BIOLOGICAL OVERVIEW

Reorganization of the cytoskeleton is regarded as a crucial intermediary step in translation of extracellular cues to cellular responses. Wiskott-Aldrich syndrome in humans is an X-linked hematopoietic disease that manifests itself in platelet deficiency and a compromised immune system. Analysis of hematopoietic cells from affected individuals reveals that mutations in the Wiskott-Aldrich syndrome protein (WASP) result in structural and functional abnormalities in the cell cortex, consistent with the suggestion that WASP is involved with regulation of the actin-rich cortical cytoskeleton. Members of the WASP family are key elements that link signal transduction pathways and the actin-based cytoskeleton. WASP proteins serve as a common platform, bringing together components of signal transduction pathways with cellular machinery that promotes actin polymerization and microfilament reorganization. Execution of this program in proximity to the cell surface can then lead to formation of protrusive, actin-based membrane structures in response to various cues. Signaling molecules with which WASP proteins associate include the activated, GTP-bound form of the CDC42 GTPase (see Drosophila Cdc42), membrane phosphoinositides, and Src homology 3 (SH3) domain proteins, which function in tyrosine kinase-based signaling. The cytoskeletal elements involved are monomeric actin and the Arp2/3 complex (see Drosophila Arp2/3 component Suppressor of profilin 2), an evolutionarily conserved complex of seven proteins that acts as a potent nucleator of nascent microfilaments and can bring about the formation of extensive dendritic microfilament networks (Ben-Yaacov, 2001 and references therein).

Mammalian species possess at least two closely related WASP homologs. In humans these include the prototype WASP, first described as the affected protein in the Wiskott-Aldrich syndrome (WAS) blood disorder, and the more generally expressed neuronal WASP (N-WASP). A variety of studies have suggested key cellular roles for members of the WASP protein family. In addition to repeated demonstrations and analyses of their ability to relay CDC42-based signaling to the actin cytoskeleton, WASP proteins have been shown to participate in the actin-based motility of both intracellular pathogens and endogenous membrane vesicles. Assessments of WASP protein function in vivo, on the basis of mutations in the structural genes, have been possible in several settings. WAS and X-linked thrombocytopenia arise in individuals bearing a wide spectrum of mutations in the gene encoding human WASP. These potentially debilitating diseases result from malfunctioning of hematopoietic cells, particularly platelets. A generally similar disorder has been described for a mouse knockout model of WAS. Structural abnormalities of the cell surface and underlying cortical cytoskeleton are commonly considered as primary causes of the various manifestations of WAS. Mutations in bee1/las17p, which encodes a WASP-related protein in yeast, result in disruption of cortical actin patch formation, upholding an evolutionarily conserved role related to proper organization of the cortical cytoskeleton (Ben-Yaacov, 2001 and references therein).

Mutations in WASp, the Drosophila homolog, reveal an essential requirement for this gene in implementation of cell fate decisions during adult and embryonic sensory organ development. SCAR/WAVE, a second Arp2/3 complex activator acts as the main Arp2/3 regulator during axonal growth, oogenesis and eye development. Phenotypic analysis of WASp mutant animals demonstrates a bias toward neuronal differentiation, at the expense of other cell types, resulting from improper execution of the program of asymmetric cell divisions, which underlie sensory organ development. Generation of two similar daughter cells after division of the sensory organ precursor cell constitutes a prominent defect in the WASp sensory organ lineage. The asymmetric segregation of key elements such as Numb is unaffected during this division, despite the misassignment of cell fates. The requirement for WASp extends to additional cell fate decisions in lineages of the embryonic central nervous system and mesoderm. The nature of the WASp mutant phenotypes, coupled with genetic interaction studies, identifies an essential role for WASp in lineage decisions mediated by the Notch signaling pathway (Ben-Yaacov, 2001).

In general, WASp flies do not display any gross morphological abnormalities. However, mutant flies exhibit a pronounced lack of neurosensory bristles, external manifestations of sensory organs stereotypically positioned just underneath the entire cuticle of the adult fly. The compound eye of the fly is composed of hundreds of individual facets (ommatidia), each of which is associated with a single cuticular sensory organ which forms during the first 2 d of development of the pupal retina and gives rise to a single bristle. Loss of interommatidial bristles is a particularly penetrant and reproducible manifestation of the WASp mutant phenotype, therefore observations concentrated on sensory organ development in this tissue to follow the process in greater detail. Selection of single sensory organ precursor (SOP) cells from within a competent proneural cell cluster constitutes an initial step in development of Drosophila adult sensory organs. The SOP selection process was examined at 3 h after puparium formation (APF) by staining dissected retinas for the A101 enhancer trap marker, which is expressed in SOPs immediately after their selection from within the proneural cluster. The A101 staining pattern of retinas derived from WASp mutants fully resembles that of wild-type, suggesting that events at the proneural stage are not affected by mutations in WASp and that sensory organ development is properly initiated in the mutant animals (Ben-Yaacov, 2001).

