6;7) (q16.2;p15.3) translocation associated with a mild form of Saethre-Chotzen syndrome. The rearrangement had occurred approximately 5 kb 3' of the human TWIST locus and has deleted 518 bp of chromosome 7. Potential exon sequences flanking the chromosome 7 translocation breakpoint did not hit known genes in database searches. The chromosome rearrangement downstream of TWIST is compatible with the notion that this is a Saethre-Chotzen syndrome gene and implies loss of function of one allele by a positional effect as a possible mechanism of mutation to evoke the syndrome (Krebs, 1997).

Mutations in the Fgfr1-Fgfr3 and Twist genes are known to cause craniosynostosis (premature fusion of the cranial sutures leading to skull deformity), the former by constitutive activation and the latter by haploinsufficiency. Although clinically achieving the same end result, the premature fusion of the calvarial (skull) bones, it is not known whether these genes lie in the same or independent pathways during calvarial bone development and later in suture closure. Fgfr2c is expressed at the osteogenic fronts of the developing calvarial bones. When FGF is applied via beads to the osteogenic fronts, suture closure is accelerated. In order to investigate the role of FGF signaling during mouse calvarial bone and suture development, detailed expression analysis of the splicing variants of Fgfr1-Fgfr3 and Fgfr4, as well as their potential ligand Fgf2 were performed. The IIIc splice variants of Fgfr1-Fgfr3 as well as the IIIb variant of Fgfr2 are expressed by differentiating osteoblasts at the osteogenic fronts (E15). In comparison to Fgf9, Fgf2 shows a more restricted expression pattern, being primarily expressed in the sutural mesenchyme between the osteogenic fronts. A detailed expression analysis has been carried out of the helix-loop-helix factors (HLH) Twist and Id1 during calvaria and suture development (E10-P6). Twist and Id1 are expressed by early preosteoblasts, in patterns that overlap those of the FGF ligands, but as these cells differentiate their expression dramatically decreases. signaling pathways were further studied in vitro, in E15 mouse calvarial explants. Beads soaked in FGF2 induce Twist and inhibit Bsp, a marker of functioning osteoblasts. Meanwhile, BMP2 upregulates Id1. Id1 is a dominant negative HLH protein thought to inhibit basic HLH proteins such as Twist. In Twist+/- mice, Fgfr2 protein expression is altered (Rice, 2000).

It is proposed that FGFs have functions at several stages of osteoblast differentiation. FGF2 has both inhibitory and stimulatory effects on osteoblast activity and evidence is presented that the inhibitory effects may be via a Twist regulated pathway. In line with Twist having a negative regulatory effect on osteoblast differentiation, the Twist mutation causing craniosynostosis is thought to be a loss-of-function mutation. Thus, Twist would appear to be upstream of FGFR/FGF signaling, though whether it is inhibitory or stimulatory cannot yet be definitively concluded. FGF may also act at a later stage in osteoblast differentiation, with both excess FGF and overactivation of FGF receptors causing an acceleration of suture closure. It is known that Id inhibits bHLH factors such as Twist, and that BMP2 induces osteoblast maturation. BMP2 is shown to stimulate Id and it is therefore postulated that the effects of BMP2 on osteoblast differentiation may be via Id's inhibition of Twist, thereby promoting cell differentiation instead of proliferation. However, it is known that overexpression of Id decreases the activity of the osteocalcin promoter, and that BMP can cause an increase in calvarial mesenchymal tissue volume. BMP2 and Id may therefore also act independently, stimulating osteoblast proliferation (Rice, 2000).

A novel bHLH protein gene Mesp2 (for mesoderm posterior 2) has been isolated that cross-hybridizes with Mesp1 expressed in the early mouse mesoderm. Mesp1 and Mesp2 are related the Twist and Nautilus protein families. Mesp2 is expressed in the rostral presomitic mesoderm, but down-regulated immediately after the formation of the segmented somites. To determine the function of MesP2 protein in somitogenesis, Mesp2-deficient mice were generated by gene targeting. The homozygous Mesp2 (-/-) mice die shortly after birth and have fused vertebral columns and dorsal root ganglia, with impaired sclerotomal polarity. The earliest defect in the homozygous embryos is a lack of segmented somites. The disruption of the metameric features, altered expression of Mox-1, Pax-1, and Dll1 (Drosophila homolog: Delta), and lack of expression of Notch1, Notch2, and FGFR1, suggest that MesP2 controls sclerotomal polarity by regulating the signaling systems mediated by Notch-Delta and FGF, which are essential for segmentation (Saga, 1997).

