Abl tyrosine kinase

DEVELOPMENTAL BIOLOGY

Effects of Mutation or Deletion (part 2/2)

Profilin and the Abl tyrosine kinase are required for motor axon outgrowth in the Drosophila embryo

The ability of neuronal growth cones to be guided by extracellular cues requires intimate communication between signal transduction systems and the dynamic actin-based cytoskeleton at the leading edge. Profilin (Drosophila Chickadee), a small, actin-binding protein, has been proposed to be a regulator of the cell motility machinery at leading edge membranes. However, any requirement it may have in the developing nervous system has been unknown. Profilin associates with members of the Enabled family of proteins, suggesting that Profilin might link Abl function to the cytoskeleton. In a genetic screen in Drosophila to identify genes required for the correct navigation and outgrowth of motoneuron growth cones two alleles of a stranded (sand) mutation were recovered in which motor growth cones arrest before reaching their final targets. The molecular genetic analysis reveals that stranded alleles are zygotic lethal mutations in chickadee. In vitro experiments confirm that axon extension is impaired in Profilin mutants. Moreover, phenotypic comparisons and genetic interactions between chic and abl mutants support the notion that Profilin and Abl cooperate to promote axon extension. Genetic analysis in Drosophila has been used to demonstrate that mutations in Profilin (chickadee) and Abl (abl) display an identical growth cone arrest phenotype for axons of intersegmental nerve b (ISNb). Moreover, the phenotype of a double mutant suggests that these components function together to control axonal outgrowth (Wills, 1999a).

Although axon outgrowth defects are observed in all motor pathways in chic alleles, supporting a general role for Profilin in all aspects of axon outgrowth, these defects arise only late in development, when motor growth cones must navigate certain key choice points. Since Profilin is expressed abundantly during oogenesis, it is possible that a maternal supply of Profilin protein can function in the absence of zygotic Profilin for the initial stages of embryogenesis, thus masking a more general role for Profilin. To test whether maternally supplied Profilin function might modulate the zygotic phenotype of chic^sand mutants, the phenotypes of embryos were compared from mothers with one or two copies of the wild-type Profilin gene. Comparing embryos with the same zygotic genotype (chic^sand/Df), it was found that those derived from mothers carrying a duplication at the Profilin locus show substantial rescue of the stranded axonal phenotype. Thus, doubling the supply of maternally expressed Profilin rescues the chic mutant zygote through embryonic stage 17. The failure of motor growth cones at particular locations in the periphery of chic^sand embryos is likely to reflect regions with the highest requirement for Profilin function (Wills, 1999a).

The choice point regions where chic mutant growth cones frequently arrest correspond to locations where growth cones typically slow down, become more complex in shape, and probe their environment. Although it is thought that the failure of these mutant growth cones reflects an intrinsic deficit in forward locomotion, it is also possible that chic mutant growth cones are unable to respond to specific extrinsic guidance cues. Therefore, the context dependence of chic mutant axon outgrowth was assessed by measuring outgrowth from dissected nerve cords in vitro. In these experiments, mutant and wild-type nerve cords were removed from embryos at embryonic stage 16 and cultured on poly-L-lysine-coated coverslips. Under these in vitro conditions, regenerating axons begin extending from transected motor nerve roots and central nervous system (CNS) longitudinal connectives after 3 hr and continue to extend for the next 8-9 hr, long after growth cones would have arrested in mutant embryos. In this assay system, homozygous chic^sand nerve cords produce Fasciclin II- (Fas II-) positive neurite fascicles, but the mutant axons do not extend as far as wild-type controls. Neurite fascicles extend 50 µm on average from wild-type nerve cords, while chic mutant fascicles extended 32 µm on average, only 64% of wild-type growth. These results support the conclusion that Profilin plays a role in general axonal extension, regardless of the environment, and suggest that the axonal defects in chic embryos result from loss of Profilin function in the mutant axons and not in the surrounding tissue (i.e., a cell autonomous function for Profilin in motoneurons) (Wills, 1999a).

The discovery that Profilin interacts biochemically with members of the Enabled family suggests that Profilin might provide a link between the Abl tyrosine kinase pathway and the actin cytoskeleton. Although axonal phenotypes have been observed in ena mutants, previous studies employing general axon markers have identified defects only when abl mutations were combined with mutations in other axon guidance genes. However, using the mAb 1D4 antibody to examine motor pathways during late embryonic development (stage 17), a previously unappreciated growth cone arrest phenotype was observed in the ISNb projection of abl homozygous mutant embryos that is essentially identical to the ISNb phenotype of chic mutants. In abl mutants, ISNb axons frequently stop at contact with muscle 13 and/or the adjacent muscle 30, failing to reach the distal target muscle 12. Less frequently, abl mutant ISNb axons stop earlier at contacts with muscles 14 or 28; such defects are rare in wild-type controls but less penetrant in abl mutants than in strong chic backgrounds. Other peripheral motor axon pathways appear normal in abl mutants, as assessed with mAb 1D4. Abl was shown to require an active kinase domain to function normally in ISNb development (Wills, 1999a).

In addition to the requirement for Profilin and Abl in ISNb development, both components are also necessary for the accurate formation of axon pathways within the CNS. Staining of chic or abl mutant embryos with mAb 1D4 reveals similar disorganization in the parallel longitudinal fascicles of Fas II-positive axons on either side of the CNS midline. Although the prevalent phenotype observed in both single mutant genotypes is mild, a range of defects can be seen, from mild, to intermediate, to extreme. The prevalent defects are not likely to be a product of alterations in CNS cell fates, since patterning in strong chic^sand mutants has been assessed with several different antibody probes. In mildly effected embryos, longitudinal pathways are often diverted, causing fusions and/or breaks in these fascicles; occasionally, inappropriate midline crossing can be seen. In embryos with intermediate effects, Fas II-positive axons often cross the midline barrier, in addition to the collapse of longitudinal fascicles. In embryos with extreme effects, axonal connections between segments along the anterio-posterior axis are often absent, consistent with a major failure in axonal extension; this extreme phenotype is very rare in chic or abl single mutants. In such extreme cases, defects are also observed in muscle patterning; such defects have been found in abl mutants (Wills, 1999a and references).

The similarity between the chic and abl phenotypes, both in the CNS and periphery, raised the question of whether these genes cooperate in axonal development. To determine if the function of Profilin is sensitive to the amount of Abl, as expected for components in the same pathway, chic homozygous embryos that lack one allele of abl were tested. Two-fold reduction of Abl function in the chic background results in a dramatic shift in the distribution of CNS axon phenotypes; in these embryos, the extreme phenotype increases 10-fold in comparison to chic mutants alone. This dose-sensitive genetic interaction suggests that Profilin and Abl cooperate in the same overall process (Wills, 1999a).

The guanine nucleotide exchange factor trio mediates axonal development in the Drosophila embryo

The ISNb and longitudinal pathway defects observed in trio mutants are similar to those of phenotypes observed in embryos mutant for the Abl tyrosine kinase. Previous analysis has shown that a partial reduction in Abl function suppresses the bypass phenotype caused by mutations in the RPTP Dlar, implying an antagonistic relationship between kinase and phosphatase. To address trio function at this ISNb choice point, ISNb pathfinding was examined for dosage-sensitive interactions between Dlar and trio. In strong zygotic Dlar mutants, ISNb bypass was observed at a moderate frequency (18.4%, A2-A7 hemisegments). However, partial reduction of trio activity in this Dlar background enhances the ISNb bypass ~2-fold. Although this potentiation disagrees with a simple model in which trio and Abl function together to oppose phosphatase signaling, it is consistent with the observation that neural expression of Drac1^N17 enhances the frequency of bypass in Dlar mutants. Thus, although trio may collaborate with Abl at the CNS midline, it rather appears to cooperate with Dlar and Drac1 during ISNb ventral target entry. The absence of bypass phenotypes in trio single mutants is likely to reflect the existence of additional inputs to Rac family GTPases that would be susceptible to the Drac1^N17 dominant-negative effect (Bateman, 2000).

Dosage-sensitive, reciprocal genetic interactions between the Abl tyrosine kinase and the putative GEF trio reveal trios role in axon pathfinding

Dosage-sensitive genetic interactions between trio and Abl have been documented. A number of observations support the interaction of Abl and Trio in a common regulatory network. First, the dosage-sensitive genetic interactions between trio and Abl are reciprocal, as assayed by either viability or CNS architecture. Heterozygous mutations in trio worsen the Abl mutant phenotype, while heterozygous mutations in Abl worsen trio mutant phenotypes. A background of compromised signaling (Abl¹/Abl⁴, Df(3L)FpaI/trio^M89, or trio^P0368/10/trio^M89) is enhanced by reduced activity of another member of this network (Liebl, 2000).

