Integrin signaling is required for maintenance and proliferation of intestinal stem cells in Drosophila

Tissue-specific stem cells are maintained by both local secreted signals and cell adhesion molecules that position the stem cells in the niche microenvironment. In the Drosophila midgut, multipotent intestinal stem cells (ISCs) are located basally along a thin layer of basement membrane that composed of extracellular matrix (ECM), which separates ISCs from the surrounding visceral musculature: the muscle cells constitute a regulatory niche for ISCs by producing multiple secreted signals that directly regulate ISC maintenance and proliferation. This study shows that integrin-mediated cell adhesion, which connects the ECM and intracellular cytoskeleton, is required for ISC anchorage to the basement membrane. Specifically, the alpha-integrin subunits including alphaPS1 encoded by mew and alphaPS3 encoded by scb, and the beta-integrin subunit encoded by mys are richly expressed in ISCs and are required for the maintenance, rather than their survival or multiple lineage differentiation. Furthermore, ISC maintenance also requires the intercellular and intracellular integrin signaling components including Talin, Integrin-linked kinase (Ilk), and the ligand, Laminin A. Notably, integrin mutant ISCs are also less proliferative, and genetic interaction studies suggest that proper integrin signaling is a prerequisite for ISC proliferation in response to various proliferative signals and for the initiation of intestinal hyperplasia after loss of adenomatous polyposis coli (Apc). These studies suggest that integrin not only functions to anchor ISCs to the basement membrane, but also serves as an essential element for ISC proliferation during normal homeostasis and in response to oncogenic mutations (Lin, 2013).

Effects of Mutation or Deletion

The Myospheroid cytoplasmic tail is sufficient to direct the localization of MYS to muscle termini. This localization does not occur via an association with structures set up by endogenous integrins, since it can occur even in the absence of the MYS protein. The subcellular localization of alphaPS2betaPS integrin is not dependent on any other interaction between the muscles and the tendon cells. In embryos that lack the segmental tendon cells, due to a mutation removing engrailed and invected, alphaPS2betaPS is still localized to the muscle termini even though the ventral longitudinal muscles are not attached to the epidermis, but instead are attached end to end. Thus the alphaPS2betaPS integrin can be localized by an intracellular mechanism within the muscles. These results challenge the view that the extracellular environment via integrins is required for the organization of the cytoskeleton and the resultant polarity (Martin-Bermudo, 1996).

The integrins are a family of transmembrane heterodimeric proteins that mediate adhesive interactions and participate in signaling across the plasma membrane. The conserved and charged membrane proximal portion of vertebrate beta1 subunit cytoplasmic tail is important in the association of integrins to the cytoskeleton. This region has been implicated in the binding of signaling molecules and cytoskeletal proteins (e.g. focal adhesion kinase, paxillin and alpha-actinin). In this study the functional significance of the cytoplasmic domains of the (alpha)PS1, (alpha)PS2 and ßPS subunits of the Drosophila Position Specific (PS) integrin family were examined by analyzing the relationship between cytoplasmic domain structure and function in the context of a developing organism. By examining the ability of ßPS molecules lacking the cytoplasmic domain to rescue embryonic abnormalities associated with PS integrin loss, it was found that although many embryonic events require the ßPS cytoplasmic domain, this portion of the molecule is not required for at least two processes requiring PS integrins: formation of midgut constrictions and maintaining germband integrity. The cytoplasmic domain is required for migration of midgut primordia along visceral mesoderm, elongation of the midgut into a convoluted tube, and somatic muscle attachment. Mutant proteins affecting four highly conserved amino acid residues in the cytoplasmic tail function with different efficiencies during embryonic development, suggesting that interaction of PS integrins with cytoplasmic ligands is developmentally modulated during embryogenesis (Li, 1998).

The ability of (alpha)PS1 and (alpha)PS2 to function without their cytoplasmic domains was also analyzed. By analyzing the ability of transgenes producing truncated (alpha)PS molecules to rescue abnormalities associated with integrin loss, it was found that the cytoplasmic tail of (alpha)PS2 is essential for both embryonic and postembryonic processes, while this portion of (alpha)PS1 is not required for function in the wing and in the retina. (alpha)PS2 is not required to maintain attachment of the germband germlayers. Temperature-shift experiments suggest roles for the (alpha)PS2 cytoplasmic domain in signaling events occurring in the developing wing. Under low expression conditions blisters are found on approximately 12% of (alpha)PS2 cytoplasmic domain null mutant wings, while in high expression conditions, significantly fewer wing blisters are observed. The extracellular and transmembrane domains of (alpha)PS1 are sufficient for function in retinal maintenance (Li, 1998).

Broad-Complex transcription factors regulate muscle attachment in Drosophila. In Brc mutants of the rbp complementation group, dorsoventral indirect flight muscles (DVM) are largely absent and the dorsal longitudinal indirect flight muscles, tergotrochanteral muscles (TTM), and remaining DVM often select incorrect attachment sites. One striking aspect of the rbp phenotype is that dorsal attachments are defective while ventral attachments seem normal. This phenomenon has also been observed in the TTM of bendless and myospheroid mutant adults. It is also noted that all the susceptible attachment sites are derived from the wing imaginal disc, while the unaffected sites are derived from leg discs. Analysis of the Derailed receptor tyrosine kinase suggests that muscles use different molecules for making attachments at dorsal and ventral ends of their fibers. Derailed is localized to the ventral ends of a subset of embryonic muscles; derailed mutations cause ectopic ventral attachment of these muscles. The stripe dorso-longitudinal indirect flight muscle phenotype resembles the rbp DVM phenotype, in that normal initial development is followed by degeneration and disappearance. Although stripe does not appear to be under rbp control at head eversion, the two genes may regulate overlapping subsets of downstream targets (Sandstrom, 1997).

In the embryo, PS1 (alphaPS1 betaPS) is found on the surface of the epidermis and endoderm, while PS2 (alphaPS2 betaPS) is restricted to the mesoderm. The integrins are concentrated at the sites where the somatic muscles attach to the epidermis and at the interface between the visceral mesoderm and the endoderm. In myospheroid mutant embryos, which lack the betaPS subunit, the adhesion between the mesoderm and the other cell layers fails (Brown, 1993).

inflated is the gene encoding the alphaPS2 subunit of the PS1 integrin of Drosophila. A comparison of the null inflated phenotype with that of the locus that encodes the betaPS subunit myospheroid reveals that while the betaPS subunit is required for the adhesion of the epidermis along the dorsal midline, the alphaPS2 subunit is not. In if mutant embryos, the muscles remain attached to the other cell layers significantly longer than in a mys mutant embryo. This shows that the alphaPS2 betaPS integrin only contributes part of the adhesive activity at the sites of PS integrin adhesion, and rules out a model where PS integrin function occurs solely by the direct interaction of the two PS integrins (Brown, 1994).

multiple edematous wings (mew) is the gene encoding the alphaPS1 subunit of the PS1 integrin of Drosophila. In contrast to if (alphaPS2) and mys (betaPS) mutants, most mutant mew embryos hatch, only to die as larvae. Mutant mew embryos display abnormal gut morphogenesis but, unlike mys or if embryos, there is no evidence of defects in the somatic muscles. Thus, the complementary distributions of PS1 (alphaPS1 betaPS) and PS2 (alphaPS2 betaPS) integrin on tendon cells and muscle, respectively, do not reflect equivalent requirements at the myotendinous junction. Dorsal herniation, characteristic of the mys lethal phenotype, is not observed in mew or in mew/if double mutants. Clonal analysis experiments indicate that eye morphogenesis is disrupted in mew clones, but if clones show relatively normal eye morphology. Adult wings display blisters around large dorsal (but not ventral) mew clones. In contrast to dorsal mys clones, small mew patches do not necessarily display morphogenetic abnormalities. Thus, another integrin in addition to PS1 appears to function on the dorsal wing surface (Brower, 1995).

The Drosophila position-specific (PS) antigens are homologous to the vertebrate fibronectin receptor family, or integrins. A Drosophila gene required for embryonic morphogenesis, l(1)myospheroid, codes for a product homologous to the beta subunit of the vertebrate integrins. l(1)myospheroid mutants die during embryogenesis. These mutants lack the beta subunit of the PS antigens. In the absence of the beta subunit in mutant embryos, the PS alpha subunits are not expressed on the cell surface. It is concluded that the l(1)myospheroid phenotype represents the lack-of-function phenotype for these Drosophila integrins. In wild-type embryos, PS antigens are found at the interface between mesoderm and ectoderm, and later mainly at muscle attachment sites in the epidermis and gut. Together these results indicate that during embryogenesis, Drosophila integrins are used to attach mesoderm to ectoderm, and are required for the proper assembly of the extracellular matrix and for muscle attachment (Leptin, 1989).

Nine alleles have been characterized of myospheroid, which encodes the beta PS subunit of the Drosophila PS integrins. The mysXB87, mysXN101 and mysXR04 genes yield restriction digest patterns similar to those seen for wild-type chromosomes, however the mys1 and mysXG43 genes contain detectable deletions. mys1, mysXB87 and mysXG43 make little or no stable protein product, and genetically behave as strong lethal alleles. For the mysXN101 mutation, protein product is seen on immunoblots and a reduced amount of beta PS protein is seen at the muscle attachment sites of embryos. This mutant protein retains some wild-type function, as revealed by complementation tests with weak alleles. Protein is also seen on immunoblots from mysXR04 embryos; this allele behaves as an antimorph, being more deleterious in some crosses than the complete deficiency of the locus. mysts2 and mysnj42 are typically lethal in various combinations with other alleles at high temperatures only, but even at high physiological temperatures, neither appears to eliminate gene function completely. The complementation behaviors of mysts1 and mysts3 are quite unusual and suggest that these mutations involve regulatory phenomena. For mysts3, the data are most easily explained by postulating transvection effects at the locus. The results for mysts1 are less straightforward, but point to the possibility of a chromosome pairing-dependent negative interaction (Bunch, 1992).

Southern blot analysis of heterozygous scab adults using the complete alpha probe reveals DNA polymorphisms in 6 alleles. Mutant embryos homozygous for l19 or 1035 alleles of scab, sometimes undergo a secondary closure after the initial failure of dorsal closure. Unlike myospheroid (mys) mutant embryos, whose muscles detach from the body wall, scb embryos show vigorous muscular movements of the cuticle; some of those undergoing secondary dorsal closure actually do hatch. In addition to dorsal closure, another phenotype observed in embryos lacking betaPS, but not observed in embryos null for alphaPS1, alphaPS2 or both, is a twisting of the germ band. Time-lapsed videomicroscopy reveals that during germ-band extension in scb X5, scb X6, and scb 2 embryos, and in embryos deficient for scb, the germ band twists laterally rather than extending dorsally, as in wild-type embryos, so that the ventral side of the posterior midgut is visible from the lateral side of the embryo. As in myospheroid mutant embryos, proper orientation of the germ band is recovered by the completion of germ-band extension (Stark, 1997).

The strong expression of Scab in the dorsal vessel led to an examination of scab embryos for defects in this tissue. Defects in the dorsal vessel have not been reported for myospheroid mutant embryos, although a detachment of alary muscles from the heart (posterior dorsal vessel) and its failure to mature at late stages has been identified in embryos lacking zygotic, but not maternal betaPS. myospheroid mutant embryos lacking both maternal and zygotic betaPS were examined to preclude the possibility of maternal rescue. The heart and dorsal vessel form from two types of cells: the external pericardial cells and the internal cardioblasts. myospheroid minus, scb X5, scb X6, scb 2 and scb-deficient embryos were all stained with antibodies that recognize pericardial cells. Embryos from scb and deficiency lines show mislocalization of the pericardial cells, which normally organize in a line along the edge of the dorsal cuticle, and appear to have fewer of these cells in this area than wild type at the same stage. A similar (although more severe) defect could be seen in mys- embryos: the pericardial cells appear to dissociate, migrate randomly and are sparse. The increase in severity of the defect suggests the possibility that either alphaPS1 or alphaPS2 may function in this process as well. The defect in mys- is also strikingly similar to that published for laminin A mutations (Yarnitzky, 1995).

The forces that connect the genetic program of development to morphogenesis in Drosophila have been investigated. Focus was placed on dorsal closure, a powerful model system for development and wound healing. The bulk of progress toward closure is driven by contractility in supracellular 'purse strings' and in the amnioserosa, whereas adhesion-mediated zipping coordinates the forces produced by the purse strings and is essential only for the end stages. Quantitative modeling was applied to show that these forces, generated in distinct cells, are coordinated in space and synchronized in time. Modeling of wild-type and mutant phenotypes is predictive; although closure in myospheroid mutants ultimately fails when the cell sheets rip themselves apart, the analysis indicates that ßPS integrin has an earlier, important role in zipping (Hutson, 2003).

Integrins modulate the Egfr signaling pathway to regulate tendon cell differentiation in the Drosophila embryo

Changes in the extracellular matrix (ECM) govern the differentiation of many cell types during embryogenesis. Integrins are cell matrix receptors that play a major role in cell-ECM adhesion and in transmitting signals from the ECM inside the cell to regulate gene expression. In this paper, it is shown that the PS integrins are required at the muscle attachment sites of the Drosophila embryo to regulate tendon cell differentiation. The analysis of the requirements of the individual alpha subunits, alphaPS1 and alphaPS2, demonstrates that both PS1 and PS2 integrins are involved in this process. In the absence of PS integrin function, the expression of tendon cell-specific genes such as stripe and beta1 tubulin is not maintained. In addition, embryos lacking the PS integrins also exhibit reduced levels of activated MAPK. This reduction is probably due to a downregulation of the epidermal growth factor receptor (Egfr) pathway, since an activated form of the Egfr can rescue the phenotype of embryos mutant for the PS integrins. Furthermore, the levels of the Egfr ligand Vein at the muscle attachment sites are reduced in PS mutant embryos. Altogether, these results lead to a model in which integrin-mediated adhesion plays a role in regulating tendon cell differentiation by modulating the activity of the Egfr pathway at the level of its ligand Vein (Martin-Bermudo, 2000a).

Cell culture experiments have shown that integrins can regulate activation of the Egfr pathway at the level of the ligand, or at the level of the receptor. In the first case, integrins can regulate ligand activity through modulation of the composition and assembly of the ECM. There is increasing evidence suggesting that the binding of growth factors to the extracellular matrix is a major mechanism regulating growth factor activity. The largest group of ECM proteins that interact with growth factors include the heparan sulfates, which are the major components of the basement membrane -- indeed integrins contribute to the stabilization of the epidermal basement membrane. Integrins can also exert control on the Egfr pathway at the level of the receptor. In this scenario, the adhesion sites formed upon integrin activation (focal adhesions) can serve as recruitment points that bring together structural and signaling proteins, thus enhancing their ability to interact with the right partner, and therefore to be activated. Indeed clustering of integrins results in co-clustering of epidermal growth factor receptor molecules leading to receptor activation, and enhanced EGF-dependent activation of MAPK. In another example, integrins can also enhance the efficiency of signal transduction between the Egfr and MAPK by promoting the recruitment and activation of Raf (Martin-Bermudo, 2000a).

The data presented here supports a model by which the PS integrins regulate Egfr signaling pathway at the level of its ligand Vein. This regulation involves the ability of the PS integrins to organize the tendon matrix and the basement membrane at the basal surface of muscles and tendon cells. In fact, integrin function in regulating assembly of the ECM, rather than integrin signaling, has been shown to be crucial in keratinocyte differentiation. In the absence of integrin function the levels of Vein at the muscle attachment sites are decreased compared to wild type. Therefore, it is proposed that integrins are required for the proper assembly of the basement membrane and the tendon matrix, which in turn regulates Vein activity. A role for PS2 in matrix assembly is in agreement with results showing a requirement for alpha3beta1 integrin in mediating assembly of basement membrane between the epidermis and the dermis in mice. Furthermore, results in Drosophila showing that defects in tendon cell-specific gene expression are stronger when both integrins are eliminated are consistent with data showing that the failure in assembly of the matrix is more severe in embryos lacking both PS1 and PS2 integrins than in single mutants. The basement membrane and the tendon matrix could then regulate Vein activity in different ways. (1) They could promote a higher affinity of Vein for the Egfr. In fact, heparan sulfate has been reported to promote high-affinity binding of the fibroblast growth factor 2 (FGF2) and hepatocyte growth factor (HGF) to their receptors. (2) They could also direct the movement of Vein by limiting its diffusion. This could be a mechanism for muscles to specifically transmit signals to those epidermal cells that are in contact with the same matrix, the tendon cells. (3) They could promote the accumulation or clustering of Vein to specific levels required for the activation of its receptor. And finally, (4) binding of integrins to the ECM might either protect Vein from proteolysis or lead to the production of proteolytic enzymes that release Vein from the tendon matrix and activate it. Several of these mechanisms could be operating at the same time. Thus, organization and assembly of the tendon matrix via the PS integrins would ensure the localized production and concentration of an active ligand for the Egfr at the muscle attachment sites (Martin-Bermudo, 2000a).

In addition, or alternatively, the PS integrins could be required in the tendon cells to regulate Egfr function at the level of the receptor. At the muscle attachment sites of the Drosophila embryo, there are special cell junctions, called hemiadheren junctions (HAJs), which form between the ends of the muscles and the basal surface of the tendon cells in opposing pairs. HAJs are organized sites of membrane-cytoskeletal linkage which have been proposed to recruit integrins. It is worth mentioning here that although PS2 has been shown to be expressed only in the muscles, loss of PS2 integrin function affects adhesion of both muscle and epidermal HAJs. This can explain why lack of PS2 alone leads to a reduction in the expression of tendon cell-specific genes. At this level for integrin modulation of the Egfr signaling, a first step requires that epidermal HAJs act as recruitment centers for the Egfr or other signaling molecules, in the same way as focal adhesions. In this case, the detachment of the epidermal HAJs from the matrix found in embryos lacking the integrins results in the disorganization of these adhesion centers leading to a failure to cluster the Egfr and/or signaling molecules, and therefore, to activate the Egfr pathway. In this scenario it is also possible that integrins and the Egfr activate parallel pathways needed to reach the threshold level of MAPK activation, required for optimal transcription of tendon cell-specific genes (Martin-Bermudo, 2000a).

This is consistent with the results presented here where over activation of the Egfr pathway can compensate for lack of integrin function. Thus, integrin-mediated cell adhesion might produce a long-lasting activation of MAPK, which cooperates with the fast and short stimulation of MAPK normally induced by activation of growth factor pathways. Experiments were performed to try to determine the relative roles of integrin adhesion versus signaling in modulating the Egfr pathway in the process of tendon cell differentiation. One of the best characterized integrin signaling events involves tyrosine phosphorylation of the focal adhesion kinase, FAK. This pathway can be mimicked by clustering the cytoplasmic domain of the betaPS subunit. It has been shown previously that clustering of the cytoplasmic tail of the bPS subunit is sufficient to initiate a signaling pathway that regulates gene expression in the Drosophila midgut. However, this signaling pathway is found to be insufficient to regulate tendon cell differentiation in the embryo. These results suggest that integrin-mediated adhesion, rather than signaling, is required to regulate tendon cell differentiation. Some experiments have shown that clustering of the cytoplasmic domain of the beta subunits does not fully mimic integrin signaling, the alpha subunits are also important and, in some cases sufficient. A pathway from integrins to MAPK has been identifed that is mediated by interactions between the transmembrane and/or extracellular domains of the alpha subunit and the adaptor protein Shc. The pathway from integrins to MAPK is alpha subunit specific, being alpha5 and alphav, which belong to the same family as the alphaPS2, the alpha subunits that signal through Shc. Therefore, it still remains possible that PS2 integrin requirements to regulate tendon cell differentiation include a signaling function through Shc (Martin-Bermudo, 2000a).

zipper nonmuscle Myosin-II functions downstream of PS2 integrin in Drosophila myogenesis and is necessary for myofibril formation

Nonmuscle myosin-II is a key motor protein that drives cell shape change and cell movement. The function of nonmuscle myosin-II has been analyzed during Drosophila embryonic myogenesis. Nonmuscle myosin-II and the adhesion molecule, PS2 integrin (Myospheroid), colocalize at the developing muscle termini. In the paradigm emerging from cultured fibroblasts, nonmuscle actomyosin-II contractility, mediated by the small GTPase Rho, is required to cluster integrins at focal adhesions. In direct opposition to this model, it has been found that neither nonmuscle myosin-II nor RhoA appear to function in PS2 clustering. Instead, PS2 integrin is required for the maintenance of nonmuscle myosin-II localization and the cytoplasmic tail of the ßPS integrin subunit is capable of mediating this PS2 integrin function. Embryos that lack zygotic expression of nonmuscle myosin-II fail to form striated myofibrils. In keeping with this, a PS2 mutant that specifically disrupts myofibril formation is unable to mediate proper localization of nonmuscle myosin-II at the muscle termini. In contrast, embryos that lack RhoA function do generate striated muscles. Finally, nonmuscle myosin-II localizes to the Z-line in mature larval muscle. It is suggested that nonmuscle myosin-II functions at the muscle termini and the Z-line as an actin crosslinker and acts to maintain the structural integrity of the sarcomere (Bloor, 2001).

The myogenic function of nonmuscle myosin-II has been analyzed by using Drosophila genetics to manipulate the levels of nonmuscle myosin-II heavy chain, PS2 integrin, and RhoA GTPase in vivo in the developing larval muscles. Both nonmuscle myosin-II and PS2 colocalize at muscle termini. However, in contrast to models based on cultured fibroblasts, there is no evidence for either nonmuscle myosin-II or RhoA function in PS2 clustering. Instead, the maintenance of nonmuscle myosin-II localization at muscle termini is dependent on the presence of PS2 integrin and the cytoplasmic tail of the ßPS integrin subunit is sufficient for this. Further, nonmuscle myosin-II maintenance at the muscle termini is compromised in ifSEF, a ßPS2 integrin subunit mutant that specifically disrupts myofibril formation. Through the analysis of actin distribution in the musculature of living wild-type and mutant embryos, it has been demonstrated that RhoA-independent nonmuscle myosin-II function is required for the proper sarcomeric organization of the muscle cytoskeleton. Finally, since nonmuscle myosin-II localizes to the Z-line in late larval muscle, it has been suggested that nonmuscle myosin-II functions at both the muscle termini and the Z-line to maintain the structural integrity of the sarcomere (Bloor, 2001).

