BIOLOGICAL OVERVIEW

spalt major (salm) and forkhead (fkd) both influence the development of the fly's gut. Both are expressed prior to invagination in anterior and posterior anlagen of the gut. The gut of the fly is formed during gastrulation by the invagination or inturning of anterior and posterior gut primordia. Endoderm forms the main tissue of the gut, with some contribution from contiguous ectoderm. In fkh mutants, ectodermal parts of the foregut develop as head structures, while Abdominal-B/salm double mutants develop thoracic structures instead of head structures. Thus fkh and salm have opposing actions in fate determination: fkh changes cell fate to a more posterior morphology while salm influences change to a more anterior morphology. salm appears to have a role in restricting trunk development outside of the mesothorax (Casanova, 1989).

Ectopic expression of Antennapedia induces antenna to leg transformations. Under this circumstance salm expression is entirely repressed by Antp activity, so that leg structures form in place of antennal structures (Wagner-Bernholz, 1991). How does ectopic Antp result in antenna-leg transformation? Current thinking suggests that since Salm represses tsh and ANTP enhances tsh, ectopic Antp provides a double dose, enhancing tsh at the same time it represses salm. This remains conjecture, since these activities have not been observed in imaginal discs (Kühnlein, 1994).

Salm could be both a transcriptional activator and repressor, as it has glutamine - rich regions which are thought to function as activator sequences and proline - and alanine - rich regions required for repressor function.

Dpp receptor levels contribute to shaping the Dpp morphogen gradient in the Drosophila wing imaginal disc via regulation of spalt

spalt and Optomotor blind are targets of Dpp, acting through the Dpp receptor Thick veins (Tkv) in the wing imaginal disc. Axis formation in the Drosophila wing depends on the localized expression of the secreted signaling molecule Decapentaplegic. Dpp acts directly at a distance to specify discrete spatial domains, suggesting that it functions as a morphogen. Expression levels of the Dpp receptor thick veins are not uniform along the anterior-posterior axis of the wing imaginal disc. tkv is expressed at low levels in the center of the disc and at higher levels toward the edges of the disc. Although tkv levels are low in the center of the disc, clonal analysis has shown that tkv activity is stringently required in this region for growth and for target gene expression. Receptor levels are low where Dpp induces its targets Spalt and Omb in the wing pouch. Receptor levels increase in cells farther from the source of Dpp in the lateral regions of the disc (Lecuit, 1998).

Evidence is presented that Dpp signaling negatively regulates tkv expression and that the level of receptor influences the effective range of the Dpp gradient. High levels of tkv sensitize cells to low levels of Dpp and also appear to limit the movement of Dpp outside the wing pouch. Thus receptor levels help to shape the Dpp gradient. It was asked whether Dpp signaling regulates tkv expression by examining the effects of clones of cells expressing Dpp at lateral positions in the disc where the level of tkv is normally high. Dpp-expressing clones were marked indirectly by their ability to induce ectopic Spalt expression. tkv transcript levels are reduced where Spalt is misexpressed, suggesting that Dpp can act at a distance to repress tkv expression These results suggest that the reduced levels of tkv transcript in the center of the disc are due to downregulation by Dpp acting at a distance (Lecuit, 1998).

Are the reduced levels of the Tkv expression important for the formation of the Dpp activity gradient? To address this, an examination was made of the effects on the expression of the Dpp-target genes Spalt and Omb in clones of cells that overexpress wild-type Tkv. Tkv-expressing clones well inside the endogenous domains show little effect on either Spalt or Omb expression. Clones near the edge of the endogenous Spalt domain show increased Spalt expression and those near the edge of the Omb domain show elevated Omb expression. Tkv-expressing clones located outside but near the endogenous Spalt domain show ectopic induction of Spalt. Clones located farther from the Spalt domain do not show ectopic activation of Spalt. Together, these observations suggest that overexpressing Tkv can increase the sensitivity of cells to low levels of Dpp (Lecuit, 1998).