Adult sensory organs are composed of clusters of four distinct cell types, which form after several rounds of asymmetric division from a single SOP. The SOP (also referred to as the pI cell) divides to produce the intermediary pIIa and pIIb cells. pIIa will give rise, upon division, to the bristle secreting trichogen and accompanying socket cell (tormogen), which form the externally visible portion of the sensory organ. Division of pIIb produces a third intermediary cell, pIIIb, which will divide again to generate the enervating neuron and supporting sheath cell (thecogen), both of which reside at a subepidermal level. The second product of the pIIb division is a glial cell, which moves away from the four-cell cluster as the sensory organ forms. Although the events surrounding precursor selection appear to proceed normally in WASp mutants, a different picture emerges when mutant retinas are examined at 30 h APF, by which time the major stages of sensory organ development are completed. The transcription factor Cut (Ct) localizes to all nuclei of external sensory organ cells, including those that form in the pupal retina. Sensory organ cells in WASp mutant retinas properly express the Ct marker, but are abnormally distributed in large clusters, in contrast to the very regular four-cell formations seen in wild-type. The availability of differentially expressed nuclear markers allows the different cell types present in the developing sensory organ to be distinguished. To study the retinal differentiation pattern, Shaven (Sv), a marker of both the sheath and bristle shaft cells, was used. Double staining of retinas dissected 30-h APF reveals that Sv, which is normally expressed in half of the mature Ct-expressing sensory organ cells, is detected in only a small minority (<10%) of such cells in the mutant. A drastic reduction is also observed in the proportion of sensory organ cells that accumulate high levels of Suppressor-of-Hairless (Su[H]), a bristle socket cell marker. In contrast, the neuronal nuclear marker Elav, which is normally restricted to the single neuron of each four-cell cluster, is found in the vast majority of mutant sensory organ cells. A similar phenotype of excess neurons and a near absence of bristle shaft, bristle socket, and sheath cells is observed during sensory organ development in the notum of WASp mutant pupae as well (Ben-Yaacov, 2000).

Taken together, these observations provide a basis for the bristle-loss phenotype of WASp mutant flies. Although the program of sensory organ development is properly set in motion, execution of the sensory organ differentiation process is defective, leading to a predominance of neurons at the expense of nonneuronal cell types. Sensory organ phenotypes of this kind have been described for mutations in a variety of Drosophila genetic loci. In particular, elements of the signaling pathway involving the Notch receptor are thought to control cell fate decisions that assure proper differentiation of sensory organ cells into distinct cell types. The WASp mutant phenotypes are consistent with a particular scenario of cell fate transformations during the asymmetric cell divisions that produce the mature sensory organ. Transformation of pIIa to a pIIb fate accounts for the absence of a bristle shaft/socket lineage, resulting in a smooth adult cuticle phenotype. The apparent generation of two pIIb cells in each lineage, followed by a second, sheath-to-neuron transformation, constitutes a basis for the observed neuronal excess and vastly reduced numbers of sheath cells (Ben-Yaacov, 2000).

The relatively late stage in development at which a zygotic WASp mutant phenotype is observed raises the issue of whether WASp function is essential only during metamorphosis and development of the adult fly. One possibility is that a maternal contribution of WASp masks a requirement during embryogenesis. To address this matter, the FLP-DFS technique was used to produce WASp^- female germline clones, thereby eliminating any contribution of WASp gene products from a maternal source. The fate of embryos derived from WASp^- germline clones is dependent on the genetic makeup of the paternal contribution. Embryos lacking both maternal and zygotic sources of WASp (referred to herein as WASp^mat/zyg embryos) do not survive, indicating an essential requirement for WASp during the course of embryogenesis. In contrast, eggs fertilized with WASp⁺ sperm develop normally, and give rise to viable and fertile adults, indicating that zygotic WASp function can overcome the lack of a maternal contribution (Ben-Yaacov, 2000).

Cuticle preparations of WASp^mat/zyg embryos, which are completely devoid of WASp function, are normal, implying that WASp is not generally required for morphogenesis of the embryo. However, a more detailed examination reveals essential roles for WASp in key cell fate decisions during Drosophila embryonic development. Based on the zygotic adult phenotype, development of sensory organs was examined in WASp^mat/zyg embryos. The sensory organs of the embryonic peripheral nervous system (PNS) form in the ectoderm during stages 10-13 of embryogenesis, in a segmentally reiterated pattern. Development of these structures, which are composed of single neurons and various nonneuronal support cells, follows the general guidelines of adult sensory organ development: selection of single SOPs from within a competent proneural cluster followed by a limited number of asymmetric divisions and N-dependent differentiation of distinct cell types (Ben-Yaacov, 2000).