When fibroblasts in culture attach to ECM substrates through their cell surface integrin receptors, they dramatically redistribute these receptors such that they become clustered at focal adhesions. Integrin ligand binding also induces the actin cytoskeleton to rearrange into stress fibers, which terminate at focal adhesions and connect to the cytoplasmic domains of the clustered integrins. Pharmacological inhibitors of contractility block this complex cellular response, providing evidence that nonmuscle myosin-II driven contractility is essential for integrin clustering. At least a subset of nonmuscle myosin-II based contractility is dependent on RhoGTPase-mediated phosphorylation events. This suggests a model for focal adhesion and stress fiber formation in which, when diffuse cell surface integrins bind ECM ligand, they associate with actin filaments and activate Rho. In turn, Rho activates nonmuscle myosin-II, driving the formation of nonmuscle myosin-II filaments and increasing contractility. This increases the tension exerted on the actin cytoskeleton causing actin filaments to bundle and align into stress fibers (Bloor, 2001).

Bundling drives actin-associated cell surface integrins into clusters and focal adhesions are formed. The localization of PS2 integrin at Drosophila muscle termini is an ideal system in which to test this model in vivo. Indeed, consistent with the possibility that RhoA-mediated nonmuscle myosin-II contractility drives PS2 integrin clustering in the larval musculature, genetic evidence implicates RhoA-dependent activation of nonmuscle myosin-II in multiple morphogenetic pathways during Drosophila development. Furthermore, previous studies have shown that an uncharacterized intracellular mechanism is capable of driving integrin localization to the muscle termini. Despite this, this study shows that genetic depletion of either nonmuscle myosin-II or RhoA fails to disrupt PS2 localization (Bloor, 2001).

Maternally contributed RNAs and proteins support the early stages of Drosophila embryogenesis. Both zip and RhoA are maternally expressed and, in the absence of zygotic expression, this maternal contribution is sufficient for development to proceed normally until stage 14. Subsequent to this, depletion of maternal gene product results in defects in epidermal morphogenesis. Since the localization of PS2 integrin to the muscle termini occurs during stages 15 and 16, it seems likely that maternally contributed gene product is depleted from zip and RhoA mutant embryos prior to PS2 clustering. Thus, the presence of PS2 at muscle termini in zip and RhoA mutants suggests that neither nonmuscle myosin-II nor RhoA are required for PS2 clustering. However, the absolute amount of these gene products required to localize PS2 may be lower than that required for continued epidermal morphogenesis. Alternatively, these gene products may perdure longer in the mesoderm than in the developing epidermis (Bloor, 2001).

It is possible to eliminate the maternal contribution of a gene by generating mutant clones in the female germline. However, germline clones of zip and RhoA null mutations fail to make eggs. A novel technique was used to address the role of maternal nonmuscle myosin-II in PS2 localization. p127-l(2)gl, a nonmuscle myosin-II heavy chain binding protein, was overexpressed in a zip2 mutant background. The p127-l(2)gl protein binds and sequesters maternal non-muscle myosin-II heavy chain, effectively titrating the available levels of nonmuscle myosin-II and antagonizing its function. In these embryos, the epidermal morphogenesis defects associated with zip2 mutations are enhanced; however, muscle attachment and PS2 localization are unaffected. Muscle abnormalities are, however, observed in these and in zip zygotic null embryos: a variable subset of ventral muscles is deleted. Interestingly the affected muscles, VA1, VA2, and VA3, are derived from two muscle progenitors that arise from the same cluster of mesodermal cells. One progenitor divides to produce the muscle founder cells for VA1 and VA2. Subsequently, the other progenitor divides to produce the VA3 muscle founder and an adult muscle founder cell. The lineage and temporal relationships between these cells are reflected in the frequency at which these muscles are deleted: VA3 is more commonly deleted than VA1 and VA2, which are always either both present or both deleted. Although it is unclear whether these are the last progenitors to divide, it seems likely that defects in nonmuscle myosin-II-dependent cytokinesis are the basis of this phenotype. The fact that depletion of nonmuscle myosin-II can affect myogenic events occurring during stages 11 and 12 without affecting the localization of PS2 that occurs during stages 15 and 16 further supports the contention that nonmuscle myosin-II is not required for PS2 localization (Bloor, 2001).

An alternative approach to the analysis of gene function is the ectopic expression of dominant negative constructs. Although the interpretations of such experiments are not always unambiguous, ectopic expression of the dominant negative RhoAN19 construct has implicated RhoA function in biological processes not revealed by maternal and zygotic mutational analyses. For example, driving RhoAN19 expression in the early mesoderm disrupts invagination of this tissue, phenocopying embryos that lack both maternal and zygotic DRhoGEF2 expression. In this study, the 24B GAL4 driver was used to express UAS-RhoAN19 in the mesoderm during later stages of development (Bloor, 2001).

24B-driven expression can be detected by stage 10 and strong expression occurs from stage 13 onward. Thus, this approach will compromise endogenous RhoA function during late stages of mesoderm development. Indeed 24B-driven UAS-RhoAN19 expression affects the development of the visceral mesoderm and causes defects in somatic muscle patterning similar to those seen in zip mutant embryos. However, no defects were detected in somatic muscle attachment or PS2 localization in these embryos. Thus, while it is not absolutely certain that maternally supplied nonmuscle myosin-II or RhoA do not contribute to PS2 localization, these experiments provide strong evidence against this possibility (Bloor, 2001).

To exert tension on an underlying substrate, a cell must form a strong transmembrane connection between the substrate and its contractile cytoskeleton. At Drosophila muscle attachments this connection is mediated by localized PS2 integrin. As such, these structures superficially resemble focal adhesions, a point further emphasized by the localization of the focal adhesion protein integrin-linked kinase (ILK) to these sites. In contrast to the recruitment of intracellular proteins to focal adhesions, it has been shown that the initial localization of nonmuscle myosin-II to Drosophila muscle termini is not dependent on integrin clustering. Similarly, the localization of PAK and ILK are both independent of PS2 integrin. Furthermore, PS2 is also not required for the formation of an electron-dense hemiadherens junction at the muscle termini. However, among the proteins known to localize to the muscle termini, nonmuscle myosin-II is so far unique in that only it is dependent on PS2 integrin for its continued localization at these sites (Bloor, 2001).

Interestingly, the localization of nonmuscle myosin-II becomes PS2-dependent at the same developmental stage at which the PS2-dependence of muscle adhesion becomes apparent. When integrins bind ECM ligand, they are thought to undergo a conformational change that displaces the cytoplasmic tail of the integrin alpha subunit, exposing protein binding sites on the cytoplasmic tail of the ß subunit, a process known as outside-in signaling. It is possible that the accessibility of the ßPS cytoplasmic tail for protein-protein binding regulates PS2-dependent nonmuscle myosin-II localization. This is supported by the observation that, in the absence of endogenous PS2, the ßPS cytoplasmic tail is sufficient to keep nonmuscle myosin-II at the muscle termini. Excitingly, biochemical studies show that peptides derived from the cytoplasmic tail of the vertebrate integrin ß3 subunit are able to interact with the tail of nonmuscle myosin-II. In addition, this interaction has recently been demonstrated in cultured cells, suggesting that the maintenance of nonmuscle myosin-II localization might occur through a direct binding reaction with the ßPS cytoplasmic tail (Bloor, 2001).

PS2 integrin is required in muscle both for attachment to the epidermis and for the generation of sarcomeric ultrastructure. These data suggest that the sarcomeric function of PS2 is due, at least in part, to its role in maintaining nonmuscle myosin-II at the muscle termini. A possible function for nonmuscle myosin-II at the muscle termini is to physically link PS2 integrin to the muscle cytoskeleton by directly binding both PS2 and actin. One prediction of this model is that, in the absence of nonmuscle myosin-II, the muscle cytoskeleton will detach from the muscle termini, but that PS2 will continue to mediate adhesion of the muscle sarcolemma to the tendon matrix of the muscle-attachment site. Such a phenotype is observed in embryos mutant for ilk, the gene that encodes ILK. However, this disconnection of the muscle cytoskeleton from the muscle termini is clearly distinct from that observed in the muscles of zip mutant embryos, where muscle actin remains connected to the muscle termini, but fails to organize into sarcomeres. While this does not rule out a role for nonmuscle myosin-II in connecting PS2 and actin, it certainly argues against nonmuscle myosin-II being the primary component of this link (Bloor, 2001).

An alternative function for nonmuscle myosin-II is suggested by the demonstration that nonmuscle myosin-II localizes to the Z-line in the somatic muscles of third instar larvae. Both the Z-line and the muscle termini are ultrastructural elements that transmit tensile stress during muscle contraction. It is possible that nonmuscle myosin-II might function as an actin-crosslinking protein at these sites to help maintain their structural integrity. This role is generally assumed to be a function of alpha-actinin, an actin crosslinking protein that is the major component of muscle termini and Z-lines in both vertebrates and invertebrates. Mutations in the single Drosophila alpha-actinin gene do cause terminal defects and sarcomeric abnormalities. Surprisingly though, embryos that lack both maternal and zygotic alpha-actinin expression hatch and lethality does not occur until the end of the first larval instar. Furthermore, the organization of actin into a striated pattern of I-bands is unaffected in alpha-actinin mutants. This implicates other actin-binding proteins in the maintenance of actin organization at the muscle termini and Z-line and the data suggest that nonmuscle myosin-II maybe one such protein (Bloor, 2001).

There is no clear model of how nonmuscle myosin-II might fulfill this function. Intriguingly, myosin heads from a single filament have been shown to be able to bind parallel actin thin filaments of opposite orientation. One speculation is that nonmuscle myosin-II might be capable of such behavior, but how such cross-links would occur at cellular levels of ATP is not clear. One possibility is that nonmuscle myosin-II in the muscle termini and Z-line is in a 'catch' muscle state in which tension is maintained without turnover of ATP. By this scenario, nonmuscle myosin-II would not function in contraction, but would serve as an effective actin crosslinker. A noncontractile function for nonmuscle myosin-II in myofibrillogenesis would explain why RhoA appears to have no role in this process. It is interesting to note that, in Dictyostelium, a contraction-independent function of non-muscle myosin-II has been shown to be important for the generation of cortical tension (Bloor, 2001).

Finally, PS2 has been shown to be present at the sarcolemma above the Z-line in cultured Drosophila myotubes. It is therefore possible that PS2 somehow functions to maintain nonmuscle myosin-II within the Z-line. Indeed, adhesion between Z-line-associated hemiadherens junctions and the muscle basement membrane fails in the absence of PS2. This suggests that the integrity of the sarcomeric muscle cytoskeleton requires it to be connected to the ECM at both the muscle termini and the Z-line and that this connection is mediated by nonmuscle myosin-II and PS2 integrin (Bloor, 2001).

Integrins and midgut migration

Cell migration during embryogenesis involves two populations of cells: the migrating cells and the underlying cells that provide the substratum for migration. The formation of the Drosophila larval midgut involves the migration of the primordial midgut cells along a visceral mesoderm substratum. Integrin adhesion receptors are required in both populations of cells for normal rates of migration. In the absence of the PS integrins, the visceral mesoderm is disorganized, the primordial midgut cells do not display their normal motile appearance and their migration is delayed by 2 hours. Removing PS integrin function from the visceral mesoderm alone results in visceral mesoderm disorganization, but only causes a modest delay in migration and does not affect the appearance of the migrating cells. Removing PS integrin function from the migrating cells causes as severe a delay in migration as the complete loss of PS integrin function. The functions of PS1 and PS2 are specific in the two tissues, endoderm and mesoderm, since one cannot substitute for the other. In addition, there is a partial redundancy in the function of the two PS integrins expressed in the endoderm: PS1 (alphaPS1betaPS) and PS3 (alphaPS3betaPS: alpha PS3 is Scab), since loss of just one alpha subunit in the midgut results in either a modest delay (alphaPS1) or no effect (alphaPS3). Thus, the migration of primordial midgut cells along a substratum provided by the visceral mesoderm requires the coordinated function of all three PS integrins. The PS2 integrin is required in the visceral mesoderm for the formation of an optimal substratum for migration. The PS1 and PS3 integrins are required in the endodermal cells for their migration over the visceral mesoderm. These two integrins are partially redundant in the midgut, but PS3 cannot fully replace PS1 integrin function. The roles of small GTPases in promoting migration of the primordial midgut cells were also examined. Dominant negative versions of Rac and Cdc42 cause a very similar defect in migration as does the loss of integrins, while dominant negative versions of Rho and Ras have no effect. Thus integrins are involved in mediating migration by creating an optimal substratum for adhesion, adhering to that substratum and possibly by activating Rac and Cdc42 (Martin-Bermudo, 1999b).

The separate loss of either alphaPS1 or alphaPS2 does not produce the same migration phenotype as loss of betaPS. However, loss of the PS2 integrin does produce the same defects in the organization of the visceral mesoderm as loss of all PS integrins. There are at least two possible explanations for this result: (1) either the alphaPS subunit is sufficient to mediate migration, or (2) there is another a subunit that can partially compensate for the loss of alphaPS1 or alphaPS2. In order to distinguish between these possibilities, embryos mutant for both alphaPS1 and alphaPS2 were examined. If the first explanation were correct then one would expect the double mutant phenotype to be as strong as the loss of betaPS. Endodermal cells from alphaPS1alphaPS2 minus embryos show a greater delay in cell migration than that observed in the single alpha subunit mutants. This can be seen when only one or two cells are found in contact in the middle in the double mutant embryos at a time when several cells have met in the single mutant embryos. Nevertheless, in this double mutant combination, the endodermal cells send projections and the delay in cell migration (1 hour) is still less severe than that observed in betaPS minus embryos. This situation is different from the role played by the integrins in muscle attachment, where the loss of PS1 and PS2 integrins is equivalent to the loss of all PS integrins. Therefore, these results suggest that additional alpha subunits exist that can compensate for the loss of alphaPS1 and alphaPS2. These alpha subunits could be required in the endoderm, in the visceral mesoderm or in both tissues (Martin-Bermudo, 1999b).

Since loss of PS2 appears to affect the formation of the visceral mesoderm substratum as strongly as loss of the betaPS subunit, and this does not substantially delay migration, a model is favored in which other alpha subunits are required in the endoderm. Since alphaPS3 (Scab) is also expressed in the endoderm, it is an obvious candidate for a second alpha required in the endoderm for cell migration. Therefore, as the next step in analyzing the role of integrins in midgut migration, the specific requirements of the betaPS subunit in the two different tissues was examined, as was the role of alphaPS3 (Martin-Bermudo, 1999b).

Analysis of embryos mutant for alphaPS3 supports previous evidence indicating that there is no defect in midgut migration, showing that the PS3 integrin is not essential for midgut migration. However, it is possible that there is some redundancy between the two alpha subunits expressed in the endoderm, so that alphaPS1 compensates for the lack of alphaPS3 and vice versa. To test this, embryos lacking both PS1 and PS3 integrins were generated and midgut migration was examined. Because alphaPS1 and alphaPS3 genes are on different chromosomes, it was difficult to use the GAL4 system to express any marker, so an enhancer trap line, 258, was used that is expressed in the endodermal cells as they migrate. This is a nuclear marker that allows identification of the position of the endodermal cells and therefore the monitoring of their migration, although it does not reveal the morphology of the cells. In the absence of both alphaPS1 and alphaPS3 subunits, the defect in the migration of the midgut is as strong as in the absence of the betaPS subunit, consisting of a delay of approximately 2 hours. Embryos doubly mutant for alphaPS3 and alphaPS2 subunits, were examined and no difference in midgut development was found as compared to the loss of alphaPS2: this shows that there is no overlap in function of the PS2 and PS3 integrins. In conclusion, the main requirement for PS integrins during migration is provided by the combination of PS1 and PS3 integrins. To investigate whether PS3 is required to regulate gene expression in the midgut, the patterns of expressions of the two integrin target genes were examined. Loss of PS3 function does not cause any change in their expression. Thus, PS1 function in migration can also be partially performed by the PS3 integrin (there is a small delay in the absence of PS1), while PS1 function in the regulating gene expression or adhesion to the visceral mesoderm cannot be performed by PS3 (Martin-Bermudo, 1999b).

Two integrin ß subunits are encoded in the Drosophila genome. The ßPS subunit is widely expressed and heterodimers containing this subunit are required for many developmental processes. The second ß subunit, βInt-ν (ßNeu), is a divergent integrin expressed primarily in the midgut endoderm. To elucidate its function, null mutations in the gene encoding βInt-ν were generated. βInt-ν is not required for viability or fertility, and overall the mutant flies are normal in appearance. However, βInt-ν function could be observed in the absence of ßPS. Consistent with its expression, removal of βInt-ν only enhanced the phenotype of ßPS in the developing midgut. In embryos lacking the zygotic contribution of ßPS, loss of βInt-ν resulted in enhanced separation between the midgut and the surrounding visceral mesoderm. In the absence of both maternal and zygotic ßPS, a delay in midgut migration was observed, but removing ßNu as well blocked migration completely. These results demonstrate that the second ß subunit can partially compensate for loss of ßPS integrins, and that integrins are essential for migration of the primordial midgut cells. The two ß subunits mediate midgut migration by distinct mechanisms: one that requires talin and one that does not. Other examples of developmental cell migration, such as that of the primordial germ cells, occurred normally in the absence of integrins. Having generated the tools to eliminate integrin function completely, it was confirmed that Drosophila integrins do not control proliferation as they do in mammals, and have identified alphaPS3 as a heterodimeric partner for βInt-ν (Devenport, 2004).

This genetic analysis of the second of the two ß integrin subunits encoded in the Drosophila genome has resulted in three main findings. (1) The ßν integrin subunit makes a minor contribution to development and viability of Drosophila. ßν integrins can contribute to the development of the tissue in which it is highly expressed, but only when the function of the other ß integrin subunit, ßPS was reduced or eliminated. (2) The alphaPS3ßν (Scabßν) integrin (possibly in combination with other ßν-containing integrins) is the missing receptor that mediates primordial midgut cell migration in the absence of ßPS-containing integrins, and thus the migration of these cells over the visceral mesoderm substrate is completely integrin dependent. (3) It was possible to eliminate concerns that the ßν subunit might be compensating for the absence of the ßPS subunit in a variety of developmental events that apparently occur normally in the absence of ßPS. By removing the maternal and zygotic contributions of both integrin ß subunits, embryos completely lacking integrin function have been generated for the first time. The only difference noted in the development of these embryos, in comparison with those lacking ßPS alone, was the complete failure of the primordial midgut cells to move. Notably, a second migration event occurring at the same time, the migration of the primordial germ cells, occurred normally without integrins. Although every developmental event during embryogenesis was not exhaustively examined, what is important is that the genetic tools have been generated to allow the examination of developmental events in the complete absence of integrins. By also analyzing clones of cells lacking both ß integrin subunits, it was confirmed that in Drosophila integrins are not required for proliferation of epithelial cells, nor for the initial establishment of apicobasal polarity within epithelial layers (Devenport, 2004).

One of the key questions to emerge from this work is what changes ~2 hours after the time when primordial midgut migration is normally initiated, so that ßν can now mediate migration? It seems likely that the developmental change that permits ßν integrin-dependent migration is the synthesis of an essential protein or proteins. It is not just the timing of ßν synthesis itself that is regulated, because expression of ßν earlier than normal did not alleviate the delay in migration, even though expression of ßPS with the same approach did successfully restore migration at the normal time in embryos lacking endogenous ßPS. Therefore, at least one additional protein is required for ßν-mediated migration. This protein could have one of several possible functions: an alpha integrin subunit heterodimeric partner; an extracellular matrix ligand; an intracellular protein required to link integrins to the cytoskeleton, vital for cell movement, and/or a regulator of any of these proteins. The alphaPS3 subunit is known to be present in the midgut at the early stages of migration because eliminating it along with the alphaPS1 subunit results in delayed migration, so the possibility that alphaPS3 is a limiting factor for ßν-mediated migration can be ruled out. Curiously, eliminating the alphaPS1 and alphaPS3 subunits zygotically did not block migration completely, as might be expected if alphaPS3 is the major alpha subunit partner for ßν. Either there is a substantial maternal contribution of the alphaPS3 subunit or perhaps the alphaPS4 or alphaPS5 subunits also function in the midgut (Devenport, 2004).

Evidence to suggest that integrins containing the two ß subunits mediate migration by interacting with different cytoplasmic linker proteins came from an analysis of the role of talin in midgut migration. Talin binds directly to integrins, with more than one binding site, and the detailed nature of one crucial interaction has been characterized at high resolution. The defect in midgut migration caused by the loss of talin resembles the defect caused by the loss of ßPS rather than both ß subunits. This is consistent with the divergence of the ßν cytoplasmic tail, particularly a lack of conservation of a tryptophan that makes a key contact with talin in ß3. Therefore, it seems likely that ßν makes interactions with an alternate cytoplasmic linker, and it may be that the synthesis of this protein is what permits ßν-dependent migration. One candidate protein, filamin 1, appears to be ruled out by the observation that its localization is not ßν dependent (Devenport, 2004).

