slow border cells


Promoter Structure

A cannonical TATA box is found 20-25 bp upstream of the P transcriptional start site, while no obvious TATA box is found near the P1 start site. Footprinting analysis has identified a C/EBP-binding site between the P2 TATA box and the P2 transcriptional start site. The C/EBP-binding site is conserved in D. virilis (Rorth, 1992).

Transcriptional Regulation

During Drosophila oogenesis, border cells perform a stereotypic migration. Slbo, a C/EBP transcription factor, is required for this migration. Drosophila Stat92E has been identified in a screen for gain-of-function suppressors of the slbo mutant phenotype. By clonal analysis for Stat92E and hop mutants it has been found that the JAK/STAT pathway is required in border cells for their migration. The activating ligand for the pathway, Unpaired, is expressed in polar cells. Polar cells are specialized cells that can induce border cell fate in anterior follicle cells. On its own, ectopic expression of Unpaired can induce ectopic expression of border cell markers, including Slbo. However, Stat92E mutant cells still express normal levels of Slbo protein, thus Stat92E must regulate other targets critical for border cell migration (Beccari, 2002).

Egg chambers are initially formed in the germarium, which also contains the germ line and somatic stem cells. When an egg chamber buds off from the germarium, few somatic follicle cells show any specialization. The specialized cells are the two polar cells present at each end of an egg chamber (follicle) and the interfollicular stalk cells. Polar cells can be identified by expression of specific molecular markers in early as well as late egg chambers. Polar and stalk cells stop dividing before exit from the germarium. The remaining follicle cells continue to proliferate, and differentiate after stage 6 when the egg chamber has its final complement of 700-1000 follicle cells (Beccari, 2002).

From stage 9 of oogenesis, both marker gene expression and cell shape can be used to distinguish different follicle cell populations at the anterior end of the egg chamber. Most anterior are the border cells, consisting of the polar cells themselves plus about six immediately adjacent cells ('outer' border cells). The border cells delaminate from the follicular epithelium as a tight cluster at the beginning of stage 9 and migrate between the nurse cells to the oocyte. Adjacent to the border cells are the stretch cells, which stretch very thin to cover the nurse cells. Finally the centripetal cells, which later will cover the anterior end of the egg, are located between the stretch cells and the columnar 'main body' follicle cells. Anterior follicle cell fates are repressed in the corresponding follicle cells at the posterior end of the egg chamber by prior Gurken-EGFR signaling from the oocyte (Beccari, 2002).

Production of ectopic polar cells by exposing early egg chambers to increased Hedgehog expression appears sufficient to induce ectopic migrating border cells at stage 9. A slbo-lacZ enhancer trap is induced in extra migrating clusters at stage 9. Similar ectopic border cell clusters have been observed in egg chambers with clones of follicle cells mutant for costal2, a negative regulator of the Hedgehog signal transduction pathway. Thus the presence of polar cells, and absence of posteriorizing signal from the oocyte, may be sufficient for the induction of border cells at the appropriate developmental stage. What signals from polar cells may be responsible for induction of border cell fate in adjacent follicle cells? There is good evidence that Upd is a key signal from polar cells: Upd is specifically expressed in polar cells and acts non cell autonomously; ectopic expression of Upd induces two border cell markers; and the JAK/STAT pathway is required in border cells. Previous studies of the JAK/ STAT pathway in Drosophila have indicated that Upd expression induces Stat92E activation through the JAK kinase Hop and that the effects of Upd can be explained in this manner. Ectopic expression of Upd induces ectopic expression of Slbo. Since the JAK/STAT pathway is required in border cells and thus must be active there, Upd regulated Stat92E may normally contribute to Slbo up-regulation in border cells (Beccari, 2002).

However, given that the Stat92E mutant affected border cell migration without affecting Slbo expression, the JAK/STAT pathway may not be required for Slbo expression. One reasonable explanation is that another signal from polar cells contributes to activation of Slbo in border cells. This upstream signal may by itself be required for Slbo expression or may act redundantly with Stat92E. The additional signal may be a novel effect of Upd, not mediated by the JAK/STAT pathway. However, given the inability of Upd to convert stretch cells to border cells, it is thought a different signal is more likely. Irrespective of its potential effect on Slbo, the effect of Stat92E mutant clones shows that other targets of Stat92E must be critical for border cell migration (Beccari, 2002).

Just as ectopic Slbo expression is not sufficient to convert other cells into border cells, Upd misexpression and ectopic activation of Stat92E is also not sufficient to convert stretch cells into migrating border cells. In the latter situation, the stretch cells experience both Stat92E activation and Slbo expression. The stretch cells nevertheless do not assume border cell fate. This has several possible explanations. The signal invoked above as upstream regulator of Slbo may, in addition to Slbo, have other target genes required for migration which are not being induced by Upd. Alternatively, there may be yet another signal from polar cells that is required for border cell differentiation. While one of these two explanations is favored, there are other possibilities. The stretch cells may already have been specified at the time of ectopic Upd expression, and thus be refractive to additional inductive signals. Also, when functional ectopic border cells are induced by extra polar cells, the timing and levels of signals to adjacent cells are likely to be relatively normal. This may not be the case when Upd is ectopically expressed (Beccari, 2002).

In addition to the spatial signal described above, a temporal signal must turn on expression of Slbo and other markers at the right stage. Upd and other polar cell markers are expressed in polar cells from earlier stages. Yet normal polar cells, or Hedgehog-induced ectopic polar cells, only induce border cells and border cell markers at stage 9. The temporal signal(s) may either modify polar cell signals to make them functional at the right stage, or act directly on border, stretch and centripetal cells to influence expression of target genes. Given the early expression of Upd and given that marker genes are induced in follicle cells with somewhat different temporal profiles, the latter scenario is favored. Two known candidates for supplying temporal signals are late Delta/Notch signaling and hormonal regulation (ecdysone). Analysis of a temperature-sensitive Notch allele has shown that Notch was required for Slbo expression. It has recently been shown that signaling by germ line Delta to Notch in follicle cells is required for proper differentiation of all follicle cells after stage 6. Although required for differentiation, the direct effect of Delta/Notch signaling at stage 6 is unlikely to explain the onset of Slbo expression at stage 8/9. But a cascade of events initiated at stage 6 might indirectly lead to expression of differentiation markers 16-24 h later. There is also evidence that some stage specific gene expression in egg chambers is regulated by the hormone ecdysone. In addition, the ecdysone receptor, EcR and its partner, Usp, appear to be required for border cell migration. One experiment in this study has suggested that ecdysone regulates timing of border cell migration, but apparently not timing of Slbo expression. Hormone application requires additional ectopic expression of Slbo to induce premature border cell migration. Thus the temporal regulation of anterior follicle cell differentiation may also have multiple components. Given that the stages of oogenesis are well-studied, this will be an interesting system in which to determine how temporal and spatial regulation of differentiation is coordinated (Beccari, 2002).

Targets of Activity

The JAK/STAT signaling pathway, renowned for its effects on cell proliferation and survival, is constitutively active in various human cancers, including ovarian. JAK (Hopscotch) and STAT are required to convert the border cells in the Drosophila ovary from stationary, epithelial cells to migratory, invasive cells. The ligand for this pathway, Unpaired (Upd: Outstretched), is expressed by two central cells within the migratory cell cluster. Mutations in upd or jak cause defects in migration and a reduction in the number of cells recruited to the cluster. Ectopic expression of either Upd or JAK is sufficient to induce extra epithelial cells to migrate. Thus, a localized signal activates the JAK/STAT pathway in neighboring epithelial cells, causing them to become invasive (Silver, 2001).