The proneural marker Achaete (Ac) is transiently expressed in SOPs and their progeny during the initial stages of embryonic sensory organ determination. The Ac staining pattern in WASp^mat/zyg embryos resembles that of wild-type. Although this observation suggests that the early steps of sensory organ development proceed normally in the mutants, lack of WASp function has a clearly deleterious effect on the subsequent maturation of embryonic sensory organs. When stained with anti-Elav or with mAb 22C10, which recognizes a neuronal membrane-associated antigen, later-stage WASp^mat/zyg embryos present an obvious excess of neurons. Quantitative assessments by nuclear and cell counts suggest a near-doubling of neurons in the mutant embryos. Thus, for instance, as many as 25 neurons are commonly found in the combined l and d clusters of abdominal segments, which normally contain 14 neurons. As was observed in the developing adult retina, neuronal excess in the embryonic PNS comes at the expense of nonneuronal support cells. Far fewer cells express A1-2-29, a shaft and socket cell marker. Similar reductions are observed in the number of cells expressing Su(H), which specifically labels socket cells of external sensory organs. These observations are readily explained by pIIa to pIIb cell fate transformations during embryonic sensory organ development, the suggested basis for the adult bristle-loss phenotype. However, not all nonneuronal cell types are affected to the same degree in WASp^mat/zyg mutants. Only mild reductions in staining of the sheath cell fate marker Prospero are observed, suggesting a lesser requirement for WASp during the neuron/sheath cell fate decision in the embryonic PNS (Ben-Yaacov, 2000).

Attempts were made to determine whether a requirement for WASp function exists in additional settings, in which execution of lineage and cell fate decisions has been shown to rely on the N pathway. This issue was examined in an embryonic neuroblast lineage decision in the developing central nervous system (CNS). A pair of neurons designated RP2 develops in a specific position of each and every segment of the embryonic CNS. The RP2 neurons are distinguishable from the RP2-sib pair, which derive from a common progenitor, by expression of markers such as the segmentation protein Even-skipped (Eve). Wild-type RP2 neurons express Eve in a persistent fashion, whereas RP2-sib neurons do so only transiently. Loss-of-function mutations in N, and in other genes that show N-like mutant phenotypes, result in a RP2-sib to RP2 fate transformation, so that in each segment all four neurons of this lineage express Eve. A similar duplication of persistent Eve-expressing neurons is characteristic of WASp^mat/zyg embryos (Ben-Yaacov, 2000).

A second process studied involves the N-dependent mesodermal lineage decision made between future pericardial (PC) and DA1 muscle founder cells, all of which derive from a common progenitor. In wild-type embryos, both cell types, which form in neighboring but distinct positions, express Even-skipped, but only the DA1 founders express the Kruppel (Kr) marker. An apparent bias towards the PC cell fate in the mesoderm of WASp^mat/zyg embryos is observed after staining with these markers. A marked reduction in the number of Eve- and Kr-expressing DA1 cells is coupled with an apparent increase in the number of Eve-expressing cells, present at the position normally occupied by PC cells. The mesodermal WASp phenotype is exceptional, since it resembles N gain-of-function phenotypes observed in this tissue, adding a level of complexity to interpretations of WASp function. In conclusion, the characterization of embryonic WASp mutant phenotypes strongly implies an essential involvement of WASp in various N-dependent lineage and cell fate decisions, throughout Drosophila development (Ben-Yaacov, 2000).

The requirement for WASp function in N-dependent cell fate decisions prompted a search for genetic interactions between WASp and N pathway elements, making use of the WASp adult bristle-loss phenotype. Although the N pathway is involved in a wide variety of cell fate decisions during fly development, use of conditional mutant alleles has been successful in demonstrating that loss-of-function mutations in N itself and in its ligands results in PNS neuronal preponderance and associated phenotypes, in both embryos and adults, including the pIIa-to-pIIb and sheath-to-neuron transformations suggested for WASp. A WASp;N double mutant was constructed using the temperature-sensitive N^ts1 allele. At 25°C, N^ts1 flies display a wild-type morphology, including a normal array of neurosensory bristles. Introducing this very mild N hypomorphic genotype into a WASp mutant background results in a strong enhancement of the WASp bristle-loss phenotype. Double mutant flies lack practically all bristles on regions of the cuticle such as the thorax, which is only partially affected by the WASp mutation alone (Ben-Yaacov, 2000).

In contrast to the enhancement achieved by reducing N function, significant suppression of the WASp bristle-loss phenotype can be observed when activity of the N pathway is even moderately elevated. The neurosensory bristle pattern of WASp mutant flies, which also lack one copy of the established N antagonist Hairless (H), is close to wild-type in appearance. These flies eclose normally. Similarly, a significant, if somewhat less dramatic rescue of the WASp phenotype is obtained using a gain-of-function allele of the N receptor itself. A transgenic construct (N^int.hs), in which the constitutively active, intracellular portion of N is expressed under the control of a heat-shock promoter, was introduced into a WASp mutant background. Mild (29°C) heat treatment of such flies, which has no noticeable effect on N^int.hs flies, when combined with a WASp mutation leads to significant restoration of the bristle pattern, particularly in abdominal segments. Sensitive genetic interactions can thus be demonstrated between WASp and elements of the N pathway, raising the possibility of a common functional framework (Ben-Yaacov, 2000).