Tkv was overexpressed to assess the consequences of broadly elevating Tkv expression levels in the central region of the disc. Wings with elevated Tkv expression are reduced in size. The effect is stronger in the posterior compartment, with the region between veins 4 and 5 being more reduced than the region between veins 2 and 3. The region between veins 3 and 4 is relatively normal, possibly because the size of this intervein region is specified directly by Hedgehog, not by Dpp. These observations suggest that the long-range activity of Dpp in the vein 2-3 and 4-5 regions is compromised by overexpression of the Dpp receptor. Overexpressing Tkv strongly reduces the size of the Spalt domain in the Posterior (P) compartment. The effect on Spalt expression is much stronger in the P compartment than in the A compartment and the Spalt domains in both compartments appear to be less graded at their edges than in wild type. The effective range of the Dpp activity gradient appears to be limited to a few cells in the P compartment in mid- and late-third instar discs. This suggests that overexpression of receptor can limit the spread of Dpp in the P compartment. These observations suggest that high levels of the receptor might sequester ligand and limit its movement across the wing disc. The difference observe between A and P compartments when Tkv is overexpressed probably reflects the fact that cells originating in the Dpp expression domain can contribute to formation of a large part of the anterior compartment, but not to the posterior compartment. Thus cells originating in the Dpp domain could ‘carry’ Dpp protein away from the source as they and their progeny are displaced by addition of new cells (the displacement process can be directly visualized by lineage tracing cells originating in the dpp-expression domain). It is concluded that artificially high levels of Thick veins outside the wing pouch appear to limit the spread of Dpp and thereby modulate the shape of the ligand gradient. In addition, the level of Tkv expression modulates the sensitivity of cells to Dpp. Thus regulation of receptor levels by Dpp modulates the shape of the Dpp gradient (Lecuit, 1998).

Function of the spalt/spalt-related gene complex in positioning the veins in the Drosophila wing

Spalt and Spalt-related regulate the vein-specific expression of the transcription factors of the knirps and iroquois gene complexes, delimiting their domains of expression in the wing pouch. The effects of spalt/spalt-related mutations on knirps and iroquois expression are cell-autonomous, suggesting that they could be direct. The regulation of iroquois involves transcriptional repression by Spalt and Spalt-related, whereas the regulation of knirps involves a combination of transcriptional activation and repression mediated by the same genes. It is suggested that the regulation of the iroquois and knirps gene complexes by Spalt and Spalt-related translates the Decapentaplegic morphogenetic gradient into precisely spaced pattern elements (de Celis, 2000).

Although the development of the four longitudinal veins of Drosophila (L2-L5) involves the same signaling systems, and vein cells show the identical type of differentiation, there are several characteristics that distinguish one vein from another. Thus, the veins L2 and proximal L4 differentiate predominantly in the ventral wing surface, whereas the veins L3, L5 and distal L4 do so in the dorsal surface. Furthermore, several genes are required for the formation of individual veins, suggesting that each vein is individually specified. Vein-specific genes include the transcription factors of the iroquois gene complex (iro-C), which are only expressed and required in L3 and L5, and the transcription factors of the knirps gene complex (kni-C), which are only expressed and required in L2. Vein-specific genes could be part of a combinatorial code of signals that activate a common vein-differentiation program in different parts of the wing. In addition, vein specific genes could also confer individual qualities to each longitudinal vein (de Celis, 2000 and references therein).

The presumptive region of the wing blade in the wing disc, the wing pouch, is subdivided in alternating vein and intervein territories. The veins correspond to four longitudinal stripes of cells where several genes, such as argos, veinlet and Delta, are expressed. These genes belong to the Egfr (argos and veinlet) and Notch (Delta) signaling pathways, and their activities are required for the formation of veins of appropriate thickness. The interveins are characterized by the expression of blistered (bs), the Drosophila homolog to the human serum responsive factor; bs activity prevents the formation of vein tissue. The expression of E(spl)m beta, a Notch-downstream gene, is present in the wing pouch in broad domains that correspond to most interveins, and it is excluded from the developing veins (de Celis, 2000 and references therein).