The functions observed for ßν are seen only in the absence of ßPS, so it is still unclear what ßν does under normal, wild-type conditions. Although a relatively divergent integrin, other insect genomes that have been sequenced also contain an orthologue of the ßν gene, suggesting that it does have a function worthy of retaining through evolution. One possibility is that ßν contributes to the architecture of the midgut in way that is not obvious by appearance but that makes an important contribution to the physiology of this organ. If the flies lacking ßν integrin are unable to digest their food as well as their wild-type counterparts, this would probably make these flies less competitive in the wild (Devenport, 2004).

Creating the tools to eliminate all integrin heterodimers allowed the addressing of whether some integrin functions that are well-established in vertebrates and cell culture are important in Drosophila. For example, cell cycle progression in cultured mammalian cells is absolutely dependent on their adhesion to an extracellular matrix and, consequently, cells in suspension will arrest their cell cycle. Integrins are the major adhesion molecules that control this phenomenon called 'anchorage-dependent growth'. However, the developmental relevance of this phenomenon is not completely clear because in an intact organism most cell types, other than those of the circulatory system, will never be in suspension. Genetic experiments in mouse are beginning to address this issue and have demonstrated that integrins are important for proliferation during the development of a number of tissues including skin, bone and embryonic ectodermal ridge cells. Whether integrins are required for cells to divide in Drosophila was tested, and, surprisingly, they are not. By comparing the size of clones of cells generated by mitotic recombination, it was found that double integrin mutant epithelial cells are able to proliferate at approximately the same rate as their siblings that just lack ßν. Thus, it appears that integrins have adapted an additional function in regulating cell proliferation during the evolution of the vertebrate lineage. Perhaps this additional level of cell cycle regulation arose with the massive increase in the number of cell divisions that mammals undergo throughout their development. It is clearly advantageous for cells to arrest proliferation in the absence of adhesion to prevent the growth of metastatic tumours, so perhaps another, non-integrin adhesion molecule is permissive for growth in Drosophila epithelia (Devenport, 2004).

There is a substantial amount of data to suggest that integrins play a role in epithelial architecture and polarity, but the data are conflicting regarding the precise roles. Experiments using 3D epithelial cysts demonstrated a role for integrins in orienting the direction of polarity, but not for establishing distinct apical and lateral membrane domains. However, experiments in vivo suggest that integrins maybe required for the initial establishment of polarity. For example, epiblasts derived from laminin-/- or ß1 integrin-/- ES cells fail to form a polarized epithelium or a proamiotic cavity. Furthermore, laminin mutants in C. elegans show numerous polarity defects, such as non-basal integrin adhesions and ectopic adherens junctions. Advantage was taken of the double ß integrin mutants to address the role of integrins in Drosophila epithelia; they are important for the maintenance, but not the establishment of polarity in secondary epithelia. In embryos lacking ßν and zygotic ßPS, the integrity of the endoderm is severely compromised, although its polarity is initially established. Furthermore, when integrin heterodimers were eliminated in clones of cells, the initial distribution of apicolateral markers in the follicular epithelium was observed to be normal. However, epithelial integrity was compromised because cells eventually rounded up and formed multiple layers, a common phenotype seen in mutants that disrupt polarity or epithelial integrity. Thus, cells cannot retain their polarity without basal adhesion, but integrins are not necessary to set up the establishment of apical and lateral domains in Drosophila (Devenport, 2004).

Although the roles of ßν during Drosophila morphogenesis are found to be minor, the important findings of this study are twofold. (1) The presence of another integrin ß subunit in Drosophila had required cautious interpretation of previous genetic analyses of PS integrin mutants. Therefore, it was important to establish whether redundancy by ßν masked any important roles for the PS integrins in Drosophila. However, this study has shown that removal of ßν does not enhance ßPS mutant phenotypes in the muscle, CNS or embryonic epidermis. It can now be said with confidence that in tissues other that the midgut, ßPS-containing integrins are the primary players in cell-matrix adhesion in Drosophila. Hence, for all practical purposes, germline depletion of ßPS essentially eliminates integrin function in all embryonic tissues other than the midgut. (2) It is now established that Drosophila integrins are important for some but not all of the functions that they mediate in vertebrates. Although important for muscle cell fusion, establishment of epithelial polarity and cell proliferation in mammals, integrins are not required for these processes in Drosophila (Devenport, 2004).

Integrins and axon guidance

Three position-specific (PS) integrins are expressed at the Drosophila neuromuscular junction (NMJ): a beta subunit (betaPS known as Myospheroid), expressed in both presynaptic and postsynaptic membranes, and two alpha subunits (alphaPS1, alphaPS2), expressed at least in the postsynaptic membrane. PS integrins appear at postembryonic NMJs coincident with the onset of rapid morphological growth and terminal type-specific differentiation, and are restricted to type I synaptic boutons, which mediate fast, excitatory glutamatergic transmission. Two distinctive hypomorphic mutant alleles of the beta subunit gene myospheroid (mysb9 and mysts1), differentially affect betaPS protein expression at the synapse to produce distinctive alterations in NMJ branching, bouton formation, synaptic architecture and the specificity of synapse formation on target cells. The mysb9 mutation alters betaPS localization to cause a striking reduction in NMJ branching, bouton size/number and the formation of aberrant 'mini-boutons'; this may represent a developmentally arrested state. The mysts1 mutation strongly reduces betaPS expression to cause the opposite phenotype of excessive synaptic sprouting and morphological growth. NMJ function in these mutant conditions is altered in line with the severity of the morphological aberrations. Consistent with these mutant phenotypes, transgenic overexpression of the betaPS protein with a heat-shock construct or tissue-specific GAL4 drivers causes a reduction in synaptic branching and bouton number. It is concluded that betaPS integrin at the postembryonic NMJ is a critical determinant of morphological growth and synaptic specificity. These data provide the first genetic evidence of a functional role for integrins at the postembryonic synapse (Beumer, 1999).

At the Drosophila larval NMJ, motor axons contact specific target muscles and form characteristic arborizations at defined locations on the muscle surface. These arborizations consist of branches forming numerous varicosities (boutons) falling into two primary classes: glutamatergic type I boutons (2-5 mm diameter) innervate all muscles and mediate fast, excitatory transmission; octopaminergic type II boutons (~1 mm diameter) are found on a subset of muscles, and appear to have a primarily modulatory function. Growth at all arbors is activity dependent and a number of molecules involved in regulating this morphological plasticity have been identified. The possible function of integrins in morphological synaptic plasticity in this system becomes apparent with the isolation of Volado, an alpha integrin subunit required for Drosophila memory. Additionally, expression of one subunit, alphaPS2, at the NMJ has been reported. To assay synaptic integrin expression, larval preparations were probed with specific antibodies against alphaPS1, alphaPS2, betaPS and betannu integrin subunits. The alphaPS1, alphaPS2 and betaPS subunits are clearly detectable; only the betanu subunit does not appear to be present in the NMJ. PS integrin expression is restricted to type I boutons in a halo surrounding an unlabeled center, suggesting primarily postsynaptic localization. In double-labeled confocal sections, alphaPS and betaPS integrin staining overlaps anti-DLG (discs-large), a primarily postsynaptic marker, although the PS integrin expression appears to be less uniform than the anti-DLG staining. alphaPS1 integrin staining is consistently fainter and less extensive than alphaPS2 and betaPS integrin staining. In addition to the intense betaPS localization at the NMJ, fainter staining is observed throughout the muscle. This is interpreted to be sarcomeric expression of betaPS dimerized with alphaPS2 (Beumer, 1999).

In order to localize PS integrins expression more precisely, mature larval NMJs were probed with anti-PS integrin antibodies and embedded for ultrastructural analysis. Antibody staining against alphaPS1, alphaPS2 and betaPS integrin subunits all show expression throughout the subsynaptic reticulum (SSR), a system of postsynaptic foldings of the muscle membrane that surrounds the presynaptic boutons in type I arbors. These results clearly confirm that all three PS integrins occur at least postsynaptically. In additional experiments, pre-embedding-anti-betaPS staining can be achieved in the absence of detergent, thus preserving the cell membranes. In these specimens, unopposed presynaptic membranes often reveal staining deposits suggesting that at least betaPS integrin is expressed in the presynaptic membrane in addition to the postsynaptic SSR. Developmentally, all three PS integrins become detectable at the NMJ in late first or early second instar larvae, although others have reported earlier expression in the CNS. No expression in embryos or hatching first instar larvae could be detected in the peripheral nerves or NMJ, although the expected expression at the muscle attachment sites is clearly visible for all three subunits. Thus, PS integrin expression becomes concentrated at NMJs only after they have been differentiated and stabilized during embryogenesis. This late onset of expression coincides with the appearance of the SSR, the period when different morphological classes of NMJs first appear and the terminals first become subject to measurable activity-dependent morphological plasticity (Beumer, 1999).

Because errors in pathfinding have been reported in animals lacking alpha integrin subunits, third instar larvae with altered integrin expression at the NMJ were examined for aberrations that might be the result of either errors in pathfinding or in synapse remodeling or maturation. For this analysis, the complex, muti-innervated muscle 12 NMJ was analyzed. Three types of gross patterning abnormalities, designated as backbranched, pulled away or multiple insertions, were noted and quantified in a blind study. The first of these phenotypes, backbranched, consists of at least one arboreal branch leaving muscle 12 and branching back onto muscle 13 from a site away from the point of nerve entry onto muscle 12. This architecture is very rare in normal animals but occurs at abnormally high levels in all mys mutant animals; the error is most common in mysb9 and mysb9 /mysXG43 (both greater than 30%). The second phenotype, pulled away, is characterized by boutons suspended in the space between muscles 12/13, and defasciculation, or branching, beginning before the nerve has actually reached the point of insertion on the muscle. This phenotype is rare in controls but common in all genotypes with severely reduced betaPS expression; mysts1/mysXG43 mutant larvae show a 50% incidence of this defect. Finally, the third phenotype, multiple insertions, is characterized by two or more separate and distinct nerve entry points on muscle 12. In these junctions, both nerves can frequently be traced back to a point of separation above muscle 13, though occasionally the separation occurs after the nerve has left muscle 13. The appearance of multiple nerve entry points is rare in controls and is not significantly increased in mysb9 and mysb9 /mysXG43 mutants. However, multiple nerve entry points are a common consequence of severe betaPS loss in both mysts1 and mysts1/mysXG43 animals. It is concluded that alterations in betaPS expression also exert effects on the pattern of innervation and synaptic specificity during NMJ development (Beumer, 1999).

These observations demonstrate that the level of PS integrins at postembryonic NMJs is a critical determinant of morphological sprouting, growth and bouton differentiation. (1) Moderate decrease in the detection of betaPS expression at the NMJ accompanied by an increase in extrasynaptic muscle expression (in mysb9) results in undergrowth and compaction of the terminal, loss of boutons, a decrease in overall bouton size and the proliferation of a new class of 'mini-boutons'. (2) In contrast, severe reduction in the detection of betaPS expression at the NMJ, unaccompanied by a detectable increase in extrasynaptic expression (in mysts1), results in an increase in sprouting and morphological growth to cause overgrowth and an increase in terminal complexity. (3) Overexpression of betaPS, generalized or nerve/muscle-specific, reduces branching complexity and bouton numbers. In agreement with the mysb9 observations, early betaPS overexpression in the muscle also increases the frequency of mini-boutons. Thus, the subcellular distribution of betaPS in both the motor neuron and its muscle partner is an important determinant of morphological development at the postembryonic NMJ (Beumer, 1999).

The identity of the mini-boutons is of particular interest. Mini-boutons are found very abundantly in mysb9 and betaPS-overexpressing terminals, but are also observed at a low frequency in wild-type NMJs. Molecularly, these structures show the hallmarks of type I boutons but are many-fold smaller. An exciting recent study (Zito, 1999) using confocal imagery of live synapses has suggested that these structures may represent immature type I boutons and/or new terminal branches. Molecular labeling, presented in the Beumer study (1999), defines these structures as nascent type I boutons. It appears, therefore, that decreased betaPS integrin expression triggers the differentiation process for these bouton precursors, but that some subsequent condition required for maturation is not being met. These observations suggest that proliferation of these immature type I boutons is correlated with extrasynaptic expression of betaPS. Thus, it is suggested that bouton formation is developmentally arrested when presented with the incorrect expression of betaPS in the target membrane, leading to a failure of terminal bouton maturation (Beumer, 1999).

Calcium/calmodulin dependent kinase II (CaMKII), PDZ-domain scaffolding protein Discs-large (Dlg), immunoglobin superfamily cell adhesion molecule Fasciclin 2 (Fas2) and the position specific (PS) integrin receptors, including ßPS (Myospheroid) and its alpha partners (alphaPS1, alphaPS2, PS3/alpha Volado), are all known to regulate the postembryonic development of synaptic terminal arborization at the Drosophila neuromuscular junction (NMJ). Recent work has shown that Dlg and Fas2 function together to modulate activity-dependent synaptic development and that this role is regulated by activation of CaMKII. PS integrins function upstream of CaMKII in the development of synaptic architecture at the NMJ. ßPS integrin physically associates with the synaptic complex anchored by the Dlg scaffolding protein, which contains CaMKII and Fas2. This study demonstrates an alteration of the Fas2 molecular cascade in integrin regulatory mutants, as a result of CaMKII/integrin interactions. Regulatory ßPS integrin mutations increase the expression and synaptic localization of Fas2. Synaptic structural defects in ßPS integrin mutants are rescued by transgenic overexpression of CaMKII (proximal in pathway) or genetic reduction of Fas2 (distal in pathway). These studies demonstrate that ßPS integrins act through CaMKII activation to control the localization of synaptic proteins involved in the development of NMJ synaptic morphology (Beumer, 2002).

ßPS integrin is a transmembrane receptor present in both pre- and post-synaptic membranes at the larval NMJ. Dlg is a synaptic scaffolding protein associated with both pre-and post-synaptic membranes and is also involved in the regulation of synaptic morphology, through the localization of diverse transmembrane proteins (including Fas2). Studies were undertaken to determine whether ßPS integrin associates with the DLG complex to tie together these disparate receptor components in a common molecular machine. In confocal analyses, ßPS integrin and Dlg co-localize at the larval NMJ. Dlg clearly has less extensive expression, more tightly localized at NMJ boutons, whereas ßPS is more extensive through the subsynaptic reticulum (SSR) and also localized at extrasynaptic sites in the muscle, including the muscle sarcomere and attachment sites. However, both proteins are most intensively expressed at the NMJ presynaptic/postsynaptic interface, where they co-localize. Therefore, tests were performed to determine if ßPS integrins form part of the synaptic complex linked by Dlg. Protein immunoprecipitation assays were performed using rabbit anti-DLG to probe Oregon R head extracts. Dlg antibodies clearly co-immunoprecipitate Dlg and ßPS integrin protein, consistent with co-localization observed in confocal analysis. Inspection of the ratio of both bound proteins and proteins not bound to beads (immunoprecipitation versus flow through lanes) indicates that a large portion of the ßPS integrin protein associates with a complex containing Dlg. The fact that ßPS was found to co-immunoprecipitate with the complex mediated by Dlg provides support for integrins existing in a synaptic complex with Fas2 and CaMKII at the synapse (Beumer, 2002).

To assay the mys requirement in synaptic development comprehensively, two regulatory alleles of mys with opposing structural phenotypes were compared and contrasted: mysb9, which causes NMJ undergrowth, and mysts1, which causes NMJ overgrowth. In both mutants, a correlation was observed between the morphological alterations and synaptic function, in contrast to the homeostasis seen in Fas2 mutant synapses but consistent with the parallel alterations seen in CaMKII-inhibited synapses. However, the correlation between synaptic structural and functional alterations in mys mutants is not striking, and it is clear that integrins primarily mediate architectural regulation. Therefore, the focus in this study was exclusively on the mechanism(s) by which integrins modulate NMJ structural development. This article presents molecular and genetic evidence that strongly support the hypothesis that synaptic integrin receptors containing ßPS modulate CaMKII activation on both sides of the synaptic cleft and, through CaMKII, control both the expression and the synaptic localization of Fas2 at the synapse. The primary experimental results supporting these conclusions are detailed below (Beumer, 2002).

(1) Transgenic expression of CaMKII is sufficient to completely rescue all synaptic structure defects in mys mutants. Genetically increasing CaMKII in postsynaptic muscle, but not presynaptic neuron, completely rescues the structural undergrowth of the mysb9 integrin mutant, whereas it is necessary to elevate CaMKII both pre- and post-synaptically to rescue structural overgrowth in the mysts1 mutant. These results are consistent with the postsynaptic mislocalization of ßPS integrin in the mysb9 mutant, as opposed to the global loss of synaptic integrin in the mysts1 mutant. These data indicate that coordinate regulation of CaMKII in the muscle and motoneuron is necessary for proper development of synaptic architecture (Beumer, 2002).

(2) At the distal end of the cascade, both the expression and synaptic localization of Fas2 are increased in mys mutants, although the extent of Fas2 misregulation is significantly different between the two regulatory mutants. Importantly, both the NMJ overgrowth (mysts1) and undergrowth (mysb9) phenotypes are rescued towards wild-type structure by genetically reducing the amount of Fas2 available at the synapse to near normal levels. In mysts1 mutants, correcting for Fas2 level suppresses the synaptic overgrowth, while in mysb9 mutants, correcting the Fas2 level suppresses the synaptic undergrowth. These results support the conclusion that Fas2 is centrally involved in mediating synaptic growth, but suggest that the Fas2-mediated mechanism is more complex than previously thought (Beumer, 2002).

The results demonstrate that integrins regulate morphological growth at the postembryonic NMJ through activation of CaMKII in both pre- and post-synaptic compartments. One important target of CaMKII is Fas2 and it is clear that regulation of Fas2 expression and localization is an important component of integrin regulation. However, the modulation of Fas2 levels alone is unlikely to account fully for the alterations in synaptic architecture in integrin mutants. In particular, both mysts1 and mysb9 upregulate synaptic Fas2 levels, albeit to different degrees, yet show opposite alterations in synaptic growth. Moreover, reducing Fas2 levels rescues both under- and overgrowth defects, but the rescue is not perfect (Beumer, 2002).

Precise control of Fas2 levels finely tunes morphological development at the NMJ. Different degrees of reduced Fas2 expression can either facilitate or inhibit the growth/maintenance of the NMJ, and reduced Fas2 expression has been demonstrated in other overgrowth mutants. However, overexpression of Fas2 in specific muscles drives increased NMJ elaboration/bouton differentiation and selective preference for a high-expressing muscle over a low-expressing muscle. How can this complexity be interpreted? One likely explanation is interaction between Fas2 and other developmental pathways regulated by CaMKII. To date, the only known Fas2-independent regulation of synaptic morphology involves a deubiquitinating protease encoded by fat facets (faf), and a putative ubiquitin ligase encoded by highwire (hiw), which have been shown to work together to modulate the degree of NMJ growth. Loss-of-function hiw mutants display NMJ structural overgrowth, importantly without a concomitant decrease in Fas2. Indeed, overexpression of Fas2 cannot suppress the overgrowth seen in hiw mutants. These observations are consistent with the overgrowth combined with overexpression of Fas2 observed in mysts1 mutants in this study. However, any putative interaction between Fas2-dependent and -independent mechanisms of morphological growth regulation at the Drosophila NMJ synapse remain to be elucidated. It is expected that further study will reveal an additional structural control mechanism regulated by CaMKII, acting in parallel to and interacting with Fas2. The ubiquination mechanism discussed here is one candidate mechanism, but as CaMKII is known to have many synaptic targets, it is clearly not the only candidate (Beumer, 2002).

In summary, it is concluded that architectural developmental defects observed in NMJ synapses mutant for ßPS integrin are due to the loss of the ability to regulate synaptic CaMKII properly. One function of CaMKII is to phosphorylate the scaffolding protein Dlg, and thus regulate the proteins synaptically localized by this scaffold. Fas2 is the central known output of this regulatory cascade. Loss of this regulation is central to the mys mutant defects in the postembryonic elaboration of NMJ structure. It is clear from this study, however, that regulation of Fas2 localization via CaMKII is only one facet of how integrins function at the synapse (Beumer, 2002).

Mechanisms that regulate axon branch stability are largely unknown. Genome-wide analyses of Rho GTPase activating protein (RhoGAP) function in Drosophila using RNA interference has identified p190 RhoGAP as essential for axon stability in mushroom body neurons, the olfactory learning and memory center. RhoGAP inactivation leads to axon branch retraction, a phenotype mimicked by activation of GTPase RhoA and its effector kinase Drok and modulated by the level and phosphorylation of myosin regulatory light chain. Thus, there exists a retraction pathway from RhoA to myosin in maturing neurons, which is normally repressed by RhoGAP. Local regulation of RhoGAP could control the structural plasticity of neurons. Indeed, genetic evidence supports negative regulation of RhoGAP by integrin and Src, both implicated in neural plasticity (Billuart, 2001).

Tests were performed to see if integrin could regulate p190 activity in MB neurons, since mammalian studies suggest potential links between integrin and p190, as well as integrin and Src. Integrins function as heterodimers of one alpha and one ß subunit. Drosophila has five genes for integrin alpha subunits, including one, volado (vol), also called scab, which is preferentially expressed in MB neurons and, when mutated, results in a short-term memory defect. No significant modification of the p190 RNAi phenotype was observed in flies heterozygous for a null mutation of vol (vol4). This may be because vol is not dosage sensitive or because it functions redundantly with other integrin alpha subunits in regulating p190 activity. Only two genes encode integrin ß subunits in the fly, ßPS and ßnu; ßnu may not associate with an alpha subunit. The myospheroid (mys) gene, encoding ßPS, shows robust genetic interaction with p190. In flies with one wild-type copy of mys, the p190 RNAi phenotype is markedly suppressed for each of the three mys alleles tested, suggesting that p190 is negatively regulated by integrin (Billuart, 2001).