The established cellular roles of mammalian WASP proteins has prompted a consideration of instances of cytoskeletal involvement in N-based signaling, to try and reveal the mechanistic basis of WASp function during Drosophila development. numb is considered a key regulator of sensory organ development, acting as an antagonist of N signaling in this tissue. During all cell divisions in the sensory organ lineage, Numb protein segregates into only one of the two progeny cells, thereby ensuring that the lateral inhibition mediated by N signaling is unidirectional, and providing a basis for assumption of distinct cell fates. Significantly, asymmetric distribution of Numb and other elements requires an intact microfilament-based cytoskeleton, suggesting a possible site of action for WASp and associated factors. This possibility was first addressed by determining and comparing the distribution and segregation of both Numb and the associated Partner of Numb (Pon) protein during division of the pI (SOP) cell in wild-type and mutant tissue. In this study antibodies were used to follow endogenous Numb and an ectopically expressed Pon-GFP chimera, previously shown to mimic the asymmetric distribution of the endogenous Pon protein during pI divisions. During metaphase and anaphase of the wild-type pI division, which is aligned along the anterior-posterior axis of the fly, Numb and Pon colocalize and form a crescent at the anterior cortex of the cell, directly above one of the poles of the mitotic spindle. This asymmetric distribution ensures that the proteins segregate only to the anterior pIIb cell at telophase. All aspects of the process are properly executed in WASp mutant animals, including alignment of the spindle with the body axis, colocalization of Numb and Pon to an anterior crescent, and strictly unequal segregation of Numb and Pon into the anterior pIIb cell (Ben-Yaacov, 2000).

To further demonstrate that the cell fate transformations observed in WASp mutant animals cannot be attributed to improper segregation and partitioning of Numb and Pon, Pon distribution was followed as was the fate of the two cells derived from the asymmetric division of pI in living pupae. A sensory organ-specific GAL4 driver, neu-GAL4, was used to express a Pon-GFP chimeric protein in pupal sensory organs, and time-lapse recordings of developing thoracic microchaete were carried out on both wild-type and WASp mutant animals expressing this construct. Normally, the pupal pI cell divides within the plane of the epithelium to generate the anterior pIIb and the posterior pIIa cells, which are aligned along the fly's antero-posterior axis. Pon-GFP forms a crescent at the anterior pole of pI, and subsequently segregates asymmetrically into pIIb. The pIIb cell divides perpendicularly to the plane of the epithelium to generate a small glial cell and the pIIIb cell. During this division, Pon-GFP forms a basal crescent and is asymmetrically distributed into the basal glial cell. Finally, pIIa divides within the plane of the epithelium to generate two cells of equal size, the future bristle shaft and bristle socket cells. In pIIa, as in pI, Pon-GFP forms an anterior crescent and segregates unequally into the anterior shaft cell (Ben-Yaacov, 2000 and references therein).

In WASp mutants, all aspects of the pI division match those seen in wild-type animals. The division generates an anterior-posterior pair of daughter cells, since Pon-GFP forms an anterior crescent within pI and segregates asymmetrically into the anterior cell. However, from this stage on the events of sensory organ differentiation in WASp differ substantially from those observed in wild-type. The first indication of an altered developmental progression is a randomization of the cell division pattern. In contrast to the strictly ordered sequence of divisions observed in the wild-type lineage [in which the anterior pIIb cell (which inherits Pon-GFP) always divides before the posterior pIIa cell], either of the two pI daughter cells in the mutant may divide after pI. In the time-lapse analysis presented in this study, the posterior ('pIIb') cell divides first. A second, striking distinction from wild-type is that the divisions of both pI daughter cells are morphologically identical, and resemble the pattern seen in wild-type pIIb. Both the anterior and posterior cell divisions are nonplanar, and generate two daughter cells of different sizes. In both 'pIIb' cells, Pon-GFP forms a basal crescent and segregates into the small basal cell. These observations conclusively demonstrate that the two progeny of the pI division in WASp mutant animals assume a similar, pIIb-like fate, but this cell fate transformation cannot be attributed to improper partitioning and segregation of Numb and Pon (Ben-Yaacov, 2000).