The formation of the veins progresses during pupal development, when the expression of the transcription factor ventral veinless is activated in all presumptive vein territories. At this stage, it is possible to distinguish three different domains of expression in each presumptive vein territory, consisting of a central stripe where Egfr signaling is active, and two adjacent stripes where Notch signaling is active. The cells where Egfr signaling is active will differentiate as vein tissue in the adult wing during normal development. The anterior and posterior lateral stripes can be visualized by the expression of the Notch-downstream gene E(spl)m beta and in these cells vein differentiation is prevented. When Notch signaling is compromised, most of these cells differentiate as vein, suggesting that each group of central and associated lateral stripes constitute a vein-competent region (de Celis, 2000 and references therein).

The expressions of Salm and Salr in the wing pouch occur in the same domain, and the Salm and Salr expression territory as Salm/Salr domain. This domain is related to the position of veins. Thus, the limit of detectable Salm protein in the posterior compartment is adjacent to the domain of E(spl)m beta expression and bs downregulation in the L5 territory, and therefore it is placed several cells anterior to the L5 vein. In the anterior compartment, the Salm/Salr expression domain includes a broad territory of low levels of E(spl)m beta expression where the levels of Salm/Salr protein are very low. This territory includes a region from where, based on the expression of other markers, the vein L2 differentiates. The L2 vein can also be recognized because of a subtle but consistent reduction in the levels of bs expression. In the center of the wing pouch high levels of Salm/Salr are detected in two stripes of cells that correspond to the veins L3 and L4, and lower levels are present anterior and posterior to these veins, respectively. In conclusion, the expression of salm/salr in the wing pouch includes the L2 vein and extends to the anterior edge of the L5 territory (de Celis, 2000).

The heterogeneity in the levels of Salm/Salr could have a functional significance during vein patterning, and suggests that other factors, in addition to Dpp, participate in the regulation of salm/salr in the wing pouch. The expressions of Kni (L2) and Iro (L3 and L5) are also related to developing vein regions. Iro proteins are localized in L3 and L5, and are present in the vein and in the associated stripes of E(spl)m beta expression, both during imaginal and pupal development. In the larval disc, Kni is expressed in a domain broader than the vein L2 that corresponds to the region where E(spl)m beta is expressed at low levels. These observations indicate that Iro and Kni are expressed in vein competent regions, and that the loss of L2 and L3/L5 veins in kni and iro mutants, respectively, is due to failures in the specification of the corresponding vein competent region (de Celis, 2000).

The spatial relationships between the distribution of Kni, Iro and Salm were examined directly using appropriate antibodies. Kni is expressed within the anterior edge of the Salm/Salr expression domain, in the region where Salm/Salr are detected at lower levels. This differs from a previous report that placed the limit of Salm expression adjacent to but not overlapping with kni expression. Iro expression in the L5 vein competent region is, in contrast, immediately adjacent to the posterior limit of Salm expression. Thus, each individual vein expresses a unique combination of transcription factors (vein L2: Kni-C + Salm/Salr; veinL3: Iro + Salm/Salr; vein L4: Salm/Salr; L5: Iro) that are required for its formation and could confer individual characteristics to each longitudinal vein (de Celis, 2000).