In order to gain further insight into the mechanism by which STAT regulates border cell migration, the expression of a number of proteins that are highly expressed in border cells was examined, some of which are also required for migration. The first gene identified as playing a critical role in border cell migration was slow border cells (slbo). slbo encodes Drosophila C/EBP, a basic region/leucine zipper transcription factor. Slbo protein expression is undetectable in stat mutant border cells, which were identified using a positive clone marking system known as MARCM. This result was confirmed by examining several additional proteins, the expression of which is reduced in slbo mutant border cells. In wild-type stage 8 and 9 egg chambers, FAK expression is upregulated in migratory border cells. Border cells that lack stat exhibit reduced levels of FAK. In wild-type stage 9 egg chambers, DE-cadherin (Shotgun) is enriched throughout the border cell cluster and is expressed to the highest level in the polar cells. Stat mutant border cells exhibit reduced DE-cadherin expression compared to wild-type border cells of the same cluster. The polar cells, though mutant, do not show reduced DE-cadherin staining, which is also true of slbo mutants. Additional downstream targets of slbo, including PZ6356 and zinc finger transcription factor jing, are also reduced in stat mosaic clones. Thus, even the few mutant cells that are recruited to the cluster fail to express many border cell proteins required for migration. The effect is specific because expression of Taiman, a protein that is required for border cell migration but is independent of the slbo pathway, was not altered. Mosaic clusters containing a mixture of wild-type and mutant cells show variable migration defects. On average, the extent of migration is proportional to the number of wild-type cells in the cluster (Silver, 2001).

Egg chambers from females heterozygous for any of the stat alleles have a semi-dominant border cell migration phenotype. Advantage was taken of this slight haploinsufficiency to test for dominant genetic interactions with other genes required for border cell migration. Dominant genetic interactions were observed with slbo, hop, and upd alleles. A mutation in the gene coding for DE-cadherin, shotgun, also exhibited a dominant interaction with stat. These interactions appeared to be specific, since stat does not interact with other known border cell migration genes, such as tai, jing, or PZ6356 (Silver, 2001).

The slbo locus encodes Drosophila C/EBP, a basic region-leucine zipper transcriptional regulator. It is intriguing to note that mammalian C/EBPß is expressed in ovarian carcinomas, and its expression increases dramatically with malignancy. STAT3 is constitutively active in ovarian carcinomas, and while it is not known whether STAT3 activates expression of C/EBPß in these cells, it could be that this relationship has been conserved in evolution (Silver, 2001).

In addition to Slbo, expression of each of its known target genes was reduced in stat mutant cells. The contributions of several of these target genes to cell migration is known. For example, dynamic regulation of DE-Cadherin plays a critical role in promoting migration by providing optimal adhesion with the nurse cells. Furthermore, FAK is essential for the migration of numerous mammalian cell types, while jing encodes a zinc finger transcription factor that cooperates with Slbo in regulating border cell migration (Silver, 2001).

Both loss-of-function and gain-of-function of JAK/STAT pathway activity are detrimental to border cell migration. Interestingly, this is also true for Slbo, since slbo mutants show border cell migration defects and overexpression of slbo also impedes migration. This similarity lends further support to the proposal that STAT exerts at least part of its effect on migration by regulating Slbo (Silver, 2001).

Two transcriptional regulatory pathways have been identified that control the invasive behavior of the border cells in vivo. In addition to expression of slbo and its targets, border cell migration requires a global hormonal signal in the form of ecdysone. This global hormonal signal appears to function in a slbo-independent manner, since the expression of neither slbo nor its targets is reduced when ecdysone signaling is compromised, and no genetic interaction has been observed between mutations affecting the ecdysone response and slbo or stat. Taken together with the putative guidance signal PVF-1 and the data presented here, these results indicate that border cell migration requires the integration of at least three signals. The global hormonal signal coordinates multiple events that occur at stage 9, including border cell migration, and PVF-1 contributes to the directional cue for the border cells. Finally, the local paracrine signal through JAK/STAT is necessary to define the population of cells capable of responding to the other signals by detaching from the epithelium and invading the nurse cell cluster. Of these three signals, only the signal through the JAK/STAT pathway is spatially restricted to the migratory population (Silver, 2001).

SLBO contains a basic region/leucine zipper (bZIP) DNA-binding domain very similar to that of mammalian C/EBP and the purified C/EBPs bound to DNA with the same sequence specificity. Among the DNA sequences that DmC/EBP binds with high affinity is a conserved site within the promoter of the DmC/EBP gene itself (Rorth, 1992).

Timely initiation of border cell migration requires SLBO. One target of SLBO in the control of border cell migration is the FGF receptor homolog encoded by the breathless locus. btl expression in the ovary is border cell-specific, beginning just prior to the migration; this expression is reduced in slbo mutants. btl mutations dominantly enhance the border cell migration defects found in weak slbo alleles. Furthermore, SLBO-independent btl expression is able to rescue the migration defects of hypomorphic slbo alleles. Purified SLBO binds eight sites in the btl 5' flanking region by DNAse I footprinting. Taken together these results suggest that btl is a direct target for SLBO, and key in the regulation of border cell migration (Murphy, 1995).

Negative results may often prove instructive. A single copy of the o-r enhancer from yolk protein genes directs female- and fat body-specific transcription. It consists of four protein-binding sites: dsxA, which binds male (DSXM) and female (DSXF) proteins encoded by the doublesex gene; aef1, which binds the AEF1 repressor; bzip1, which binds the DmC/EBP activator encoded by the slbo gene; and ref1, which binds an unknown activator. DSXF activates from dsxA by sterically excluding AEF1 repressor from the aef1 site and synergistically activating transcription together with a protein at bzip1. Sex specificity in fat bodies arises from the opposite effect of DSXM, a repression of protein activity at bzip1. Tissue specificity is regulated by all four DNA sites. Separately, bzip1 and ref1 activate transcription in ovarian somatic cells and all nongonadal tissues, respectively, whereas together they activate only in fat bodies. The aef1 site represses ectopic transcription in ovaries; dsxA antirepresses this activity in fat bodies. Thus, in the organism, ref1 and bzip1 act combinatorially to direct the fundamental tissue specificity, aef1 and dsxA modulate this tissue specificity, and dsxA adds sex specificity. However, SLBO does not appear to be the bZIP factor regulating the bzip1 site. SLBO is not found in adult fat bodies, and altered SLBO levels do not alter the fat body expression of Yp genes (An, 1995a).

Transcription of the Drosophila yolk protein (Yp) genes is regulated by the somatic sex determination pathway. A gene at the bottom of this pathway, doublesex, encodes the female-specific DSXF and male-specific DSXM proteins that bind to and regulate transcription from several sites in the Yp genes. Site-directed mutagenesis, protein binding and germline transformation experiments have been carried out that identify and characterize the activity of a single binding site (dsxA) for the Doublesex proteins and two binding sites for other regulatory proteins. The fat body enhancer (FBE), located between -197 and -332 of yp transcriptional start, has three bzip binding sites that bind Slbo non-cooperatively. A single copy of the three sites is sufficient to direct the sex and fat body specificities of Yp transcription. The sites form an enhancer with two strongly synergistic enhancer elements. One element (22 bp) consists of dsxA and an overlapping site, bzip1, that binds the DmC/EBP (Slbo) protein, a member of the bZIP family of transcriptional activators. bzip1 is the strongest of the three FBE binding sites for Slbo and dsxA is the strongest FBE binding site for DSXM and DSXF. Overlapping these two elements is a binding site for Aef1, the only FBE binding site for AEF1, a Drosophila repressor protein. The other element is an 11 bp binding site (ref1) for an unknown protein. Tissue-specific activation requires strong cooperation between the ref1 site and the bzip1 or dsxA sites. Sex specificity is regulated exclusively by the dsxA site which connects the sex determination pathway to the target gene through DSXM repression and DSXF activation. The dsxA site is not necessary for tissue specificity. The bzip1 site activates transcription in both sexes, but does not appear to be necessary for tissue specificity (An, 1995b).