Runx2 is necessary and sufficient for osteoblast differentiation, yet its expression precedes the appearance of osteoblasts by 4 days. Twist proteins transiently inhibit Runx2 function during skeletogenesis. Twist-1 and -2 are expressed in Runx2-expressing cells throughout the skeleton early during development, and osteoblast-specific gene expression occurs only after their expression decreases. Double heterozygotes for Twist-1 and Runx2 deletion have none of the skull abnormalities observed in Runx2+/- mice; a Twist-2 null background rescues the clavicle phenotype of Runx2+/- mice, and Twist-1 or -2 deficiency leads to premature osteoblast differentiation. Furthermore, Twist-1 overexpression inhibits osteoblast differentiation without affecting Runx2 expression. Twist proteins' antiosteogenic function is mediated by a novel domain, the Twist box, which interacts with the Runx2 DNA binding domain to inhibit its function. In vivo mutagenesis confirms the antiosteogenic function of the Twist box. Thus, relief of inhibition by Twist proteins is a mandatory event precluding osteoblast differentiation (Bialek, 2004).

Heterozygous loss of Twist1 function causes coronal synostosis in both mice and humans. In mice this phenotype is associated with a defect in the neural crest-mesoderm boundary within the coronal suture, as well as with a reduction in the expression of ephrin A2 (Efna2), ephrin A4 (Efna4) and EphA4 in the coronal suture. Mutations in human EFNA4 are a cause of non-syndromic coronal synostosis. This study investigated the cellular mechanisms by which Twist1, acting through Eph-ephrin signaling, regulates coronal suture development. EphA4 mutant mice exhibit defects in the coronal suture and neural crest-mesoderm boundary that phenocopy those of Twist1+/- mice. Further, it was demonstrated that Twist1 and EphA4 interact genetically: EphA4 expression in the coronal suture is reduced in Twist1 mutants, and compound Twist1-EphA4 heterozygotes have suture defects of greater severity than those of individual heterozygotes. Thus, EphA4 is a Twist1 effector in coronal suture development. Finally, by DiI labeling of migratory osteogenic precursor cells that contribute to the frontal and parietal bones, it was shown that Twist1 and EphA4 are required for the exclusion of such cells from the coronal suture. It is suggested that the failure of this process in Twist1 and EphA4 mutants is the cause of craniosynostosis (Ting, 2009).

Twist-1 is a PPARδ-inducible, negative-feedback regulator of PGC-1α in brown fat metabolism

Brown fat is specialized for energy expenditure, a process that is principally controlled by the transcriptional coactivator PGC-1α (Drosophila homolog: Spargel). This study describes a molecular network important for PGC-1α function and brown fat metabolism. Twist-1 is selectively expressed in adipose tissue, interacts with PGC-1α, and is recruited to the promoters of PGC-1α's target genes to suppress mitochondrial metabolism and uncoupling. In vivo, transgenic mice expressing twist-1 in the adipose tissue are prone to high-fat-diet-induced obesity, whereas twist-1 heterozygous knockout mice are obesity resistant. These phenotypes are attributed to their altered mitochondrial metabolism in the brown fat. Interestingly, the nuclear receptor PPARγ not only mediates the actions of PGC-1α but also regulates twist-1 expression, suggesting a negative-feedback regulatory mechanism. These findings reveal an unexpected physiological role for twist-1 in the maintenance of energy homeostasis and have important implications for understanding metabolic control and metabolic diseases (Pan, 2009).

Twist homolog and neural crest development

Loss of Twist function in the cranial mesenchyme of the mouse embryo causes failure of closure of the cephalic neural tube and malformation of the branchial arches. In the Twist-/- embryo, the expression of molecular markers that signify dorsal forebrain tissues is either absent or reduced, but those associated with ventral tissues display expanded domains of expression. Dorsoventral organization of the mid- and hind-brain and the anterior-posterior pattern of the neural tube are not affected. In the Twist-/- embryo, neural crest cells stray from the subectodermal migratory path and the late-migrating subpopulation invades the cell-free zone separating streams of cells going to the first and second branchial arches. Cell transplantation studies reveal that Twist activity is required in the cranial mesenchyme for directing the migration of the neural crest cells, as well as in the neural crest cells within the first branchial arch, to achieve correct localization. Twist is also required for the proper differentiation of the first arch tissues into bone, muscle, and teeth (Soo, 2002).