As further evidence for the involvement of Abl and Trio in a common signaling network, the Abl and trio homozygous mutant phenotypes show a synergistic interaction. Neither the Abl mutant background nor the trio mutant background have dramatic phenotypic consequences on CNS architecture. Phenotypes similar to those reported here have been observed in a variety of trio mutant combinations (Awasaki, 2000; Bateman, 2000). However, combining these two backgrounds to generate trio, Abl homozygous mutant embryos results in dramatic disruption of the CNS scaffolding. Taken alone, this synthetic enhancement may represent common or independent signaling pathways involving Abl and Trio. However, combined with the dosage-sensitive interactions between Abl and trio observed, it is likely that these molecules are involved in overlapping or interdependent networks. Similar synergistic effects between Abl and fax, and Abl and dab, have been reported. In addition to the reciprocal genetic enhancement between Abl and trio, a null allele of fax (fax^M7) greatly worsens the trio hypomorphic mutant's viability, while a dab null allele weakly modifies this background (Liebl, 2000).

The ena gene was identified through its ability to suppress the Abl mutant phenotype. Since reductions in ena compensate for the absence of Abl, it has been hypothesized that a precise balance between Abl and Enabled activity is required for viability. Similar to the genetic interaction between Abl and ena, heterozygous mutations in ena can partially alleviate the trio mutant phenotype. One interpretation of this interaction is that a balance between trio and ena is required, and Trio may possess a biochemical function that is antagonistic to Enabled's. Since neural enriched isoforms of Mena, the murine homolog of Enabled, are believed to be involved in filopodia formation to extend the growth cone, a potential antagonistic role for Trio is the retraction of growth cones. Drosophila Trio's second DH domain stimulates the formation of stress fibers in REF-52 cells (Newsome, 2000). In neurons, the formation of similar actin-myosin contractile filaments leads to neurite retraction. Therefore, a balance between the biochemical activities of Enabled and Drosophila Trio may be required for a proper balance between extension and retraction of the growth cone in response to attractive and repulsive pathfinding cues (Liebl, 2000).

four-jointed interacts with dachs, abelson and enabled and feeds back onto the Notch pathway to affect growth and segmentation in the Drosophila leg

The molecular basis of segmentation and regional growth during morphogenesis of Drosophila legs is poorly understood. four-jointed is not only required for these processes, but also can direct ectopic growth and joint initiation when its normal pattern of expression is disturbed. These effects are non-autonomous, consistent with the demonstration of both transmembrane and secreted forms of the protein in vivo. The similarities between four-jointed and Notch phenotypes led to an investigation of the relationships between these pathways. Surprisingly, it was found that although four-jointed expression is regulated downstream of Notch activation, four-jointed can induce expression of the Notch ligands, Serrate and Delta, and may thereby participate in a feedback loop with the Notch signaling pathway. four-jointed interacts with abelson, enabled and dachs, which suggests that one target of four-jointed signaling is the actin cytoskeleton. Thus, four-jointed may bridge the gap between the signals that direct morphogenesis and those that carry it out (Buckles, 2001).

Similar mutant phenotypes may indicate that the genes causing them may act in the same molecular pathway. dachs and abl mutant phenotypes imitate those of fj, thus both of these genes are attractive candidates for the fj signaling pathway. A major substrate for Abl kinase activity is the Ena gene product. Ena homozygotes are embryonic lethal and imaginal phenotypes are not known. However, Abl and Ena appear to function in the same pathway in Drosophila. Finally, given the molecular epistatic interactions observed between fj and Ser, whether fj and Ser interact genetically was tested (Buckles, 2001).

To test the relationships of these genes, two hypomorphic alleles of fj, fj⁴ and fj^N7 were used. The majority of legs of fj^N7 flies retain partial joints of a ball and socket morphology at the juncture between T2 and T3, while fj⁴ produces larger partial joints or complete joints at the T2/3 boundary. Introduction of one mutant copy of dachs, abl, ena, or Ser into these backgrounds significantly increases the severity of the hypomorphic phenotypes, while each of these genes by itself is wholly recessive in the leg. Thus, dachs, abl, ena, and Ser act as dominant enhancers of fj, suggesting these genes may be part of a common pathway or network (Buckles, 2001).

Loss-of-function abl alleles are recessive, and heterozygous abl flies have normal leg morphology. However, when one copy of abl was removed in a strong fj background, the complete or partial loss of an additional tarsal joint at T1/2 on one or more legs was observed in one third of the animals. A wild-type abl transgene (Tnabl⁺) can restore this joint, confirming that abl is indeed responsible for the interaction. To test if kinase activity is required for abl activity at this joint, a transgene with an inactive kinase (Tnabl^K-N) was inserted into the same genetic background. This transgene is unable to rescue the interaction, suggesting that abl kinase activity is indeed required. These results suggest that abl and fj participate in redundant pathways in the leg. Moreover, they demonstrate that fj is required at segment boundaries other than T2/3, which is the only boundary lost in fj null mutants. More widespread activity is consistent with the expression of fj at additional segment boundaries in the leg. These results reveal a concealed function for abl in leg morphogenesis (Buckles, 2001).

Abl kinase function partially masks a requirement for fj at the T1/2 segment boundary. While fj is expressed at all tarsal segment boundaries, it appears to be required for segmentation only at T2/3, although rare instances of partial loss of T1/2 have been observed. However, in the absence of one copy of abl, additional loss of the T1/2 boundary is seen in a third of all animals. This is interpreted to mean that additional information, modified by or working through abl, acts together with fj to elaborate that boundary. Most abl homozygous flies have normal legs, although occasional severe truncations of the legs in individual abl flies are observed. In addition, over-expression of abl causes tarsal truncations and segment fusions. A likely target of Abl activity is Ena. However, Abl is not the only tyrosine kinase that phosphorylates Ena, and this multiple regulation may in part explain the variability of abl loss-of-function phenotypes. A critical test of whether Ena is indeed part of a common pathway for the morphogenetic work of segmentation at all leg segment boundaries will be to examine loss-of-function ena clones in the leg. Alternatively, the common pathway at each segment boundary may be the actin cytoskeleton itself, with multiple inputs providing redundancy, and overlapping position-specific regulators competing or cooperating to regulate the state of actin dynamics in each cell. Indeed, in the larger sense, Fj itself may serve to coordinate information provided by multiple signaling pathways (Notch, Jak/Stat, and Wingless, which all regulate fj) with alterations in the actin cytoskeleton that ultimately have morphological consequences (Buckles, 2001).

Notch regulates abl signaling in axon guidance

Abl is an axonal tyrosine kinase that has yet to be clearly linked to a receptor; Notch is a receptor for which the signaling pathway remains incompletely understood. Genes that interact synergistically with Abl are collectively termed HDA loci (haploinsufficient, dependent on Abl). Similar synergistic genetic interactions are often found in genes whose products interact directly, such as the different constituents of multiprotein complexes. While it has not been shown directly that HDA loci encode proteins that associate with Abl, the sequence of Dab makes it a good candidate to bind to the Abl SH2 domain, and indeed the mouse Dab homolog binds to mouse Abl in vitro and to the closely related SRC SH2 domain in vivo. Similarly, the Abl-interacting gene ena is thought to encode a direct substrate of the Abl kinase. Notch and Abl mutations are shown to interact synergistically to produce synthetic lethality and defects in axon extension. These axonal aberrations cannot be accounted for on the basis of changes in cell identity, as the Notch/abl interaction is shown not to cause neurogenic or anti-myogenic phenotypes. Notch is shown to be present in the growth cones of extending axons, and the Abl accessory protein Disabled binds to a signaling domain of Notch in vitro. It is therefore speculated that Disabled and Abl may play a role in Notch signaling in Drosophila axons, perhaps by binding to the Notch intracellular domain (Giniger, 1998).

The gross morphology of the nervous system is typically normal in N/abl embryos, but specific axon tracts fail to develop. Axonal defects are observed in all of the nerve tracts that are known to require Notch, i.e., the CNS longitudinal tracts between neuromeres and the lateral portion of the ISN. In contrast, longitudinal tracts within each neuromere and commissural tracts appear normal, as does the dorsal and ventral portions of the ISN. The penetrance (fraction of embryos affected) and expressivity (number of affected hemisegments per affected animal) of the N/abl axonal phenotype depend on the particular combination of alleles used (Giniger, 1998).