Epithelial to mesenchymal transitions and cell migration are important features of embryonic development and tumor metastasis. A systematic genetic approach has been applied to study the border cells in the Drosophila ovary, as a simple model for these cellular behaviors. Expression of the basic-region/leucine zipper transcription factor, C/EBP, is required for the border cells to initiate their migration. A second nuclear factor, named Jing (which means 'still'), is required for initiation of border cell migration. The jing locus was identified in a screen for mutations that cause border cell migration defects in mosaic clones. The jing mutant phenotype resembles that of slbo mutations, which disrupt the Drosophila C/EBP gene, but is distinct from other classes of border cell migration mutants. Expression of a jing-lacZ reporter in border cells requires C/EBP. Moreover, expression of jing from a heat-inducible promoter rescues the border cell migration defects of hypomorphic slbo mutants. The Jing protein is most closely related to a mouse protein, AEBP2, which was identified on the basis of its ability to bind a small regulatory sequence within the adipocyte AP2 gene to which mammalian C/EBP also binds. It is proposed that the need to coordinate cell differentiation with nutritional status may be the link between mammalian adipocytes and Drosophila border cells that led to the conservation of C/EBP and AEBP2 (Liu, 2001).

Jing exhibits 50% amino acid identity with AEBP2 within the zinc finger motifs and 20% identity C-terminal to the zinc fingers (BLAST E value of 1e-34). After AEBP2, the most similar proteins are several members of the GLI family of zinc finger transcription factors. However GLI proteins typically contain five zinc finger motifs. Members of the GLI family of proteins, which includes Drosophila CI, are only 25% identical to Jing within the zinc fingers and do not exhibit homology outside of these motifs (Liu, 2001).

What are the similarities between control of border cell migration and adipocyte differentiation? Undoubtedly many genes are required for an epithelial cell to become motile; therefore it is striking that the jing locus encodes a protein with such a clear functional connection to C/EBP. Evolution, it seems, has conserved functional networks of transcriptional regulators, rather than individual genes. Mammalian proteins related to Jing and Slbo appear to be involved in the regulation of adipocyte differentiation, a process that bears little obvious resemblance to border cell migration. One similarity between adipocytes and border cells is that both cell types appear to coordinate their differentiation with nutritional status of the organism. In the case of adipocytes, at least two independent transcriptional regulatory pathways appear to be required. One pathway is the C/EBP pathway, which may also require AEBP2. A second pathway requires the activity of PPARgamma, a steroid hormone receptor-like molecule that is activated by circulating fatty acids whose levels depend upon diet. Drosophila ovarian development also responds to nutritional signals. Flies emerge from the pupal case with ovaries that contain only immature egg chambers. Further progression of oogenesis requires that the flies consume a rich diet. Application of lipophilic hormones, such as juvenile hormone or ecdysone, can bypass this requirement. The ecdysone receptor is required in the border cells for their migration. Thus border cell migration, like adipocyte differentiation, requires a hormonal signal, which reflects nutritional status, to be integrated with an intrinsic developmental program. It is this latter program that appears to be mediated by the C/EBP, AEBP2 and the Drosophila proteins Slbo and Jing (Liu, 2001 and references therein).

Analysis of cell migration using whole-genome expression profiling of migratory cells in the Drosophila ovary

Cell migration contributes to normal development and homeostasis as well as to pathological processes such as inflammation and tumor metastasis. Previous genetic screens have revealed signaling pathways that govern follicle cell migrations in the Drosophila ovary, but few downstream targets of the critical transcriptional regulators have been identified. To characterize the gene expression profile of two migratory cell populations and identify Slbo targets, border cells and centripetal cells expressing the mouse CD8 antigen were purified and whole-genome microarray analysis was carried out. Genes predicted to control actin dynamics and the endocytic and secretory pathways were overrepresented in the migratory cell transcriptome. Mutations in five genes, including ttk, failed to complement previously isolated mutations that cause cell migration defects in mosaic clones. Functional analysis revealed a role for the Notch-activating protease Kuzbanian in border cell migration and identified Tie receptor tyrosine kinase as a guidance receptor for the border cells (Wang, 2006).

Gene expression profiling of migratory cells in the Drosophila ovary has allowed comparison of the global patterns of gene expression of developmentally regulated cell movements to that previously reported for invasive carcinoma cells. Of 30 genes that encode motility-associated proteins that were identified as upregulated in invasive breast carcinoma relative to the primary tumor, 23 have easily identified Drosophila homologs. Of these, 11 (48%) were identified as upregulated in migratory follicle cells in the current analysis. This seems noteworthy given that the cells derive from different organisms and different tissues. In contrast, only one of the cytoskeleton-associated, migratory cell-enriched genes was identified out of the top 419 genes upregulated in the adult Drosophila eye (Wang, 2006).

Finding a large number of genes that are differentially expressed in a microarray analysis can make it difficult to decide which individual genes merit additional, detailed study. One approach to limiting the number of genes in an analysis is to use stringent fold-change cutoffs. However, it is not clear that this is the best way to derive biologically meaningful information from large data sets. An alternative approach was used, employing a sensitive method to reveal a large number of differentially expressed genes and then separating the large data set into smaller sets by using gene ontology with GO Slim. This allowed discernment, in a relatively unbiased manner: genes that encode cytoskeletal proteins and proteins associated with the secretory and endosomal pathways were overrepresented in the migratory cell-enriched genes compared to the genome as a whole, providing a rationale for the selection of smaller, functionally related subsets of genes for further study (Wang, 2006).

The overrepresentation of cytoskeleton-associated gene products among the migratory cell-enriched genes is interesting to consider in light of the striking morphology of border cells during their migration. One, or occasionally two, cells at the front of the cluster extend a long dominant protrusion that can be up to 50 μm long. This may be a common morphology for cells migrating in vivo, since it has also been observed for cells of the rostral migratory stream and neural crest cells. It seems reasonable to propose that this extended morphology may require special regulation of the cytoskeleton: such regulation might differ in some respects from that of the broad, flat lamellae and ruffles formed by cells cultured on two-dimensional surfaces. For example, longer parallel bundles of F-actin are probably required to create and maintain long protrusions such as those observed in border cells (Wang, 2006).

Although the general idea that proteins associated with the cytoskeleton are important in migratory cells is not surprising, this analysis leads to generation of hypotheses regarding specific genes. For example, two proteins known to promote long, parallel actin bundles are among the migratory cell-enriched genes, including tropomyosin and fascin. Loss of function of fascin (encoded by the gene singed) does not result in a discernible border cell migration defect; however, this may be because of redundancy with tropomyosin. Similarly, loss of the filamin-like protein encoded by the cheerio locus causes a mild border cell migration defect. The microarray analysis reveals that another filamin-like protein (Jitterbug) is expressed at a higher level in the migratory follicle cell population. The microarray data therefore can guide the development of specific, testable hypotheses concerning possible gene redundancies (Wang, 2006).

In addition to proteins with well-characterized functions in actin dynamics, such as actin and actin-related proteins, a number of genes emerged from the microarray analysis that encode proteins with motifs or domains that suggest a specific role in regulating the actin cytoskeleton, but which have not yet been characterized at all. These include Rexin, a protein composed of three SH3 domains, and CG31352, which encodes a protein composed of three LIM domains and a motif resembling the villin headpiece. Mammalian homologs of these proteins exist but have not been characterized. It will therefore be of interest to determine if these genes and their products represent evolutionarily conserved, but previously unrecognized, contributors to cell motility (Wang, 2006).

Genes encoding proteins associated with the endoplasmic reticulum, Golgi apparatus, cytoplasmic vesicles, and endosomes were significantly overrepresented among the migratory cell-enriched genes compared to the genome as a whole. This observation suggests that border cells have a special need for dynamic trafficking of proteins to and from the cell surface. It has been proposed that dynamic cell-cell adhesion between border cells and nurse cells is required for the cells first to gain traction and then to translocate, and that this may involve high rates of turnover of membrane proteins such as E-cadherin. Moreover, it is clear that several receptor molecules such as Domeless, PVR, and EGFR are present at lower concentrations on the surfaces of the border cells as compared to other follicle cells. Therefore, it seems likely that there is a high rate of movement of these proteins onto and off of the plasma membrane. All of this traffic would likely require an upregulation of proteins functioning in the secretory and endocytic pathways. Consistent with this hypothesis, it has been shown that multivesicular bodies are markedly more prevalent in migrating border cells as compared to other follicle cells in the egg chamber (Wang, 2006).