WASp mutant phenotypes generally resemble those described for positive mediators of N signaling, whereas mutations in the N antagonist numb are distinct and opposite in character. Thus, adult sensory organ development in the absence of numb function leads to formation of multiple sockets, since both progeny of the pI division in this case assume a pIIa fate, and the subsequent division is characterized by shaft-to-socket transformations. The opposite effects on cell fate provide an opportunity to determine an epistatic relationship between WASp and numb. This issue was examined by producing clones of numb cells in a WASp mutant background. A powerful system for producing mutant clones in derivatives of the eye imaginal disc that includes the cuticle of the adult head capsule, has recently been described. This system has been successfully adapted for the study of numb and other regulators of sensory organ formation. Using this adaptation, large numb head clones, in which the multiple socket phenotype characteristic of numb was observed throughout the head cuticle, were consistently produced. When such clones are made in animals hemizygous for WASp alleles, multiple sockets are rarely observed, while the WASp smooth head cuticle phenotype predominates. These observations demonstrate that WASp is epistatic to numb, i.e., a requirement for WASp during adult sensory organ formation persists even in the absence of numb gene function. This finding is consistent with the normal segregation of Numb and Pon-GFP in WASp mutants, with both observations suggesting that WASp is not involved in localization of asymmetrically localized components, but rather provides a function further downstream (Ben-Yaacov, 2000).

These observations suggest that WASp function is required for establishing cell fate during the asymmetric cell division stage, subsequent to the initial determination of sensory organs. Indeed, by monitoring sensory organ development in living tissue, the transformation of the intermediate pIIa cell to a pIIb cell fate has been conclusively demonstrated, and additional observations strongly imply a subsequent sheath-to-neuron cell fate transformation in this lineage. These findings imply a specific role for WASp during sensory organ formation, in the context of cell fate determination via asymmetric division (Ben-Yaacov, 2000).

The challenge of research into the functions of WASp, the Drosophila WASP homolog, is to elucidate the manner in which the established cellular functions of WASP proteins can be united with the role of WASp in generation of cell fate diversity during Drosophila development. A possible hint comes from the particular developmental processes in which WASp function is required. In addition to the requirement during a specific phase of embryonic and adult sensory organ development, roles have been established for WASp in cell fate decisions encompassing aspects of lineage determination in the embryonic CNS and mesoderm. This subset of N-dependent processes has been singled out previously due to significant functional requirements for the genes sanpodo (spdo) and the N antagonist numb. Mutations in spdo result in embryonic phenotypes highly reminiscent of N loss-of-function circumstances in these tissues, whereas impairments to numb lead to the opposite phenotypic effects. The striking similarities in functional requirements have lead to a proposal that numb, spdo, and WASp mediate N signaling within a common mechanistic framework. The nature of this framework is currently unclear and is a matter for speculation. spdo encodes a Drosophila homolog of vertebrate Tropomodulin, a microfilament pointed-end capping protein, suggesting a possible biochemical basis for cooperative function with WASp. However, it should be noted that whereas mutations in both spdo and WASp result in a bias toward a neuronal cell fate in the embryonic PNS and in duplication of RP2 neuroblasts, these mutations have opposite effects on the PC cell/DA1 muscle decision in the embryonic mesoderm, imparting a degree of complexity to the potential functional association between these elements. The requirement for an intact cellular microfilament array in establishing asymmetric localization of Numb and other factors suggested an attractive target for WASp function. However, the data strongly argue against a role for WASp in influencing the cytoskeletal basis of Numb localization, since both Numb and the associated factor Pon are properly localized in WASp mutants. Therefore, the manner in which the presumed disruptions to cytoskeletal organization resulting from mutations in WASp adversely affect the N pathway remains an open question. One avenue which should be considered, in light of recent findings, is the association of endocytosis with both N-based signaling and WASP cellular functions. Substantial genetic and biochemical evidence implies a crucial involvement of ligand-mediated endocytosis in N signal transduction during various developmental processes, including sensory organ formation. Parallel studies have fostered a growing appreciation for WASP protein function in linking endocytic mechanisms with the microfilament-based cytoskeleton, suggesting an intriguing cellular context in which WASp may exert an influence over the N signaling pathway (Ben-Yaacov, 2000).

The involvement of WASp in execution of cell fate decisions during fly development may well have implications for the manner in which mammalian WASP function is perceived. It is worthwhile to note in this context that roles for mammalian N homologs in lineage decisions of hematopoietic cells have been described. However, it is unclear whether the existing data support cell fate defects as an explanation for the human WAS phenotype. The full spectrum of hematopoietic cell types are found in the blood of WAS patients and the pleiotropic phenotypes described appear consistent with general abnormalities in cellular structure, rather than with defects in programs of tissue differentiation. Still, it may be too early to draw parallels between the invertebrate and mammalian systems, particularly since specific functional requirements for N-WASP, the ubiquitously expressed mammalian WASP, are yet to be described (Ben-Yaacov, 2000 and references therein).