The flat bones of the vertebrate skull vault develop from two migratory mesenchymal cell populations, the cranial neural crest and paraxial mesoderm. At the onset of skull vault development, these mesenchymal cells emigrate from their sites of origin to positions between the ectoderm and the developing cerebral hemispheres. There they combine, proliferate and differentiate along an osteogenic pathway. Anomalies in skull vault development are relatively common in humans. One such anomaly is familial calvarial foramina, persistent unossified areas within the skull vault. Mutations in MSX2 and TWIST are known to cause calvarial foramina in humans. Little is known of the cellular and developmental processes underlying this defect. Neither is it known whether MSX2 and TWIST function in the same or distinct pathways. The origin of the calvarial foramen defect in Msx2 mutant mice was traced to a group of skeletogenic mesenchyme cells that compose the frontal bone rudiment. This cell population is reduced not because of apoptosis or deficient migration of neural crest-derived precursor cells, but because of defects in its differentiation and proliferation. In addition heterozygous loss of Twist function causes a foramen in the skull vault similar to that caused by loss of Msx2 function. Both the quantity and proliferation of the frontal bone skeletogenic mesenchyme are reduced in Msx2-Twist double mutants compared with individual mutants. Thus Msx2 and Twist cooperate in the control of the differentiation and proliferation of skeletogenic mesenchyme. Molecular epistasis analysis suggests that Msx2 and Twist do not act in tandem to control osteoblast differentiation, but function at the same epistatic level (Ishii, 2003).

Vertebrate Twist homologs: Transcriptional regulation

In Drosophila, the Dorsal protein establishes the embryonic dorso-ventral axis during development. The vertebrate homolog of Dorsal, nuclear factor-kappa B (NF-kappaB), is vital for the formation of the proximo-distal organizer of the developing limb bud known as the apical ectodermal ridge (AER). c-rel mRNA is first detected in the chick limb bud at stage 15/16, before the appearance of the AER. Expression remains strong within the distal compartment during limb bud outgrowth. As the digits form, c-rel mRNA levels begin to decrease within the mesenchyme, persisting only in regions adjacent to the cartilage anlage. By stage 34, c-rel message is no longer detected. Transcription of the NF-kappaB proto-oncogene c-rel is regulated, in part, during morphogenesis of the limb bud by AER-derived signals such as fibroblast growth factors. Interruption of NF-kappaB activity using viral-mediated delivery of an inhibitor results in a highly dysmorphic AER, reduction in overall limb size, loss of distal elements and reversal in the direction of limb outgrowth. Inhibition of NF-kappaB activity in limb mesenchyme leads to a reduction in expression of Sonic hedgehog and Twist but derepresses expression of the bone morphogenetic protein-4 gene. These results are the first evidence that vertebrate NF-kappaB proteins act to transmit growth factor signals between the ectoderm and the underlying mesenchyme during embryonic limb formation. It is thought that the function of the kappaB factors is to modulate Twist gene expression during development (Bushdid, 1998).

Vertebrate Twist homologs: Interaction with transcriptional co-activators

Histone acetyltransferases (HATs) play a critical role in transcriptional control by relieving the repressive effects of chromatin, and yet how HATs themselves are regulated remains largely unknown. Here, it is shown that Twist directly binds two independent HAT domains of acetyltransferases, p300 and p300/CBP-associated factor (PCAF), and directly regulates their HAT activities. Twist strongly binds the C-terminal fragment (amino acids 1257-2414) of p300 spanning the HAT domain as well as the CH3 domain. Further deletion reveals that this interaction requires the CH3 domain (compare 1572-2414 and 1869-2414), which is known to interact with other proteins. Of particular interest, Twist retains an interaction with a HAT domain even in the absence of the CH3 domain (1257-1572). Twist also binds the N terminus of p300 (1-566 and 1-744), although these interactions are 5- to 10-fold weaker than those with the CH3 and HAT domains. Twist shows strong binding to PCAF. Intriguingly, experiments using a series of PCAF internal deletion mutants reveal that this interaction required the presence of the intact HAT domain and bromodomain. Thus, Twist interacts independently with the HAT domains of two different proteins, p300 and PCAF, suggesting that Twist may recognize common motifs present in these HAT domains. The N terminus of Twist is a primary domain interacting with both acetyltransferases, and the same domain is required for the inhibition of p300-dependent transcription by Twist. Taken together, these findings support the view that Twist suppresses the coactivator functions of p300 and PCAF through physical interactions mediated by the N terminus of Twist. Adenovirus E1A protein mimics the effects of Twist by inhibiting the HAT activities of p300 and PCAF. These findings establish a cogent argument for considering the HAT domains as a direct target for acetyltransferase regulation by both a cellular transcription factor and a viral oncoprotein (Hamamori, 1999).