Large mitotic clones of cell homozygous for a deficiency including salm and salr result in reorganizations to the venation pattern that are manifest both within and outside the salm/salr domain of expression. To characterize more specifically the roles of salm/salr in vein patterning, the effects of small salm and salr double mutant clones on vein development were analyzed. The phenotype of these clones depends on their position in the wing blade. All such clones that span L2 result in the elimination of this vein. However, cells where salm but not salr is mutant form normal L2 tissue. In addition these salm;salr mutant clones can also differentiate ectopic stretches of L2 when they are localized posterior to the normal L2 vein. This result suggests that salm and salr are not equivalent in their roles in L2 development, salm being more effective in suppressing and salr in promoting L2 differentiation. Small salm/salr mutant clones localized between the veins L2 and L3 and between L4 and L5 cause autonomous formation of ectopic vein tissue, irrespective of the wing surface where the clones appear. The differentiation of L3 and L4 is not affected in these clones, but mutant cells close to either of these veins tend to contribute to the vein, resulting in the displacement of L3 and L4. salm mutant clones in the posterior compartment result in similar but weaker phenotypes, suggesting that both salm and salr antagonize some vein promoting factor/s in the L2/L3 and L4/L5 interveins. Clones of salm/salr in the anterior compartment can also differentiate ectopic sensilla characteristic of L3, indicating that these genes repress other attributes typical of the L3 territory. Finally, most salm/salr mutant clones localized in the region between L3 and L4 differentiate normal intervein tissue or displace the L3 vein, suggesting that other factors in addition to salm/salr antagonize vein formation here. A good candidate for a vein-suppressing factor in the L3/L4 intervein is the gene knot, which encodes a transcription factor that is specifically expressed in response to Hh signaling in the L3/L4 intervein (de Celis, 2000).

The contrasting effects of salm/salr on the formation of specific veins (promoting L2 and suppressing L3, L4 and L5) indicate that these genes could both stimulate and antagonize the expression or activity of other vein-promoting genes. Thus, there is no evidence to indicate that salm and salr regulate vein differentiation directly; rather, they appear to influence vein development indirectly through regulating the expression of other genes that define individual veins. This regulation would require low levels of Salm/Salr to promote L2 and higher levels of Salm/Salr to inhibit L3 and L5 development. Two good candidates to be regulated by Salm/Salr are the kni-C and iro-C, because they are expressed in L2 and L3/L5, respectively. Furthermore, the function of kni-C and iro-C is required for the formation of these veins (de Celis, 2000).

To characterize the relationships of salm/salr with kni and iro-C, the effects of salm/salr mutant clones on kni and iro expression were examined. The expression of kni is eliminated in salm/salr clones that overlap the domain of Kni expression. These effects are cell-autonomous and can be observed in very small salm/salr clones. This suggests that, in contrast to the non-autonomous effect of Sal on kni, Salm/Salr regulate kni expression in a cell-autonomous fashion. All salm/salr clones localized between veins L2/L3 and L4/L5 are associated with ectopic expression of Iro proteins. Again, this effect is strictly cell-autonomous, suggesting that the repression of iro-C genes by Salm/Salr could be direct. Ectopic expression of Iro in salm/salr mutant clones is not observed in regions close to the dorso-ventral boundary, presumably because in this region iro-C expression is repressed by wingless. The effects of salm/salr on kni-C and iro-C expression have also been analyzed in experiments in which salm and salr are expressed ectopically using the GAL4 system. Widespread expression of salm or salr in the wing blade eliminates L2 and L5, and prevents expression, respectively, of Kni and Iro in the L2 and L5 territories of the corresponding imaginal discs. The expression of Iro in L3, which depends on Hh activity, is not affected by removal salm/salr functions and only slightly decreased by their ectopic expression. Taken together, these observations indicate that Salm/Salr negatively regulates iro-C expression in cells not exposed to Hh protein, and suggests that precise levels of Salm/Salr proteins are needed to activate kni expression in L2 (de Celis, 2000).