Activation of Arp2/3 by WASp is essential for the endocytosis of Delta only during cytokinesis in Drosophila

The actin nucleator Arp2/3 generates pushing forces in response to signals integrated by SCAR and WASp. In Drosophila, the activation of Arp2/3 by WASp is specifically required for Notch signaling following asymmetric cell division. How WASp and Arp2/3 regulate Notch activity and why receptor activation requires WASp and Arp2/3 only in the context of intra-lineage fate decisions are unclear. This study found that WASp, but not SCAR, is required for Notch activation soon after division of the sensory organ precursor cell. Conversely, SCAR, but not WASp, is required to expand the cell-cell contact between the two SOP daughters. Thus, these two activities of Arp2/3 can be uncoupled. Using a time-resolved endocytosis assay, it was shown that WASp and Arp2/3 are required for the endocytosis of Dl only during cytokinesis. It is proposed that WASp-Arp2/3 provides an extra pushing force that is specifically required for the efficient endocytosis of Dl during cytokinesis (Trylinski, 2019).

In animal cells, a thin cortex of actin filaments is dynamically regulated to produce the force required for basic cellular processes, such as motility, cytokinesis, and endocytosis. This regulation involves the nucleation of branched actin filaments by the actin-related proteins 2/3 (Arp2/3) complex (Goley, 2006, Pollard, 2007, Rotty, 2013). By itself, Arp2/3 is weakly active, and nucleation-promoting factors (NPFs) are needed to stimulate its nucleation activity. Thus, when and where actin-based pushing forces are produced in the cell depends on the localization and activity of the NPFs. Wiskott-Aldrich syndrome protein (WASP) family proteins are the best-studied NPFs. These are usually maintained in an autoinhibited state and can be activated at the membrane by small GTPases. Three WASP family members are known in Drosophila: WASp, SCAR/WAVE (suppressor of cyclic AMP repressor/WASp-family verpolin-homologous protein), and WASH (WASp and SCAR homolog). Genetic analysis indicates that SCAR is the primary NPF in Drosophila, since the loss of SCAR activity leads to developmental and cellular defects that are similar to those seen upon the disruption of Arp2/3 activity, whereas WASH has a non-essential function during oogenesis, and WASp is only required for specific Notch-mediated fate decisions following asymmetric cell divisions in muscle, brain, and sensory organ lineages. This function of WASp is mediated by Arp2/3, since the loss of the Arp3 and Arpc1 subunits of the Arp2/3 complex leads to WASp-like cell fate defects. How WASp and Arp2/3 regulate Notch signaling is unclear. In addition, given the ubiquitous expression of WASp and the functional pleiotropy of Notch, it is unclear why WASp is only required for Notch signaling in the context of asymmetric cell division (Trylinski, 2019).

Notch receptor activation requires a pulling force to expose an otherwise buried cleavage site in the extracellular domain of Notch, the cleavage of which eventually produces the Notch intracellular domain (NICD). Previous studies have shown that endocytosis of the Notch ligands provides a strong enough pulling force to direct receptor activation. Since WASp and Arp2/3 are known to increase the efficiency of endocytosis by nucleating branched filaments shortly after membrane ingression begins (i.e., when high forces are required), it is conceivable that WASp-stimulated Arp2/3 activity may facilitate receptor activation by regulating the endocytosis of the Notch ligand Delta (Dl). However, it was reported that the endocytosis did not depend on Arp3, clearly arguing against this model. It was proposed that Arp2/3 may instead regulate the transport of endocytosed Dl back to the apical membrane, where it would activate Notch. This model, however, is not supported by a recent photo-tracking analysis of fluorescent Notch receptors, showing that signaling takes place along the lateral membrane following asymmetric division. NICD was produced during cytokinesis from a subset of Notch receptors that are located basal to the midbody (Trylinski, 2017). Thus, how WASp-Arp2/3 positively regulates Notch signaling is not known (Trylinski, 2019).

Early loss of Notch signaling in WASp and Arp3 mutants did not merely result from a defect in pIIa-pIIb contact expansion at cytokinesis. Contact expansion involves the activation of Arp2/3 by Rac and SCAR, but not by WASp, and SCAR and Rac are dispensable for Notch activation during cytokinesis and pIIa specification. Thus, Arp2/3 has separable functions in contact expansion and Notch signaling at cytokinesis. Instead, this detailed analysis of the endocytosis of Dl revealed that WASp is required for the efficient endocytosis of Dl during cytokinesis, but not afterward. This specific requirement of WASp and Arp2/3 for endocytosis during cytokinesis only may explain its specific requirement in Notch-mediated intra-lineage decision (Trylinski, 2019).

This study showed that Arp2/3 has two separable activities in asymmetric cell divisions: Arp2/3 promotes the rapid expansion of the new cell-cell contact and stimulates the endocytosis of Dl from this cell-cell contact to regulate intra-lineage fate decisions by Notch. These two activities involve distinct NPFs. SCAR, downstream of Rac, promotes the formation of a dense F-actin network around the midbody to generate a force that regulates cell-cell contact between sister cells and facilitates withdrawal of the membranes of the neighboring cells. SCAR, however, is largely dispensable for Notch receptor activation, suggesting that the force required for contact expansion is not key for Notch receptor activation. In contrast, WASp is required for Notch signaling but is dispensable for contact expansion during cytokinesis. While these two functions of Arp2/3 are separable, a functional interplay is possible, if not likely. For instance, Rac and SCAR may facilitate the activity of WASp in Dl endocytosis through the recruitment of Arp2/3 along the pIIa-pIIb interface (Trylinski, 2019).