E1A has been shown to bind the CH3 domain of p300/CBP and to displace PCAF from this domain. The effect of E1A has been interpreted as a simple competition between E1A and PCAF for the CH3 domain. The present study adds a further level of complexity by demonstrating that E1A and Twist may exert their inhibition not only by physically disrupting the p300-PCAF complex formation but also through suppression of their enzymatic activities. The interaction of Twist at the CH3 domain raises the intriguing possibility that Twist might also prevent PCAF association with p300/CBP by competing with PCAF for the common CH3 domain. These two mechanisms may not necessarily work simultaneously, and cells would have exquisite control mechanisms that determine how these two mechanisms of p300 and PCAF regulation may be differentially utilized in a given situation. Individual histone acetyltransferases have distinct roles. For instance, myogenic transcription and differentiation are dependent on the HAT activity of PCAF but not on that of p300/CBP. Similar observations are made in other systems, indicating that the transcriptional activities of the HAT domains of p300 and PCAF are highly promoter dependent. The dual inhibitory mechanisms involving the HAT inhibition as well as the competitive displacement of cofactors would allow E1A and possibly Twist to regulate a broad range of transcriptional activators that are differentially dependent on p300 and PCAF and their HAT activities (Hamamori, 1999 and references).

Twist homologs: DNA binding specificity

Basic helix-loop-helix (bHLH) proteins perform a wide variety of biological functions. Most bHLH proteins recognize the consensus DNA sequence CAN NTG (the E-box consensus sequence is in bold) via the DNA-binding basic region (BR) but acquire further functional specificity by preferring distinct internal and flanking bases. In addition, induction of myogenesis by MyoD-related bHLH proteins depends on myogenic basic region and BR-HLH junction residues, both of which are unessential for binding to a muscle-specific site, implying that their BRs may be involved in other critical interactions. An investigation has been carried out to see whether the myogenic residues influence DNA sequence recognition and how MyoD, Twist, and their E2A partner proteins (Daughterless in Drosophila) prefer distinct CAN NTG sites. In MyoD, the myogenic BR residues establish specificity for particular CAN NTG sites indirectly, by influencing the conformation through which the BR helix binds DNA. An analysis of DNA binding by BR and junction mutants suggests that an appropriate BR-DNA conformation is necessary but not sufficient for myogenesis, supporting the model that additional interactions with this region are important. The sequence specificities of E2A and Twist proteins require the corresponding BR residues. In addition, mechanisms that position the BR allow E2A to prefer distinct half-sites as a heterodimer with MyoD or Twist, indicating that the E2A BR can be directed toward different targets by dimerization with different partners. These findings indicate that E2A and its partner bHLH proteins bind to CAN NTG sites by adopting particular preferred BR-DNA conformations, from which they derive differences in sequence recognition that can be important for functional specificity (Kophengnavong, 2000).

In part, the specificity with which bHLH proteins function derives from preferential recognition of different classes of CAN NTG sites by different bHLH protein subgroups. The HLH segment consists of a parallel, left-handed, four-helix bundle. The BR is unstructured in solution but when bound to DNA, it extends N terminally from the HLH segment as a helix that crosses the major groove. Crystallographic analyses have revealed some differences in how these proteins bind DNA. For example, in Myc family and related bHLH proteins, an arginine residue at BR position 13 specifies recognition of CACGTG sites by contacting bases in the center. However, it still is not understood how bHLH proteins that have a different amino acid at BR position 13 bind preferentially to distinct CAN NTG sites or how bHLH proteins establish differences in flanking sequence selectivity that can be of biological importance (Kophengnavong, 2000 and references therein).