In principle, the axonal defects observe in mature N/abl embryos could reflect a failure either to form axon tracts or to maintain them. Moreover, if the defect is in the initial development of the axon, it could be due to the absence of required substratum cells, the absence or improper identity of the neurons themselves, or else the failure of the actual guidance machinery of the growth cone. To discriminate among these possibilities, the development of pioneer neurons and substratum cells were examined directly for affected axon tracts. The initial extension of pioneer axons were examined in N/abl embryos. Consistent with the terminal phenotype, the combined MP fascicle, the first to form between successive neuromeres, is obviously aberrant from a very early stage (st 13). In contrast, both the anterior and posterior commisures appear to develop normally, as do the longitudinal tracts within the neuromeres. The substratum cells for affected axon tracts were examined. The MP fascicle projects between neuromeres on a specific Fasciclin II-expressing glial cell, LG5, and this is present in affected hemisegments. In the PNS, the direct cellular substratum for ISN extension in the lateral part of the embryo is a cluster of lateral peritracheal cells that lie along the trachea. Examination of stalled motor axons in an N/abl embryo shows that the nerve frays and stalls precisely as it attempts to grow along the trachea. Since substratum cells for affected axon tracts are present in N/abl embryos, the pioneer neurons themselves were examined. The positions and cell body morphologies of the sensory neurons in the PNS provide sensitive assays for the identities of these cells, and these typically appear to be wild type. The neuron aCC that pioneers the ISN and innervates the most dorsal muscle (muscle 1) is readily apparent in N/abl embryos. The neurons that pioneer the MP fascicle within the CNS are MP1, pCC, dMP2, and vMP2, and cells whose positions and axonal morphologies are appropriate for these cells can be seen in affected hemisegments of N/abl embryos (Giniger, 1998).

The observation of morphologically normal pioneer neurons and substratum cells in N/abl embryos is surprising, since perturbation of cell identity seems a priori to be the simplest explanation for the axonal defects in these embryos. Molecular markers for the development of affected pioneer neurons were therefore examined to determine whether their identities were disturbed in some more subtle way. The homeobox proteins Ftz and Eve are expressed in the pioneer neurons aCC and pCC, and changes in the expression of either protein disrupts the guidance of some axons. Eve expression is wild type in these cells in 98% of hemisegments of N^ts/Abl embryos and >85% of hemisegments of DfN⁸/+;abl/abl embryos. Ftz is wild type in these cells in 98% of hemisegments of N^ts/Abl embryos (Giniger, 1998).

Particularly telling tests of neuronal identity in the CNS of N/abl embryos are provided by analysis of Eve expression in the neuron RP2 and Odd protein expression in the pioneers of the MP fascicle. Notch controls the identities and projections of RP2 and of the MP2 progeny cells, and the effect of Notch on the fates of these cells can be assayed by their expression of Eve and Odd, respectively. These are, however, among the neurons whose axons are also affected by the Notch/abl interaction. Thus, if the axonal defects observed in N/abl embryos are due to Notch-dependent alterations of cell identity, one should be able to detect precisely these alterations by assaying the expression of Eve and Odd. In wild-type embryos, Eve is expressed in RP2 but not in its sibling cell (RP2sib). Eve expression is wild type in RP2 and RP2sib in 98% of hemisegments of N^ts/Abl embryos and >88% of hemisegments of DfN⁸/+;abl/abl embryos. At the time that the MP fascicle is pioneered, Odd protein is expressed in the MP1 and dMP2 neurons but not in vMP2 (the sibling cell to dMP2). Notch is responsible for differentiating the fates of dMP2 and vMP2. 97% of st 13/14 N^ts/Abl hemisegments have the proper pattern of Odd-expressing cells. Moreover, upon double staining a N/abl embryo with anti-Odd and anti-FasII, appropriate Odd staining is observed even in a hemisegment in which the MP fascicle has failed to develop. This argues directly against the model that the failure to form the MP fascicle in N/abl embryos arises from a Notch-dependent transformation in the identities of the dMP2 and vMP2 pioneer neurons. Similarly, the notion that PNS axonal defects in N/abl embryos might arise from a Notch-dependent interconversion of identity between sensory neurons and their sibling glia is inconsistent with the observation that the peripheral pattern of 22C10 expression (a marker for PNS neurons: see Futsch) and of Pros expression (a marker for sense organ glia) is generally unaffected by the N/abl interaction (Giniger, 1998).

The experiments above suggest that most of the axonal defects in Notch/abl embryos cannot be accounted for on the basis of observed transformations of pioneer neuron identity. The converse question was therefore asked: whether Notch-dependent transformations of pioneer neuron identity are sufficient to produce axonal defects like those observed in N/abl embryos. Indeed they are not. Embryos were prepared that were N^ts1;elav-GAL4;UAS-Notch and they were shifted to restrictive temperature in mid-embryogenesis. In these embryos, the endogenous Notch is inactivated by the temperature shift after the completion of neuroblast segregation but during the time when neuronal identities are still being specified and prior to axonogenesis. The GAL4 system then restores wild-type Notch to each neuron at about the time it begins to extend its axon, after its identity has been decided. The characteristic pattern of Notch-dependent axonal defects are found in >90% of N^ts1 embryos subjected to a standard temperature shift protocol. In contrast, the axon scaffold of the CNS is rescued to wild type or nearly wild type in >80% of N^ts1;elav-GAL4;UAS-Notch embryos. As assayed by staining with anti-FasII, 49% of embryos show rescue of longitudinal tracts in all hemisegments and 32% of embryos show residual defects in just a single hemisegment. In only 19% of cases do N^ts1;elav-GAL4;UAS-Notch embryos have CNS axonal aberrations that overlap in severity with those observed in the N^ts control. By monitoring the expression of Odd and Eve, it was verified that expression of wild-type Notch via elav-GAL4 does not rescue Notch-dependent defects in cell identity. In temperature-shifted N^ts embryos, 5.7 ± 1.6 Odd-positive neurons are found per neuromere, versus 4 Odd⁺ cells in wild type. By comparison, the number of Odd⁺ cells found in N^ts1;elav-GAL4;UAS-Notch embryos is 5.1 ± 1.2. Analogous results were found for the Notch-dependent transformation of RP2sib to RP2, as assayed with anti-Eve. These data show directly that the Notch-dependent perturbations of cell identity induced in temperature-shifted N^ts embryos are not sufficient to produce the axonal defects observed in these embryos. They therefore provide strong evidence that the requirement for Notch in axon patterning reflects a function of the protein at the time of axon outgrowth, genetically separable from the role of Notch in the establishment of cell identity (Giniger, 1998).

Abl is localized to developing axons: it is thought that Abl works in the axon directly to control cytoskeletal organization and function. Might Notch also act in the axon to control axon extension directly? Notch is known to be present in mature nerves, but its presence in developing nerves, and specifically in growth cones, has not been investigated. Since Notch expression in substratum cells interferes with visualizing growth cones in situ, the localization of Notch protein was examined in primary Drosophila neurons cultured in vitro. Primary fly embryo neurons were differentiated in culture and analyzed either by indirect immunofluorescence with anti-HRP, to characterize neuronal morphologies, or with anti-Notch. All samples were also labeled with anti-Elav, to verify that the cells being examined were neurons. Notch protein is clearly detected on the entire cell surface, including extending axons, and on a variety of bulbous, spiked, and flattened structures at the tips of axons, which have the appearance of growth cones. To test further whether the Notch-containing structures at the ends of axons are bona fide growth cones, cell preparations were double labeled for Notch and for a known growth cone marker, kinesin-ß-galactosidase. Notch protein is present on the growth cones of axons extending in culture (Giniger, 1998).

What might be the physical basis of the N/abl genetic interaction? It is unlikely that the absence of Abl is affecting Notch protein levels, since Western analysis of extracts from homozygous abl^- females detects wild-type amounts of Notch protein. Such a mechanism would be expected to alter Notch-dependent cell identities as well as cell morphologies, and this does not generally occur. Might Abl bind Notch directly? This seems unlikely. While Abl contains a variety of protein interaction domains, Notch does not resemble its known ligands. It has recently been shown that the Drosophila Numb protein includes a PTB domain that binds two sites in the intracellular domain of Notch, even when Notch is not phosphorylated. Recalling that the Abl-interacting gene disabled includes a PTB domain closely related to the Numb PTB, and which like Numb can bind to nonphosphorylated targets, a test was performed to discover whether Dab can bind the intracellular domain of Notch in vitro (Giniger, 1998).