Finally, MB neuroblast clones homozygous for mys1 were examined using the MARCM system. Twelve of 52 neuroblast clones exhibited obvious dorsal lobe axon overextension not seen in control flies. These include overextension of thin axon bundles near the tip of dorsal lobe similar to those seen in Drok2 clones, or overextension of a large portion of the dorsal axons similar to those seen in MB neurons overexpressing p190. This experiment indicates that integrins are essential negative regulators of axon extension in MB neurons (Billuart, 2001).

Integrins are concentrated within growth cones, but their contribution to axon extension and pathfinding is unclear. Genetic lesion of individual integrins does not stop growth cone extension or motility, but does increase axon defasciculation and axon tract displacement. In this study, a dosage-dependent phenotypic interaction is documented between genes for the integrins, their ligands, and the midline growth cone repellent, Slit, but not for the midline attractant, Netrin. Longitudinal tract axons in Drosophila embryos doubly heterozygous for slit and an integrin gene, encoding alphaPS1, alphaPS2, alphaPS3, or ßPS1, take ectopic trajectories across the midline of the CNS. Drosophila doubly heterozygous for slit and the genes encoding the integrin ligands Laminin A and Tiggrin reveal similar errors in midline axon guidance. It is proposed that the strength of adhesive signaling from integrins influences the threshold of response by growth cones to repellent axon guidance cues (Stevens, 2002).

Axon fasciculation and guidance were studied in the CNS of embryos mutant for the ß integrin gene myospheroid (mys) and three alpha integrins, alphaPS1 (mew), alphaPS2 (if), and alphaPS3/4 (scb). It is not known whether the scb locus affects alphaPS3 or alphaPS4 or both, because the alphaPS4 locus is separated from alphaPS3 by 259 bp. Variation in penetrance of mutant phenotype was observed with all integrin alleles. The three longitudinal fascicles are intact in loss-of-function mutations in mys, which encodes the only ß integrin expressed in the CNS; however, midline fusions of the most medial tract are seen in some segments. Axon tract structure is least disrupted in mew mutant embryos. No midline guidance errors are seen; however, the longitudinal fascicles appeared to be thinner, with occasional defasciculation. if mutant embryos are similar in phenotype to mew, but also reveal a low frequency of midline crossover errors. Midline fusions as well as transient merging of lateral fascicles are seen in scb mutant embryos (Stevens, 2002).

Axon tract fasciculation appears normal in embryos heterozygous for mutations in one integrin gene and also in embryos heterozygous for mutations in two integrin genes. However, all four integrin mutations reveal a semidominant phenotype when doubly heterozygous with slit. In all instances, the frequency of midline guidance errors is increased over the levels seen in homozygous integrin mutants. Apart from midline axon crossings in one-third of the segments, the LT appeared normal in mys/+;sli/+ embryos. Less than 10% of segments revealed midline guidance errors in mew/+;sli/+ embryos. In contrast, 40% and 58% of segments had midline guidance errors in if/+;sli/+ and scb/sli embryos. Only in scb/sli double heterozygotes are the middle and most lateral axon tracts affected, indicating a strong phenotypic interaction between scab and slit (Stevens, 2002).

Midline guidance phenotypes are observed in integrin homozygotes that are also haplosufficient for slit. In particular, the frequency of midline crossing is much higher in mew/Y;sli/+ and if/Y;sli/+ mutants and also involves more lateral axon tracts (Stevens, 2002).

Thus, a reduced level of expression of the genes for four integrins (alphaPS1, alphaPS2, alphaPS3/4, and ßPS1) or two integrin ligands (Tiggrin and Laminin) increases the probability that CNS axons make pathfinding errors when slit expression is reduced. Expression of the integrins Tiggrin and Laminin A has been demonstrated in the CNS. Integrin expression is not localized and may be expressed in both glia and neurons. Overexpression of alphaPS3 or Laminin A in motoneurons affects axon guidance. Loss of function of the integrins disrupts axon fasciculation and longitudinal axon fascicle placement in the embryonic nerve cord but does not clearly affect axon guidance. These observations have been extended in this study, with different alleles of the integrins, demonstrating a similar function for alphaPS3/4, and revealing axon fascicle phenotypes for loss of function of integrin ligands Tiggrin and Laminin A. The mutant phenotypes share common elements: mild phenotypes show wavy axon tracts and reduced Fas II labeling between segments, whereas severe phenotypes include defasciculation and fascicle displacement, including midline axon guidance errors. The integrins have different extracellular ligands. Therefore, the integrins contribute similarly to axon tract integrity, independent of the ligand that they bind (Stevens, 2002).

Integrins are concentrated in the growth cones of Drosophila axons, and their ligands are uniformly distributed over pathways of axon extension. Integrins facilitate the growth of axons by providing a link between the ECM and the cytoskeleton of the growth cone. Defasciculation and guidance errors seen in integrin mutants reflect decreased adhesion to the ECM and a lower threshold to errors in guidance. Axon extension is not impaired in Drosophila integrin mutants, although it is possible that a maternal contribution of integrin may mask this requirement (Stevens, 2002).

Part of the adhesive function of integrins is to activate intracellular signals that alter motility and axon outgrowth. During ligand binding, ß integrins may activate Focal Adhesion Kinase and Rho, which stabilize actin structures, permit actin filament growth, and facilitate the formation of focal adhesions. Intracellular signals can also modify adhesiveness by altering the affinity of integrins for their ligands. These signals can combine to cluster integrins and strengthen their attachment to the ECM. This increased adhesiveness can act in opposition to factors that decrease adhesion and axon extension, such as myelin or aggrecan. Furthermore, integrin signaling can influence other adhesion systems active in the growth cone (Stevens, 2002).

Similarly, integrin function alters the sensitivity of growth cones to repellent signaling by Slit. When slit expression and integrin function are both reduced, growth cones are more likely to respond to attractive guidance from the midline. Signals from axon guidance receptors promote growth cone reorientation and remodeling of the growth cone cytoskeleton. Integrin adhesion promotes local stabilization of cytoskeletal links to the ECM, which antagonizes growth cone reorientation. The threshold of response of growth cones to axon guidance signals is therefore regulated by the ability of guidance signals to reduce the stability of ECM to cytoskeletal linkages. This threshold may be reached at axon guidance choice points, including the segment boundary and the commissures, where clustering of errors in axon guidance occur. The efficacy of guidance signals may be reduced between choice points by local increases in ECM affinity (Stevens, 2002).

Integrin-ligand affinity and integrin-cytoskeletal linkages are logical targets of axon guidance signals. Further studies of the targets of integrin and guidance signals should reveal how growth cones integrate information from the ECM (Stevens, 2002).

Effects of Mutation: Integrins and oogenesis

Regulation of actin structures is instrumental in maintaining proper cytoarchitecture in many tissues. In the follicular epithelium of Drosophila ovaries, a system of actin filaments is coordinated across the basal surface of cells encircling the oocyte. These filaments have been postulated to regulate oocyte elongation; however, the molecular components that control this cytoskeletal array are not yet understood. The receptor tyrosine phosphatase (RPTP) Dlar and integrins are involved in organizing basal actin filaments in follicle cells. Mutations in Dlar and the common ß-integrin subunit mys cause a failure in oocyte elongation, which is correlated with a loss of proper actin filament organization. Immunolocalization shows that early in oogenesis Dlar is polarized to membranes where filaments terminate but becomes generally distributed late in development, at which time ß-integrin and Enabled specifically associate with actin filament terminals. Rescue experiments point to the early period of polar Dlar localization as critical for its function. Furthermore, clonal analysis shows that loss of Dlar or mys influences actin filament polarity in wild-type cells that surround mutant tissues, suggesting that communication between neighboring cells regulates cytoskeletal organization. Two integrin alpha subunits encoded by mew and if are required for proper oocyte elongation, implying that multiple components of the ECM are instructive in coordinating actin fiber polarity. It is concluded that Dlar cooperates with integrins to coordinate actin filaments at the basal surface of the follicular epithelium. This is the first direct demonstration of an RPTP's influence on the actin cytoskeleton (Bateman, 2001).

Wild-type oocytes stained with anti-ß-integrin show intense labeling at the basal surface throughout development, consistent with a strong association with basal actin structures and the underlying laminin-rich ECM. Higher magnification views of stage 7-8 follicles show that ß-integrin staining is somewhat diffuse at the basal surface, with significant staining over cell-cell contacts and where actin filaments terminate. Often, integrin staining is observed along an individual actin filament, reflecting a close association with the actin cytoskeleton. A similar pattern of staining is observed for the ECM component laminin, a ligand for PS1 integrin heterodimers. As with Dlar, integrin staining is intense at cell-cell contacts where 3 cells meet at this early stage. However, unlike the pattern observed for Dlar, ß-integrin remains localized to the terminals of actin bundles until late stages of oogenesis, primarily highlighting cell membranes parallel to the A-P axis. This continued association of ß-integrin with actin bundles throughout development is consistent with a role in actin filament formation and maintenance, as observed between stress fibers and focal adhesions (FAs) in cultured cells (Bateman, 2001).

The coupling of integrins to the cytoskeleton is mediated by a group of intracellular proteins that bind the cytoplasmic tails of receptors and interact with actin filaments. For example, members of the Vasodilator-stimulated phosphoprotein (VASP) family localize to FAs, where they are thought to regulate actin assembly. To explore the relationship between stress fibers and basal actin filaments of follicle cells, oocytes were stained with antibodies to the Drosophila VASP homolog Enabled (Ena). At early stages (stage 7-8), Ena is relatively diffuse throughout the cytoplasm, with little obvious concentration at cell membranes. However, at stages 10-12, significant Ena staining at actin filament terminals is observed, coinciding with the period of specific ß-integrin association with filaments. Staining was also observed along actin filaments that traverse the cell, consistent with the VASP family's proposed role in actin regulation. The association of actin bundles with multiple FA markers supports the hypothesis that the basal cytoskeleton of Drosophila follicle cells is analogous to stress fibers observed in vitro (Bateman, 2001).

Due to the association between actin filaments and ß-integrin in late-stage oocytes, it was asked if integrin staining was affected in cells within and surrounding Dlar mutant clones. Indeed, cells lacking Dlar show improper localization of integrin associated with a loss of F-actin polarity. ß-integrin staining remains highest at the filament terminals of mutant cells regardless of their orientation, resulting in a loss of staining at membranes parallel to the A-P axis. In some cases, ß-integrin staining shows no restriction to opposite sides of a cell as it does in the wild-type but instead appears generally distributed around the cell border (Bateman, 2001).

Because strong defects in ß-integrin localization are only observed in cells with actin defects, it is unclear whether these errors reflect a direct effect of Dlar on integrins or whether integrin localization is dependent upon prior orientation of F-actin by Dlar. To address this issue, it was asked whether integrins determine the polarity of basal actin filaments, which would suggest an upstream instructive role rather than a downstream dependence. Strong integrin mutants do not survive to adulthood, therefore, clones of mutant tissue were created. A screen for genes required in follicular development identified a mutation (968) with a round-egg phenotype indistinguishable from Dlar mutants; mapping data placed the insertion near the myospheroid (mys) locus encoding ß-integrin. The 968 mutant was shown to be allelic to mys by its failure to complement the allele mysnj42 and by loss of ß-integrin staining within 968 clones. As observed with Dlar, Texas Red-phalloidin staining of oocytes carrying mys968 clones shows disruptions in basal actin fiber polarity within mutant cells and in wild-type cells beyond the clonal boundary. Follicle cell clones of the independent allele mys10 also result in round eggs, confirming that integrins are required in follicle cells for elongation of the oocyte (Bateman, 2001).

Drosophila integrins function as heterodimers of a common ß subunit (mys) with a series of alpha subunits, creating functional receptors with different binding properties. For example, the alpha integrin encoded by multiple edematous wings (mew) is a receptor for laminin when combined with ß-integrin, while the alpha-integrin encoded by inflated (if) confers specificity to ECM components carrying the amino acid motif RGD. To determine if these ligands are relevant to integrin function in follicle cells, clones of if and mew were created. Rounded eggs were observed in oocytes with clones of either subunit, implying that both laminin and RGD components of the ECM are involved in mediating follicle cell control of oocyte shape (Bateman, 2001).

The similarity in phenotypes caused by loss of either Dlar or integrins suggests a cooperative relationship in organizing F-actin. To explore this, whether a genetic relationship exists between Dlar and mys was investigated. It was reasoned that the mild penetrance of round eggs in Dlar mutants may be sensitive to the dosage of genes in the same pathway. Indeed, removal of half of the gene dosage of mys in a Dlar null background causes a substantial increase in the penetrance of the round egg phenotype. For example, Dlar5.5/Dlar13.2 females show defective rounding in 14.1% of their stage 14 oocytes, while in mys1/+;Dlar5.5/Dlar13.2 females, the penetrance is increased nearly 4-fold to 48.7%. This enhancement is not due to changes in the overall fitness of mutant flies since Dlar mutants carrying the balancer TM6B, which decreases the viability of Dlar escapers, show little change in mutant oocyte penetrance (22.0% round stage 14 oocytes). Thus, genetic interactions support a model wherein Dlar and integrins cooperate in organizing basal actin filaments (Bateman, 2001).

Effects of Mutation: Integrins and trachea

Laminin is a common ligand for integrins in vertebrates. Since Scab mRNA is expressed in parts of the trachea throughout embryogenesis, antibodies to examine the trachea in wild-type, mys-, scb X5, scb X6, scb 2 and both scb deficiencies. To explore further the potential of laminin as a ligand, lamA 9-32 mutant embryos were also examined for tracheal defects. Embryos from each of the mutant stocks have significant gaps in the dorsal trunk of the trachea, which are not present in wild-type embryos. Due to the strong expression of alphaPS3 mRNA in the salivary glands throughout development, the glands of wild-type, mys - , scb X5 , scb X6 and scb 2 embryos were examined using a reporter gene. mys and scb embryos frequently show one gland to be misshapen and smaller than the other. The salivary gland is sometimes shifted closer to the midline. In conclusion, several defects are seen in scb mutant embryos, all of which are shared with mys and some with lamA; among them are defects previously reported in mys, but not in multiple edematous wings (aPS1) or inflated (aPS2) mutations. Noteworthy is the fact that these defects arise in areas where Scab is strongly expressed (Stark, 1997).

Migration of the Drosophila tracheal cells relies on cues provided by nearby cells; however, little is known about how these signals specify a migratory path. The role of cell surface proteins in the definition of such a pathway has been examined. The PS1 integrin is required in the tracheal cells of the visceral branch, whereas the PS2 integrin is required in the visceral mesoderm; both integrins are necessary for the spreading of the visceral branch over its substratum. This is the first identification of a cell surface molecule with expression restricted to a subset of tracheal cells that all migrate in a given direction. Expression of PS1 in the visceral branch is regulated by the genes that direct tracheal cell migration, showing that integrin expression is part of the cell-fate program that they specify. These results support a model in which signal transduction determines the tracheal migratory pathways by regulating the expression of cell surface proteins, which in turn interact with surface molecules on the surrounding cell population (Boube, 2001).

Tracheal cells migrate over the visceral mesoderm to form the visceral branches, the ramifications of the tracheal tree that transport oxygen to the gut. They arise as small bulges in some of the centrally located pairs of tracheal pits (TP). In particular, visceral branches arise from six pairs of tracheal pits, TP2 and TP4 to TP8. A distinctive feature of these branches is their direction of migration; whereas formation of the other primary branches is guided by branchless (bnl) expression in small clusters of epidermal cells, the outgrowth of the visceral branch is directed toward clusters of bnl-expressing cells in the mesoderm. At stage 11, the tracheal pits are located just over the clusters of the fat body precursor cells that are organized in metameric primordia. It is between these discrete regions of fat body precursors that the cells of the visceral branch contact the cells of the visceral mesoderm. Next, the cells of most visceral branches migrate anteriorly along the visceral mesoderm, with the exception of some cells of the branch derived from TP8, which migrate posteriorly. As development of the gut progresses, the visceral branches remain in contact with the visceral mesoderm and follow the convolutions that occur during formation of gut constrictions. Subsequently, fine branches form at positions in which visceral branches contact the gut (Boube, 2001).

Only the tracheal cells that will form the visceral branch establish a distinct association with the visceral mesoderm. A mechanism to explain such a specific association could involve particular cell surface proteins being expressed in a complementary way in both the visceral mesoderm substrate and the migrating tracheal cells. A similar scenario exists in which the visceral mesoderm serves as a substrate for the migrating primordial midgut cells. In this case, the migration requires a combination of PS1 and PS3 integrins in the migrating cells, and the PS2 integrin in the visceral mesoderm substratum. Integrins are a class of cell surface molecules formed as heterodimers of two single-pass type I transmembrane proteins, an alpha and a ß subunit. The Drosophila position-specific (PS) integrins are most similar to the ß1 family of vertebrate integrins, and one ßPS subunit is likely to form heterodimers with all five of the alphaPS subunits encoded by the Drosophila genome. As with the majority of integrins, the PS integrins are thought to function by binding ligands within the extracellular matrix. To see whether integrins might be playing a role in tracheal migration, their expression in the tracheal cells was examined. Around stage 11, the alphaPS1 mRNA is specifically expressed in those tracheal cells that give rise to the visceral branch, although weak expression is detected in the dorsal cells. Localized expression in the visceral branch is more clearly evident at stage 12, during germ-band retraction, when the main branches appear as protrusions. alphaPS1 mRNA expression is maintained during stage 13 as the visceral branch grows, and is lost by stage 14. Whether there is expression of the alphaPS2 subunit in the tracheal cells was examined, and no expression was detected (Boube, 2001).

The specific expression of the alphaPS1 subunit prompted an analysis of its role in the development of the visceral branch. The alphaPS1 subunit is encoded by the multiple edematous wings (mew) gene. In mew mutant embryos the visceral branches fail to develop properly. Although the visceral branches migrate normally toward the gut and reach and contact the visceral mesoderm, they do not migrate normally along the visceral mesoderm. At a stage when the cells of the visceral branches have migrated along the visceral mesoderm in wild-type embryos, in mew mutant embryos, the cells of the visceral branches remain at their initial points of contact with the visceral mesoderm. Some delayed migration can also be observed at later stages. Other aspects of visceral branch development are not affected in mew mutant embryos; for instance, the mispositioned visceral branches form fine branches later in development (Boube, 2001).

Since the PS2 integrin is expressed in the visceral mesoderm (the substrate of visceral branch migration), it was asked whether PS2 integrin is also required for visceral branch formation. The alphaPS2 subunit is encoded by the inflated (if) gene and in if mutant embryos, visceral branch formation is also impaired. As is the case for mew mutant embryos, the visceral branches of if mutant embryos reach the visceral mesoderm, but fail to migrate over it. However, the defects are usually milder than in mew mutant embryos. Later branching is not affected. The defect in visceral branch formation in if mutant embryos is not due to a failure of bnl expression, as it appears at the right position in the visceral mesoderm (Boube, 2001).

Because both alphaPS1 and alphaPS2 subunits form heterodimers with the ßPS subunit, mutations in the myospheroid (mys) gene encoding this subunit will also simultaneously eliminate the PS1 and PS2 integrins. The ßPS subunit is maternally deposited in the embryo, and, therefore, to examine the effect of complete loss of the ßPS subunit, embryos were generated lacking both maternal and zygotic mys expression by making germ-line clones (referred to as mys mutant embryos). All mys mutant embryos exhibit a failure in the migration of the cells of the visceral branches along the visceral mesoderm similar to those observed in the alphaPS mutants. However, in mys mutant embryos, some visceral branches detach from the visceral mesoderm, suggestive of a more extensive failure in the establishment of cell-cell interactions than in mutants for individual alphaPS subunits. Some mys- embryos also show gaps in the dorsal trunk, the more prominent branch of the tracheal tree, and defects in the dorsal branches. Thus, the loss of all PS integrins produces stronger migration defects than the loss of alphaPS1 or alphaPS2 individually (Boube, 2001).

The above results suggest that PS1 (alphaPS1ßPS) integrin is required in the migrating cells of the visceral branches, and PS2 (alphaPS2ßPS) integrin, in the visceral mesoderm substrate. If this is the case, it should be possible to rescue the visceral branch phenotype of mutant embryos for each alpha subunit by expressing the appropriate subunit specifically in the cell population in which it is required. Expression of alphaPS1 just in the tracheal cells can rescue the visceral branch phenotype of the mew (alphaPS1-) mutant embryos, although it causes some defects in the other branches. Similarly, the visceral branch phenotype of if (alphaPS2-) mutant embryos can be rescued by targeted expression of alphaPS2 just in the mesodermal cells. These results rule out the possibility that the defects in the visceral branches observed in integrin mutants are due to secondary consequences of disruption of other tissues. Furthermore, they also show that migration of the visceral branches requires the coordinated functions of PS1 integrin in the visceral branch and PS2 integrin in the visceral mesoderm (Boube, 2001).

Whether the complementary expression of these two integrins is essential for tracheal migration was tested. The specific requirement of PS1 in the tracheal cells of the visceral branch was analyzed by assessing whether the PS2 integrin could fulfill the role of PS1. Expression of alphaPS2 in the tracheal cells does not rescue the defects in visceral branch migration caused by the lack of alphaPS1 in mew mutant embryos. This result indicates that PS1 function in particular, rather than PS integrin function in general, is required in the cells of the visceral branch for them to migrate along the visceral mesoderm. In contrast, expression of alphaPS1 in place of alphaPS2 in the mesoderm does substantially rescue the visceral branch phenotype caused by lack of alphaPS2. This finding suggests that either integrin can assemble a suitable substrate for PS1-dependent tracheal migration (Boube, 2001).

Previous clonal analysis has shown that cell lineage does not play an important role in the specification of the distinct identity of the different tracheal cells. Instead, different signaling pathways act to spatially subdivide the tracheal placode by inducing groups of cells to migrate in a particular direction and acquire a specific morphology. This inductive process appears to be mediated by the differential activation of a set of transcription factors in discrete groups of tracheal cells. Therefore, whether these transcription factors regulate the specific expression of the alphaPS1 subunit was examined (Boube, 2001).