Relatively little is known about the mechanisms governing centripetal cell migration. Border cells and centripetal cells take two different paths, but they arrive at the same place. Both cell types express Slbo and require E-cadherin, Rac, and myosin II for their respective movements. Thus, mechanical aspects of these two migrations may be similar. However, there are also differences in the migrations of these two cell types. Border cells completely exit the follicle cell epithelium during their migration down the center of the egg chamber. Centripetal cells, in contrast, stay connected to the outer follicle layer. In addition, the directions of the two migrations are quite different. Whereas border cells migrate posteriorly, centripetal cells migrate symmetrically toward the center, orthogonal to the path of border cell migration. Therefore, the cues that direct the two migrations must be different. Consistent with this, none of the known border cell guidance receptors is required for centripetal cell migration. In addition, the border cells and centripetal cells initiate migration at distinct times: the border cells complete their migration before the centripetal cells begin. The gene expression profile presented in this study provides a wealth of candidate genes to test for effects on centripetal cell migration and to flush out the similarities and differences between border cell and centripetal cell migration (Wang, 2006).

One goal on this study in determining the gene expression profile of border cells was to facilitate the molecular identification of genes corresponding to the mutations that cause border cell migration defects in mosaic clones. Five genes in the microarray lists, gliotactin, tramtrack, catsup, latheo, and zipper, were matched to mutant lines identified in mosaic screens. The next challenge will be to elucidate precisely how each of these genes contributes to border cell migration (Wang, 2006).

In addition to facilitating the identification of genes that cause cell migration defects in mosaic clones, the gene expression profile can identify genes that would be unlikely to be identified in such genetic screens. For example, all of the known guidance factors for border cell migration produce either no defect when mutated individually (ligands for the EGF receptor) or quite mild defects (PVF1). Their contributions become much more obvious when multiple mutations are combined. Therefore, it is a challenge to identify this class of proteins by using conventional forward genetics. In the gene expression profile reported here, an uncharacterized receptor tyrosine kinase was found to be expressed at higher levels in migratory cells and to be an Slbo target. Expression of a putatively dominant-negative form of this receptor exacerbated the migration defects associated with loss of PVR alone or loss of PVR and EGFR, implicating this receptor in guidance of border cell migration. Therefore, the expression profile has provided a source of candidate genes that would be difficult or impossible to identify by other methods (Wang, 2006).

Systematic analysis of the transcriptional switch inducing migration of border cells

Cell migration within a natural context is tightly controlled, often by specific transcription factors. However, the switch from stationary to migratory behavior is poorly understood. Border cells perform a spatially and temporally controlled invasive migration during Drosophila oogenesis. Slbo, a C/EBP family transcriptional activator, is required for them to become migratory. Wild-type and slbo mutant border cells as well as nonmigratory follicle cells were purified and comparative whole-genome expression profiling was performed, followed by functional tests of the contributions of identified targets to migration. About 300 genes were significantly upregulated in border cells, many dependent on Slbo. Among these, the microtubule regulator Stathmin was strongly upregulated and was required for normal migration. Actin cytoskeleton regulators were also induced, including, surprisingly, a large cluster of 'muscle-specific' genes. It is concluded that Slbo induces multiple cytoskeletal effectors, and that each contributes to the behavioral changes in border cells (Borghese, 2006).

Only one of the identified cytoskeletal regulators is known to affect microtubules, namely, Stathmin. Mammalian Stathmin/Op18 protein is well characterized. It binds to microtubules and promotes depolymerization by sequestration of tubulin dimers or direct action at microtubule ends. Interestingly, the activity of Stathmin can be regulated by phosphorylation in response to signaling or cell cycle phases. Drosophila Stathmin appears to have similar biochemical features. The availability of an antibody directed against Drosophila Stathmin allowed analysis of protein levels in situ. As expected, the level of Stathmin was higher in border cells than follicle cells. When analyzing slbo mutant border cells, a clear difference was observed between the inner polar cells and the outer border cells. The outer border cells are the migratory cells and require Slbo expression. In these cells, Stathmin expression was undetectable in the absence of Slbo, indicating a very strong dependence on Slbo. In contrast, Stathmin was still expressed in mutant polar cells, explaining why only a moderate reduction of stathmin mRNA levels was seen in whole border cell clusters (Borghese, 2006).

To analyze the function of Stathmin in border cells, stathmin mutants were generated. This was done by imprecise excision of a P element located immediately upstream of the stathmin C transcript. A mutant deleting only the stathmin C isoform (stathminexC), leaving stathmin A and B intact, was homozygous viable and had no effect on border cell migration. A mutant deleting the complete stathmin locus (stathminL27) and four adjacent genes (including Arc-p20, a component of the Arp2/3 complex) was homozygous lethal, and clones of stathminL27 mutant border cells were unable to migrate. Both the lethality and the migration block were rescued by reintroducing ubiquitously expressed stathmin and Arc-p20 at the same time. Reintroducing Arc-p20 alone did not rescue border cell migration, indicating that stathmin is essential for this process. To interfere with stathmin upregulation at the time of migration, a functional stathmin “hairpin”-RNAi construct was expressed in the sensitized stathminexC/stathminL27 background. By using the slbo-GAL4 driver, stathmin RNAi expression could be could specifically targeted to outer border cells right before and during migration. This strongly decreased the amount of Stathmin protein in border cells and caused significant delays in migration. The delays in migration could be reversed by driving higher levels of stathmin expression from a UAS construct. These results identify Stathmin as an important regulator downstream of Slbo. To test whether lack of Stathmin was solely responsible for the slbo phenotype, Stathmin was overexpressed in the slbo mutant background. Migration was not restored, indicating that additional genes downstream of Slbo must also be important (Borghese, 2006).

Singed is an actin-bundling protein related to Fascin, highly expressed in border cells. Fascin is important for the formation of cell protrusions and has been implicated in the control of cell migration, also in vivo. It was confirmed by clonal analysis that Singed protein levels are regulated by Slbo. Despite the strong and regulated expression, migration is normal in border cells mutant for singed. The strongest allele of singed available was used, but it retained a low level of protein expression. In addition, functional overlap may exist between actin regulators. Quail is an actin binding protein of the villin family, and its function in the germline of the ovary genetically overlaps with that of Singed. quail mRNA is also upregulated in border cells relative to follicle cells, and Quail protein is detected in border cells. Quail is structurally similar to Gelsolin, which was also upregulated in border cells, as well as the Gelsolin-related FliI, which was not detectably expressed. However, Gelsolin is enriched in polar cells rather than the migratory outer border cells. As for singed, no migration defects were observed in quail mutant border cells, nor in cells mutant for quail and only one functional copy of singed or vice versa. It was not possible to recover clones of border cells simultaneously mutant for both singed and quail, which is likely to reflect a functional overlap between the two genes at an earlier stage. The simultaneous upregulation of redundant actin regulators may reflect a genetically robust approach to changing the actin cytoskeleton in border cells (Borghese, 2006).