Many bHLH proteins that lack R13, including MyoD and other E2A partners, can bind to similar DNA sequences in vitro but they act on different tissue-specific genes. Cooperative or inhibitory relationships with other transcriptional regulators might contribute to this specificity, but it is not likely to derive entirely from other lineage-specific factors, because MyoD can induce myogenesis in many different cell types. Initiation of myogenesis by MyoD and other myogenic bHLH proteins depends on three residues that are located within the BR and the BR-HLH junction (A5, T6, and K15). These residues, which are referred to in this study as myogenic are not essential for binding a muscle-specific site in vitro or in vivo, suggesting that they are involved in other critical interactions. These interactions have been proposed to involve distinct cofactors and the unmasking of an activation domain in MyoD or the myogenic cofactor MEF2. In the MyoD-DNA structure, K15 is oriented away from the DNA, but A5 and T6 face the major groove and could not contact other proteins directly. However, the latter two residues could influence protein-protein interactions indirectly, by affecting how the BR helix is positioned on the DNA. Although substitutions at these positions might not substantially impair binding to particular CAN NTG sites, it is important to determine whether they might have more subtle influences on sequence specificity that could reflect conformational effects (Kophengnavong, 2000 and references therein).

The myogenic residues A5 and T6 establish the characteristic MyoD sequence preference, which includes a CAGCTG core. Individual substitutions at these BR positions simultaneously alter preferences for multiple bases that MyoD does not contact directly, indicating that these preferences are determined indirectly, by how the BR helix is positioned on the DNA. This mechanism is distinct from the standard model for sequence specificity, in which preferred bases are contacted directly. The corresponding BR residues are also required for the sequence preferences of E2A proteins, which can recognize either of two distinct half-sites depending on their dimerization partner. E2A homodimers and E2A-MyoD heterodimers bind to asymmetric sites that include a CACCTG core. In contrast, as a heterodimer with the bHLH protein Twist, E2A binds preferentially to half of the symmetric sequence CATATG. The preference of E2A for the former asymmetric sites depends not only on the BR sequence but also on BR positioning that involves the junction region. An analysis of DNA binding by MyoD and E2A junction and BR mutants indicates that a MyoD-like sequence specificity is associated with, but not sufficient for, myogenesis. This supports the model that the BR-junction region is also involved in other critical interactions. The results suggest that E2A and its partner bHLH proteins bind DNA by adopting a limited number of preferred BR conformations, each of which is associated with a characteristic DNA sequence preference. They also indicate that binding of cofactors to the MyoD BR might be influenced by how it is positioned on the DNA and are consistent with the idea that relatively subtle differences in binding sequence recognition can modulate bHLH protein activity in vivo (Kophengnavong, 2000).

Twist transcriptional targets

During Drosophila embryogenesis, the Dorsal transcription factor activates the expression of twist, a transcription factor required for mesoderm formation. The mammalian twist proteins (twist-1 and -2), are induced by a cytokine signaling pathway that requires the dorsal-related protein RelA, a member of the NF-kappaB family of transcription factors. Twist-1 and -2 repress cytokine gene expression through interaction with RelA. Mice homozygous for a twist-2 null allele or doubly heterozygous for twist-1 and -2 alleles show elevated expression of proinflammatory cytokines, resulting in perinatal death from cachexia. These findings reveal an evolutionarily conserved signaling circuit in which twist proteins regulate cytokine signaling by establishing a negative feedback loop that represses the NF-kappaB-dependent cytokine pathway (Šošić, 2003).

In Drosophila, the NF-κB-like transcription factor Dorsal activates twist expression through binding to κB sites in the twist promoter. To determine if this aspect of twist regulation might be conserved in vertebrates, whether twist-1 and -2 are regulated by TNFα, a well-known activator of the NF-κB pathway, was tested. Indeed, stimulation of immortalized mouse fibroblasts with TNFα evokes an increase in expression of twist-1 and -2 transcripts. Twist genes are also induced by TNFα in 10T1/2 and NIH3T3 cells. The IκBα and TNFα genes, which are known transcriptional targets of NF-κB, are also induced by TNFα stimulation (Šošić, 2003).