The transcription factors knirps (kni) and knirps-related (knrl) are expressed in overlapping patterns during tracheal development in which they share redundant functions. They are both required in two phases of tracheal development. The initial tracheal expression of kni/knrl appears around stage 10 and their activity is required early in the tracheal placode for primary branching outgrowth. Subsequently, kni/knrl expression becomes restricted to some branches, among them the visceral branches, and they are required for the directed outgrowth of these tracheal branches. Since kni and knrl share redundant functions, an examination was made of mutant embryos homozygous for a deficiency that uncovers both genes and whose effect on tracheal development is due solely to the lack of kni and knrl. In this mutant background, it was found that alphaPS1 is no longer expressed in the tracheal cells, indicating that these genes act as positive regulators of alphaPS1. To further assess the specific role of the late expression of kni/knrl in the expression of alphaPS1 in the visceral branch, a mutant combination was examined that provides the segmentation and early tracheal kni expression, but not the later branch-specific expression of kni. In these mutant embryos, there are also defects in visceral branch development and expression of alphaPS1 is either absent or very much reduced. Thus, the late phase of kni/knrl branch-specific expression is specifically required for the proper expression of alphaPS1 in the cells of the visceral branches. However, kni/knrl are not sufficient to drive alphaPS1 expression, as they are expressed in other branches that do not express alphaPS1. Similarly, the alphaPS1 gene is not the only target of these transcription factors, since GAL4-driven expression of alphaPS1 is not able to rescue the phenotype of the deficiency (Boube, 2001).

Another transcription factor that has an important role in specifying cell fate within the tracheal branches is encoded by spalt (sal). The expression of sal is restricted to the dorsal cells in the developing tracheal tree. By stage 12, sal is expressed only in the dorsal parts of the tracheal metameres, that is, in the cells of the dorsal trunk and the dorsal branches. By stages 14 and 15, sal expression declines in the cells of the dorsal branches, but remains high in the cells of the dorsal trunk. In sal mutant embryos there is ectopic expression of alphaPS1 in the cells of the dorsal trunk and in the dorsal branches, indicating that sal activity is required to inhibit alphaPS1 expression in the dorsal tracheal structures. In summary, these results indicate that alphaPS1 expression is restricted to the cells of the visceral branches by the direct or indirect activities of the transcription factors that spatially subdivide the tracheal placode. The fact that alphaPS1 expression is regulated by the same genes that specify a particular cell fate, rather than independently, suggests that integrin expression is an important part of the cell fate decision (Boube, 2001).

To investigate the functional significance of the restricted expression of PS1, whether it is important that the other tracheal branches do not express the PS1 integrin was examined. The expression of alphaPS1 construct was induced in all of the tracheal cells and several defects were found in tracheal development, mainly in the dorsal branches. In wild-type embryos, each dorsal branch grows toward the dorsal midline, where it meets and fuses with the dorsal branch coming from the contralateral hemi segment. However, in embryos with ubiquitous tracheal alphaPS1 expression, many dorsal branches fail to fuse with their contralateral counterpart or fuse with an inappropriate branch. Occasional interruptions are seen in the dorsal trunk of these embryos. However, no redirection of migration is observed in this situation. These results suggest that different groups of tracheal cells need to express different distinct adhesion molecules to migrate on the correct substrate. If this is the case, one would expect changes in the migratory behavior of these cells to be accompanied by changes in the expression of the adhesion molecules. A change in the path of migration of some tracheal cells has been observed in particular mutant combinations of the wingless (wg) pathway. The wg pathway is required for the formation of the dorsal trunk, and upon ubiquitous tracheal activation of the wg pathway, cells that would normally contribute to the visceral branch appear to adopt a dorsal trunk fate and change their path of migration accordingly. As a result, the dorsal trunk becomes larger and fewer cells migrate along the visceral mesoderm. In these mutant embryos, fewer cells express alphaPS1. Thus, the signaling pathways that mediate the choice between alternative paths of migration regulate the appropriate expression of adhesion molecules in subsets of the tracheal cells (Boube, 2001).

It has been shown that the alternative migratory pathways of the tracheal cells are associated with distinct subsets of mesodermal cells. This study provides the first identification of cell surface receptors that mediate these specific interactions. These results support a model in which signaling by transduction pathways specifies the particular migratory pathways of tracheal cells by regulating a precise array of adhesion proteins such as integrins at their surface. This can be a very general mechanism to regulate integrin expression in distinct subpopulations of cells within a wider field. For instance, many genes operating in the morphogenesis of the insect tracheal system and in the vertebrate lung are functionally conserved despite the important differences that exist between the two systems. In this scenario, it is worth noting that distinct integrin subunits are differentially expressed in the cells involved in mesenchymal/epithelial interactions during lung development. Thus, as in the case of the Drosophila tracheal system, control of cell surface molecules by signaling pathways in other systems may be involved in the mechanism conferring the specificity of cell-cell adhesions in migration and in other biological processes (Boube, 2001).

Epithelial tubes that compose many organs are typically long lasting, except under specific developmental and physiological conditions when network remodeling occurs. Although there has been progress elucidating mechanisms of tube formation, little is known of the mechanisms that maintain tubes and destabilize them during network remodeling. This study describes Drosophila tendrils mutations that compromise maintenance of tracheal terminal branches, fine gauge tubes formed by tracheal terminal cells that ramify on and adhere tightly to tissues in order to supply them with oxygen. Homozygous tendrils terminal cell clones have fewer terminal branches than normal but individual branches contain multiple convoluted lumens. The phenotype arises late in development: terminal branches bud and form lumens normally early in development, but during larval life lumens become convoluted and mature branches degenerate. Their lumens, however, are retained in the remaining branches, resulting in the distinctive multi-lumen phenotype. Mapping and molecular studies demonstrate that tendrils is allelic to rhea, which encodes Drosophila talin, a large cytoskeletal protein that links integrins to the cytoskeleton. Terminal cells mutant for myospheroid, the major Drosophila ß-integrin, or doubly mutant for multiple edematous wings and inflated α-integrins, also show the tendrils phenotype, and localization of myospheroid ß-integrin protein is disrupted in tendrils mutant terminal cells. The results provide evidence that integrin-talin adhesion complexes are necessary to maintain tracheal terminal branches and luminal organization. Similar complexes may stabilize other tubular networks and may be targeted for inactivation during network remodeling events (Levi, 2006; full text of article).

Integrins and wing development

The defects observed in the gynandromorphs of myospheroid demonstrate widespread requirements for PS integrins during development especially in ventrally derived structures, which also show strong expression of betaPS integrin. Smaller myospheroid clones induced during larval development result in blister and vein defects in the wings and aberrant development of photoreceptor cells, demonstrating roles for PS integrins during development of both wings and eyes. PS integrins are required for the close apposition of the dorsal and ventral wing epithelia and for the proper arrangement of photoreceptor cells. However, many other differentiation processes proceed normally in the absence of integrins containing Myospheroid (Zusman, 1990).

blistered, coding for Drosophila serum response factor is required during metamorphosis for the initiation of intervein development and the concomitant inhibition of vein development. Integrin mutants interact with blistered mutants to increase the frequency of intervein blisters but do not typically enhance vein defects (Fristrom 1994)

Morphogenesis of the Drosophila wing depends on a series of cell-cell and cell-extracellular matrix interactions. During pupal wing development, two secreted proteins, encoded by the short gastrulation (sog) and decapentaplegic (dpp) genes, vie to position wing veins in the center of broad provein territories. Expression of the Bmp4 homolog dpp in vein cells is counteracted by expression of the secreted Bmp antagonist sog in intervein cells, which results in the formation of straight veins of precise width. A screen was performed for genetic interactions between sog and genes encoding a variety of extracellular components and interactions were uncovered between sog and myospheroid (mys), multiple edematous wing (mew) and scab (scb), which encode ßPS, alphaPS1 and alphaPS3 integrin subunits, respectively. Clonal analysis reveals that integrin mutations affect the trajectory of veins inside the provein domain and/or their width and that misexpression of sog can alter the behavior of cells in such clones. In addition, a low molecular weight form of Sog protein has been shown to bind to alphaPS1ßPS. Sog can diffuse from its intervein site of production into adjacent provein domains, but only on the dorsal surface of the wing, where Sog interacts functionally with integrins. Finally, it has been shown that Sog diffusion into pro-vein regions and the reticular pattern of extracellular Sog distribution in wild-type wings requires mys and mew function. It is proposed that integrins act by binding and possibly regulating the activity/availability of different forms of Sog during pupal development through an adhesion independent mechanism (Araujo, 2003).

Because diffusion of Sog into provein domains is restricted to the dorsal surface of the wing where integrins interact with Sog, it was asked whether integrins play a role in regulating the distribution of Sog protein on the dorsal surface of pupal wings. Marked mys minus or mew minus clones were generated in an otherwise wild-type background and Sog staining was examined. In control wings, double-labeling with anti-ß integrin and anti-Sog antisera confirmed that the dorsally restricted pattern of reticular Sog staining extends beyond ß-integrin staining into provein domains. By contrast, Sog staining has a patchy intracellular appearance in dorsal mys minus clones, and is excluded from wild-type provein cells on the dorsal surface that are adjacent to mys minus clones. In such cases where mys minus clones are located on one side of a provein domain, Sog is still able to enter the provein region from the opposite mys+ side of the same vein. These results demonstrate that mys is required for diffusion or transport of Sog into the vein competent domain. Consistent with the observation that only dorsally located integrin minus clones can alter the course of veins, it was found that only dorsal mys minus clones modify Sog distribution in the pupal wing. Similarly altered Sog staining was observed within mew minus clones on the dorsal wing surface resulting in punctate rather than reticular staining and lack of Sog diffusion into the provein region. These results demonstrate that the ßPS and alphaPS1 integrins play an important role in determining the distribution of Sog protein in the pupal wing (Araujo, 2003).

In this study, three primary lines of evidence are provided that integrins play an important role in regulating Bmp signaling in provein regions of the pupal wing: (1) integrin minus clones generated on the dorsal surface of the wing alter the trajectory and/or width of adjacent veins; (2) a truncated form of Sog present in pupal wings binds to alphaPS1; (3) diffusion of Sog into provein domains, which is restricted to the dorsal surface of the wing, depends on integrin function. Cumulatively, these results strongly suggest that the ability of Sog to diffuse or to be transported into provein regions on the dorsal surface depends on an interaction with integrins (Araujo, 2003).

Consistent with Sog interacting genetically with integrins to alter the course of veins on the dorsal surface of the wing, it was found that the alphaPS1 and ßPS-integrins are required for the diffusion or transport of Sog from dorsal intervein cells where sog mRNA is expressed into adjacent provein regions. Since alphaPS1 binds Sog, this physical interaction may contribute to regulating the distribution of Sog. The 8B anti-Sog antiserum used in this study, that recognizes Sog protein in intervein cells and inside the provein domain, detects an epitope located near the second cystein repeat (CR2). Consequently, Sog fragments that diffuse or that are delivered to provein cells must be either full length, which weakly binds to alphaPS1 in co-immunoprecipitation experiments, or fragments that contain CR2. The truncated Supersog-like fragment that binds strongly to alphaPS1 in coimmunoprecipitation experiments should not be recognized by the 8B antiserum. Therefore, integrins may differentially regulate the distribution of Sog fragments on the dorsal surface of the pupal wing, restraining the movement of broad spectrum Bmp inhibitory Sog fragments (such as Supersog-like molecules) and allowing or mediating transport of other fragments to provein cells, such as full-length Sog, which also has a vein inhibitory function. Unfortunately, it is not possible currently to examine the diffusion of Supersog-like fragments directly because the 8A antiserum is not suitable for staining pupal wings. These findings suggest that integrins regulate the delivery or diffusion of active Sog protein from intervein cells into the vein competent domain. In contrast to the dorsally restricted functions of integrins required for vein development, the adhesive functions of integrins depends on subunits functioning on both surfaces of the wing (Araujo, 2003).

There are several possible mechanisms by which interaction with integrins could modulate Sog activity in pupal wings. Elevated sog expression results in vein truncation, while misexpression of dpp induces ectopic veins, indicating that sog restricts vein formation by opposing Bmp signals emanating from the center of the vein (Yu, 1996). One possibility is that such a Bmp inhibitory form(s) of Sog must interact with integrins in order to diffuse or be transported into provein domains (on the dorsal surface of the wing only). This hypothesis would be consistent with the finding that veins appear to be attracted to integrin minus clones. Such a vein repulsive form(s) of Sog would presumably act as a Bmp antagonist (Araujo, 2003).

According to the simple model in which integrins are essential for delivering a Bmp inhibitory form of Sog to provein cells, one would expect that integrin minus and sog minus clones would generate similar phenotypes in which veins deviated towards the mutant clones and/or broadened within them. However, sog minus clones induce meandering of veins (Yu, 1996) -- veins show only a weak tendency to track along the outside of sog minus clones, in contrast to integrin minus clones, which bend or widen veins in a more dramatic fashion. One possible explanation for the differences between the sog minus and integrin minus phenotypes is that there are several different endogenous forms of Sog in pupal wings, which might exert opposing activities. If multiple Sog fragments exert effects on vein development, some providing repulsive and others attractive activities on vein formation, the differences between the behaviors of sog- and integrin- clones could be explained by a repulsive (Bmp inhibitory) form(s) of Sog selectively requiring an interaction with integrins. The possibility that a positive Bmp-promoting activity of Sog might also be present that acts as a vein attractant has precedent in that a positive Sog activity has been proposed to explain a requirement for Sog in activating expression of the Dpp target gene race in early embryos. Structure/function studies of Sog have also revealed a potential Bmp promoting form of Sog, which is longer than Supersog forms. According to this model, altering the balance between repulsive and attractive Sog activities would generate different vein phenotypes. In the total absence of sog, both repulsive and attractive activities would be lost, generating a mild meandering vein phenotype in which neither attraction nor repulsion clearly dominates, as is observed in sog minus clones (Yu, 1996). If an interaction with integrins were required only for production or delivery of Bmp inhibitory forms of Sog into the vein, then integrin minus clones, which still contain the Bmp-activating forms of Sog, could exert a net attractive influence on veins, leading to more pronounced deviation of veins toward the clones. This hypothesis is consistent with vein phenotypes observed associated with integrin minus clones that cross over veins or run along both sides of the vein, such as narrowing, bending and wandering of veins; these phenotypes are similar to those observed in correspondingly located sog minus clones. The existence of different Sog fragments bearing opposing activities would also explain the different phenotypes obtained upon ectopic Sog expression in some sogEP lines (Yu, 1996), such as sporadic ectopic vein material between L3 and L2 and meandering L2 veins in addition to vein loss in other areas (Araujo, 2003).

Another possible explanation for the differences between the sog minus and integrin minus phenotypes is that integrins may regulate the activity of extracellular signals in addition to Sog. One hint of such an activity is that when a scb minus clone falls within the provein area, the vein splits around the border of the clone in a cell autonomous fashion. Since this later phenotype is enhanced by ectopic sog expression in veins (e.g., in a sogEP background), alphaPS3 may normally promote Bmp signaling within the vein. Although the identity of such potential targets is unknown, candidates would include Bmps (e.g. Dpp or Gbb) or Bmp receptors. Further analysis will be needed to explain the basis for the different behaviors of sog minus and integrin minus clones, as well as the variations observed in different integrin minus clones (Araujo, 2003).

In summary, it is proposed that Sog fragments with differential activities may regulate vein formation. The vein bending phenotype observed in the absence of alphaPS1 would result from a remaining attractive Sog activity that outweighs the activity of a repulsive form of Sog, which can no longer be delivered from intervein cells. Since ßPS integrin forms heterodimers with both alphaPS1 and alphaPS3 mys would be expected to be required for the activity of both alphaPS chains. Consistent with this expectation, the phenotype of mys minus clones (i.e., broad poorly defined veins) resembles a hybrid of those observed for mew minus and scb minus clones (Araujo, 2003).

Endocytosis has been shown to play an important role in the establishment of Bmp activity gradients. The endocytic pathway has been implicated in transport of Dpp between cells by transcytosis during larval wing development. During early embryonic development, formation of a Sog protein gradient in dorsal regions also relies on the action of Dynamin, although in this preblastoderm context, it has been proposed that endocytosis limits the dorsal diffusion of Sog, which is essential for the partitioning of the dorsal ectoderm into epidermis and amnioserosa. Vertebrate alpha3ß Integrin (E. deRobertis, personal communication to Araujo, 2003) has been shown to bind to the Xenopus Sog counterpart Chordin in vitro, leading to endocytosis of Chordin (Araujo, 2003).

Although the possible regulation of Sog endocytosis by integrins is not addressed in this current study, the altered distribution of Sog within integrin minus clones is suggestive of such a role. Reticular Sog staining, which outlines the cell perimeter is lost in integrin minus clones on the dorsal surface, leaving only a punctate intracellular staining. This mis-localization of Sog implicates integrins in internalizing and/or trafficking of Sog to the cell surface. Because appropriately located integrin minus clones also block the accumulation of Sog in adjacent pro-vein domains, the observed defects in Sog distribution between the surface and the cytoplasm may underlie the failure to deliver Sog to vein competent cells. The endocytic pathway could promote the transport of Sog to pro-vein cells by a mechanism similar to that proposed to be involved in the transport of Dpp along the AP axis during larval stages. Alternatively, endocytosis could function to limit Sog diffusion as is the case during embryogenesis. According to this latter scenario, integrins would normally prevent or reduce Sog endocytosis because integrins are necessary for delivery of Sog to pro-vein cells. Integrins have been shown to play a direct role in endocytosis of viral particles and in mediating membrane traffic through the endocytic cycle. Indirect mechanisms for integrin-mediated endocytosis may also exist that would not involve endocytosis of the integrin receptor itself, but of other components that regulate Sog trafficking. Further analysis will be necessary to investigate whether Drosophila integrins regulate delivery of Sog to endocytic vesicles or transport of Sog through the endocytic pathway to adjacent cells (Araujo, 2003).

The modulatory effect of integrins on Sog activity described in this paper are likely to be mediated by dpp and/or gbb signaling because existing evidence indicates that Sog is a dedicated modulator of Bmp signaling. In addition, the phenocritical period for mys and sog interaction coincides with that for interaction between sog and dpp (Yu, 1996). The existence cannot be excluded of an additional role of integrins in regulating vein formation through another pathway, such as the Egf and Notch pathways, which have been shown to exert important roles on vein development. However, the integrin minus clonal phenotypes described in this manuscript are observed only on the dorsal surface and all known components of the Egfr pathway promote vein development on both surfaces of the wing (Araujo, 2003).

mysnj42 and scb1 were found suppress the thickened vein phenotype of tkv1 mutants, which raises the possibility of a direct interaction between integrins and a Bmp receptor involved in wing vein development. The vein splitting and vein thickening scb minus clonal phenotypes are reminiscent of tkv mutant phenotypes, which derive from a positive requirement for Bmp signaling for vein formation inside the vein competent domain and a negative ligand titrating function that limits the range of Bmp diffusion into the intervein territory adjacent to the provein domain. The fact that scb is expressed in both vein and intervein territories is consistent with a dual action of scb. Additional experiments will be necessary to investigate whether scb plays a direct role in modulating Bmp receptor activity (Araujo, 2003).

Diffusion of putative growth factors and the shaping of their activity gradients have been the focus of intense interest since Alan Turing formulated the concept of morphogens (Turing, 1952: see 'The chemical basis of morphogenesis'. Philos. Trans. R. Soc. Lond. B Biol. Soc. 237: 37-72). Recently, several groups have described mechanisms to explain how soluble factors can create morphogen gradients. These include: (1) degradative proteolysis and a retrieval role for endocytosis in creating the early embryonic Sog gradient; (2) regulated endocytosis of wingless extracellular transport of Wg in membrane bound argosomes; (3) planar transcytosis (as is required for Dpp movement in the wing imaginal disc), and (4) the formation of thin cell extensions (cytonemes) that deliver Dpp over several rows of cells. Protein-protein interactions in the extracellular milieu, such as those described here, may also be capable of modulating the magnitude and spatial pattern of Bmp activity, working independently or in conjunction with other mechanisms (Araujo, 2003).

Integrin-dependent apposition of Drosophila extraembryonic membranes promotes morphogenesis and prevents anoikis

Two extraembryonic tissues form early in Drosophila development. One, the amnioserosa, has been implicated in the morphogenetic processes of germ band retraction and dorsal closure. The developmental role of the other, the yolk sac, is obscure. By using live-imaging techniques, intimate interactions are reported between the amnioserosa and the yolk sac during germ band retraction and dorsal closure. These tissue interactions fail in a subset of myospheroid (mys: ßPS integrin) mutant embryos, leading to failure of germ band retraction and dorsal closure. The Drosophila homolog of mammalian basigin (EMMPRIN , CD147) -- an integrin-associated transmembrane glycoprotein -- is highly enriched in the extraembryonic tissues. Strong dominant genetic interactions between basigin and mys mutations cause severe defects in dorsal closure, consistent with basigin functioning together with ßPS integrin in extraembryonic membrane apposition. During normal development, JNK signaling is upregulated in the amnioserosa, as midgut closure disrupts contact with the yolk sac. Subsequently, the amnioserosal epithelium degenerates in a process that is independent of the reaper, hid, and grim cell death genes. In mys mutants that fail to establish contact between the extraembryonic membranes, the amnioserosa undergoes premature disintegration and death. It is concluded that intimate apposition of the amnioserosa and yolk sac prevents anoikis of the amnioserosa. Survival of the amnioserosa is essential for germ band retraction and dorsal closure. It is hypothesized that during normal development, loss of integrin-dependent contact between the extraembryonic tissues results in JNK-dependent amnioserosal disintegration and death, thus representing an example of developmentally programmed anoikis (Reed, 2004).