A rather surprising finding of this global expression analysis was that the remaining genes encoding cytoskeleton-associated proteins and upregulated in border cells in a slbo-dependent manner were all “muscle specific”. This included a complete palate of structural genes: muscle actin (57B), muscle myosin heavy chain and light chains, tropomyosin 2 (tm2), troponins, and the calponin-related mp20. The muscle-specific expression has been shown for this group of genes in Drosophila embryos as well as mature muscles. For tropomyosin 2, a GFP gene trap allele was available and, and this allele confirmed expression in border cells as well as in the muscle sheath. The expression profiling indicated that border cells also express the corresponding non-muscle forms such as actin42A, zipper (myosin heavy chain), and sqh (myosin light chain), but at the same level as in follicle cells. The nonmuscle proteins are generally required for many cellular processes, including, where tested, migration of border cells. This raised the question of why this large cluster of muscle-specific structural genes would be turned on in border cells as well. To address this, migration was analyzed of border cells mutant for individual muscle genes for which mutants were available (mhc, mlc2, upheld=troponinT and tm2). Since mhc and mlc2 are essential genes, this was done by clonal analysis. No defects were seen in border cells mutant for mlc2, upheld, or tm2, but clear migration defects were observed in border cells mutant for mhc (mhc1 or mhc3). Thus, while not all of the muscle structural genes are required for border cell migration, at least muscle Mhc expression contributes to effective migration (Borghese, 2006).

Given that both muscle and nonmuscle forms of the same cytoskeletal proteins have a role in border cell migration, their functions are likely to be different. In agreement with this, no genetic interactions were observed between mutants affecting muscle and nonmuscle forms of myosin heavy or light chains. There is precedence for such nonoverlapping functions. For example, Zipper has a unique role in developing muscle cells, which contain plenty of muscle myosin heavy chain. In mammalian cells, different myosin heavy chain isoforms can have distinct subcellular localization. Also, the actin proteins, despite having few amino acid differences, are functionally distinct in vivo (Borghese, 2006).

The muscle gene expression program activated in migratory border cells extended beyond structural genes to regulatory genes. One such gene was bent, encoding a very large titin-like molecule with a myosin light chain kinase domain. Being essential but on the fourth chromosome, bent was not amenable to standard clonal analysis. Genes required for myoblast fusion were also identified, namely, rols/antisocial and rost. mbc, which encodes a DOCK180 family Rac GEF and is required for myoblast fusion, has a role in border cell migration downstream of the PVR guidance receptor. Mbc protein interacts physically with the presumed adaptor protein Rols. Clonal analysis with a strong (likely complete loss-of-function) allele of rols showed defects in border cell migration, suggesting that Mbc and Rols might act together during migration as well. The defect was milder than for mbc, implying that Mbc activity might not be completely dependent on Rols. For the small transmembrane protein Rost, no useful mutants were available. It was also noted that a very closely related and adjacent gene, CG13101, was similarly regulated in border cells and might overlap rost function. Thus, at least mbc and rols function in border cells as well as in muscle (myoblast fusion). Activation of a broad “muscle-specific” gene expression program in border cells may reflect a requirement for a specific subset of the genes within this program (Borghese, 2006).

Previous unbiased genetic approaches to identify genes important for border cell migration have largely identified transcription factors or inducing signals. Changes in cell fate can alter cell behavior dramatically without affecting cell survival, thus still allowing analysis of the mutant cells. The transcription factors themselves often show differential expression. In addition to Slbo, the posttranslationally regulated transcription factor STAT, which is important for border cell migration, was also upregulated in border cells (1.6-fold). The transcription factors that were upregulated in border cells and had mutants available for effects on border cell migration were also tested. aop/yan transcripts were increased 1.9-fold in border cells. In a PiggyBac transposon-based clonal screen for border cell migration defects, an insertion was identified in aop. Complementation analysis confirmed the gene assignment, and quantification of the phenotype showed a clear effect of aop on border cell migration. As expected, border cell migration was strongly affected, but, in addition, clones were rare and morphological abnormalities were seen in other follicle cells as well as in germline cells. Thus, aop may affect the behavior of multiple cell types in the ovary. Another transcription factor, vrille, was also upregulated (over 2-fold). vrille has been implicated in signaling, circadian rhythm, and cellular morphogenesis, but border cells mutant for vrille were largely unaffected and experienced only subtle delays (Borghese, 2006).

The most border cell-enriched RNA encoding a transcription factor, apart from Slbo, was Six4 (4.5-fold). Six4 expression in border cells was confirmed by in situ analysis. Drosophila Six4 is the homeodomain transcription factor most related to mammalian Six4 and Six5. Six family proteins act in complex with proteins of the Eya (Eyes absent) family. eya transcripts were also 2.7-fold enriched in border cells relative to follicle cells of the same stage, and Eya was expressed in a pattern similar to that of six4. Both six4 and eya were expressed in earlier-stage follicle cells as well, and eya has been shown to function at these stages to repress polar cell fate. Follicle cells mutant for six4 expressed a polar cell marker (Fas3) and were functional polar cells, as determined by the ability to induce surrounding anterior follicle cells to become Slbo-positive, migratory border cells. This suggested that Six4 cooperates with Eya in repressing polar cell fate. It had been indicated that Six proteins affect nuclear localization of their Eya partner. The six4 mutant allowed testing this in an in vivo context. Although six4 mutant cells were transformed to functional polar cells, Eya protein was not absent as in the endogenous polar cells, showing that Eya accumulation was independently regulated. However, Eya protein was partially relocalized to the cytoplasm of six4 mutant cells, supporting the hypothesis that Six4 and Eya interact in vivo. Since six4 and eya are both upregulated in outer border cells when they migrate, they are likely to act together in this process as well. However, their earlier roles precludes straightforward loss-of-function analyses in border cells, since “border cell clusters” consisting only of six4 or eya mutant cells are not functional simply because polar cells do not migrate on their own. Overexpression of HA-tagged six4 in border cells interfered with migration, as found for transcription factors required in border cells slbo (Borghese, 2006).

The expression of Six4 in border cells may contribute to activation of the muscle gene program described above. The conserved muscle transcription factor Mef2, an activator of muscle actin and myosin expression, was not detected in border cells by expression profiling or by antibody staining, nor were Twist and Nautilus/MyoD. Six4 is required for development of muscle and other mesodermal tissues in Drosophila. Mutants of C. elegans Unc-39, belonging to the Six4/5 family, also affect muscle/mesodermal differentiation as well as directed cell migration. Mammalian Six5, also called myotonic dystrophy-associated homeodomain protein (DMAHP), has been analyzed due to its contribution to DM1, and Six4/5 affect normal muscle development. Another transcription factor complex that might contribute to the activation of the muscle program is that of MAL-D (or MRTF) and SRF (serum response factor). The MRTF/SRF complex plays an important role in muscle development in mammals and directly regulates muscle (structural) genes. MAL-D/SRF plays a crucial role in border cell migration and this complex acts to strengthen the cytoskeleton of invasive border cells in response to perceived tension. This mode of regulation makes MAL-D/SRF activity in border cells indirectly dependent on Slbo, which could be responsible for the apparent regulation of the muscle gene cluster by Slbo. The possibility cannot be excluded that Slbo might affect muscle genes directly; the mammalian C/EBP transcription factors are known to regulate different differentiation-specific genes in different contexts (Borghese, 2006).

This study analyzed overall gene expression changes resulting from a transcriptional switch that induces invasive migratory behavior in vivo. The major goal of the analysis was to identify transcriptional changes that directly affect cell behavior and make the cells move. The results indicate that regulation of both the actin cytoskeleton and the microtubule cytoskeleton, likely coordinated regulation, is important for this transition. Identifying Stathmin as an important regulator downstream of Slbo in border cells indicates that microtubule dynamics are critical for border cell migration. Key questions are now how microtubule dynamics affect the process, and whether Stathmin activity is regulated. Two recent findings suggest that Stathmin may be a more general regulator of cell migration: Stathmin-microtubule interactions are spatially regulated in migrating cells in culture, and Stathmin upregulation may promote migration and metastasis of sarcoma cells. The actin cytoskeleton is clearly crucial for cell migration and is controlled by many regulators. The upregulated modulators identified in this study were different from those identified in a whole-genome study of tumor cells selected, in vivo, to be highly motile. There are obviously many differences between these studies; for one, a normal transition to migratory behavior may differ from unrestrained, high motility. The activation of a “muscle-specific” program in migratory border cells was unexpected and provides an intriguing connection between these cells that move and the specialized cells that move an animal (muscle). Overall, the analysis of actin regulators indicates that this is a robust system, with many effectors coregulated, even by one transcription factor. Genetically, this is reflected by minor defects in individual “effector” mutants despite absolute dependence on the transcriptional switch. Further analysis in other systems, and subsequent comparisons, will reveal to what extent the gene expression program employed by border cells to become migratory is a general one (Borghese, 2006).