Three experiments demonstrate that the PTB domain of Drosophila Disabled binds directly to the intracellular domain of Notch in vitro. (1) First, beads bearing a glutathione S-transferase (GST) fusion of the Dab PTB domain were incubated in an extract of total embryo protein. Western analysis of the protein bound by Dab shows that GST-Dab selects Notch protein out of an embryo lysate, whereas GST alone binds only a small amount of Notch nonspecifically. (2) It was next asked what portion of Notch is recognized by Dab. Four protein fragments, each of which represents a distinct functional domain from the intracellular tail of Notch, were expressed. These are the RAM23 region (amino acids 1766-1896), the ankyrin repeats (amino acids 1896-2109), the PEST/OPA region (amino acids 2262-2606), and the notchoid region (amino acids 2612-2703). The four proteins were translated in vitro in reticulocyte lysates and assayed for binding to GST-Dab as above. Of the four Notch domains, only the RAM23 peptide binds to GST-Dab, while none of the four bind to GST alone. This pattern is similar but not identical to the pattern of Notch binding to the Numb PTB: like Dab, the Numb PTB binds to the Notch RAM23 domain but not to the ankyrin repeats or PEST/OPA region. Unlike Dab, Numb does bind to the notchoid domain. (3) Finally, to determine whether the Dab-Notch interaction is direct, a stable and soluble N-terminal fragment of the Notch intracellular domain (amino acids 1767-2235) was purifed from bacteria and its binding to the purified Dab PTB domain was assayed. The Notch intracellular domain is precipitated by GST-Dab beads but not by GST alone, demonstrating that the purified Dab PTB domain can bind directly to purified Notch intracellular domain in vitro (Giniger, 1998).

The data above demonstrate that Notch interacts genetically with Abl and biochemically with Disabled. These results beg the question whether Notch interacts genetically with disabled. Since isolated Dab alleles were not available, their genetic interactions with Notch could not be tested directly. It can be asked, however, whether flies that are triply heterozygous for all three mutations, Notch, abl, and Dab, display any synthetic phenotypes. Flies were constructed that are both heterozygous for a strong Notch allele (N⁸ or N^55e11) and for one of two unrelated chromosomes that bear strong mutations of both Abl and Dab. All pairwise combinations cause defects in eye development, giving rise to flies with rough eyes reminiscent of the defective eyes observed in Abl homozygotes (Giniger, 1998).

Given the genetic and biochemical evidence for Abl-Dab interaction, it is attractive to speculate that Dab may act as an adaptor protein that links Notch to Abl in response to a signal from Delta. Recruitment of Abl by Notch would in turn engage the actin cytoskeleton via mechanisms similar to those that have been studied in vertebrate systems. The notion that Notch may use distinct signaling pathways to control different downstream events is consistent with analysis of other signaling receptors. For example, receptor tyrosine kinases typically bind and activate a complex array of intracellular signaling proteins upon ligand induction, and different downstream signaling pathways are often responsible for different aspects of the induced phenotype. Finally, there is extensive precedent for receptors that control cell fate in some developmental contexts and cell motility or axon extension in others (Giniger, 1998 and references).

The central problem in axon guidance is to understand how guidance signals interact to determine where an axon will grow. A specific axon guidance decision in Drosophila embryos has been investigated, the sharp inward turn taken by the ISNb motor nerve to approach its muscle targets. This turn requires Notch and its ligand Delta. Delta is expressed on cells adjacent to the ISNb turning point, and it is known from previous work that Notch is present on axonal growth cones, suggesting that Delta and Notch might provide a guidance signal to ISNb. To induce the turning of ISNb axons, Notch interacts genetically with multiple components of a signal transduction pathway that includes the Abl tyrosine kinase and its affiliated accessory proteins. In contrast, genetic interaction experiments fail to provide evidence for a major role of the 'canonical' Notch/Su(H) signaling pathway in this process. It is suggested that the Notch/Abl interaction promotes the turning of ISNb axons by attenuating the Abl-dependent adhesion of ISNb axons to their substratum, thus releasing the axons to respond to attraction from target muscles (Crowner, 2003).

The receptor Notch is present on axons and growth cones and is required for extension of some early-growing 'pioneer' axons in the fly embryo. More recently, later functions of Notch in axon patterning have been investigated, using a temperature-sensitive Notch allele (Notch^ts1) to remove Notch activity well after most embryonic neuronal identities have been specified. In temperature-shifted mutant embryos, it has been found that ISNb axons reach their targets via an aberrant bypass trajectory, in which ISNb axons remain associated with the ISN. All Notch^ts embryos display the bypass phenotype, with 31% of hemisegments affected. Raising the temperature 1 hr earlier in development increases the expressivity of the ISNb bypass phenotype to 73% of hemisegments. Wild-type embryos subjected to the same temperature protocol, or N^ts embryos maintained at 25°, displayed few if any defects in ISNb defasciculation. Despite the aberrant pathfinding in Notch^ts embryos, formation of neuromuscular synapses to ventral longitudinal muscles occur as efficiently in temperature-shifted mutant embryos as in similarly treated wild-types (Crowner, 2003).

The site of Notch activation was localized by examining the Notch ligand, Delta. Temperature shifts of a temperature-sensitive combination of Delta alleles (Dl^6B37/Dl^via1) produced an ISNb bypass phenotype indistinguishable from that induced by Notch^ts. The Delta^ts mutant combination is not as 'tight' as Notch^ts1, so the reduced expressivity of the Delta phenotype relative to that of Notch is not surprising. Antibody staining revealed that at the time when ISNb is pioneered, Delta is expressed on cells very near to the first choice point, most prominently on the ganglionic branch of the trachea. This tracheal branch develops prior to ISNb outgrowth, the ISN grows in close association with the trachea, and ISNb axons separate from the ISN at that point where they first contact the trachea. The highest tracheal acccumulation of Delta protein is on the apical surface of the cells, in the tracheal lumen; however, Delta protein is also found on the basal surface of tracheal cells, available for interaction with ISNb axons. Adding back wild-type Delta to the trachea of temperature-shifted Delta^ts embryos (with btl-GAL4) rescues the Delta ISNb bypass phenotype. btl-GAL4 is expressed in midline glial cells in addition to tracheal cells; however, midline expression of Delta does not rescue ISNb trajectory in Delta^ts. Staining with an anti-tracheal antibody demonstrates that the ganglionic tracheal branch develops normally in temperature-shifted N^ts embryos (Crowner, 2003).

While tracheal expression of Delta is sufficient to restore ISNb defasciculation, ISNb still defasciculates properly in btl mutant embryos that lack trachea. Delta protein, however, is also detectable on nontracheal cells that abut the first choice point, and it is postulated that the Delta on these other cells might act redundantly with that on the trachea to provide a defasciculation signal for ISNb axons. The positions of these cells are consistent with some of them being peripheral glia, and indeed some of these cells label with Repo, a marker for glial cell nuclei. Embryos lacking glia show a low frequency of ISNb bypasses, and this frequency was substantially enhanced in embryos that simultaneously lack the trachea, consistent with the notion that both glia and trachea contribute to the defasciculation of ISNb at the first choice point (Crowner, 2003).

Two signaling pathways have been described for Notch. Notch controls cell fate and differentiation by a mechanism whereby a fragment of the receptor is released by proteolysis to transit to the nucleus as part of a transcriptional activation complex [the 'Su(H)/mam pathway'] and it controls axon patterning by regulating a signal transduction pathway defined by the Abl tyrosine kinase and its accessory genes, fax, dab, nrt, trio, and ena (the 'Notch/abl pathway') (Giniger, 1998). Genetic interaction experiments were performed; every mutant tested in the abl pathway displayed dominant genetic interactions with Notch in ISNb pathfinding. Removal of just one copy of the abl, neurotactin (nrt), or trio genes from Notch^ts embryos significantly suppresses the Notch bypass phenotype, and simultaneous reduction of abl and nrt suppresses the Notch phenotype more effectively than either heterozygous mutation by itself. Conversely, increasing Abl activity either by removing one copy of the abl antagonist, enabled, or by overexpression of abl significantly enhances the expressivity of the Notch bypass phenotype. Moreover, a bypass phenotype produced by overexpression of abl in wild-type embryos is suppressed by co-overexpression of Notch. In addition, it was found that reduction of Notch activity suppresses the ISNb zygotic mutant phenotype of abl homozygotes, mirroring the suppression of the Notch phenotype by reduction of abl documented above (Crowner, 2003).