In Drosophila, the role of extraembryonic tissues in regulating embryonic development has only recently begun to be appreciated . Two cell types that arise at the Drosophila cellular blastoderm stage are extraembryonic (i.e., do not contribute to the mature embryo). The first, the amnioserosa, is an epithelium derived from the dorsalmost region of the blastoderm. The second, the yolk sac, originates during cellularization of the blastoderm: membrane fusion basal to the blastoderm nuclei forms both the basal membrane of each somatic cell and a single continuous plasma membrane -- the yolk sac membrane -- that envelops the yolk. Within the yolk syncytium, there are some 200 nuclei; thus, the yolk sac is a large, membrane bound, multinucleate cell (Reed, 2004).

The amnioserosa plays a key role in germ band retraction and dorsal closure. It is likely to function both in cell signaling and in generating the forces that drive these morphogenetic processes. The role of the yolk sac during development has remained obscure. The expression of several genes in the yolk nuclei, including serpent, sisterlessA, D-ret, forkhead, and those encoding imaginal disc growth factors (IDGFs), suggests that the yolk sac may play important roles in processes other than nutrition. The developmental defects produced by loss-of-function alleles of sisterlessA, which is expressed exclusively in the yolk nuclei from blastoderm stages on, have led to speculation that the yolk may play a role in morphogenesis. However, the functions of the yolk sac in morphogenesis, if any, are unknown (Reed, 2004).

Physical interaction of the amnioserosa and yolk sac has been shown to play a crucial role in both germ band retraction and dorsal closure of the embryo. βPS integrin mediates extraembryonic membrane interactions that are required for survival of the amnioserosa. Anoikis of the amnioserosa occurs during normal development after closure of the midgut disrupts integrin-dependent apposition of the amnioserosa and yolk sac. In mys mutants, failure to establish apposition of extraembryonic membranes leads to premature anoikis of the amnioserosa. A possible role for JNK signaling and the reaper/hid/grim cell death genes in amnioserosal anoikis during normal development was investigated (Reed, 2004).

In fixed, sectioned material it can be seen that as germ band retraction commences, there is a gap between the amnioserosa and the yolk sac membrane. Membrane projections from both the basal side of the amnioserosa and the dorsal region of the yolk sac can be seen to penetrate this space. This space is enriched in glycoconjugates as assayed by ruthenium red staining. Since the bulk of the extracellular matrix is not laid down at this developmental stage, these polysaccharides may be associated with transmembrane glycoproteins rather than an elaborate extracellular matrix (ECM) per se (Reed, 2004).

Live imaging of germ band retraction and dorsal closure has revealed that contacts between the yolk sac membrane and the amnioserosa initiate at the beginning of germ band retraction and are remarkably dynamic. Imaging was carried out by using combinations of three different GFP fusion proteins that serve as markers of the F actin-based cytoskeleton (actin-GFP); the amnioserosal and yolk sac membranes (DE-cadherin-GFP), and G289, a homozygous viable PTT line that reports basigin expression as a basigin-GFP fusion protein (Reed, 2004).

The initial, transient contacts between the amnioserosa and the yolk sac membrane, referred to here as phase I interactions, occur as germ band retraction initiates and are accomplished by two classes of cellular extensions: filopodia that emanate from the amnioserosa and contact the yolk sac membrane, and membrane bound projections emanating from the yolk sac, which contact the amnioserosa (marked by basigin-GFP). Their lack of stable association with their target cells and their highly dynamic character suggest that neither the amnioserosal nor the yolk sac projections generate the mechanical forces that drive morphogenesis. Instead these projections may facilitate a chemosensory or signaling function between the amnioserosa and yolk sac membrane (Reed, 2004).

The intimate and persistent interaction between the amnioserosa and yolk sac -- phase II -- initiates in the dorsal-anterior region of the amnioserosa. This contact is maintained and further contact is established in an anterior-to-posterior direction as retraction progresses. Close apposition of the amnioserosa and yolk sac membranes persists during dorsal closure (Reed, 2004).

In mammals, basigin has been reported to be expressed and to function in extraembryonic tissues during early development, when it is required for embryo implantation. Basigin also functions in retinal epithelial morphogenesis. Since Drosophila Basigin is highly enriched on the extraembryonic membranes prior to and during their close apposition, attention was directed to the structure and function of Drosophila Basigin (Reed, 2004).

The Drosophila basigin transcription unit (CG31605, FBgn0051605) encodes multiple transcript variants. The transcripts encode two distinct protein isoforms: a long, 298 amino acid (aa) isoform and a short, 265 aa isoform. The long and short isoforms differ only at their amino and carboxy termini: the first 50 aa of the long form are substituted by 25 aa in the short form; the long form also has an 8 aa carboxy-terminal extension. The distinct N-terminal regions each contain their own unique transmembrane domains and signal peptide cleavage sites. The long isoform's N-terminal region is glycine rich. Database searches show that long and short isoforms also exist for human basigin (Reed, 2004).

Drosophila and mammalian basigin exhibit strong conservation of immunoglobulin (Ig) domain organization, location of predicted O linked glycosylation sites, as well as extracellular and cytoplasmic tail length. Both mammalian and Drosophila basigin have two extracellular Ig domains, the C-terminal of which appears to be representative of a 'primordial' Ig domain. There is an additional, more C-terminal 50 amino acid stretch of conservation, which will be referred to as the 'basibox' and which includes the predicted transmembrane domain. One of the defining features of the basibox is a glutamic acid residue in the middle of the transmembrane domain. The basibox is 52%-54% identical between Drosophila and vertebrates; the central 27 amino acids show 78%-81% identity (Reed, 2004).

There are multiple P element inserts in or near the basigin gene. One, the NP6293 GAL4 P element insertion, is in the 5'UTR of a predicted basigin transcript. This insertion causes leaky postembryonic lethality when homozygous and is referred to here as bsgNP6293. Homozygous bsgNP6293 embryos show no defects in germ band retraction and dorsal closure. A P element insert that causes male sterility has been referred to as gelded (Castrillon; 1993; Reed, 2004).

Basigin and integrins associate physically in mammals, possibly through direct contacts between basigin and the β1 integrin subunit. In Drosophila there is a single β integrin, called βPS integrin, which is encoded by the myospheroid (mys) gene. mys1 mutants show germ band retraction and dorsal closure defects (Reed, 2004).

Basigin and βPS integrin mutants show striking dominant genetic interactions: while bsgNP6293 mutants show no defects in dorsal closure and mys1 mutant embryos show only weak dorsal closure defects -- evidenced by a small dorsal hole -- mys1 mutant embryos from females in which the dose of the basigin gene is reduced by 50% show a striking increase in the size of the dorsal hole, while double mutant embryos show an even greater increase in dorsal hole size. The dominant genetic interaction of bsg and mys mutants is consistent with the possibility that basigin and integrin proteins interact physically in Drosophila (Reed, 2004).

Live imaging shows that those mys1 mutant embryos that fail germ band retraction exhibit apparently normal phase I interactions (for example, yolk sac projections are produced and contact the amnioserosa). However, phase II membrane apposition fails completely. Most striking is a failure of the dorsal-anterior region of the amnioserosa to initiate contact with the yolk sac membrane. In those mys1 mutant embryos that complete germ band retraction, there is failure to maintain the apposition of the amnioserosa and yolk sac membrane, with subsequent high penetrance failure of dorsal closure (Reed, 2004).

In summary, phase II membrane intimacy is compromised in mys1 mutants, implicating βPS integrin in the close apposition of amnioserosal and yolk sac membranes. The failure of both germ band retraction and dorsal closure in mys1 mutants suggests that close apposition of the extraembryonic membranes is required for these morphogenetic processes. The strong enhancement of mys1 dorsal closure defects by bsgNP6293 mutants suggests that Basigin functions together with βPS integrin in these morphogenetic processes. Anterior-to-posterior 'zipping up' of the membranes may generate forces that help push the germ band posteriorly. Alternatively, the role of integrin-dependent membrane apposition may be indirect, promoting survival of the amnioserosa, which in turn directs retraction and closure via signaling and/or physical contacts (Reed, 2004).

In wild-type embryos, the concomitant closure of the dorsal epidermis and midgut abrogate apposition of the amnioserosa and yolk sac. It was therefore asked when during normal development the amnioserosa loses integrity and dies. It has been shown, by using live imaging, that a small subset of the amnioserosal cells drop out of the epithelium prior to completion of closure. However, live imaging of the majority of amnioserosal cells (which remain in the epithelium) after dorsal closure has not been attempted previously (Reed, 2004).

Therefore, embryos in which amnioserosal cells had been specifically labeled were live-imaged, thus definitively addressing the fate of the amnioserosa after dorsal closure: the amnioserosa invaginates to form a tube-like structure with its perimeter cells aligning on the dorsal side of the tube, beneath the dorsal midline of the embryo. Over a period of 2-3 hr, individual nonperimeter cells round up and are extruded from the tube. Finally, the amnioserosal perimeter cells also dissociate. As amnioserosal cells are extruded, they are rapidly engulfed by hemocytes, which thus become GFP positive. These results are fully consistent with those inferred from analysis of fixed sectioned embryos (Reed, 2004).

It is possible to visualize a subset of the amnioserosal cells as acridine orange positive either before they leave the tube or shortly thereafter. Both acridine orange staining and engulfment by hemocytes are hallmarks of dying cells. To determine whether death of amnioserosal cells might be reaper dependent, it was asked whether reaper expression could be visualized in the amnioserosal cells prior to or after extrusion. No reaper-expressing cells were detected. To further test whether amnioserosal cell death might be reaper dependent, the H99 deficiency [Df(3L)H99] was used; this deficiency removes the reaper, head involution defective (hid), and grim genes, and the amnioserosa with anti-HNT antibody was visualized. If amnioserosal death were reaper dependent, one would expect HNT-positive cells to persist in H99 mutants when compared with wild-type. Such persistence does not occur. While it is conceivable that HNT expression is downregulated in a persistent amnioserosa, the simplest interpretation of these data is that death of the amnioserosa is reaper independent. This conclusion is consistent with the recent suggestion that Drosophila embryos have a caspase-independent cell engulfment system, which is still operative in H99 mutants (Reed, 2004).

It has been shown that loss of integrin-dependent contact between cells and the extracellular matrix leads to cell death, a process referred to as anoikis. Anoikis is promoted by the Jun amino-terminal kinase (JNK) pathway. Previous analyses have shown that JNK signaling in the amnioserosa is downregulated prior to dorsal closure. In those analyses, puckered-lacZ expression was used as a read-out of JNK signaling, and it was shown that relocation of JUN and FOS proteins from the nucleus to the cytoplasm of amnioserosal cells correlates with downregulation of JNK signaling. While JNK signaling is downregulated in the amnioserosa prior to dorsal closure, JNK signaling is upregulated in this tissue as dorsal closure approaches completion. Thus, reactivation of JNK signaling in the amnioserosa follows loss of integrin-dependent apposition of the amnioserosa and yolk sac membrane and precedes amnioserosal disintegration and death. These data are consistent with the hypothesis that midgut closure disrupts integrin-dependent apposition of the amnioserosa and yolk sac, thus inducing JNK signaling in the amnioserosa and its subsequent anoikis (Reed, 2004).

Therefore, in the Drosophila embryo, intimate apposition of the extraembryonic membranes is integrin dependent and promotes the integrity and survival of the amnioserosa. During normal development, closure of the midgut abrogates contact between the amnioserosa and yolk sac. JNK signaling is then upregulated in the amnioserosa, which subsquently disintegrates and dies, consistent with this being an example of developmentally programmed anoikis. In a subset of mys (βPS integrin) mutant embryos, apposition of the extraembryonic membranes never occurs, and the amnioserosa undergoes premature anoikis. The strong genetic interaction of mys and basigin mutants is consistent with the known physical interaction of these molecules in mammals (Berditchevski, 1997) and suggests that basigin might act together with integrin to promote extraembryonic membrane interaction and to prevent anoikis of the amnioserosa. Failure of germ band retraction and dorsal closure occurs in integrin mutants and is greatly enhanced when basigin levels are reduced. Together, these results suggest that extraembryonic membrane interaction promotes survival of the amnioserosa, which in turn directs germ band retraction and dorsal closure through physical contacts and/or intercellular signaling (Reed, 2004).

The hypothesis that amnioserosal anoikis is triggered during normal development by loss of integrin-mediated contact with the yolk sac membrane allows several testable predictions: (1) that in mutants in which the amnioserosa undergoes premature apoptosis prior to germ band retraction (e.g., hindsight), phase II apposition of the amnioserosa and yolk sac membrane may fail; (2) that premature amnioserosal apoptosis in these mutants is a consequence, rather than a cause of loss of amnioserosal epithelial integrity; (3) that the amnioserosa may persist in mutants lacking a midgut or in those defective for midgut closure (Reed, 2004).

It remains to be determined whether disintegration and death of the amnioserosa during normal development is caused solely by loss of contact with the yolk sac (i.e., is nonautonomously induced) versus whether signals from cell types other than the yolk -- or even an amnioserosa-autonomous program -- also play a role. For example, it is possible that upregulation of JNK signaling in the amnioserosa is independent of loss of contact with the yolk sac. Analysis of mutants lacking a midgut provide a test of this possibility: if disintegration and death of the amnioserosa occur even when apposition with the yolk sac is maintained, signals from other cell types or amnioserosa-autonomous processes would be implicated (Reed, 2004).

The specific role of JNK signaling in amnioserosal anoikis is difficult to assess because downregulation of JNK signaling in the amnioserosa and up-regulation of JNK signaling in the leading edge of the epidermis are required for dorsal closure. Thus JNK pathway mutants stall morphogenesis prior to dorsal closure, making it impossible to assess a possible later role. Expression of dominant-negative JNK specifically in the amnioserosa only later in development, when closure is almost complete, will be necessary to rigorously test the role of JNK activation in amnioserosal anoikis (Reed, 2004).

All of the data presented above support the hypothesis that phase II amnioserosa-yolk sac membrane association is necessary for maintenance of the amnioserosal epithelium and, thus, the morphogenetic processes of germ band retraction and dorsal closure. However, the role of the transient phase I interaction is less clear. It is unlikely that the phase I interactions play a role in generation of the forces that lead to close apposition of these extraembryonic membranes. It seems more likely that the transient interactions play a role in communication between the yolk sac and the amnioserosa. The ecdysone receptor and active ecdysteroids are reported to be present in the amnioserosa and required for germ band retraction (Kozlova, 2003). Expression of a dominant-negative form of the ecdysone receptor worsens germ band retraction defects in mys (βPS integrin) mutants (Kozlova, 2003). Furthermore, it has been speculated that enzymes residing in the yolk might participate in conversion of ecdysone to its active forms (Kozlova, 2003). Dynamic invaginations of the yolk sac membrane, which dive into the yolk mass and transiently contact the yolk spheres, have been observed. Thus, one tantalizing possibility is that these invaginations transport active forms of ecdysone -- as well as other key signaling molecules -- from the yolk spheres to the yolk sac membrane. Phase I amnioserosa-yolk membrane contacts and/or phase II intimate membrane apposition might subsequently bring these molecules to the amnioserosa (Reed, 2004).

It is concluded that the extraembryonic tissues of Drosophila play a crucial role in directing embryonic morphogenesis. Close apposition of the yolk sac membrane and the basal cell membranes of the amnioserosa is dependent on βPS integrin. This intimate membrane association is required to promote survival and to prevent anoikis of the amnioserosa. The amnioserosa then directs germ band retraction and dorsal closure through physical contacts and/or signaling. Disintegration and death of the amnioserosa after closure of the epidermis and midgut correlates with upregulation of JNK signaling in the amnioserosa, is independent of reaper/hid/grim function, and is likely to represent the first example of developmentally programmed anoikis in Drosophila (Reed, 2004).

Basigin interacts with integrin to affect cellular architecture

Basigin, an IgG family glycoprotein found on the surface of human metastatic tumors, stimulates fibroblasts to secrete matrix metalloproteases that remodel the extracellular matrix. Using Drosophila melanogaster, intracellular, matrix metalloprotease-independent, roles for basigin have been identified. Specifically, Basigin, interacting with integrin, is required for normal cell architecture in some cell types. Basigin promotes cytoskeletal rearrangements and the formation of lamellipodia in cultured insect cells. Loss of basigin from photoreceptors leads to misplaced nuclei, rough ER and mitochondria, as well as to swollen axon terminals. These changes in intracellular structure suggest cytoskeletal disruptions. These defects can be rescued by either fly or mouse Basigin. Basigin and integrin colocalize to cultured cells and to the visual system. Basigin-mediated changes in the architecture of cultured cells require integrin binding activity. Basigin and integrin interact genetically to affect cell structure in the animal, possibly by forming complexes at cell contacts that help organize internal cell structure (Curtin, 2005).

The two P-element insertions in bsg are located 1145 bp from the start of transcription for the D-basigin 265 protein isoform. Homozygous mutant animals from both lines died after the second larval instar with only 3% of mutant larvae living to the third instar. The insertions failed to complement each other. Because this P-insertion did not interrupt the coding portion of the gene, animals carrying this mutation may have produced some functioning protein. To generate a more severe allele, the P-element (P1478) was mobilized; such mobilization occasionally caused loss of genetic material near the insertion site. Two hundred excision lines were establised in which the P-element was missing; 182 were viable, indicating a clean excision of the P-element, whereas 18 were homozygous lethal and failed to complement the original P-element allele. By DNA blot analysis, two excision lines, bsgΔ265 and excision number 64, were shown to be missing ~4 kb, including the first coding exon for the D-basigin 265 protein. Both lines showed high embryonic lethality with 75%-80% of the animals dying as embryos. Those embryos that did hatch died within the first day and were small, lethargic and uncoordinated (Curtin, 2005).

Effects of D-basigin on placement of internal cellular organelles in photoreceptors werre examined. Because the mutations are embryonic lethal, mosaic animals were made in which D-basigin protein expression was missing only in the eye and invariably missing from photoreceptor neurons. Such mosaics were generated by expression of FLP recombinase from the eye-specific promoter of the eyeless gene (ey). Eyeless-FLP mediates recombination in the eye between chromosome arms bearing engineered copies of the FLP binding sites (FRTs) near their centromeres. A chromosome arm bearing a bsg mutation was recombined with a chromosome arm bearing the cell death gene hid expressed specifically in all photoreceptors. After recombination and chromosome segregation, only photoreceptors that inherit two copies of mutant bsg survive to repopulate the eye; bsg eyes were almost normal in size (Curtin, 2005).

Photoreceptor nuclei were visualized with an antibody against Elav, a neuron-specific nuclear protein. Normally, photoreceptor nuclei lie in tight rows across the eye, so that any mislocalization is readily detected. The nuclei of the R1-R6 photoreceptors lie in the apical region of the retina. The nuclei of the R7 photoreceptors are just proximal to those of R1-R6 and the R8 nuclei lie near the basement membrane of the retina (Curtin, 2005).

Photoreceptor nuclei of mosaic flies mutant in the eye for the hypomorphic P1096 allele, which encodes a nuclear ß-gal, were visualized with anti-ß-gal. Most nuclei were properly located, although a few nuclei were misplaced. Similar results were seen for these mosaics with anti-elav. In mosaics that are mutant in the eye for the bsgΔ265 excision allele, Elav immunolabeling revealed that 16-50% of photoreceptor nuclei were mislocalized. Nuclei were counted as misplaced only if they were obviously located between the normal position for R7 and the normal position for R8, in the region of the eye where no nuclei are usually located. Thus nuclei that were slightly displaced were not counted. Sections from a total of 18 animals were counted (10,250 nuclei). Although the range of nuclear misplacements per fly was 16-50%, most animals fell within the lower end of this range, the average number of misplaced nuclei, pooling data from all animals, being 22% (Curtin, 2005).

The nuclear placement defect was rescued by expressing D-basigin 265. Nuclear placement was counted in 12 animals that were mutant in the eye for bsgΔ265, but also contained a bsgΔ265 transgene that expressed D-basigin 265 in photoreceptors, and only 1% of misplaced nuclei were found. Expression of the mouse basigin gene in photoreceptors also rescued the nuclear misplacement with only 1.5% of nuclei misplaced in a total of 12 animals counted. Thus despite limited sequence homology, mouse basigin can promote the formation of normal cell architecture in flies (Curtin, 2005).

Photoreceptors R1-R6 terminate in the lamina, or first optic neuropile. Laminas were examined in which only the photoreceptors are mutant for bsgΔ265 (i.e. the postsynaptic lamina neurons and glia are wild type). Rough endoplasmic reticulum (rER) was found misplaced into the mutant photoreceptor axon terminals. Normally rER, which is continuous with the nuclear membrane, is confined distally to the photoreceptor cell body in the overlying retina. Its more proximal displacement into the photoreceptor terminal in the lamina accords with the more proximal location of many R1-R6 nuclei. In addition to misplaced nuclei, mitochondria were also misplaced. The mitochondria accumulated in excessive numbers in the distal portion of the photoreceptor terminals, but were absent from the proximal portion of the terminals, where they are also normally found. In addition to misplaced organelles, bsgΔ265 mutant photoreceptors showed a clear increase in axon terminal size, with profiles that were >80% larger in cross-sectional area compared to the control, a difference that was significant. None of these defects was seen in control animals in which non-mutant chromosomes were recombined. On the whole, these defects, misplaced internal organelles and enlarged terminals, suggest global disruptions in cell structure in bsgΔ265 mutant cells, probably due to alterations in the cytoskeleton (Curtin, 2005).

Colocalization of integrin and D-basigin was found in the retina. In addition, studies of integrin gene mutants have reported that the R8 nuclei are sometimes misplaced, descending beneath the basement membrane. This was somtimes seen in bsgΔ265 mosaics and therefore genetic interactions were examined between bsgΔ265 and integrin genes with respect to nuclear placement (Curtin, 2005).