To determine whether the induction of twist expression by TNFα was dependent on NF-κB, twist-1 and -2 expression was measured in immortalized fibroblasts derived from p65-/- mice. Twist-1 and -2 are not inducible in p65-/- fibroblasts. Expression of TNFα and IκBα is also reduced in mutant cells, consistent with their known dependence upon NF-κB. These findings demonstrate that twist genes are induced by TNFα in an NF-κB-dependent manner, like other well-characterized TNFα-responsive genes (Šošić, 2003).

The concept of modularity in evolutionary biology is firmly established. Modules are shared in unrelated processes and are continuously moved during evolution, leading to the co-option of genes into new regulatory circuits. The NF-κB-twist partnership is an example of a module that has been conserved from insects to mammals and is utilized during two unrelated biological processes: dorsoventral patterning and the immune response. In both flies and mammals, NF-κB activates twist expression during dorsoventral patterning. During the immune response, NF-κB also induces the expression of peptides with antimicrobial properties. This study shows that in mammals twist has been adopted to negatively modulate NF-κB mediated cytokine activation, thus bridging two branches of NF-κB pathway and completing a negative feedback loop (Šošić, 2003).

Remarkably, a similar negative feedback loop also exists in the Drosophila NF-κB pathway. Upstream of spätzle, in that pathway, is the Easter protease. Mutations in any of the downstream genes required for activity of the pathway lead to an increase in the amount of activated Easter. Similarly, upon blockage of this pathway downstream of spätzle, its active form is accumulated in the embryo. Since mutations in dorsal led to increased Easter activation, it has been proposed that the initial component of the feedback loop is a transcriptional target of Dorsal. The results of this study raise the possibility that the function of twist as a key component of a negative feedback loop in this pathway may have been evolutionarily conserved from flies to mammals (Šošić, 2003).

Twist and apoptosis

Oncogene activation increases susceptibility to apoptosis. Thus, tumorigenesis must depend, in part, on compensating mutations that protect from programmed cell death. A functional screen for cDNAs that could counteract the proapoptotic effects of the myc oncogene has identified two related bHLH family members, Twist and Dermo1. Twist and Dermo1 are quite similar (greater than 90% identity) in the bHLH and carboxy-terminal domains. The amino termini are less closely related; Dermo1 lacks a glycine-rich region that is present in Twist. Both of these proteins inhibit oncogene- and p53-dependent cell death. Twist expression bypasses p53-induced growth arrest. These effects correlate with an ability of Twist to interfere with activation of a p53-dependent reporter and to impair induction of p53 target genes in response to DNA damage. An underlying explanation for this observation may be provided by the ability of Twist to reduce expression of the ARF tumor suppressor (ARF is one of two isoforms of INK4a/ARF: ARF binds to and sequesters MDM2, a p53 ligase, permitting the functioning of p53). Thus, Twist may affect p53 indirectly through modulation of the ARF/MDM2/p53 pathway. Consistent with a role as a potential oncoprotein, Twist expression promotes colony formation of E1A/ras-transformed mouse embryo fibroblasts (MEFs) in soft agar. Furthermore, Twist is inappropriately expressed in 50% of rhabdomyosarcomas, a tumor that arises from skeletal muscle precursors that fail to differentiate. Twist is known to block myogenic differentiation. Thus, Twist may play multiple roles in the formation of rhabdomyosarcomas, halting terminal differentiation, inhibiting apoptosis, and interfering with the p53 tumor-suppressor pathway (Maestro, 1999).

Recent evidence suggests that oncogenes such as myc and E1A sensitize cells to p53-dependent cell death, at least in part, through their effects on the ARF tumor suppressor. ARF is an upstream regulator of p53 that acts by affecting the localization and activity of MDM2. Expression of either E1A or Myc in primary MEFs provokes substantial increases in ARF mRNA levels, leading, in turn, to activation of the p53 pathway and to consequent induction of downstream targets such as p21 and MDM2. The p53 pathway fails to respond to E1A or Myc in ARF-null cells; this suggests ARF is in a position to act as a key mediator of homeostatic responses to oncogene expression. Therefore, it was asked whether Twist expression had any effect on ARF. C8 cells that have been engineered to ectopically express Twist show a dramatic reduction in ARF mRNA, when compared with control (LacZ-expressing) cells. This down-regulation is striking considering that loss of p53 function, such as is observed in the Twist-expressing cells, normally results in substantial increases in the abundance of the ARF transcript (Maestro, 1999 and references therein).