Before the present study, WASp-mediated activation of Arp2/3 was thought to regulate the intracellular trafficking of internalized Dl, not its endocytosis. This model assumed that Dl signals at the apical membrane, which seems unlikely since it was recently shown that NICD originates from the lateral membrane during cytokinesis (Trylinski, 2017). Using a time-resolved endocytosis assay, this study showed that WASp-mediated activation of Arp2/3 is required to promote the endocytosis of Dl during cytokinesis, but not afterward. The specific time window during which the activities of WASp and Arp3 are required may explain why this requirement had previously been missed. In analogy to the role of WASp in yeast, it is proposed that WASp is recruited at sites of Dl endocytosis to form a branched actin network that provides an inward pushing force onto the invaginated membrane. This would increase the efficiency of endocytosis-hence the rate of the force-dependent activation of Notch. Accordingly, WASp would play a modulatory role, which is critical within a defined time window. Consistent with this view, the WASp mutant bristle phenotype can be suppressed by lowering the threshold for NICD levels in flies with reduced levels of the CSL co-repressor Hairless (Trylinski, 2019).

Why are WASp and Arp3 required only during cytokinesis for the endocytosis of Dl? WASp is likely to have a general function in endocytosis in Drosophila, as in other organisms, and may therefore regulate the endocytosis of many cargoes, including Dl, throughout development. Consistent with a general function of WASp, it is ubiquitously expressed and is not specifically upregulated in sensory lineages. However, the role of WASp-activated Arp2/3 in endocytosis is essential, at the organismal level, only in the context of intra-lineage decisions regulated by Notch. This specificity may be explained by when Notch signals, namely, at the end of mitosis. It is well established that clathrin-mediated endocytosis is shut down at mitosis and is progressively restored during cytokinesis. One mechanism contributing to this inhibition throughout mitosis is increased membrane tension. Since an increased requirement for actin is observed in cells in which membrane tension is high, it is speculated that the endocytosis of Dl more critically depends upon actin regulation by WASp during cytokinesis due to increased membrane tension. In other words, it is proposed that the pushing force provided by WASp-induced F-actin is needed for the efficient endocytosis of Dl to counteract the increased membrane tension associated with mitosis. Thus, while WASp and Arp2/3 likely play a general role in endocytosis, their activities become critical for the mechanical activation of Notch only when the inhibition of endocytosis needs to be overcome in late mitosis (Trylinski, 2019).

The requirement of WASp for Notch receptor activation is symmetric to those of Epsin, a conserved endocytic adaptor that helps generate the force for membrane invagination during endocytosis. Epsin is generally required for ligand endocytosis and Notch signaling in flies and mammals, with the exception of Notch-mediated intra-lineage decisions, as revealed by the development of sensory bristles in epsin mutant clones. It is speculated that the inhibition of Epsin at mitosis, possibly via its phosphorylation by CDK1/Cdc2, renders necessary the extra pushing force provided by WASp for the efficient endocytosis of Dl (Trylinski, 2019).

In summary, a model is proposed whereby the activity of WASp-Arp2/3 generally increases the efficiency of endocytosis and becomes specifically required only during cytokinesis, when Dl activates Notch to mediate intra-lineage decisions. This model may be general and apply to mammalian tissues where Notch is known to regulate intra-lineage decisions. N-WASp, the ubiquitously expressed WASp in mammals, is required for the maintenance of skin progenitor cells and hair follicle cycling in the mouse, and Notch plays a critical role in the self-renewal of skin stem cells. Whether Notch signaling is regulated by N-WASp in this context remains to be examined (Trylinski, 2019).

Genetic dissection of active forgetting in labile and consolidated memories in Drosophila

Different memory components are forgotten through distinct molecular mechanisms. In Drosophila, the activation of 2 Rho GTPases (Rac1 and Cdc42), respectively, underlies the forgetting of an early labile memory (anesthesia-sensitive memory, ASM) and a form of consolidated memory (anesthesia-resistant memory, ARM). This study dissected the molecular mechanisms that tie Rac1 and Cdc42 to the different types of memory forgetting. Two WASP family proteins, SCAR/WAVE and WASp, act downstream of Rac1 and Cdc42 separately to regulate ASM and ARM forgetting in mushroom body neurons. Arp2/3 complex, which organizes branched actin polymerization, is a canonical downstream effector of WASP family proteins. However, this study found that Arp2/3 complex is required in Cdc42/WASp-mediated ARM forgetting but not in Rac1/SCAR-mediated ASM forgetting. Instead, Rac1/SCAR may function with formin Diaphanous (Dia), a nucleator that facilitates linear actin polymerization, in ASM forgetting. The present study, complementing the previously identified Rac1/cofilin pathway that regulates actin depolymerization, suggests that Rho GTPases regulate forgetting by recruiting both actin polymerization and depolymerization pathways. Moreover, Rac1 and Cdc42 may regulate different types of memory forgetting by tapping into different actin polymerization mechanisms (Gao, 2019).