The integrin proteins expressed in the eye, αPS1 and ßPS, are encoded by genes located on the X chromosome, mew codes αPS1 integrin and mys codes ßPS integrin. Mysb45 is a viable allele and males carrying this mutation showed normal placement of photoreceptor nuclei. Mutant flies homozygous in the retina for a weak P-allele (P1096) of basigin showed occasional nuclear misplacement. To look for genetic interactions between bsgΔ265 and integrin genes, double mutants were maed by creating males that carried the mysb45 allele (coding a mutant ßPS integrin), but were also homozygous mutant only in the retina for the P1096 bsg allele. These animals showed obvious misplacement of nuclei. The average number of misplaced photoreceptor nuclei per head section, after examining at least 12 animals of each genotype, was three times higher in the double mutants than that predicted from the summed effect of the two single mutations. Mosaics doubly mutant for mysb45 and bsgΔ265 also showed a more severe photoreceptor nuclear misplacement phenotype than the sum of the two single mutations would predict; 80% of nuclei were misplaced compared with an average of 24% for bsgΔ265 and 1-2% for mysb45 (Curtin, 2005).

Some integrin gene allelic combinations also showed nuclear misplacement. Animals heterozygous for mewM6, a null allele for αPS1 integrin showed normal placement of photoreceptor nuclei. Animals heterozygous for mysb45, a ßPS1 allele, showed normal nuclear placement, similar to the mysb45 hemizygous males. However, animals heterozygous for both mewM6 and mysb45 showed 3% misplaced nuclei (Curtin, 2005).

Because mammalian basigin stimulates secretion of MMPs, the role of MMPs in the fly visual system was examined. Drosophila has two MMP genes, Mmp1 and Mmp2, both required for viability. Only Mmp2 is expressed in the developing eye (Llano, 2000; Llano, 2002; Page-McCaw, 2003). If D-basigin were acting primarily through MMP-2, then flies lacking MMP-2 in the retina should have the same phenotypes as those found in bsgΔ265 mutant retina. Using the same method previously described to make bsgΔ265 eye mosaics, flies were made that were mutant in the eye for a null Mmp2 allele, Mmp2w307* (Page-McCaw, 2003). No misplaced photoreceptor cell nuclei were seen. In case MMP1 functionally replaces MMP-2, mosaics were made that were mutant in the eye for both genes. These also showed no misplaced nuclei. Finally, no effect was seen on nuclear placement when expression of Drosophila TIMP (tissue specific inhibitors of MMPs) was driven in the eye, even though this TIMP gene has previously been reported (Page-McCaw, 2003) to block biological activity of Drosophila MMPs (Curtin, 2005).

Behavioral responses to odorants in Drosophila require nervous system expression of the β integrin gene myospheroid

Integrins are cell adhesion molecules that mediate numerous developmental processes in addition to a variety of acute physiological events. Two reports implicate a Drosophila β integrin, βPS, in olfactory behavior. To further investigate the role of integrins in Drosophila olfaction, Gal4-driven expression of RNA interference (RNAi) transgenes were used to knock down expression of myospheroid (mys), the gene that encodes βPS. Expression of mys-RNAi transgenes in the wing reduced βPS immunostaining and produced morphological defects associated with loss-of-function mutations in mys, demonstrating that this strategy knocked down mys function. Expression of mys-RNAi transgenes in the antennae, antennal lobes, and mushroom bodies via two Gal4 lines, H24 and MT14, disrupted olfactory behavior but did not alter locomotor abilities or central nervous system structure. Olfactory behavior was normal in flies that expressed mys-RNAi transgenes via other Gal4 lines that specifically targeted the antennae, the projection neurons, the mushroom bodies, bitter and sweet gustatory neurons, or Pox neuro neurons. These studies confirm that mys is important for the development or function of the Drosophila olfactory system. Additionally, these studies demonstrate that mys is required for normal behavioral responses to both aversive and attractive odorants. These results are consistent with a model in which βPS mediates events within the antennal lobes that influence odorant sensitivity (Bhandari, 2006).

Focal adhesion kinase interacts with myospheroid to control morphogenesis of the Drosophila optic stalk

Photoreceptor cell axons (R axons) innervate optic ganglia in the Drosophila brain through the tubular optic stalk. This structure consists of surface glia (SG) and forms independently of R axon projection. A screen for genes involved in optic stalk formation identified Fak56D, encoding a Drosophila homolog of mammalian focal adhesion kinase (FAK). FAK is a main component of the focal adhesion signaling that regulates various cellular events, including cell migration and morphology. Fak56D mutation causes severe disruption of the optic stalk structure. These phenotypes were completely rescued by Fak56D transgene expression in the SG cells but not in photoreceptor cells. Moreover, Fak56D genetically interacts with myospheroid, which encodes an integrin ß subunit. In addition, CdGAPr is also required for optic stalk formation and genetically interacts with Fak56D. CdGAPr encodes a GTPase-activating domain that is homologous to that of mammalian CdGAP, which functions in focal adhesion signaling. Hence the optic stalk is a simple monolayered structure that can serve as an ideal system for studying glial cell morphogenesis and the developmental role(s) of focal adhesion signaling (Murakami, 2007).

Integrin signaling regulates spindle orientation in Drosophila to preserve the follicular-epithelium monolayer

Epithelia act as important physiological barriers and as structural components of tissues and organs. In the Drosophila ovary, follicle cells envelop the germline cysts to form a monolayer epithelium. During division, the orientation of the mitotic spindle in follicle cells is such that both daughter cells remain within the same plane, and the simple structure of the follicular epithelium is thus preserved. This study shows that integrins, heterodimeric transmembrane receptors that connect the extracellular matrix to the cell's cytoskeleton, are required for maintaining the ovarian monolayer epithelium in Drosophila. Mosaic egg chambers containing integrin mutant follicle cells develop stratified epithelia at both poles. This stratification is due neither to abnormal cell proliferation nor to defects in the apical-basal polarity of the mutant cells. Instead, integrin function is required for the correct orientation of the mitotic apparatus both in mutant cells and in their immediately adjacent wild-type neighbors. Integrin-mediated signaling, rather than adhesion, is sufficient for maintaining the integrity of the follicular epithelium. These data show that integrins are necessary for preserving the simple organization of a specialized epithelium and link integrin-mediated signaling to the correct orientation of the mitotic spindle in this epithelial cell type (Fernandez-Minan, 2007).

These results demonstrate that integrin-mediated signaling is required for proper alignment of the mitotic spindle and thus for the maintenance of the follicular-epithelium monolayer. The mechanism by which integrins influence the orientation of the mitotic apparatus is still unknown, but several lines of evidence point to an interaction between the actomyosin cytoskeleton and integrin activity in this process. (1) This cytoskeleton is one of the main targets of integrin signaling, and a variety of molecules that transmit signals from activated integrins to the actin cytoskeleton have been identified; one such molecule is Talin. In addition, unconventional myosins have been shown to provide a motor-based link between integrins and the actin cytoskeleton in vertebrate cells in culture. Finally, integrins are required in epithelial follicle cells for the organization of actin filaments. (2) The class VI unconventional myosin Jaguar and myosin II are required for positioning the cell-division axis in Drosophila neuroblasts and in vertebrate cells, respectively. Hence, a model is favored whereby integrin adhesion to the ECM elicits a signal cascade in the follicle cells and this signal cascade organizes the actin cytoskeleton, either by regulating the interaction of microtubules with the actin cytoskeleton or by polarizing its activity. As a consequence, mitotic cells position their centrosomes so that the spindle is aligned parallel to the germline surface (Fernandez-Minan, 2007).

Although it has been shown that mouse keratinocytes require β1 integrin to control spindle orientation, this integrin is also necessary for the establishment of apical-basal polarity in this cell type, in contrast with the role of mys in follicle cells. These results thus provide a link between integrin activity and the orientation of cell division and may explain the aberrant behavior of certain epithelia, such as the mammary glandular epithelium, when integrin function is impaired. Interestingly, it has been reported recently that the integrin β1 tail regulates several aspects of mitosis in cells in culture; such aspects include centrosome function, the organization of the mitotic spindle, and cytokinesis (Li, 2005). Although these results do not directly implicate βPS integrin in cell proliferation, the above findings reinforce the role of integrins during mitosis (Fernandez-Minan, 2007).

Lasp anchors the Drosophila male stem cell niche and mediates spermatid individualization

Lasp family proteins contain an amino-terminal LIM domain, two actin-binding nebulin repeats and a carboxyl-terminal SH3 domain. Vertebrate Lasp-1 localizes to focal adhesions and the leading edge of migrating cells, and is required for cell migration. To assess the in vivo function of Lasp, a null mutant in Drosophila Lasp was generated. Lasp1 is homozygous viable, but male sterile. In Lasp mutants the stem cell niche is no longer anchored to the apical tip of the testis, and actin cone migration is perturbed resulting in improper spermatid individualization. Hub cell mislocalization can by phenocopied by expressing Lasp or betaPS integrin RNAi transgenes in somatic cells, and Lasp genetically interacts with betaPS integrin, demonstrating that Lasp functions together with integrins in hub cells to anchor the stem cell niche. Finally, it is shown that the stem cell niche is maintained even if it is not properly localized (Lee, 2008).

In conclusion, Lasp has an integrin-dependent function in anchoring the stem cell niche, where Lasp likely functions at the periphery of the integrin adhesion site, given its weak phenotype compared to βPS integrin. Lasp also exhibits an integrin-independent function in actin cone migration. Intriguingly, human Lasp-1 similarly localizes to focal adhesions and to the cortical actin cytoskeleton in the leading edge of migrating cells, and it is the leading edge localization that appears to mediate cell migration. Lasp function in whole animals is restricted to highly specialized tissues, suggesting that integrin adhesion sites and actin organization differ depending on tissue and function, which may be one reason for the large number of biochemically defined components of integrin adhesion sites and actin-binding proteins. Furthermore, the data show that the stem cell niche does not require a specific location to function properly, and by extension, that the male stem cell niche in Drosophila requires only one type of support cells, the hub cells (Lee, 2008).

Mesoderm migration in Drosophila is a multi-step process requiring FGF signaling and integrin activity

Migration is a complex, dynamic process that has largely been studied using qualitative or static approaches. As technology has improved, it is now possible to take quantitative approaches towards understanding cell migration using in vivo imaging and tracking analyses. In this manner, a four-step model of mesoderm migration during Drosophila gastrulation was establised: (I) mesodermal tube formation, (II) collapse of the mesoderm, (III) dorsal migration and spreading and (IV) monolayer formation. The data provide evidence that these steps are temporally distinct and that each might require different chemical inputs. To support this, the role was analyzed of fibroblast growth factor (FGF) signaling, in particular the function of two Drosophila FGF ligands, Pyramus and Thisbe, during mesoderm migration. It was determined that FGF signaling through both ligands controls movements in the radial direction. Thisbe is required for the initial collapse of the mesoderm onto the ectoderm, whereas both Pyramus and Thisbe are required for monolayer formation. In addition, it was uncovered that the GTPase Rap1 regulates radial movement of cells and localization of the beta-integrin subunit, Myospheroid, which is also required for monolayer formation. These analyses suggest that distinct signals influence particular movements, since it was found that FGF signaling is involved in controlling collapse and monolayer formation but not dorsal movement, whereas integrins are required to support monolayer formation only and not earlier movements. This work demonstrates that complex cell migration is not necessarily a fluid process, but suggests instead that different types of movements are directed by distinct inputs in a stepwise manner (McMahon, 2010).

Mesoderm migration was found to be a combination of complex three-dimensional movements involving many molecular components. live imaging, coupled with quantitative analyses, is important for studying complex cell movements, as it allowed migration to be decomposed into different movement types and thus has allowed description of subtle phenotypes. First, analysis of the directional movements of mesoderm cells within wild-type embryos was extended, focusing on the temporal sequences of events. Cells were found follow a sequential and distinct set of trajectories: movement in the radial direction (tube collapse: -5 to 15 minutes, 0=onset of germband elongation), followed by movement in the angular direction (dorsal migration: 15 to 75 minutes) and ending with small intercalation movements in the radial direction (monolayer formation: 75 to 110 minutes). These movements appear temporally distinct (i.e. stepwise), and thus molecular signals controlling each process were sought (McMahon, 2010).

Which mesoderm movements were FGF-dependent were investigated and, in particular, either Ths- or Pyr-dependent. The interaction between Htl and its two ligands provides a simpler system relative to vertebrates (which exhibit over 120 receptor-ligand interactions) in which to study how and why multiple FGF ligands interact with the same receptor. Previously, it was found that FGF signaling via the Htl FGFR controls collapse of the mesodermal tube but not dorsal-directed spreading (McMahon, 2008). This study demonstrated that FGF signaling is also required for monolayer formation. In addition, distinct, non-redundant roles were defined for the FGF ligands: Ths (but not Pyr) is required for collapse of the mesodermal tube, whereas both Pyr and Ths are required for proper intercalation of mesoderm cells after dorsal spreading (McMahon, 2010).

This analysis raises questions about ligand choice during collapse and monolayer formation. Within the mesodermal tube, cells at the top require a long-range signal in order to orient towards the ectoderm during tube collapse, whereas the signals controlling intercalation during monolayer formation can be of shorter range. It is suggested that the ligands have different activities that are appropriately tuned for these processes. In fact, recent studies of the functional domains of these proteins suggest that Ths has a longer range of action than Pyr, in agreement with the analysis that Pyr does not support tube collapse, but does have a hand in monolayer formation (McMahon, 2010).

This study has demonstrated that Rap1 mutants have a similar mesoderm phenotype to the FGFR htl mutant, with defects in collapse and monolayer formation. It was not possible to establish whether Rap1 acts downstream of FGF signaling, as the complete loss of Mys in Rap1 mutants is more severe than the patchy expression of Mys seen in htl mutants. Therefore, Rap1 could be working in parallel to or downstream of FGF signaling during mesoderm migration. Rap1 has been implicated in several morphogenetic events during Drosophila gastrulation and probably interacts with many different signaling pathways. Further study of Rap1, along with other GTPases, will shed light onto their role during mesoderm migration, how they interact with one another and what signaling pathways control them (McMahon, 2010).

Focus was placed on the more specific phenotype of mys mutants, as its localization is affected in htl mutants and it exhibits a monolayer defect that is similar to pyr and ths mutants. Integrins are important for cell adhesion, so it is not surprising that cells fail to make stable contact with the ectoderm through intercalation in mys mutants. However, some cells do contribute to monolayer formation in the absence of Mys, implying that other adhesion molecules are involved in maintaining contact between the mesoderm and ectoderm. These other adhesion molecules might be activated downstream of FGF signaling as the htl mutant monolayer phenotype is more severe than the mys mutant. Discovering the downstream targets of Htl, which might regulate cell adhesion properties, will help to shed light on the mechanisms supporting collapse of the mesodermal tube (which is not dependent on Mys) and monolayer formation (which is Mys-dependent) (McMahon, 2010).

Cell protrusions, such as filopodia, are important for sensing chemoattractants and polarizing movement during migration. Previous studies have focused on protrusive activity at the leading edge during mesoderm migration in Drosophila and shown that these protrusions are FGF-dependent. In this study, it was found that protrusions exist in all mesoderm cells, not just the leading edge, and that these protrusions also extend into the ectoderm (McMahon, 2010).

The study demonstrates that FGF signaling, as well as integrin activity, is required to support protrusive activity into the ectoderm; this is a potential mechanism by which FGF signaling and Mys could control movement toward the ectoderm during monolayer formation. The function of protrusions at the leading edge remains unclear, as they appear to be reduced in pyr and mys mutants, but migration in the dorsal direction still occurs in both mutant backgrounds. One interpretation is that FGF and Mys are important for generalized protrusive activity and that extensive protrusions are required for intercalation but not dorsal migration (McMahon, 2010).

Based on this study, it is proposed that mesoderm migration is a stepwise process, with each event requiring different molecular cues to achieve collective migration. Invagination of the mesoderm is the first step in this process and is dependent on Snail, Twist, Concertina, Fog and several other genes. Next, collapse of the mesoderm tube onto the ectoderm requires Htl activation via Ths. Rap1 might be involved in this process as well but the phenotype of Rap1 mutants is complex and it is unclear which phenotypes are primary defects (McMahon, 2010).

Following collapse, mesoderm cells spread dorsally by an unknown mechanism. Dorsal migration is unaffected in pyr and ths mutants and occurs in all cells that contact the ectoderm in htl mutants, implying that FGF signaling is, at most, indirectly involved in this step owing to the earlier tube collapse defect (McMahon, 2008). Whether dorsal migration requires chemoattractive signals or whether the cells simply move in this direction because it is the area of least resistance remains unclear (McMahon, 2010).

Finally, after dorsal spreading is complete, any remaining cells not contacting the ectoderm intercalate to form a monolayer. This process is controlled by a combination of both Pyr and Ths interacting through Htl and also by Rap1 and Mys. In other systems, intercalation can lead to changes in the properties of the cell collective, for instance, lengthening of a body plan. However, this study has shown that dorsal migration and spreading are not a result of intercalation, as intercalation occurs after spreading has finished (McMahon, 2010).

Coordination of these signals to control collective migration enables the mesoderm to form a symmetrical structure, which is essential for embryo survival. This model begins to address the question of how hundreds of cells move in concerted fashion and is relevant for a generalized understanding of embryogenesis and organogenesis. It was found that mesoderm migration is accomplished through sequential movements in different directions, implying that collective migration might be best achieved by distinct phases of movement (McMahon, 2010).

Identification of integrin beta subunit mutations that alter affinity for extracellular matrix ligand

Over 50 mutations in the Drosophila βPS integrin subunit that alter integrin function in situ were examined for their ability to bind a soluble monovalent ligand, TWOW-1. Surprisingly, very few of the mutations, which were selected for conditional lethality in the fly, reduce the ligand binding ability of the integrin. The most prevalent class of mutations activates the integrin heterodimer. These findings emphasize the importance of integrin affinity regulation and point out how molecular interactions throughout the integrin molecule are important in keeping the integrin in a low affinity state. Mutations strongly support the controversial deadbolt hypothesis, where the CD loop in the β tail domain acts to restrain the I domain in the inactive, bent conformation. Site-directed mutations in the cytoplasmic domains of βPS and αPS2C (inflated) reveal different effects on ligand binding from those observed for humN αIIbβ3 integrins and identify for the first time a cytoplasmic cysteine residue, conserved in three human integrins, as being important in affinity regulation. In the fly, it was found that genetic interactions of the βPS mutations with reduction in talin function are consistent with the integrin affinity differences measured in cells. Additionally, these genetic interactions report on increased and decreased integrin functions that do not result in affinity changes in the PS2C integrin measured in cultured cells (Kendell, 2011).

Integrins are necessary for the development and maintenance of the glial layers in the Drosophila peripheral nerve

Peripheral nerve development involves multiple classes of glia that cooperate to form overlapping glial layers paired with the deposition of a surrounding extracellular matrix (ECM). The formation of this tubular structure protects the ensheathed axons from physical and pathogenic damage and from changes in the ionic environment. Integrins, a major family of ECM receptors, play a number of roles in the development of myelinating Schwann cells, one class of glia ensheathing the peripheral nerves of vertebrates. However, the identity and the role of the integrin complexes utilized by the other classes of peripheral nerve glia have not been determined in any animal. This study shows that, in the peripheral nerves of Drosophila melanogaster, two integrin complexes (αPS2βPS and αPS3βPS) are expressed in the different glial layers and form adhesion complexes with integrin-linked kinase and Talin. Knockdown of the common beta subunit (βPS) using inducible RNAi in all glial cells results in lethality and glial defects. Analysis of integrin complex function in specific glial layers showed that loss of βPS in the outermost layer (the perineurial glia) results in a failure to wrap the nerve, a phenotype similar to that of Matrix metalloproteinase 2-mediated degradation of the ECM. Knockdown of βPS integrin in the innermost wrapping glia causes a loss of glial processes around axons. Together, these data suggest that integrins are employed in different glial layers to mediate the development and maintenance of the protective glial sheath in Drosophila peripheral nerves (Xie, 2011).

Peripheral nerves in vertebrates and Drosophila are organized in similar ways, with central axons wrapped by an inner class of glia that are surrounded in turn by layers of external glia. The glia and the surrounding ECM establish a tubular sheath to protect axons from physical damage and pathogens. This study shows that at least two integrin heterodimers are expressed in these different glial layers and are localized at focal adhesions with Ilk and Talin. αPS2βPS integrin is prevalent in the PG and the αPS3βPS integrin is more prevalent in the wrapping glia (WG). Since βPS2 integrin is expressed mostly in the outermost PG and can bind ligands that contain the tripeptide RGD sequence, it most likely functions by binding to ECM ligands in the NL. The majority of αPS3 integrin is expressed by the internal SPG and WG and αPS3 integrin has been shown to interact with laminins in Drosophila. However, it is possible that the integrin complex in the WG might have other ligands and might mediate direct cell-cell interactions, since a pronounced basal lamina associated with the peripheral axons was not detected and Mmp2 expression had no effect in the internal regions of the peripheral nerves (Xie, 2011).

Peripherial glia (PG) form the outermost glial layer in the Drosophila nervous system but their origin and function are not well understood, even though they were identified some time ago. Drosophila PG are structurally similar to their vertebrate counterparts and have been proposed to have similar roles. Vertebrate perineurial cells and the associated collagen fibers provide important mechanical support to nerves and their development might rely on ECM-mediated signals, as β1 integrin is found in the perineurium. However, little is known about the role of integrins in perineurial cells (Xie, 2011).