Spalt mediates an evolutionarily conserved switch to fibrillar muscle fate in insects

Flying insects oscillate their wings at high frequencies of up to 1,000 Hz and produce large mechanical forces of 80 W per kilogram of muscle. They utilize a pair of perpendicularly oriented indirect flight muscles that contain fibrillar, stretch-activated myofibres. In contrast, all other, more slowly contracting, insect body muscles have a tubular muscle morphology. This study has identified the transcription factor Spalt major (Salm) as a master regulator of fibrillar flight muscle fate in Drosophila. salm is necessary and sufficient to induce fibrillar muscle fate. salm switches the entire transcriptional program from tubular to fibrillar fate by regulating the expression and splicing of key sarcomeric components specific to each muscle type. Spalt function is conserved in insects evolutionarily separated by 280 million years. It is proposed that Spalt proteins switch myofibres from tubular to fibrillar fate during development, a function potentially conserved in the vertebrate heart—a stretch-activated muscle sharing features with insect flight muscle (Schönbauer, 2011).

To generate fast wing oscillations, both indirect flight muscle (IFM) units are attached to the thoracic exoskeleton. The contraction of one unit, the dorsal-longitudinal flight muscles (DLMs), deforms the thorax and moves the wings down; simultaneously it stretches and hence activates the second IFM unit, the dorsoventral flight muscles (DVMs), which moves the wings up again, generating an oscillatory movement of thorax and wings at high frequency. IFMs have a unique fibrillar organization to achieve these asynchronous, stretch-activated contractions (Schönbauer, 2011).

A genome-wide RNA interference (RNAi) screen was performed for muscle morphogenesis in Drosophila, and a function was identified for salm in IFM development (Schnorrer, 2010). The conserved Spalt family of transcription factors has two members in Drosophila, spalt major (salm) and spalt related (salr). RNAi knockdown of salm in muscle leads to viable but flightless animals with a reduced number of DLMs. Detailed analysis of the actin cytoskeleton revealed a striking change in fibre organization in salm knockdown IFMs: instead of the fibrillar IFM morphology with distinct, unaligned myofibrils and nuclei located between the fibrils, these muscles show a tubular morphology normally found in leg muscle, with aligned myofibrils and nuclei located in the tube centre. Leg muscles are normal in salm knockdown flies (Schönbauer, 2011).

Adult muscles develop in pupae by fusion of undifferentiated adult muscle progenitors (AMPs). DLMs form by fusion of AMPs with three larval templates, inducing their splitting into the six DLMs at 14 h after pupa formation (APF) (at 27°C). This splitting is inhibited in salm knockdown pupae. In wild-type DLMs, myofibrils start to assemble at 30 h APF with characteristically spaced nuclei between the fibrils and distinct, unaligned fibrils visible by 45 h. Leg myoblasts fuse and form tubular fibres with aligned filaments and nuclei located within the tube. In salm knockdown IFMs, distinct fibrils never form; instead, a tubular organization similar to leg muscles develops. Together, this evidence shows that salm is required to initiate IFM-specific muscle fate (Schönbauer, 2011).

To investigate the mechanism of how salm determines IFM identity, salm expression was analyzed. Salm is specifically expressed in adult IFMs, lost in salm knockdown and absent from leg muscles. At 12 h APF Salm is present in the DLM templates to which the AMPs fuse. This expression increases after template splitting at 24 h and is lost in salm knockdown IFMs. Using a GAL4-reporter line salm expression was detected in the templates from 8 h APF onwards throughout IFM development. With the same line, it was confirmed that salm is absent in developing leg muscles, consistent with the idea that salm selects fibrillar muscle fate (Schönbauer, 2011).