There are 3 major findings. First, 2 WASP family proteins, SCAR/WAVE and WASp, act as downstream effectors of Rac1-mediated ASM forgetting and Cdc42-mediated ARM forgetting, respectively. Second, although the Arp2/3 complex is a well-established effector that links activation of WASP family proteins to actin polymerization, it is only required in Cdc42/WASp-mediated ARM forgetting. Instead, formin Dia functions together with Rac1/SCAR in ASM forgetting. Third, feeding inhibitors of the Arp2/3 complex and Dia to fruit flies led to rather specific effects on ASM and ARM forgetting, raising the possibility of developing drugs on these molecular targets to treat memory-related diseases (Gao, 2019).

The effect of Rac1 on ASM forgetting has been tied to the activation of an actin depolymerization regulator cofilin presumably through a PAK/LIMK phosphorylation cascade. However, actin dynamics is a balanced play that requires continuous turnover between polymerization and depolymerization. It is not known whether signaling pathways regulating actin polymerization also play a role. There are 3 families of proteins that nucleate and promote actin polymerization, Arp2/3 complex, WH2-domain proteins, and formin. The finding that Arp2/3 complex and formin Dia function in ARM and ASM forgetting suggests that both actin polymerization and depolymerization pathways contribute to forgetting. How Arp2/3 complex and Dia separately contribute to ARM and ASM forgetting remains an open question. It is yet to be determined whether these proteins have different expression or subcellular locations in the MB neurons. However, it is interesting that Arp2/3 complex and formins are specialized in different types of actin polymerization (Gao, 2019).

In a working model, Cdc42 activates Arp2/3 complex via a canonical pathway (Cdc42/WASp/Arp2/3 complex), while Rac1-mediated ASM forgetting depends on SCAR/WAVE complex. This complex, in addition to SCAR/WAVE, includes at least 4 other members: Sra-1, Abi, HSPC300, and Kette. These additional members are thought to hold SCAR/WAVE in the complex in an inactive state, until GTP-bound Rac1 binds to Sra-1 and relieves the inhibition. On the other hand, the intact complex is essential for the stability of the SCAR/WAVE protein as well (i.e., failure to keep the intact complex can lead to SCAR degradation). This latter effect may explain the observation that RNAi knockdown of SCAR complex members has the same effect on inhibiting forgetting as the knockdown of SCAR. As a WASP family protein, SCAR/WAVE is able to associate with and activate Arp2/3 complex through its C-terminal region. However, RNAi knockdown of Arp2 and Arp3 and pharmacological inhibition of Arp2/3 complex specifically affects ARM forgetting, while no effects on ASM retention were observed. It is therefore proposed that Rac1/SCAR may function through Arp2/3 complex-independent mechanisms. SCAR/WAVE complex is reported to physically associates with Dia through one of its members, Abi, to regulate actin dynamics. Behavioral characterization of Dia knockdown and overexpression, as well as the genetic epistasis experiment, support the idea that Dia could be downstream of Rac1/SCAR in ASM forgetting. Details about the functional coordination between SCAR/WAVE and Dia therefore await further clarification (Gao, 2019).

GENE STRUCTURE

Genomic size - 6.5 kb

Exons - 7

PROTEIN STRUCTURE

Amino Acids - 527

Structural Domains

Drosophila WASp is ~35% identical to mammalian WASPs. Sequence similarity is particularly apparent within the recognized functional and structural domains of WASP proteins. Indeed, WASp binds both (GTP-bound) CDC42 and cytoskeletal elements, implying conservation of biochemical function. No additional homologs were identified in searches of the recently published sequence of the entire Drosophila genome, suggesting that WASp is the sole bona fide WAS gene family homolog in Drosophila. The major structural and functional domains of WASP proteins are found in the Drosophila protein. These include: the NH₂-terminal WH1 membrane-interacting domain; the CDC42 GTPase binding domain; the proline-rich SH3-binding domain; monomeric actin-binding domains homologous to yeast verprolin, and two COOH-terminal domains: a cofilin-homologous domain and an acidic tail that are responsible for Arp2/3 complex binding. Drosophila WASp binds the activated form of CDC42 in a blot overlay assay. A strong interaction between WASp and GTP-CDC42 and weak binding to GTP-Rac are observed. Binding to GTP-Rho could not be detected (Ben-Yaacov, 2001).

date revised: 5 August 2021

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

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

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