Protein Interactions

In vitro, SLBO and mammalian C/EBP form functional heterodimers; however, since there is no evidence for a family of Drosophila C/EBPs, SLBO may function as a homodimer in vivo (Rorth, 1992).

A transgenic rescue assay was used to determine which molecular functions of the SLBO protein are required for it to fulfill its essential role during development. Chimeric proteins that contain the SLBO basic region, a heterologous zipper, and a heterologous activation domain could functionally substitute for slbo. Mammalian C/EBPs are also functional in Drosophila. In contrast, 9 of 25 single amino acid substitutions in the basic region disrupt biological function. Thus, the conserved basic region specifies SLBO activity in the organism (Rorth, 1994).

Regulation of Slbo Degradation

The C/EBP transcription factor, Slbo, is required for migration of border cells during Drosophila oogenesis. Neither increase nor decrease of Slbo activity is tolerated in border cells. Correct protein level is in part ensured by cell type-specific regulated turnover of Slbo protein. Through genetic screening, two genes that are involved in this regulation have been identified. The Ubp64 ubiquitin hydrolase acts as a stabilizer of Slbo protein. A novel gene, tribbles, is a negative regulator of slbo in vivo. Tribbles acts by specifically targeting Slbo for rapid degradation via ubiquitination (Rørth, 2000).

Border cells are a group of six to ten cells that delaminate from the follicular epithelium and migrate as a cluster to the oocyte at a specific time during Drosophila oogenesis, called stage 9. slbo is expressed in border cells and is absolutely required for their migration. If border cell migration is perturbed, the resulting eggs cannot be fertilized; thus, slbo mutant females are sterile. Slbo protein is detected in border cells before and as they migrate, both in the centrally located anterior polar cells (APCs) and the remaining 'outer border cells.' After migration, the level of Slbo protein in border cells decreases. Another group of cells, the centripetal cells, migrate over the anterior part of the oocyte during stage 10. slbo is also expressed in centripetal cells but is not required for their migration (Rørth, 2000).

To determine which cells require slbo activity as well as the fate of individual cells with no slbo activity, a clonal analysis was carried out using a slbo null mutant. A total of 185 border cell clusters with one or more mutant cells were analyzed. If all border cells are mutant, then the cells do not move from the anterior tip. Similarly, no migration was seen in over 200 egg chambers from slbonull females (slbonull females were rescued from embryonic lethality with a transgene). If the only wild-type cell in the cluster is one (or both) of the APCs, migration is also not initiated. However, if one or two outer border cells are wild type, these initiate movement but remain associated with the mutant border cells and the cluster moves very little. Conversely, if one or two outer border cells are mutant, these mostly remain associated with the wild-type border cells and are 'dragged along' at the rear of the cluster. If one or both APCs is mutant, this does not affect migration or position of the cell. The first conclusion from these experiments is that APCs are not, by themselves, migratory cells. The APCs have separate lineage and different morphology than the remaining border cells. Also posterior polar cells (PPC) express Slbo protein but do not migrate. The second conclusion is that slbo is required in outer border cells for active migration, but not for the selective adhesion of border cells (Rørth, 2000).

Although slbo is autonomously required for each outer border cell to be actively migratory, the efficiency of cluster movement reflects the fraction of cells that are migration competent (slbo+). Similarly, altering the overall level of Slbo in the cluster using different hypomorphic slbo mutants does not result in complete cessation of movement, but in quantitative defects. Less Slbo protein results in longer delay and smaller percentage of clusters initiating migration. Complete block of border cell migration is only observed in the slbo null mutant. In a screen for gain-of-function suppressors of the slbo mutant phenotype, a slbo mutant, slbo1310, was used that allows a low level of slbo expression. This should allow for identification of suppressors that act by affecting the level of slbo activity (Rørth, 2000).

One line isolated in the suppressor screen, EP3584, was found to have an EP insertion upstream of the Ubp64 gene. To verify the activity of the gene, the Ubp64 cDNA was expressed directly using the UAS-GAL4 system and the slboGAL4 driver. Overexpression of Ubp64 suppresses the slbo phenotype significantly. Ubp64 encodes a putative ubiquitin hydrolase (Henchoz, 1996). Based on the similarity to specific ubiquitin hydrolases in yeast, it is likely that Ubp64 removes ubiquitin moieties from ubiquitinated proteins, thereby stabilizing them. The suppressor activity of EP3584 suggested that Slbo might be a target for Ubp64 deubiquitination, since stabilization of residual Slbo protein should suppress the slbo mutant phenotype. Whether Slbo protein levels are affected by Ubp64 overexpression was tested. Slbo protein levels are very low in the slbo1310 mutant but markedly increase when Ubp64 is induced. slbo transcription, reflected by the slbo-lacZ reporter gene, is unaltered by Ubp64 expression. These results suggest that increased expression of Ubp64 stabilizes Slbo protein (Rørth, 2000).

The suppressor activity of the ubiquitin hydrolase indicates that Slbo protein is targeted for degradation by ubiquitination and furthermore that Slbo might have a short half-life in vivo. To look at Slbo protein turnover in vivo, but in a situation where its transcription is under temporal control, slbo mRNA was ectopically induced by a heat shock pulse to flies carrying a HS-slbo construct. Abundant ectopic protein was detected 1 hr after HS in all follicle cells of stage 10 egg chambers, including border cells. Most follicle cells retain a high level of Slbo protein 1 hr later. However, Slbo protein is almost undetectable in border cells. By 4 hr, Slbo protein has disappear from almost all follicle cells. The difference between border cells and other follicle cells is also observed in stage 9 egg chambers. As expected, slbo RNA is detected in all follicle cells 30 min after heat shock in these experiments and declines in all cells thereafter. Since Slbo protein initially accumulates equally in all follicle cells, the difference in protein levels is best explained by differential turnover. Thus, Slbo protein appears to be selectively unstable in border cells, the cells that endogenously express the protein (Rørth, 2000).

When the same experiment was performed in a slbo mutant background, Slbo protein levels were identical in border cells and other follicle cells at different time points. This indicates that the selective instability of Slbo protein in border cells is dependent on the activity of Slbo itself. In both wild-type and slbo mutant chambers, Slbo protein has almost disappeared from all follicle cells 4 hr after heat shock. This time coincides with the time when the first border cell clusters have initiated migration in the slbo mutant background as a result of rescue by HS-slbo. Note that HS-slbo does not induce significantly precocious migration, nor migration of more than the normal number of border cells. Thus, when expressed in cells with no prior exposure to Slbo protein, the decline in Slbo levels coincides with the time when downstream events are triggered (migration of border cells). This further supports that Slbo directly or indirectly stimulates its own degradation (Rørth, 2000).

Given that Slbo turnover is specifically regulated in border cells, it might be possible to genetically identify factors important for this regulation. A screen was performed to identify genes that when overexpressed using the slboGAL4 driver, would stop border cell migration. slboGAL4 drives efficient expression in both border cells and centripetal cells. Out of 2000 EP lines, one line with this phenotype was identified: EP3519. This gene corresponds to tribbles (Rørth, 2000).