In contrast to the strong genetic interactions of Notch with abl pathway genes in ISNb development, genetic tests failed to provide evidence indicative of a major role for the Su(H)/mam pathway in the Notch-dependent control of ISNb trajectory. Reducing the dosage of either Su(H) or mam did not enhance the Notch mutant phenotype, and expression of a signal-independent Su(H) did not suppress the requirement for Notch in ISNb. While these are negative results, it is noted that mam in particular displays strong, dominant-genetic interactions with Notch in a wide variety of genetic paradigms, while expression of Su(H)-VP16 mimics many of the effects of activated Notch, for example, blocking the development of ~98% of embryonic abdominal sensory neurons when expressed in wild-type embryos under control of the ectodermal GAL4 driver, 69B. Moreover, no evidence was found for aberrations in the identities or differentiation of ISNb neurons in temperature-shifted embryos. Taken together, these data suggest a simple hypothesis for how Notch and Abl cooperate to direct the trajectory of ISNb axons. ISNb guidance is known to reflect a competition between adhesion of these axons to the ISN pathway versus attraction to the target muscle field, and Abl activity is limiting for adhesion to the ISN pathway. It is now seen that defasciculation of ISNb axons requires Notch and Delta. Since the presence of Notch in postmitotic neurons and of Delta in the trachea is sufficient to fulfill their respective roles, it seems that the relevant activation of Notch must occur in the peripheral motor axons, where these cells touch. It is further found that Notch acts by antagonizing Abl activity in ISNb patterning -- mutations in these genes mutually suppress -- and thus it is suggested that the specific turn of ISNb axons arises from ligand-dependent attenuation of Abl pathway activity by Notch. By reducing the Abl-dependent adhesion of ISNb axons to the ISN pathway, Notch evidently makes the axons competent to respond to attraction from their target muscles. Consequently, they can now dive down into the ventrolateral muscle field. It is noted that it has not yet been determined whether the action of Notch, or Abl, is autonomous to the ISNb axons themselves. Cell-cell adhesion, and particularly axon fasciculation, is necessarily a bidirectional interaction. Since all ISN axons, dorsally directed ISN axons as well as ISNb axons, come into contact with the trachea and peripheral glia at the first choice point, regulation of ISNb-ISN defasciculation by Notch and Abl could plausibly reflect the function of these genes in ISNb, in ISN, or both. It is further noted that, in this model, the essential function of Delta and Notch is to prevent ISNb axons from remaining associated with the ISN dorsal to the choice point, not to act in isolation to set uniquely the position where ISNb turns. Delta and Notch function in ISNb patterning mainly to antagonize Abl, but even the strongest abl zygotic mutant does not cause premature defasciculation, presumably reflecting the presence of yet other factors that act in parallel to prevent ISNb axons from entering the muscle field prematurely. Thus, ubiquitous expression of Delta also does not cause premature turning of ISNb (Crowner, 2003).

The mechanism by which Notch antagonizes Abl activity is not yet clear. It is noted, however, that Notch, Abl, Disabled, Fax, Trio, and Enabled are all enriched in axons, and preliminary experiments show that Notch coimmunoprecipitates with both Disabled and Trio from wild-type fly lysates. It may be, therefore, that the Notch/abl genetic interactions observed here reflect a physical complex of Notch with Abl pathway proteins. It also seems remarkable that Notch can interact with the Abl signaling pathway in two apparently opposite ways. Where axons grow along Delta-expressing substrata, as in CNS longitudinal axon tracts, Notch works cooperatively with Abl: partial loss-of-function mutations of Notch and abl interact synergistically to produce synthetic phenotypes and embryonic lethality. In contrast, in ISNb, Notch and Abl act antagonistically: their gain- and loss-of-function phenotypes mutually suppress one another. It is imagined that there must be some additional factor that determines the 'sign' of the interaction between Notch and Abl. This could be a negative signal that prevents ISNb axons from growing along the Delta-expressing trachea, or a positive signal that allows Notch-dependent growth cone motility along other Delta-expressing substrata. The capacity of a single growth cone receptor to be switched between opposite functional modes has become a common observation in recent years, with cyclic nucleotides often playing a crucial role in the process. Perhaps the interaction of Notch with Abl is 'switched' by some analogous mechanism (Crowner, 2003).

In recent years, studies of axon guidance have focused on a relatively small set of receptors that have strong, instructive effects on growth cone trajectory. It has long been clear, however, that many guidance decisions reflect quantitative integration of signals from broadly distributed factors, many of which individually have low specificity. The data reported here provide a paradigm for understanding how subtle modulation of a key signaling pathway allows a combination of relatively low-specificity cellular interactions to produce a precise axonal trajectory (Crowner, 2003).

Using Bcr-Abl to examine mechanisms by which abl kinase regulates morphogenesis in Drosophila

Signaling by the nonreceptor tyrosine kinase Abelson (Abl) plays key roles in normal development, whereas its inappropriate activation helps trigger the development of several forms of leukemia. Abl is best known for its roles in axon guidance, but Abl and its relatives also help regulate embryonic morphogenesis in epithelial tissues. This study explores the role of regulation of Abl kinase activity during development. First the subcellular localization of Abl protein and of active Abl were compared, by using a phosphospecific antibody, providing a catalog of places where Abl is activated. Next, the consequences for morphogenesis of overexpressing wild-type Abl or expressing the activated form found in leukemia, Bcr-Abl, were explored. Dose-dependent effects of elevating Abl activity were found on morphogenetic movements such as head involution and dorsal closure, on cell shape changes, on cell protrusive behavior, and on the organization of the actin cytoskeleton. Most of the effects of Abl activation parallel those caused by reduction in function of its target Enabled. Abl activation leads to changes in Enabled phosphorylation and localization, suggesting a mechanism of action. These data provide new insight into how regulated Abl activity helps direct normal development and into possible biological functions of Bcr-Abl (Stevens, 2008).

Loss-of-function mutations in abl disrupt many morphogenetic events. To further the mechanistic understanding of the roles of Abl, whether deregulated kinase activity disrupts morphogenesis waas examined. Inappropriate activation of Abl affects many of the same morphogenetic events disrupted by loss of Abl. Normally, Abl is likely to exist primarily in an inactive form. Docking with ligands for the SH2 or SH3 domains may help trigger the active conformation. Embryos are relatively resistant to overexpression of wild-type Abl. It is suspected that increasing protein levels are largely accommodated by normal regulatory mechanisms until levels become extremely high. Consistent with this, it was found that active Abl has a more restricted localization than total Abl, suggesting that Abl activation is normally restricted to the apical cell cortex and in particular to tricellular junctions, with a pool of inactive Abl in the cytoplasm. Increasing wild-type Abl levels may drive formation of more active Abl, or it may titrate negative regulators (Stevens, 2008).

Misexpression of Bcr-Abl has more drastic consequences on morphogenesis, consistent with the constitutive activation of Bcr-Abl. In some cases, effects were simply quantitatively stronger, e.g., both Abl and Bcr-Abl affected head involution and segment grooves. However, other processes such as dorsal closure were only affected by Bcr-Abl. These processes may simply be less sensitive, affected only by very high level Abl activity. Alternately, Bcr-Abl may have cell biological effects in Drosophila distinct from those of Abl (Stevens, 2008).

The best-known target of Drosophila Abl is Ena. Abl negatively regulates Ena, in part by restricting its localization. This study examined effects on embryogenesis of depleting maternal and zygotic Ena (enaM/Z). This allows evaluation of which effects of Abl activation result from negative regulation of Ena (Stevens, 2008).

Ena loss-of-function and Abl activation share striking similarities. Both disrupt head involution. In both segmentalgrooves are substantially deepened and persist long after they normally retract. Finally, both alter cell behavior during dorsal closure in similar ways: dorsal closure is significantly slowed, leading edge cells produce fewer filopodia, and epithelial cell matching and zippering are disrupted. These data are consistent with the idea that Ena is the major target of both Abl and Bcr-Abl during Drosophila morphogenesis. This mechanism of action is further supported by other data. First, reduction in Ena levels enhances effects of Bcr-Abl overexpression. Second, overexpression of GFP-Ena partially rescues the effects of Abl activation on filopodial behavior. Finally, Ena localization is regulated by Abl. In abl loss-of-function mutants, Ena accumulates inappropriately at the apical cortex of early embryos and at the leading edge during dorsal closure. In contrast, in embryos overexpressing wild-type Abl, Ena is lost from places it normally accumulates (e.g., tricellular junctions), and it localizes instead at lower levels all around the cell cortex and in the cytoplasm. These data are consistent with Ena being a key Abl target (Stevens, 2008).

Current models of Ena function provide good insight into some of the biological and cell biological effects of Abl activation. Both Ena inactivation and Abl activation led to a reduction in filopodia produced by leading edge cells and defects in the last stages of dorsal closure. These roles fit well with the role of Ena as an anti-capping protein that may also mediate filament bundling into filopodia. Reduction in Abl function leads to the formation of excess filopodia with elevated levels of Ena at the tips, further supporting this regulatory mechanism. Abl activation and Ena loss of function also have parallel effects on head involution and segmental groove formation. In both biological events a row of cells adopts an unusual localization of Ena, with elevated Ena levels and Ena planar polarized at the dorsal-ventral cell boundaries. It seems reasonable that the substantial alterations of Ena subcellular localization caused by Abl activation could interfere with Ena function in these key subsets of cells. However, it remains unclear precisely how localized Ena activity contributes to the distinctive cell shape changes of cells of the segmental grooves or head fold (Stevens, 2008).