Twist and metastasis

Metastasis is a multistep process during which cancer cells disseminate from the site of primary tumors and establish secondary tumors in distant organs. In a search for key regulators of metastasis in a murine breast tumor model, it was found that the transcription factor Twist, a master regulator of embryonic morphogenesis, plays an essential role in metastasis. Suppression of Twist expression in highly metastatic mammary carcinoma cells specifically inhibits their ability to metastasize from the mammary gland to the lung. Ectopic expression of Twist results in loss of E-cadherin-mediated cell-cell adhesion, activation of mesenchymal markers, and induction of cell motility, suggesting that Twist contributes to metastasis by promoting an epithelial-mesenchymal transition (EMT). In human breast cancers, high level of Twist expression is correlated with invasive lobular carcinoma, a highly infiltrating tumor type associated with loss of E-cadherin expression. These results establish a mechanistic link between Twist, EMT, and tumor metastasis (Yang, 2004).

Loss of E-cadherin appears to be critical to an EMT. One major mechanism for inhibiting E-cadherin expression involves silencing of E-cadherin transcription through three E-boxes in its promoter. An over 100-fold reduction of E-cadherin mRNA level is observed in HMEC cells expressing Twist. To test whether this transcriptional repression is achieved through the three E-boxes in the E-cadherin promoter, HMEC-Twist cells was transiently transfected with a reporter construct containing the luciferase gene (Luc) under the control of the human E-cadherin promoter. Indeed, Luc activity was efficiently suppressed in the HMEC-Twist cells compared to the HMEC-control cells. A similar degree of repression was also observed in the HMEC cells expressing Snail, a known repressor of E-cadherin expression. When two Luc reporter constructs in which the E-box elements had been mutated were introduced, the Luc activities were derepressed in the HMEC cells expressing Twist. These data indicate that Twist directly or indirectly causes the transcriptional repression of E-cadherin through the E-box elements on the E-cadherin promoter (Yang, 2004).

TGF-beta-Id1 signaling opposes Twist1 and promotes metastatic colonization via a mesenchymal-to-epithelial transition

ID genes are required for breast cancer colonization of the lungs, but the mechanism remains poorly understood. This study shows that Id1 expression induces a stem-like phenotype in breast cancer cells while retaining epithelial properties, contrary to the notion that cancer stem-like properties are inextricably linked to the mesenchymal state. During metastatic colonization, Id1 induces a mesenchymal-to-epithelial transition (MET), specifically in cells whose mesenchymal state is dependent on the Id1 target protein Twist1, but not at the primary site, where this state is controlled by the zinc finger protein Snail1. Knockdown of Id expression in metastasizing cells prevents MET and dramatically reduces lung colonization. Furthermore, Id1 is induced by transforming growth factor (TGF)-beta only in cells that have first undergone epithelial-to-mesenchymal transition (EMT), demonstrating that EMT is a prerequisite for subsequent Id1-induced MET during lung colonization. Collectively, these studies underscore the importance of Id-mediated phenotypic switching during distinct stages of breast cancer metastasis (Stankic, 2013).

Oncogenic BRAF disrupts thyroid morphogenesis and function via twist expression

Thyroid cancer is common, yet the sequence of alterations that promote tumor formation are incompletely understood. This study describes a novel model of thyroid carcinoma in zebrafish that reveals temporal changes due to BRAFV600E (see Drosophila Raf). Through the use of real-time in vivo imaging, disruption in thyroid follicle structure was observed to occur early in thyroid development. Combinatorial treatment using BRAF and MEK inhibitors reverses the developmental effects induced by BRAFV600E. Adult zebrafish expressing BRAFV600E in thyrocytes develop invasive carcinoma. A gene expression signature from zebrafish thyroid cancer was identified and found to be predictive of disease-free survival in patients with papillary thyroid cancer. Gene expression studies nominate TWIST2 (see Drosophila twist) as a key effector downstream of BRAF. Using CRISPR/Cas9 to genetically inactivate a TWIST2 orthologue, the effects of BRAFV600E were suppressed and thyroid morphology and hormone synthesis were restored. These data suggest that expression of TWIST2 plays a role in an early step of BRAFV600E-mediated transformation (Anelli, 2017).

back to twist Evolutionary homologs part 1/2

twist: Biological Overview | Regulation | Targets of Activity and Protein Interactions | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.