The specificity of the tribbles phenotype for border cell migration raised the possibility that tribbles might affect Slbo expression or activity, since slbo is required for border cell, but not centripetal cell migration. Slbo protein expression was examined by antibody staining. Overexpression of tribbles in border cells causes a dramatic decrease in Slbo protein levels. Slbo protein is reduced to a level similar to that seen in the slbo1310 mutant border cells, thus explaining the phenotype of tribbles overexpression. To see whether tribbles normally controls Slbo protein levels, loss-of-function mutant clones of tribbles were analyzed. Cells mutant for tribbles show slightly higher levels of Slbo protein, indicating that tribbles normally contributes to downregulating Slbo. The effect is detectable from early stage 9 to stage 10. The role of tribbles was addressed by looking at genetic interactions between tribbles and slbo. Removing one copy of tribbles suppresses the slbo phenotype 2-fold (from 5% migration to 10%). Flies homozygous mutant for tribbles are poorly viable, due to defects unrelated to slbo, but the slbo mutant phenotype is suppressed about 4-fold. Thus, both gain-of-function and loss-of-function results show that tribbles is a negative regulator of Slbo expression in migrating border cells (Rørth, 2000).

Overexpression of tribbles in border cells does not affect beta-galactosidase expression from the slbo enhancer trap, indicating that tribbles is affecting slbo at a posttranscriptional level. This was confirmed by looking at the effect of tribbles on slbo, which was expressed under control of a heterologous promoter. For example, ectopic expression of slbo in wings using vestigialGAL4 and UAS-slbo results in small wings, and this effect is suppressed by coexpressing tribbles. Thus, the effect of tribbles on slbo in vivo is independent of cell context and is posttranscriptional (Rørth, 2000).

To further analyze the effect of tribbles, Schneider cell transfections were carried out. Induced expression of tribbles decreases the level of cotransfected epitope-tagged Slbo protein. This is observed without any effect on Slbo mRNA level. Addition of lactacystin, a potent inhibitor of proteosome-mediated degradation, increases Slbo levels and partially blocks the effect of tribbles. Since lactacystin has a more pronounced effect in the presence of tribbles (7-fold versus 3.5-fold increase in Slbo level), it has been concluded that Slbo is normally degraded via the proteosome pathway fairly rapidly and that this turnover is stimulated by Tribbles. Support for this idea came from looking at addition of His-tagged ubiquitin moieties to Slbo protein. Expression of Tribbles causes an increase in higher molecular weight bands, which most likely correspond to Slbo-ubiquitin conjugates, suggesting that Tribbles stimulates Slbo ubiquitination. Finally, epitope-tagged Tribbles and Slbo proteins could be coimmunoprecipitated from cells, indicating that the proteins can physically associate, directly or indirectly. In conjunction with the in vivo data, these experiments indicate that Tribbles affects Slbo protein levels by targeting Slbo for degradation via the ubiquitin-proteosome pathway (Rørth, 2000).

One conclusion from these experiments is that Slbo protein level is under very tight control in vivo. In addition to transcriptional regulation of the slbo gene, Slbo protein is targeted for degradation via the proteosome pathway by Tribbles. Slbo protein turnover is particularly high in border cells, possibly as a negative feedback regulation. To address the importance of this tight control of protein level, attempts were made to override it by forcing overexpression of Slbo. Ectopic expression of Slbo is deleterious in most tissues. More importantly, increased Slbo level is deleterious even in cells where it is normally expressed. Overexpression of Slbo in border cells delays their migration. This effect is observed when Slbo protein levels are about 5-fold over wild-type level (shown by quantitation of immunoflouresence staining). Thus, border cells fail to migrate properly if the level of Slbo protein is too low or too high. The effect of overexpressing slbo was alleviated by coexpressing tribbles. Thus, cooverexpression of slbo and tribbles reciprocally rescues the gain-of-function phenotypes. This confirms that Tribbles stops border cells by removing Slbo. Conversely, delay in border cell migration due to moderate overexpression of Slbo is worsened by coexpressing Ubp64, confirming its role as a positive regulator of Slbo protein levels (Rørth, 2000).

The deleterious effects of Slbo overexpression could either be due to increased activity of this transcription factor or to effects such as squelching or other abnormal interactions. To see if Slbo activity is responsible for the effect, an inactive version of Slbo with a leucine to proline mutation in the leucine zipper (Slbo-LZ) was tested. Slbo-LZ protein cannot dimerize and therefore cannot bind DNA and is not able to rescue border cell migration when expressed in vivo. Because the mutant cannot dimerize, it also does not act as a dominant-negative. However, the mutant still goes to the nucleus and has functional activation domains. When overexpressed at a level similar to that of the control Slbo protein, the mutant Slbo protein does not affect border cell migration. This result suggests that the problems caused by high levels of Slbo are due to excessive activity of the transcription factor (Rørth, 2000).

It is concluded that the role of APCs may be to signal to and/or recruit the adjacent cells to become migratory outer border cells. slbo is not required for selective border cell adhesion, indicating that slbo null mutant border cells retain some border cell characteristics. Also ectopic expression of slbo by HS-slbo or the GAL4/UAS system does not convert other follicle cells into migratory border cells. Thus, slbo is not a master regulator of border cell fate. In addition, although HS-slbo can induce migration later than normal in slbo mutant egg chambers, precocious expression of slbo cannot force much earlier migration. Thus, specification of border cells must be dependent upon temporal and spatial control of one or more factor(s) in addition to slbo (Rørth, 2000).

Expression of slbo in the ovary is under tight transcriptional control, spatially and temporally. In addition, Slbo protein is rapidly degraded in a regulated fashion. Why might the organism impose this extra control on Slbo protein accumulation? One rationale is the need for efficient induction of Slbo to certain levels for border cell differentiation (inducing migration), coupled with the need to keep Slbo from overaccumulating. That the increased turnover of Slbo seen in border cells may be induced by Slbo itself (autoregulation) fits with this rationale; it may prevent Slbo levels from becoming too high. Rapid turnover may also contribute to timing the initial effect of Slbo. Although there is obvious upregulation of slbo transcription at stage 8/9, transcriptional activity of slbo reporter genes can be seen earlier in polar cells and adjacent cells. The cues regulating slbo transcription may be present at earlier stages, whereas functional activity of Slbo requires protein level above a certain threshold. Rapid turnover, in part imposed by tribbles, may keep Slbo below this threshold by preventing its accumulation early. In any case, the Slbo overexpression experiments indicate that control by proteolysis can be overridden if Slbo is sufficiently overexpressed (Rørth, 2000).

The observations raise two further questions. The first is why overaccumulation of Slbo protein is deleterious. Sustained overexpression of Slbo is deleterious to all tissues tested, including where it is normally expressed. The latter shows that Slbo levels are critical. It is assumed that the problem is elevated Slbo transcriptional activity, as indicated by the Slbo-LZ mutant result, although it is formally possible that Slbo (and not the Slbo-LZ mutant) engages in some other aberrant activity. The increased transcriptional activity could, in turn, reflect increased expression of normal target genes, or activation of inappropriate, suboptimal 'target genes'. Slbo, like other proteins of the C/EBP family, has a reasonable affinity for suboptimal target sites in vitro. Slbo may bind to low affinity sites if present at higher than normal concentration and influence transcription of genes not intended to be target genes (Rørth, 2000).

It is possible that Tribbles is part of a ubiquitin E3 complex for Slbo. E3 complexes are defined by their activity, to stimulate transfer of ubiquitin from specific ubiquitin-charged E2 complexes to substrates, and can be unrelated in sequence. It is also possible that the effect on Slbo protein levels is indirect and, for example, reflects an effect on subcellular localization. Tribbles can be located both in the nucleus and the cytoplasm; when overexpressed in border cells, it is primarily nuclear. In vivo, the GOF phenotype of tribbles was very specific, and in cell transfections, the effect of Tribbles was specific to Slbo (as well as the mammalian homolog C/EBPalpha). In conclusion, these studies of Slbo regulation have revealed an unexpected importance of transcription factor levels and precise control thereof, even for a transcription factor that acts as an apparent on/off switch and is itself transcriptionally regulated. These observations stress the importance of looking at transcription factors and targets in their natural context, with physiological levels of protein present (Rørth, 2000).