One key question is the mechanism(s) by which Abl regulates Ena. The data above and earlier loss-of-function experiments are consistent with a model in which Abl regulates Ena localization, restricting its activity to places it is essential. Abl may form a complex with Ena, keeping it in an inactive state. Consistent with this, Abl can bind Ena, Ena and active Abl colocalize to tricellular junctions and leading edge cell AJs, and Abl overexpression or Bcr-Abl misexpression leads to elevated Abl activity all around the cell cortex, disrupting normal Ena localization. Perhaps Abl docks inactive Ena at sites near where its activity will be needed. For example, Ena at leading edge cell AJs could be the source of Ena needed at the leading edge to make filopodia. Although this model is attractive, some data cannot be easily accommodated by it, e.g., Abl and active Abl both localize to the cortex of syncytial embryos, but Ena is not normally localized there, and Ena localizes there in the absence of Abl, suggesting that there may be alternate mechanisms by which Abl regulates Ena. Further experiments are needed to test these hypotheses (Stevens, 2008).

One way Abl may regulate Ena is by phosphorylation. The effects on embryogenesis of Abl and Bcr-Abl require kinase activity. However, mutating all the Ena phosphorylation sites does not lead to the 'activated' phenotype expected if this is the sole mechanism of negative regulation (i.e., mimicking abl loss-of-function); instead, it has a weak ena loss-of-function phenotype. Thus, Abl regulation of Ena involves more than direct phosphorylation. Abl may phosphorylate itself and other partners, creating or disrupting protein complexes. Consistent with this, although Ena phosphorylation sites are not conserved in its mouse homologues, mouse Abl promotes Mena phosphorylation and binds VASP. Given the clear ability of Bcr-Abl to alter Ena localization/activity in Drosophila, further exploration of Ena/VASP proteins as possible targets in mammalian cells seems warranted (Stevens, 2008).

Although many effects of Abl activation can be explained by negative regulation of Ena, a subset of the effects suggest alternate targets. For example, effects of Abl overexpression on leading edge cell behavior are more drastic than those seen in enaM/Z mutants, e.g., reduced lamellipodial activity was not seen after Ena was inactivated, and high-level Bcr-Abl expression during embryogenesis is more detrimental than Ena loss. Thus, both Abl and Bcr-Abl likely have Ena-independent effects on actin and cell behavior in Drosophila (Stevens, 2008).

One critical issue in interpreting these results is whether Bcr-Abl acts in Drosophila simply as a deregulated form of Abl, or whether it has additional effects on cell behavior. In most of assays, Bcr-Abl expression has effects similar to but stronger than those of Abl overexpression. In some cases, high-level Bcr-Abl expression affected processes not affected by high-level Abl overexpression (e.g., amnioserosa integrity), but sufficient levels of wild-type Abl overexpression may not have been achieved to mimic them. However, one striking effect of Bcr-Abl was not seen either with Abl overexpression or Ena loss-of-function: the explosive production of lamellipodia by amnioserosal cells, which normally only produce filopodia. Perhaps the Bcr part of the fusion recruits additional proteins that influence its abilities. Alternately Bcr-Abl may stimulate signaling pathways such as those of c-Jun NH2-terminal kinase or mitogen-activated protein kinase, targets of mammalian Bcr-Abl; both affect Drosophila epidermal cell shape or fates. Further studies of the mechanisms of action of Bcr-Abl in Drosophila may offer clues into additional targets of Abl (Stevens, 2008).

Although both Abl and Bcr-Abl modulate actin dynamics, the cytoskeletal response they program is complex. In fibroblasts, loss of Abl prevents ruffling in response to PDGF, whereas loss of Arg reduces lamellipodial dynamics. These data suggest that Abl regulates formation of branched actin filaments involved in lamellipodia, consistent with its ability to speed migration. Many of the current observations are consistent with this, including reduced filopodial number after Abl activation, and Bcr-Abl-triggered lamellipodia. Likewise, Drosophila Abl inhibits dendrite branching. However, in other contexts, Abl modulates actin differently. Mouse Abl and Arg maintain dendrite branching, and Abl promotes actin microspikes in fibroblasts plated on fibronectin. Both are consistent with promoting unbranched actin. Bcr-Abl expression also has distinct effects in different cell types, triggering ruffling and filopodial extension in BaF3 cells, while preventing spreading and polarization on fibronectin in dendritic cells (Stevens, 2008).

Bcr-Abl adds additional complexity. Distinct effects were seen of Bcr-Abl expressed at different levels. This dose sensitivity mimics that seen in myeloid cells expressing different levels of Bcr-Abl, which differ in adhesion to fibronectin and ability to induce tumors. A second complexity involves differences between p210 and p185. In Drosophila, p210 produced consistently stronger phenotypes and also triggered higher levels of tyrosine-phosphorylated proteins. p185 and p210 differ in their biochemical and biological activities in mammals as well, and p185 and p210 cause distinct diseases in patients, and induce different pathways of differentiation in primary bone marrow cells. However, in human cells, p185 is the more active kinase. Further exploration of these functional distinctions will help illuminate the different pathways activated by Abl and Bcr-Abl during normal development and oncogenesis (Stevens, 2008).

Thus, both Abl and Bcr-Abl have distinct and at times seemingly opposite cytoskeletal effects in different cells. Perhaps this is not surprising, given the array of cytoskeletal regulators Abl can target, including those promoting unbranched actin filaments, such as Ena/VASP, and those regulating Arp2/3 and branching, such as WASP and WASP family Verprolin-homologous protein. The choice of target may be dictated by upstream inputs regulating Abl, and the consequences for actin dynamics will depend on the suite of other regulators active in the same cell. Understanding how individual cells integrate these inputs and outputs is one challenge for the future (Stevens, 2008).

In the absence of frazzled over-expression of Abelson tyrosine kinase disrupts commissure formation and causes axons to leave the embryonic CNS

In the Drosophila embryonic nerve cord, the formation of commissures require both the chemoattractive Netrin receptor Frazzled (Fra) and the Abelson (Abl) cytoplasmic tyrosine kinase. Abl binds to the cytoplasmic domain of Fra and loss-of-function mutations in abl enhance fra-dependent commissural defects. To further test Abl's role in attractive signaling, Abl was over-expressed in Fra mutants anticipating rescue of commissures. The Gal4-UAS system was used to pan-neurally over-express Abl in homozygous fra embryos. Surprisingly, this led to a significant decrease in both posterior and anterior commissure formation and induced some commissural and longitudinal axons to project beyond the CNS/PNS border. Re-expressing wild-type Fra, or Fra mutants with a P-motif deleted, revert both commissural and exiting phenotypes, indicating that Fra is required but not a specific P-motif. This is supported by S2 cell experiments demonstrating that Abl binds to Fra independent of any specific P-motif and that Fra continues to be phosphorylated when individual P-motifs are removed. Decreasing midline repulsion by reducing Robo signaling had no effect on the Abl phenotype and the phenotypes still occur in a Netrin mutant. Pan-neural over-expression of activated Rac or Cdc42 in a fra mutant also induced a significant loss in commissures, but axons did not exit the CNS. Taken together, these data suggest that Fra activity is required to correctly regulate Abl-dependent cytoskeletal dynamics underlying commissure formation. In the absence of Fra, increased Abl activity appears to be incorrectly utilized downstream of other guidance receptors resulting in a loss of commissures and the abnormal projections of some axons beyond the CNS/PNS border (Dorsten, 2010).

Frazzled and Abelson Tyrosine kinase activity clearly cooperate during the formation of embryonic commissures. In the absence of Fra, detection of Netrin-dependent chemoattraction is compromised and many posterior commissures fail to form. Both anterior and posterior commissures are absent if fra and abl activity is lost. This presumably reflects the ability of abl mutations to interact with a second Netrin receptor, Dscam, as well as Netrin independent receptors (e.g. Turtle) known to be important for commissure formation. Finally, as most commissures are also lost when both maternal and zygotic contributions of Abl are genetically removed, it seems Abl itself is required for commissure formation. Given these different observations, it seemed plausible that over-expressing Abl in fra null embryos would improve commissure formation. However, instead of an improvement, this study clearly documented a major decrease in both anterior and posterior commissures and the induction of a novel phenotype whereby axons normally confined to the CNS now project into the periphery (AEP defects). It is worth emphasizing that these phenotypes occur even with the over-expression of a wild-type Abl transgene that retains its autoinhibitory domain and must be activated by endogenous mechanisms. These phenotypes are completely dependent on the absence of Fra but not any specific P-motif, occur in the absence of Netrins as well, and are not alleviated if Robo-dependent midline repulsion is reduced. Interestingly, the loss of commissures, but not the Axons Exiting to Periphery (AEP) defects, is also observed when activated Rac or Cdc42 GTPases are over-expressed in a homozygous fra mutant. Taken together, it is proposed that during exploration of the midline, Fra is a key regulator of Abl activity and helps determine how the cytoskeletal machinery utilizes Abl. In the absence of Fra, axon outgrowth does not simply stall; but rather, axons follow a variety of aberrant trajectories away from the midline. This suggests that Fra normally competes with several other receptor systems to dictate how Abl functions to regulate the cytoskeletal machinery. While competitors undoubtedly include other midline guidance cues, the emergence of AEP defects suggests that Fra also competes with guidance systems not normally associated with the midline. In the absence of Fra, these other receptors appear to utilize the extra Abl to alter cytoskeletal dynamics at a variety of choice points, ultimately preventing commissure formation and directing some axons out of the CNS (Dorsten, 2010).