The current results show that Drosophila PG express αPS2 and βPS integrin subunits (and to a lesser extent αPS3), which colocalize with both Ilk and Talin. Knocking down integrins disrupts perineurial wrapping, but it was not possible to distinguish whether the PG failed to initiate wrapping or failed to maintain their processes around the nerves to accommodate the growing nerve surface. However, degradation of the neural lamella (NL) by Mmp2 overexpression generates a PG wrapping phenotype similar to that of βPS RNAi. Thus, binding of βPS integrin to ligands in the ECM mediates the radial spread of PG around the tubular structure of the nerve and suggests that PG retract their membrane when integrin-ECM interaction is interrupted. Moreover, knockdown of Talin in the PG produces a similar wrapping defect. This suggests that, in the PG, integrin adhesion complexes mediate the connection between the extracellular NL and the intracellular actin cytoskeleton and are required for the initiation or maintenance of glial ensheathment (Xie, 2011).

The function of the PG in the peripheral nerve is not well understood. The PG do not generate an impermeable barrier and it is the subperineurial glia (SPG) layer that creates the blood-nerve barrier. Loss of PG ensheathment did not result in paralysis or lethality, which are signs of a disrupted blood-brain barrier. Larger molecules (~500 kDa) are blocked by the NL or the PG, perhaps mirroring the protective function of the vertebrate perineurium against pathogens. However, in Mmp2-overexpressing larvae, the affected nerves become thin and are difficult to retain intact during tissue preparation. This suggests that the NL and PG provide important mechanical support as is seen with the perineurium in mammalian nerves (Xie, 2011).

In the center of Drosophila peripheral nerves, the WG embed axons in bundles or individually within single membrane wraps, similar to the non-myelinating Schwann cells of vertebrate peripheral nerves. The current results show that WG predominantly express αPS3 and βPS integrin subunits in complexes positive for Ilk and Talin along the WG membrane. When βPS expression is knocked down, the complexity of the glial processes between the associated axons is greatly reduced. Only a few long processes and small membrane protrusions are observed around the axons, suggesting that the WG might be retracting their processes in the absence of βPS integrin. The role of integrins appears to be conserved between WG and Schwann cells in vertebrates. For example, myelinating Schwann cells lacking β1 integrin or Ilk do not extend membrane processes around axons, resulting in impaired radial sorting. The role of integrins in non-myelinating Schwann cells is not known but the current results suggest that integrins have similar functions in Drosophila and vertebrates in mediating the glial ensheathment of peripheral axons (Xie, 2011).

No clear ECM has been observed in the internal regions of Drosophila nerves by immunofluorescence analysis or transmission electron microscopy. Therefore, the integrin complex could promote WG sheath formation by mediating direct cell-cell adhesion between the glial membrane and its associated axon, or between glial membranes. A potential candidate for an integrin-interacting protein expressed on axons or glia is Neuroglian, the Drosophila L1 (Nrcam) homolog. L1 is an Ig domain transmembrane protein that is known to bind RGD-dependent integrins. Loss of the integrin-binding domain of L1 results in wrapping defects in both myelinating and non-myelinating Schwann cells. However, the presence of low levels of ECM cannot be ruled out given the weak laminin immunolabeling that was observed. Even though Mmp2 expression in the WG had no effect, it is still possible that the integrin-mediated adhesion is Mmp2 resistant. To resolve this issue, further ultrastructural and genetic studies will be required (Xie, 2011).

Cellular mechanics of germ band retraction in Drosophila

Germ band retraction involves a dramatic rearrangement of the tissues on the surface of the Drosophila embryo. As germ band retraction commences, one tissue, the germ band, wraps around another, the amnioserosa. Through retraction the two tissues move cohesively as the highly elongated cells of the amnioserosa contract and the germ band moves so it is only on one side of the embryo. To understand the mechanical drivers of this process, a series of laser ablations was designed that suggest a mechanical role for the amnioserosa. first, it was found that during mid retraction, segments in the curve of the germ band are under anisotropic tension. The largest tensions are in the direction in which the amnioserosa contracts. Second, ablating one lateral flank of the amnioserosa reduces the observed force anisotropy and leads to retraction failures. The other intact flank of amnioserosa is insufficient to drive retraction, but can support some germ band cell elongation and is thus not a full phenocopy of ush mutants. Another ablation-induced failure in retraction can phenocopy mys mutants, and does so by targeting amnioserosa cells in the same region where the mutant fails to adhere to the germ band. It is concluded that the amnioserosa must play a key, but assistive, mechanical role that aids uncurling of the germ band (Lynch, 2013).

Loss of focal adhesions in glia disrupts both glial and photoreceptor axon migration in the Drosophila visual system

Many aspects of glial development are regulated by extracellular signals, including those from the extracellular matrix (ECM). Signals from the ECM are received by cell surface receptors, including the integrin family. Previous studies have shown that Drosophila integrins form adhesion complexes with Integrin-linked kinase and talin in the peripheral nerve glia and have conserved roles in glial sheath formation. However, integrin function in other aspects of glial development is unclear. The Drosophila eye imaginal disc (ED) and optic stalk (OS) complex is an excellent model with which to study glial migration, differentiation and glia-neuron interactions. The roles of the integrin complexes was studied in these glial developmental processes during OS/eye development. The common β subunit βPS and two α subunits, αPS2 and αPS3, are located in puncta at both glia-glia and glia-ECM interfaces. Depletion of βPS integrin and talin by RNAi impaired the migration and distribution of glia within the OS resulting in morphological defects. Reduction of integrin or talin in the glia also disrupted photoreceptor axon outgrowth leading to axon stalling in the OS and ED. The neuronal defects were correlated with a disruption of the carpet glia tube paired with invasion of glia into the core of the OS and the formation of a glial cap. These results suggest that integrin-mediated extracellular signals are important for multiple aspects of glial development and non-autonomously affect axonal migration during Drosophila eye development (Xie, 2014).

This study found that OS glia express integrin complexes that play a role in the development of the glia and axons of the ED and OS. βPS integrin is located in puncta at the glial membrane and associates with Talin and ILK. These focal adhesion markers are found between the perineural glia (PG) and ECM, plus at the interfaces of the PG-CG (carpet glia) and CG-CG layers. A different distribution was found for the αPS2 and αPS3 integrins, which were concentrated at the periphery and interior of the OS, respectively. The results from RNAi-mediated knockdown revealed that these complexes play important roles in OS glial development, as knockdown led to disruption of PG and CG morphology. Specifically, the loss of βPS integrin or talin caused PG to aggregate in the distal half of the OS, resulting in an accumulation of glia in the OS. The PG formed clusters instead of a surrounding monolayer, suggesting that PG make integrin-mediated associations that maintain their distribution (Xie, 2014).

The PG migrate between the CG and the basal ECM, and loss of focal adhesions led to a disruption in PG migration, suggesting that integrin complexes on one or both surfaces play a role in mediating glia migration. The αPS2/βPS heterodimer binds ligands containing the tripeptide RGD sequence and αPS3/βPS binds laminins, so either or both could mediate adhesion of the PG to the ECM. However, it appears that depletion of the integrin complex in apposing glial layers is necessary to disrupt glial migration into the OS, as MARCM clones within the PG alone did not disrupt migration. Integrin function is conserved in mediating glial cell migration either on ECM or neighboring glial surfaces. For example, vertebrate glial studies found that integrins are involved in astrocyte, oligodendrocyte precursor and Schwann cell migration on various ECM molecules. Loss of β1-integrin in Bergmann glia leads to mislocalization, ectopic migration and disruption of process growth within the vertebrate cerebellum. ILK and CDC42 within Bergmann glia are required for the β1-integrin-dependent control of process outgrowth (Xie, 2014).

Reduction of focal adhesions disrupted the CG sheath and the integrity of the blood-nerve barrier, suggesting that maintenance of the CG tube also requires integrin-mediated adhesion. The disruption of the CG tube is similar to observations made in vertebrates, in which glial tubes are necessary for chain migration of neuroblasts along the rostral migratory stream; β1-integrin plays a role in both chain migration and maintenance of the glial tubes. However, the link between the integrin complex and the formation or stabilization of the CG tube is currently unknown (Xie, 2014).

The integrin complex appears to play a limited role in the migration of the WG into the OS. After knockdown of βPS or talin, WG with normal bipolar membrane processes were observed and WG mys1 MARCM clones had morphologies similar to control clones, suggesting that integrin signaling is not required for migration in differentiated WG. TEM analysis suggests that the WG failed to properly ensheath and segregate the bundles of photoreceptor axons, a phenotype consistent with that observed in vertebrate glia, although it is also possible that the lack of WG ensheathment is a secondary effect of axon stalling (Xie, 2014).

Loss of integrin complexes resulted in a failure of photoreceptor axons to exit the ED, navigate the OS or correctly target the optic lobe. Previously, blocking glial migration from the OS using dominant-negative Ras1 (Ras85D -- FlyBase) resulted in photoreceptor axons stalling in the ED but not the OS. The phenotype suggested that photoreceptor axons require physical contact with retinal glia to exit the ED. However, this mechanism does not seem to apply to the stalling phenotype observed with knockdown of the integrin complex. In the majority of samples, glia were still present in the ED and around the axonal stalling region, suggesting that the axon stalling phenotypes are likely to be due to a different mechanism (Xie, 2014).

It is possible that axon stalling results from a combination of glial changes affecting the multiple subtypes of OS glia and disruption of both the βPS/αPS2 and βPS/αPS3 adhesion complexes. Only the simultaneous loss of both αPS2 and αPS3 triggered the axonal phenotypes. Similarly, knockdown of βPS or talin within individual glial subtypes does not trigger axon stalling, whereas disruption of adhesion complexes in all glial subtypes does. It might be that only the repo-GAL4 driver is sufficiently strong or expressed early enough for the effective knockdown of integrin or talin to trigger the axon stalling phenotypes. However, the axon stalling phenotype can be effectively produced by delaying the expression of the RNAi with the repo-GAL4 driver until the second instar, suggesting that early expression is not key. Overall, the results suggest that it is the combined loss of the focal adhesion complex in multiple glial layers that led to the axon stalling phenotype, although the underlying mechanism is not known. It is possible that the simultaneous aggregation of the stalled PG and disruption of the CG sheath triggers axon stalling by allowing ectopic PG to enter the center of the OS or form the glial cap. The ectopic PG within the axon stalling area and in the glial cap were likely to be PG given their expression of LanB2, Apt and the lack of the WG Gli-lacZ marker. Normally, the PG migrate into the ED and differentiate into WG in the presence of photoreceptor axons. However, in the RNAi-treated OS many of the Apt-positive glia also expressed Gli-lacZ, suggesting a change in the normal differentiation pathway, perhaps owing to the premature and ectopic contact of the PG with the photoreceptors within the OS. Loss of integrins throughout the entire glial population could also lead to global changes to the ECM, as loss of integrins can alter the deposition of ECM components during epithelial morphogenesis. Although loss of integrins in the PNS does not lead to changes in the neural lamella of the peripheral nerve, it is possible that ECM changes in terms of structural integrity or the ability to recruit protein components could result in the multiple glial morphological changes (Xie, 2014).

In summary, this study has shown that glia in the OS and ED express integrins and Talin, through which they receive external signals important for PG migration, organization and CG barrier formation. The combined impact of integrin complexes on the morphology and development of both glial layers is crucial for proper axonal outgrowth through the OS and targeting in the brain (Xie, 2014).

Alpha-Spectrin and integrins act together to regulate actomyosin and columnarization, and to maintain a mono-layered follicular epithelium

This study reports the role of Spectrins during epithelia morphogenesis using the Drosophila follicular epithelium (FE). α-Spectrin and β-Spectrin are shown to be are essential to maintain a mono-layered FE, but, contrary to previous work, Spectrins are not required to control proliferation. Furthermore, spectrin mutant cells show differentiation and polarity defects only in the ectopic layers of stratified epithelia, similar to integrin mutants. These results identify α-Spectrin and integrins as novel regulators of apical constriction-independent cell elongation, as α-spectrin and integrin cells fail to columnarize. Finally, increasing and reducing the activity of the Rho1-myosin-II pathway enhances and decreases multi-layering of α-spectrin cells, respectively. Similarly, higher myosin-II activity enhances the integrin multi-layering phenotype. This work identifies a primary role for α-Spectrin in controlling cell shape, perhaps by modulating actomyosin. All together, it is suggested that a functional Spectrin-Integrin complex is essential to balance adequate forces, in order to maintain a mono-layered epithelium (Ng, 2016).

This study found that in the germline α-Spec is not a major regulator of the Hippo pathway. Mutations in hippo, β-Spec or α-Spec result in a stratified FE, but contrary to previous interpretations, and unlike Hippo, spectrins are not required for the FCs to exit mitosis. The suggestion that Spec mutant FCs over-proliferate is thought to be an over-interpretation from the multilayering phenotype, as α-Spec cells were not checked for mitotic markers in a previous report. Again unlike hippo, α-Spec mutant PFCs only show defects in differentiation when they are located in the ectopic layers of the stratified FE, and oocyte polarity is largely unaffected in mutant egg chambers. It was recently shown that a β-Spec allele with a premature stop codon at amino-acid 1046 partially phenocopies hippo, with strong defects in FE integrity, actin organization and oocyte polarity. The null β-SpecG113 mutant allele behaves similarly to α-Spec mutants, showing Hnt defects mainly in ectopic layers, but Fas3 mislocalization in monolayers. More importantly, β-SpecG113 FCCs exit mitosis properly. The differences observed between the two β-Spec alleles are likely to be due to the fact that β-SpecG113 is a null allele (Ng, 2016).

In conclusion, α-Spec and β-Spec FCCs do not phenocopy hippo mutants when the cells are part of a monolayer, and they seem to adopt a partial hippo-like differentiation phenotype only when positioned at ectopic layers, even though α-Spec and β-Spec cells never divide after S6. Thus, the main function of the spectrin cytoskeleton in FCs is not proliferation control or regulation of the Hippo pathway, although an interaction between spectrins and Hippo might occur once the FCs are within an aberrantly organized FE. The function of spectrins in FCs is in contrast with other tissues, where α- and β-Spec appear to regulate growth through Hippo (Ng, 2016).

Similar to Hippo, α-Spec and β-Spec are required for the FE to maintain a monolayer. There is an increase in the multilayering phenotype in egg chambers with large clones from S3/6 to S7/8 and 100%, respectively. Also, the presence of control cells in α-Spec mosaic epithelia aids the mutant cells to maintain a monolayer from S6, as there is a higher percentage of S7-9 egg chambers with multilayers when the FE contains large α-Spec clones than when the mutant clone is only at the posterior end. The control of FE architecture appears to be mediated by the lateral spectrin network. Loss of α-Spec seems to disrupt both lateral (α/β) and apical (α/βH) spectrin-based membrane skeleton (SBMS) in the FE, as β and βH subunits are no longer localized laterally and apically in α-Spec cells, but no multilayering was reported for βH-Spec egg chambers, in which a loss of apical α-Spec was observed, suggesting that the loss of the lateral α/β is responsible for the FE stratification. Also, βH-Spec is mislocalized in sosie mutants, but the FE architecture is maintained (Ng, 2016).

Incipient SJs are first detected between the FCs with the completion of proliferation at S6. This study shows that the localization of several SJ components is affected in α-Spec FCCs, suggesting that spectrins are required for proper SJ formation. This is further supported by other observations. (1) Fas3 localization is affected in β-Spec FCCs. (2) Neuroglian (an SJ component) is required for maintaining the stability of the FE. (3) The reduction of both α- and β-Spec leads to mislocalization of Dlg, Neuroglian and Fas2 in neuromuscular junctions. (4) It has been suggested that the SBMS and ankyrin associate with SJ components (Ng, 2016).

As the mislocalization of SJ components in Spec mutant FCCs is observed in monolayers, and thus prior to the onset of stratification, it is speculated that Spec-dependent distribution of SJ components might contribute to the Spec function in the epithelium. This idea is supported by Crumbs overexpression, which leads to defects in SJs and ZA, and multilayering of the ectoderm cells, and by dpak (Pak - FlyBase) FCs, which mislocalize Fas3 and show multilayering and columnarization defects. Furthermore, the aberrant accumulation of Fas2 at the lateral membrane of Tao FCs prevented membrane shrinking in the cuboidal-to-squamous transition. However, fas3, fas2 and cora mutant cells do not show shape defects or multilayering. Thus, if SJ components contribute to the α-Spec phenotype at all, it might be not because they are absent in α-Spec mutant cells, but because they are not properly distributed (Ng, 2016).

Transitions between squamous, cuboidal and columnar epithelial cell shapes are common during development, and contribute to the morphogenesis of tissues. This study demonstrates a cell-autonomous role for α-Spec in promoting the cuboidal-to-columnar shape transition of the FCs. It is important to point out that the FE undergoes lateral elongation without apical constriction, which might allow phenotypes to be interpreted in a simpler manner. This morphogenetic FC behavior is similar to that of vertebrate neuroepithelia, where cell elongation precedes apical constriction, and it would be interesting to study the function of Spec in the columnarization of these cells (Ng, 2016).

Although the molecular mechanism of apical constriction-independent cell elongation is unknown, a primary role for the SBMS is thought to lie in facilitating changes in cell shape, which is further supported by the cell shape defects in α-Spec gut epithelia, perhaps by contributing to the proper distribution of adhesion molecules. This function of the SBMS in membrane biology is conserved in other cells, as spectrins stabilize the plasma membrane during blastoderm cellularization, and control photoreceptor morphogenesis through the modulation of membrane domains. The spectrin cytoskeleton might also impact on FE columnarization by interacting with the actomyosin cytoskeleton. It is known that apical-basal elongation in cytoplasmic actin-binding protein drebrin E (drebrin 1) depleted human Caco2 cells is impaired, as a possible consequence of the lack of interaction between drebrin E with spectrins and actomyosin. Also, the elongation of neuroepithelial cells depends on the assembly of an actomyosin network in the apical junctional complex, regardless of whether cells are constricting or not . In Drosophila wing discs, the Rho1-Myosin II pathway at the apicolateral membrane seem to regulate the cuboidal-to-columnar shape transition, whereas in the germline, Rok and sqh mutant FCs fail to adopt a normal shape. Finally, SBMS seems to modulate cortical actomyosin contractility in the eye, and possibly in the FE. Together, these data suggest that Myosin II activity is aberrant in α-Spec mutant FCs, contributing to defects in columnarization and FE architecture (Ng, 2016).

Increasing Rho1 and Sqh activities enhances the Spec multilayering phenotype, whereas reducing Myosin II activity decreases it. In addition to this functional link between the SBMS and the Rho-Myosin pathway, this study also shows that mys cells fail to columnarize, and that an extra copy of sqh increases the mys multilayering phenotype. It has been shown that integrins regulate the Rho-Myosin pathway to induce actomyosin-generated forces. Thus, as is the case for spectrins, integrins might also control cell shape and epithelia morphogenesis by modulating the actomyosin activity (Ng, 2016).

How the SBMS and integrins might modulate actomyosin is unknown, and one possible mechanism is by regulating Myosin II activity directly. However, an alternative mechanism is proposed. Spectrins can bind F-actin, and integrins and spectrins interact with proteins involved in the association of F-actin with the membrane. Furthermore, α-Spec and integrins regulate the actin cytoskeleton through Rac. Previous studies have shown that both β-Spec and mys mutant FCs display similar defects in the basal level of F-actin, which are recapitulated in α-Spec mutant cells. Thus, any defects in actin organization in mys and Spec mutant FCs could in turn result in defects in the activity of Myosin II (Ng, 2016).

Regardless of whether integrins and spectrins regulate F-actin or myosin, or both, spectrins and integrins might act together. The SH3 domain of α-Spec interacts with Testin ortholog (Tes), a component of integrin-dependent focal adhesions, and mammalian αII-Spec stabilizes β3-integrin anchorage, suggesting α-Spec as a physical link between focal adhesions and F-actin. In the FE, this study observed that α-Spec and αPS1 colocalize in the lateral, and possibly apical, membrane. In addition, it was shown that the localization of α-Spec in mys clones, and the localization of βPS in α-Spec mutant clones, is majorly unaffected. Furthermore, expression of a constitutively active integrin that reduces multilayering of mys FCCs, failed to rescue α-Spec multilayers. Thus, it is proposed that α-Spec and integrins act independently of each other, but as part of the same functional complex regulating the actomyosin cytoskeleton and tissue architecture (Ng, 2016).

An early event following oncogenic mutations in an epithelium is the escape of the daughter cells from the monolayered epithelium, forming disorganized masses. Spindle orientation has been linked to tumor-like growth in various tissues, and this study found that there is a good correlation between spindle misorientation and 'tumor-like masses' at the FE: hippo, mys and α-Spec FCCs show misaligned spindles and severe multilayering, whereas Notch FCCs, which overproliferate, do not show multilayering or spindle orientation defects. However, perpendicular divisions alone are insufficient to promote stratification, and a mechanism, depending on lateral cell-cell adhesions, is in place to avoid multilayering as a sole consequence of spindle misorientation. It is proposed that spindle misorientation contributes to FE disorganization, but that this 'safeguard' mechanism is somehow inactive in hippo, mys and Spec mutant FCCs. What other aspect of the mutant phenotypes might then be linked to multilayering? A clue might come from the Spec mutant and mys FCCs. First, there is an increase in the α-Spec multilayers after S6, when both FCs and egg chambers undergo various morphogenetic changes. Second, the volume of the germline surrounded by large α-Spec FCCs appears smaller. And third, Myosin II activity is increased in α-Spec and mys mutant cells. In this interpretation of the results, a proper distribution of Myosin II activity in a Spec- and integrin-dependent manner allows the right amount of forces to be distributed across the membrane and the epithelium. Thus, it is possible that proper cell-cell interactions, adequate force balance and precise spindle orientation are key to maintaining a monolayered epithelium, especially upon the mechanical stress induced by morphogenesis (Ng, 2016).

myospheroid: Biological Overview | Evolutionary Homologs | Regulation | Developmental Biology | 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.