The kinase domain of Drosophila Tribbles is required for turnover of fly C/EBP during cell migration

Drosophila Tribbles (Trbl) encodes the founding member of the Trib family of kinase-like proteins that regulate cell migration, proliferation, growth and homeostasis. Trbl was identified in a misexpression screen in the ovary as an antagonist of border cell migration and acts in part by directing turnover of the C/EBP protein encoded by the gene slow border cells (slbo). The ability of mammalian Trib isoforms to promote C/EBP turnover during tissue differentiation indicates that this function is highly conserved. To better understand the role of Trbl in cell migration, specific Trbl antisera, a trbl null allele and Trbl transgenes bearing site-directed mutations were tested. Trbl is expressed at high levels in the nuclei of follicle cell epithelia and is downregulated in delaminating epithelia as expression of Slbo (C/EBP) is upregulated. This complementary pattern of expression during subsequent cell migration is achieved by negative feedback whereby slbo represses Trbl expression and trbl is necessary and sufficient to promote Slbo protein turnover. A series of point mutations that scan the conserved kinase domain of Trbl reveal that the conserved DLK catalytic loop is required for Trbl-Slbo binding and turnover, as well as for interactions between Trbl subunits, suggesting a mechanism of Trbl function (Masoner, 2013).

Mammalian Trib proteins have diverse functions (reviewed in Dobens, 2012), as transcriptional co-activators and repressors in the nucleus, and as MAP kinase kinase inhibitors and proteosome adapters in the cytoplasm. Tribs bind and direct the degradation of key regulatory proteins, in particular members of the C/EBP family, notably (1) C/EBP α, which is degraded by Trib1 to promote the formation of acute myelogenous leukemia (AML) tumors and by Trib2 during myeloid differentiation (Keeshan, 2006) and (2)C/EBPβ, which is degraded by Trib2 during differentiation of 3T3-L1 preadipocytes (Naiki, 2007). As well, C/EBPβ levels increase in Trib1 knockout mice (Yamamoto, 2007; Keeshan, 2008; Masoner, 2013 and references therein).

For Trib2, it has been demonstrated recently that a point mutation in the DLK catalytic loop motif disrupts its ability to direct turnover of C/EBP α and promote acute myelogenous leukemia (Keeshan, 2010), and this study demonstrates that the same DLK/R mutation in fly Trbl compromises its ability to direct Slbo turnover and block BC migration. It was also shown that an intact DLK catalytic motif is required for Trbl interactions both with Slbo and with Trbl, itself. These data support the notion that the conserved catalytic domain is critical for Trbl function, however it remains unclear if this mutation disrupts ATP binding-dependent protein folding or whether Tribs are functional kinases, a distinction which will be resolved ultimately by identifying bonafide Trib substrates (Masoner, 2013).

The D/NLK Trbl mutant surprisingly retains WT activity when misexpressed in the posterior wing compartment, resulting in larger cells as measured by more widely spaced bristles when compared to the anterior compartment. It is noted that a similar mutation in Trbl located just C-terminal to DLK (DLKLK/R) also retains WT activity to block cell proliferation when injected into the blastoderm epithelium. Because WT Trbl is thought to block cell division by directing String/cdc25 turnover, the data suggest another domain outside the Trbl catalytic loop must be required for String degradation (Masoner, 2013).

Several observations are worth noting regarding the trbl protein null allele and specific antisera that were analyzed. The trblD13 deletion allele described in this study is a stronger allele than those used previously, but still yields a few escaper adults in combination with a deficiency of the locus. Though trbl is not required for viability, it was found that rare escaper animals hatch late and are infertile with vestigial ovaries, suggesting trbl has unexplored roles both in larval tissue growth and cell proliferation during early oogenesis. Antisera to Trbl reveal dynamic changes in subcellular localization during late oogenesis: Trbl accumulates in FC nuclei up to stage 11 after which nuclear levels drop and Trbl accumulates at low levels in the cytoplasm. The notion that Trib localization is regulated and has functional importance is supported by observations that (1) GFP-tagged versions of Tribs are localized variously to the cytoplasm, nucleus and even the mitotic spindle (Saka, 2004), and (2) during BMP/TGF-β-signaling Tribs bind BMP receptors at the cell cortex, upon signaling are released into the cytoplasm to bind and degrade SMURFS, and subsequently translocate to the nucleus to serve as SMAD co-activators (Hegedus, 2006; Hegedus, 2007; Chan, 2007; Hua, 2011; Masoner, 2013 and references therein).

Antisera detect dynamic changes in Trbl levels during BC migration: Trbl is strongly expressed prior to delamination from the anterior epithelium and Trbl levels decrease during posteriorwards BC migration. In slbo mutant egg chambers Trbl levels are significantly higher compared to WT at all stages examined, indicating that slbo is necessary to repress Trbl. However Slbo misexpression alone is not sufficient to repress Trib expression, suggesting that cofactors mediate Slbo repression of Trbl. The observation that slbo represses Trbl stands at odds with previous work showing that Slbo turnover is reduced in slbo mutant egg chambers. While further work must be done to reconcile these observations, it is possible that slbo both represses Trbl expression and activates an unidentified Trbl activator, whose existence is implied by work in embryos. In this way, while Trbl levels increase in a slbo mutant, its ability to direct Slbo turnover could be compromised (Masoner, 2013).

Together these data indicate that Slbo and Trbl are at the core of a negative feedback loop in which Slbo represses Trbl expression and conversely Trbl (via its catalytic loop) binds and degrades Slbo. A yeast two-hybrid interaction assay was used to test the strength of Trbl interactions in the absence of Trbl-directed Slbo turnover and it was shown (1) that weak interactions occur between either Trbl-Trbl or Slbo-Slbo, while (2) comparatively stronger Trbl-Slbo interactions occur. These observations adds a key feature to the model: the proposition that weak homomeric complexes of Trbl multimers and Slbo dimers exchange preferentially for strong heteromeric Trbl-Slbo complexes to direct Slbo turnover. Such a model is consistent with observations that in some instances protein kinase dimerization leads to autoinhibited complexes and is supported by work showing that (1) purified Trbl protein expressed in E. coli can form dimers and tetramers and (2) the strength of Trbl-Trbl interaction impacts the activity of target pathways (Masoner, 2013).

Snail-related transcription factors, which direct E-cadherin-dependent cell migration in a wide range of normal and diseased tissues, are also regulated by protein turnover. Several modifications affect Snail stability, including phosphorylation by PAK and GSK3β, dephosphorylation by the small C-terminal domain phosphatase (SCP), and lysine oxidation is promoted by NFkappaB. In the latter case, NFkappaB prevents Snail phosphorylation by GSK-3 and subsequent degradation, whereas formation of a ternary complex between wild-type p53, the ubiquitin ligase Mdm2, and Snail2 promotes degradation. It is likely that during BC migration a similar level of complexity underlies Trbl effects on Slbo stability, and the proposed feedback between these genes results in oscillating levels of Slbo and corresponding fluctuating expression of Slbo target genes, notably DE-cadherin, whose turnover at the cell membrane has been demonstrated to promote proper BC migration (Masoner, 2013).

In mammals, Trb family members have been identified as tumor suppressors or oncogenes, depending on tissue context and sorting out these conflicting data may be aided from a simpler Drosophila model. Conserved interactions between Trbl and C/EBP during cell differentiation in flies and mammals suggest the possibility that interactions between Trbl and the cdc25 phosphatase String observed during cell division in fly tissue might be conserved in mammals as well. Conversely, mammalian work holds the promise to illuminate and direct further tests of Trbl function in Drosophila, notably documented interactions between mouse Trib3 and ATF4, another member of the B-Zip class of transcription factors active during pancreas β-cell differentiation, and human Trib3 with Atk kinase during insulin-target cell metabolism. Synergies between these parallel lines of investigation will shed light on the diverse roles of Trb family members in cell growth, proliferation and differentiation (Masoner, 2013 and references therein).

slow border cells: Biological Overview | Evolutionary Homologs | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

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