If salm indeed specifies fibrillar muscles, overexpressing salm in tubular muscle should switch its sarcomere organization from tubular to fibrillar. salm was ectopically expressed using Mef2-Gal4 in combination with Tub-GAL80ts, and the flies were shifted to restrictive temperature at 0 h APF, or using 1151-GAL4, which is expressed in AMPs and developing muscles until about 40 h APF. In both cases, ectopic salm expression induces a clear transformation of the tubular leg muscles into fibrillar IFM-like muscles. As a consequence, these transformed leg muscles do not function properly and flies die as pharate adults. A similar transformation was found in the abdominal muscles upon ectopic salm expression. This demonstrates that salm is sufficient to specify fibrillar muscle fate and to switch the developmental program from tubular to fibrillar fate. In trachea and eyes salm or both salm and salr are required for developmental fate decisions. However, the selection of fibrillar flight muscle fate is largely specific to salm, as knockdown of salr by RNAi does not cause a tubular transformation, and ectopic expression of salr in leg or abdominal muscle does not result in a fibrillar transformation. Consistently, a gain of the IFM-specific protein Fln was detected, together with a repression of the body-muscle-specific Mf-IsoB/D, Mlp84B and Mlp60, in salm- but not in salr-expressing leg muscle. Thus, it is concluded that salm is a master regulator of Drosophila indirect flight muscle development (Schönbauer, 2011).

As salm acts as a developmental switch, its muscle expression is restricted to IFMs. It is unclear how this precise expression is regulated. Salm is not expressed in larval AMPs; however, the larval AMPs that build the IFMs do express the transcription factor vestigial (vg). vg-null flies lack wings and halteres and have a defect in their IFMs. The morphology of vg mutant IFMs was analyzed in detail and notably the same phenotype was found as in salm knockdown IFMs. vg mutant DLMs are reduced in number and show a tubular fibre phenotype. Their leg muscles are normal, which is as expected because these flies are viable and can walk. Importantly, Salm protein is lost in vg mutant IFMs. To investigate whether vg has an additional function downstream of salm, salm was expressed using 1151-GAL4 in vg mutants, and a complete rescue of the vg IFM phenotype was found. No fibrillar transformation of leg muscles was observed, possibly because Salm levels driven with 1151-GAL4 in vg mutant legs are too low to override the leg muscle fate. Interestingly, overexpression of salr also results in some rescue of vg mutant IFMs, probably mediated by regained Salm expression. Together this demonstrates that vg is required upstream of salm for its IFM expression, and that salm does not require vg to implement the fibrillar flight muscle program (Schönbauer, 2011).

Interestingly, vg with its cofactor scalloped (sd) is not sufficient to induce fibrillar fate. Misexpression of vg and sd neither results in a fibrillar transformation nor in salm expression in leg muscles or wing disc AMPs. In contrast to vg, the Lbx1 homologue ladybird early (lbe) is specifically expressed in AMPs associated with the leg disc and can abrogate vg expression if misexpressed in the wing disc. Consistently, it was found that 1151-GAL4-driven lbe blocks Salm expression in the IFMs, leading to tubular IFM morphology. In summary, salm, but not vg, is capable of overruling the leg muscle program and determining the fibrillar muscle fate if expressed in leg myoblasts. It is proposed that in the absence of Salm the tubular fate program is initiated by default and does not necessarily require lbe, which is absent from many tubular muscles such as the abdominal muscles (Schönbauer, 2011).

To investigate further the mechanism by which salm induces and executes the fibrillar program, microarray analysis was performed of dissected wild-type IFMs and salm knockdown IFMs using two independent hairpin constructs, and of wild-type leg muscles. Notably, it was found that most known IFM-specific proteins or protein isoforms are downregulated in salm knockdown IFMs, including the IFM-specific stretch-sensitive TpnC4, Fln, Mf-IsoC, Prm-IsoC/D and Strn-Mlck-IsoE. Interestingly, vg was also identified as downregulated, suggesting that salm is required to maintain vg expression in IFMs and initiates a feed-forward loop by activating its own activator. Consistently, salm knockdown leads to a gain of body-muscle-specific proteins such as MP20, and body-wall-muscle-specific actins, TpnC41, Mlp84B, Mf-IsoB/D, Prm-IsoA and Msp300-IsoE/G. The salm-induced switch is largely transcriptional, but also changes alternative splicing, as is the case for Mf or Strn-Mlck. A number of these changes was confirmed by western blot and antibody staining. Again, salr expression is not changed in salm knockdown IFMs, arguing for a specific role of salm in IFM patterning. It was also noted that Act88F, which is enriched in IFMs as compared to leg muscles, is not changed in salm knockdown IFMs. However, Act88F is also expressed in a subset of tubular leg muscles, questioning its specific role in IFM development. Together, these data indicate that salm initiates a network of gene expression by regulating transcription and alternative splicing that switches the molecular architecture of the muscle from tubular to fibrillar morphology (Schönbauer, 2011).