The Abl gain-of-function phenotype described in this study occurs only if fra is absent. That is, commissures form correctly in a heterozygous fra mutant or when partially active Fra transgenes with a single P-motif deleted are re-expressed with Abl. In S2 cell immunoprecipitation experiments, both Abl and BcrAbl bind to the cytoplasmic tail of Fra independent of any specific P-motif. While surprising given the conservation of these P-motifs and their known importance to Fra function, the lack of P-motif specificity is consistent with genetic rescue experiments. All three P-motif deletion mutants rescue commissure formation and the AEP defects elicited by over-expression of either wild-type Abl or BcrAbl in fra embryos. The ectopic midline crossovers (fuzzy commissures) observed with only BcrAbl also depend on Fra and specifically the P3-motif. However, BcrAbl is not an endogenous Drosophila protein and, the human Bcr domain may induce neomorphic phenotypes. Because BcrAbl does not preferentially bind to the P3 motif, and wild-type Abl does not elicit crossover defects, it is now suspected that the ectopic crossovers are an example of a neomorphic phenotype, a hypothesis that will be extensively addressed in future work (Dorsten, 2010).

The physical interaction between Fra and either Ablwt or BcrAbl in immunoprecipitation assays can reflect direct or indirect association between the two proteins. It is possible that the failure to observe P-motif dependence reflects the binding of Abl to multiple P-motifs or the use of scaffold proteins associated with more then one P-motif. Given that it has been demonstrated that the cytoplasmic tail of Fra fused to glutathionine-S-transferase (GST) binds to in vitro translated Abl, direct binding of Abl to Fra is clearly possible. If so, these experiments suggest that Abl may bind to Fra in the regions between P-motifs, which is, in fact, where most of the tyrosine residues within the cytoplasmic domain of Fra reside. Moreover, in S2 cells, tyrosine phosphorylation of Fra is not affected by removal of the P1 or P2 motif and may actually increase when P3 is removed. This is intriguing as removal of the P3-motif is known to significantly affect Fra signaling in vivo and the Fra^ΔP3 transgene is the least capable of rescuing the AEP defects caused by Ablwt or BcrAbl expression. Immunoblots of Fra phosphorylation in the absence of pervanadate pretreatment also suggest the steady-state level of tyrosine phosphorylation may be relatively low, or highly dynamic. S2 cells are known to express tyrosine phosphatases that antagonize Abl activity for some substrates, and antagonistic action between Abl and several phosphatases during nerve cord development has been documented. Since tyrosine phosphorylation of vertebrate DCC is required for attractive responses, axon outgrowth and orientation of the axon, it will be important to systematically assess how Fra and Abl physically interact to regulate each other's activity during midline guidance (Dorsten, 2010).

Both of the phenotypes observed when Abl activity is elevated in a fra mutant, the loss of commissures and the exiting of CNS axons to the periphery, suggest these embryos are experiencing an excess of midline repulsion. During commissure formation, Slit dependent repulsion prevents commissural axons from crossing unless Commissureless prevents the Slit receptor Robo from accumulating on the cell surface. Before commissural axons extend towards the midline, Fra activity may help increase Comm expression so Comm levels are expected to be reduced in fra mutants leading to an increase in Robo-dependent repulsion. Since increasing Robo activity in a fra embryo is sufficient to reduce commissure formation, it was important to test whether an excess of Robo-dependent repulsion underlies the Abl over-expression phenotypes. However, introduction of one null allele of robo (Robo1) had no discernable affect on the Abl gain-of-function phenotypes, even though, in previous work, elevating Abl activity in a heterozygous robo mutant induces ectopic midline crossing errors. Given the absence of even a minimal suppression, it seems unlikely that the loss of commissures and/or AEP defects noted in mutants reflects an increase in Robo-dependent midline repulsion. While two other Robo receptors, Robo2 and Robo3, operate during midline guidance and could conceivably contribute to these Abl phenotypes, neither of these receptors have the conserved CC3 cytoplasmic domain known to be important for Abl binding to Robo1. Moreover, while certain Netrin receptors (e.g., Unc5) can also elicit a repulsive response, both commissure loss and AEP defects still occur when Abl is over-expressed in a Netrin mutant. This provides strong evidence that abnormal signaling by other Netrin receptors is not responsible for these phenotypes (Dorsten, 2010).

In terms of the AEP defects, which also point to excess repulsion, it is worth noting that axons do not leave the CNS in a commissureless mutant experiencing very high levels of Slit-dependent repulsion, nor do they appear evident in published figures of fra Dscam double mutants, or even fra Dscam abl triple mutants. While the identity of all the axons leaving the CNS has not been established, ut was confirmed that at least two subtypes of CNS axons are exiting: both FasII expressing interneurons and sema2b commissural axons. FasII axons do not leave the CNS when midline repulsion is elevated in a commissureless mutant, and while the level of Abl activity affects the trajectory of FasII interneurons, in most cases altering Abl activity leads to midline crossing errors rather than an exit from the CNS. Over-expressing Abl in a fra mutant also affects several different guidance decisions by Sema2b-expressing commissural axons. While the cues guiding these neurons are not well understood, the spectrum of defects observed both before and after they cross the midline suggest that these neurons are responding to more then just midline repulsion. Thus, if a repulsive mechanism is functioning, as initially suggested by the phenotype, the origin of the signal remains to be determined. Indeed, data using activated forms of Rac and Cdc42 suggest that the primary defect lies in the ability of growth cones to properly regulate actin dynamics underlying axon outgrowth and steering. This could involve both attractive and repulsive systems (Dorsten, 2010).

Proper axon guidance also requires concerted regulation of the cytoskeletal dynamics underlying axon outgrowth and steering. Like most guidance receptors, Fra, or its vertebrate and C. elegans homologues, is known to initiate signaling pathways affecting cytoskeletal dynamics. Abl is also a key regulator of actin dynamics in vertebrate cells and of the development of the Drosophila nervous system. Mutations in abl interact with several cytoskeletal regulators to affect axon pathway formation: kette, capulet, chicadee (Profilin), enabled and trio. Thus, in the absence of Fra-dependent regulation, does elevated Abl activity affect the cytoskeletal machinery to indirectly cause a reduction in commissures and AEP defects? This study tested this basic concept by expressing in fra mutants other key regulators of cytoskeletal dynamics known to affect midline guidance. Surprisingly, over-expression of activated Rac and Cdc42 in a fra mutant replicates the loss of commissure phenotype (but not the AEP defects) observed with Abl. The Cdc42 result is most intriguing as expression in a wild-type or heterozygous fra embryo results in fused commissures and gaps in the longitudinal connectives. Yet, upon complete removal of Fra, commissures do not form and the longitudinal connectives reform. Thus, in the absence of Fra, commissure formation, but not AEP defects, appears to be particularly sensitive to manipulation of actin-based processes. It is possible that the manipulation of Cdc42 and Rac activity in a fra mutant is affecting shared processes related to actin polymerization. For example, in vertebrate studies, Cdc42 and Abl work in parallel to regulate actin polymerization and Abl may activate Rac in response to cell adhesion. If so, the data suggest that in the absence of Fra activity these key regulators are being used by other surface receptors to regulate actin dynamics in a manner that ultimately prevents commissure formation. This is certainly consistent with the number of guidance systems that have been linked to these regulators and the scope of guidance detected defects. Minimally, the Cdc42 and Rac data continue to highlight the degree and importance of Fra-dependent regulation of cytoskeletal dynamics, especially actin-based processes, during commissure formation. Moreover, they point to a highly competitive process between multiple surface receptors and the cytoskeletal machinery where Fra is a major player. While competition between midline attractive and repulsive cues has been recognized, in the current experiments, midline repulsive activity had no affect on the Abl phenotypes. Therefore, it seems likely that Fra is competing with several other receptor systems whose presence (but not identity) has been uncovered by over-expression of Abl, Rac or Cdc42 in homozygous fra embryos. Which guidance events are being affected has not yet been determined, but a few candidates exist. In addition to fra and Dscam, loss-of-function mutations in abl interact with the cell-cell adhesion molecules neurotactin and amalgam, fasI, midline-fasciclin and turtle to reduce commissure formation and some of these are fairly ubiquitously expressed in the nerve cord. Abl has also been linked to the regulation of cell-cell adhesion molecules alone or in combination with receptor systems such as Notch (Dorsten, 2010).