Many winged insects use IFMs to move their wings at various frequencies. The IFM morphology was determined in different insect orders across an evolutionary distance of 280 million years. Calliphora was chosen as a second dipteran species, the wasp Nasonia as a hymenopteran and the beetle Tribolium as a coleopteran representative. All these species have a fibrillar organization of their IFMs and a tubular organization of their leg muscles. Salm expression in Calliphora was found to be IFM specific, indicating that the functional distinction of muscle types correlates with salm expression in dipteran species. To investigate functionally a potential role of spalt, systemic RNAi in Tribolium was used. Injection of spalt dsRNA into Tribolium larvae leads to pupae that are unable to complete metamorphosis and die as pharate adults. Histological analysis of the DLMs reveals a marked transformation to the tubular muscle morphology after spalt knockdown, as opposed to the fibrillar morphology in control injected animals. Hence, spalt is required in Tribolium, as it is in Drosophila, to specify fibrillar flight muscles, suggesting that spalt function as a regulator of fibrillar flight muscles is conserved in all insects harbouring stretch-activated indirect flight muscles (Schönbauer, 2011).

Mice and humans possess four spalt-like (SALL) genes, none of which are expressed in differentiated striated body muscles. This is not surprising, as all vertebrate body muscles harbour aligned sarcomeres that resemble the tubular insect muscles. Interestingly, SALL1 and SALL3 are both expressed in mouse and human hearts, which contain distinct unaligned myofibrils in cardiomyocytes and utilize the stretch-modulated Frank–Starling contraction mechanism. Mutations in human SALL1 cause the heart abnormalities observed in Townes–Brocks syndrome, leading to the idea that spalt function determines fibrillar stretch-activated muscle all the way up to vertebrates (Schönbauer, 2011).

The transcription factor Spalt and human homologue SALL4 induce cell invasion via the dMyc-JNK pathway in Drosophila

Cancer cell metastasis is a leading cause of mortality in cancer patients. Therefore, revealing the molecular mechanism of cancer cell invasion is of great significance for the treatment of cancer. In human patients, the hyperactivity of transcription factor Spalt-like 4 (SALL4) is sufficient to induce malignant tumorigenesis and metastasis. This study found that when ectopically expressing the Drosophila homologue spalt (sal) or human SALL4 in Drosophila, epithelial cells delaminated basally with penetration of the basal lamina and degradation of the extracellular matrix, which are essential properties of cell invasion. Further assay found that sal/SALL4 promoted cell invasion via dMyc-JNK signaling. Inhibition of the c-Jun N-terminal kinase (JNK) signaling pathway through suppressing matrix metalloprotease 1 or basket can achieve suppression of cell invasion. Moreover, expression of dMyc, a suppressor of JNK signaling, dramatically blocked cell invasion induced by sal/SALL4 in the wing disc. These findings reveal a conserved role of sal/SALL4 in invasive cell movement and link the crucial mediator of tumor invasion, the JNK pathway, to SALL4-mediated cancer progression (Sun, 2020).

Human SALL4 has been reported to be significantly elevated in metastatic cancer cells. This study provides genetic evidence for a model in which sal/SALL4 regulates cell invasiveness by dMyc-JNK signaling. The JNK pathway is an important cellular signaling pathway that regulates a variety of cellular activities relevant to tumorigenesis, such as cell migration, apoptosis and proliferation. JNK promotes the expression of Mmp1, which acts as an enzyme to degrade basement membrane and ECM components to promote tumor cell motility. Manipulation of expression of many genes can lead to cell death, cell extrusion and invasive cell migration through activation of JNK signaling. sal/SALL4 overexpression activates Mmp1 and reducing JNK can suppress cell invasion and Mmp1 level. In addition to Mmp1, some other markers in the JNK pathway such as pJNK (activated bsk) and puc showed a significant increase in expression. Promotion of cell invasion by sal/SALL4 induction was accompanied by activation of the apoptotic pathway, but it was not dependent on apoptosis because caspase inhibition did not prevent cell invasion upon sal/SALL4 expression. Therefore, the JNK pathway probably mediates the role of sal/SALL4 overexpression to regulate cell invasion through an apoptosis-independent mechanism (Sun, 2020).

The MYC gene is one of the most highly amplified oncogenes among many human cancers. For instance, in some certain cancer cells, Myc is upregulated through directly transcriptional activation by SALL4. Besides promoting cancer progression and metastasis, MYC has a bivalent role in regulating tumorigenesis and cell invasion. MYC restrains breast cancer cell motility and invasion through transcriptional silencing of integrin subunits. In Drosophila, dMyc inhibits JNK signaling in retinal progenitors to block non-autonomous glia over-migration (Tavares, 2017). The Drosophila puc gene, encoding the sole JNK-specific MAPK phosphatase and inhibitor, and its mammalian homologue Dusp10 are directly bound by Myc as shown in ChIP-sequencing data . In Drosophila tissues, direct evidence illustrates that dMyc and cMyc activate puc transcription through binding to the Myc binding-motif EB3, and consequently inhibit JNK signaling to suppress cell invasion. This study found that dMyc is repressed in sal/SALL4-expressing regions and introducing dMyc partially rescues cell invasion, indicating a repressive role of dMyc in tumor cell migration. As Sal is a transcriptional repressor in both Drosophila and human cells, it is possible that Sal/SALL4 binds to Myc and suppresses its expression because the cMyc promoter has putative binding sites that are available to Zinc finger binding. Sall2, another emerging cancer player in the Sall family, binds to the cMyc promoter region and represses cMyc expression. Thereby, sal/SALL4 may activate JNK signaling through the repression of puc, which is activated by dMyc in Drosophila (Sun, 2020).

Cell competition occurs when Myc is unevenly distributed between cells. Clones expressing high levels of Myc expand and eliminate the surrounding cells by apoptosis. On the contrary, downregulation of Myc in clones leads to their elimination. Given sal/SALL4-expressing cells are relatively lower Myc expression, it is possible that the surrounding cells with higher Myc expression become competitors and eliminate those lower Myc expression cells. Intriguingly, sal/SALL4-induced migrating cells are not dead and inhibiting cell death cannot repress sal/SALL4-induced cell invasion, so the mechanism may not be apoptosis-driven cell elimination. Previous studies found that JNK activation in surrounding wild-type cells promotes elimination of their neighboring scrib mutants by activating the PVR-ELMO/Mbc-mediated engulfment pathway, and the surrounding JNK is independent of JNK activation in mutant clones. Distinct from this, sal/SALL4-activated non-autonomous activation of JNK is dependent on JNK activation in sal/SALL4-expressing cells. Whether JNK-dependent engulfment plays a major role in sal/SALL4-mediated extrusion needs to be addressed in the future (Sun, 2020).

GENE STRUCTURE

Introns - three

Bases in 3' UTR - 1903

PROTEIN STRUCTURE

Structural Domains

In Spalt are found three widely spaced "double zinc finger" motifs, which have internally conserved sequences; in addition, a single zinc finger motif of a different sequence is found in the central region. Each C-terminal zinc finger of the double zinc fingers contains a stretch of eight conserved amino acids with the sequence KTTKGNLK, similar to the zinc finger sequences of human transcription factor PRDII-BF1 (Kühnlein, 1994).

date revised: 2 February 2023

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.