In summary, these data suggest a model whereby Fra activity initiates key signaling events that dictate when and how Abl activity is utilized during commissure formation. Rac and Cdc42 are probably also involved in this process, and, together with Abl, help regulate key aspects of actin dynamics underlying commissure formation. In the absence of Fra other midline guidance systems are still functioning well enough to form most commissures, but they are clearly sensitive to perturbation of intracellular signaling pathways regulating cytoskeletal dynamics. Thus, when Fra is removed, other guidance systems appear to recruit Abl, Rac or Cdc42 activity to misdirect axon outgrowth, ultimately preventing commissure formation and, with Abl, causing some axons to exit the CNS. Thus, in a normal embryo, Fra must be sending information that allows it to compete very well against these other guidance receptors to properly regulate axon outgrowth and steering during commissure formation. While an alteration in midline guidance decisions may also account for the AEP defects, the scope of guidance errors observed in neurons leaving the CNS suggest that increasing Abl activity could also be affecting other guidance systems not directly associated with the midline. While the identity and specific role of these guidance systems awaits discovery, the sensitivity of the CNS axon scaffold to Abl over-expression will be an important tool for identifying these competing pathways (Dorsten, 2010).

The Abelson tyrosine kinase regulates Notch endocytosis and signaling to maintain neuronal cell fate in Drosophila photoreceptors

The development of a functional organ requires coordinated programs of cell fate specification, terminal differentiation and morphogenesis. Whereas signaling mechanisms that specify individual cell fates are well documented, little is known about the pathways and molecules that maintain these fates stably as normal development proceeds or how their dysregulation may contribute to altered cell states in diseases such as cancer. In Drosophila, the tyrosine kinase Abelson (Abl) interfaces with multiple signaling pathways to direct epithelial and neuronal morphogenesis during embryonic and retinal development. This study shows that Abl is required for photoreceptor cell fate maintenance, as Abl mutant photoreceptors lose neuronal markers during late pupal stages but do not re-enter a proliferative state or undergo apoptosis. Failure to maintain the differentiated state correlates with impaired trafficking of the Notch receptor and ectopic Notch signaling, and can be suppressed by reducing the genetic dose of Notch or of its downstream transcriptional effector Suppressor of Hairless. Together, these data reveal a novel mechanism for maintaining the terminally differentiated state of Drosophila photoreceptors and suggest that neuronal fates in the fly retina retain plasticity late into development. Given the general evolutionary conservation of developmental signaling mechanisms, Abl-mediated regulation of Notch could be broadly relevant to cell fate maintenance and reprogramming during normal development, regeneration and oncogenic transformation (Xiong, 2013).

This study reports a novel requirement for the Abl nonreceptor tyrosine kinase in maintaining the terminally differentiated state of Drosophila photoreceptors. The failure to maintain expression of neuronal markers correlates with impaired trafficking and ectopic signaling by the Notch receptor. Consistent with the idea that aberrant Notch activation might drive photoreceptor dedifferentiation, Notch signaling has been shown to inhibit neuronal differentiation in many developmental contexts in all animals, and Notch is normally absent in pupal photoreceptor cells. Thus it is proposed that ectopic activation of Notch signaling provides a molecular mechanism coupling Abl loss to a program of neuronal dedifferentiation in the fly retina (Xiong, 2013).

Dedifferentiation describes a regressive process whereby a differentiated somatic cell loses its mature identity and reverts to an earlier multipotent developmental state. The observation that some Abl mutant photoreceptors not only lose neuronal marker expression, but also turn on pigment cell marker expression, raises the possibility that the former photoreceptor neurons might transdifferentiate toward a new pigment cell-like state. Genome-wide gene expression analysis of Abl mutant cells should provide a more precise molecular definition of this transition and of the final state of these cells. Experiments to determine whether the dedifferentiated or partially transdifferentiated Abl mutant cells can be redirected toward other cell fates or to re-enter the cell cycle will provide additional insight into the extent of their multipotency and plasticity (Xiong, 2013).

Previous work has shown that Abl is largely dispensable for photoreceptor cell fate specification, but then plays crucial roles throughout the elaborate morphogenetic programs that lead to rhabdomere formation and determine the spatial organization of ommatidial cells within the epithelium. Whether altered Notch signaling and dedifferentiation result directly from failed morphogenesis, or whether they reflect independent requirements for Abl later in development is an open question. The fact that ectopic expression of an activated Notch transgene at mid-pupal stages, after completion of the morphogenetic program, leads to loss of neuronal marker expression, suggests the two can be uncoupled. However Notch trafficking defects are apparent in Abl mutant photoreceptors well before morphogenesis is complete, raising the possibility that these cellular events could be tightly intertwined. Examination of neuronal marker expression and Notch signaling at mid-late pupal stages in apical polarity mutants might help elucidate the extent of molecular coupling between morphogenesis and photoreceptor fate maintenance (Xiong, 2013).

A second open question concerns the molecular mechanisms by which Abl regulates Notch trafficking to activate downstream signaling. Prior work has implicated endocytic trafficking in regulating both ligand-independent and ligand-dependent Notch signaling. Supporting the argument for the former mechanism in Abl mutant photoreceptors, it was found that Delta levels are below detection threshold at the mid-pupal stages when Notch accumulation, and presumably signaling, is highest. However, previous studies have reported that ligand-independent activation of Drosophila Notch in mutants affecting late endosome/ multivesicular body sorting results in overproliferation rather than dedifferentiation of retinal cells. One possible explanation to reconcile a model of ligand-independent Notch activation with these observations is that the endocytic pathway genes potentiate early functions of Notch in regulating cell proliferation, whereas loss of Abl affects later roles. Another non-mutually exclusive explanation is that the endocytic defects observed in Abl clones might be highly specific to Notch trafficking, an argument substantiated by lack of effect on two other cell-surface proteins Delta and Egfr, whereas the phenotypes resulting from loss of a general component of the endocytic machinery might reflect a complex disruption of multiple signaling pathways. Alternatively, the ectopic Notch signaling that results from loss of Abl could reflect a ligand-dependent response, either to undetectably low levels of Delta or to an alternate ligand such as Serrate. Analysis of the endocytic route taken by Notch in Abl mutant photoreceptors, genetic exploration of ligand-dependence versus independence, and investigation of Abl interactions with specific Notch regulators like Nedd4, Deltex and Cbl should help distinguish between the different models (Xiong, 2013).

In conclusion, these results reveal a novel requirement for the Abl tyrosine kinase in preventing Notch activation to maintain the terminally differentiated state of Drosophila photoreceptor cells. The discovery that Abl, a key morphogenetic regulator, is also required for cell fate maintenance, suggests a new molecular strategy for coordinating tissue morphogenesis with differentiation. The extent to which Abl-mediated maintenance of the differentiated cell state might be relevant to other tissues and developmental or pathogenic contexts will be an important direction for future investigation (Xiong, 2013).

Coordinate regulation of stem cell competition by Slit-Robo and JAK-STAT signaling in the Drosophila testis

Stem cells in tissues reside in and receive signals from local microenvironments called niches. Understanding how multiple signals within niches integrate to control stem cell function is challenging. The Drosophila testis stem cell niche consists of somatic hub cells that maintain both germline stem cells and somatic cyst stem cells (CySCs). This study shows a role for the axon guidance pathway Slit-Roundabout (Robo) in the testis niche. The ligand Slit is expressed specifically in hub cells while its receptor, Roundabout 2 (Robo2), is required in CySCs in order for them to compete for occupancy in the niche. CySCs also require the Slit-Robo effector Abelson tyrosine kinase (Abl) to prevent over-adhesion of CySCs to the niche, and CySCs mutant for Abl outcompete wild type CySCs for niche occupancy. Both Robo2 and Abl phenotypes can be rescued through modulation of adherens junction components, suggesting that the two work together to balance CySC adhesion levels. Interestingly, expression of Robo2 requires JAK-STAT signaling, an important maintenance pathway for both germline and cyst stem cells in the testis. This work indicates that Slit-Robo signaling affects stem cell function downstream of the JAK-STAT pathway by controlling the ability of stem cells to compete for occupancy in their niche (Stine, 2014: PubMed).

Abl tyrosine kinase Effects of mutation: back to part 1/2

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

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