canoe: Biological Overview | Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

Gene name - canoe

Synonyms -

Cytological map position - 82E4--82E8

Function - scaffolding protein

Keywords - eye, wing, Ras pathway, Notch pathway, glia

Symbol - cno

FlyBase ID: FBgn0259212

Genetic map position - 3-49

Classification - GLGF/PDZ domain

Cellular location - submembrane surface

NCBI links: Precomputed BLAST | Entrez Gene
Recent literature
Keder, A., Rives-Quinto, N., Aerne, B. L., Franco, M., Tapon, N. and Carmena, A. (2015). The Hippo Pathway Core Cassette Regulates Asymmetric Cell Division. Curr Biol 25: 2739-2750. PubMed ID: 26592338
Asymmetric cell division (ACD) is a crucial process during development, homeostasis, and cancer. Stem and progenitor cells divide asymmetrically, giving rise to two daughter cells, one of which retains the parent cell self-renewal capacity, while the other is committed to differentiation. Any imbalance in this process can induce overgrowth or even a cancer-like state. This study shows that core components of the Hippo signaling pathway, an evolutionarily conserved organ growth regulator, modulate ACD in Drosophila. Hippo pathway inactivation disrupts the asymmetric localization of ACD regulators, leading to aberrant mitotic spindle orientation and defects in the generation of unequal-sized daughter cells. The Hippo pathway downstream kinase Warts, LATS1-2 in mammals, associates with the ACD modulators Inscuteable and Bazooka in vivo and phosphorylates Canoe, the ortholog of Afadin/AF-6, in vitro. Moreover, phosphosite mutant Canoe protein fails to form apical crescents in dividing neuroblasts in vivo, and the lack of Canoe phosphorylation by Warts leads to failures of Discs Large apical localization in metaphase neuroblasts. Given the relevance of ACD in stem cells during tissue homeostasis, and the well-documented role of the Hippo pathway as a tumor suppressor, these results represent a potential route for perturbations in the Hippo signaling to induce tumorigenesis via aberrant stem cell divisions.
Rives-Quinto, N., Franco, M., de Torres-Jurado, A. and Carmena, A. (2017). canoe and scribble loss synergizes causing tumor-like overgrowth via Ras activation in neural stem cells and epithelia. Development [Epub ahead of print]. PubMed ID: 28619817
Over the past decade an intriguing connection between asymmetric cell division, stem cells and tumorigenesis has emerged. Neuroblasts, the neural stem cells of the Drosophila central nervous system, divide asymmetrically and constitute an excellent paradigm for further investigating that connection. This study shows that the simultaneous loss of the asymmetric cell division regulators Canoe (Afadin in mammals) and Scribble in neuroblast clones leads to tumor-like overgrowth through both a severe disruption of the asymmetric cell division process and a canoe loss-mediated Ras-PI3K-Akt activation. Moreover, canoe loss also interacts synergistically with scribble to promote overgrowth in epithelial tissues, here just by activating the Ras-Raf-MAPK pathway. Finally scribble functionally related genes discs large and lethal (2) giant larvae were shown to contribute to repress the Ras-MAPK signaling cascade in epithelia. Hence, this work uncovers novel cooperative interactions between all these well-conserved tumor suppressors to ensure a tight regulation of the Ras signaling pathway.

canoe(cno) has a dual character, interacting as it does with genes in both the Notch and the Ras pathways. In the Notch pathway, canoe interacts with the Notch pathway genes Notch and scabrous. Cno protein and its mammalian homolog AF-6 have also been shown to physically interact with Ras. The cno locus was defined by Jurgens (1984). Extra machrochaetes (sensory bristles) are occasionally seen in the head, notum and scutellum, and some mutants show variable but conspicous wing phenotypes such as a notched blade and the loss of a cross vein. In the eye, clones of mutant photoreceptor cells are devoid of any ommatidia, while small mutant patches have reduced numbers of photoreceptors. Ommatidia of mutant pupae often contain extra primary pigment cells. Some of these ommatidia lack a secondary pigment cell, resulting in the apparent fusion of adjacent facets (Miyamoto, 1995).

Canoe and the Notch pathway

Fused ommatidia have been reported to occur in scabrous mutants; sca mutants are also accompanied by bristle abnormalities. Although the compound eye of sca1 homozygotes contains some fused facets and others with excess or reduced numbers of photoreceptors, the majority of ommatidia have a normal complement of eight photoreceptors. When sca1 homozygotes are also heterozygous for the mutant allele cnomis1, the phenotype is dramatically enhanced: in such cases, many ommatidia are fused to one another. The eyes of double mutants for sca1 and cnomis1 are extremely rough, accompanied by giant ommatidia, formed as a result of fusion. The bristle phenotype of sca1 is also enhanced by cnomis1. Double mutant flies show supernumerary bristles. There is a genetic interaction between the two loci, also revealed by an examination of wing morphology. Double mutants have rumpled wings curved downward, with occasional shortening of the anterior cross vein. cno likewise interacts with the split allele of Notch, for eye, bristle and wing development. An interrupted wing vein in Ax1, one N allele producing an activated form of Notch protein, is dominantly suppressed by cnomis1 (Miyamoto, 1995).

Additonal interactions occur between cno and shaggy, armadillo, and myospheroid. It is intriguing that the genes encoding proteins involved in adhesive cell-cell communications are shown to interact with cno. Proteins coded for by these genes mediate the adhesion of the different cell layers in a variety of developing tissues. It is conceivable that Cno is localized to junctional complexes (Miyamoto, 1995).

Canoe and the Ras pathway

Just when it seemed that Canoe was being understood in terms of its interaction with the Notch pathway, it was also becoming clear that Canoe and its mammalian homolog AF-6 physically interact with Ras (Kuriyama, 1996). Morphological studies of the fly's compound eye suggest that induction of photoreceptors, including the R7 photoreceptor, is unaffected by cno mutations. The linear alignment of the R7 photoreceptor is disrupted in the pupal retina of cno mutants, but without exception a single R7 photoreceptor is formed in each ommatidium in cno mutant retinae (Matsuo, 1997). Since Ras signaling is involved in R7 fate determination (see Sevenless), what exactly is the role of Canoe in eye development?

It has been suggested that R7 photoreceptors occupy unusual positions in pupal discs because the mutant discs fail to form a correct mesh of pigment cells, which shape ommatidia into regular hexagonal structures. A strong mutation in cno eliminates from one to three cone cells from most ommatidia. In contrast, overexpression of cno+ results in the formation of supernumerary cone cells. These results indicate that an increase in cno+ expression promotes cone cell induction and a significant decrease in cno+ expression prevents cone cell precursors from differentiating into cone cells (Matsuo, 1997).

The involvement of Ras in non-neural retinal cell formation has attracted little attention. Freeman (1996) has provided evidence that the EGF receptor is required for the development of all retinal cells, including cone cells and pigment cells. A dominant negative form of EGF receptor prevents cell differentiation, whereas a secreted form of Spitz, the EGF receptor ligand, causes differentiation of supernumerary cells when expressed in the R7 equivalence group cells (cells from which arise a single R7 photoreceptor). Expression of a constitutively active form of Ras1, an important component in the EGF receptor signal transduction, causes a strong rough-eye phenotype. The rough-eye phenotype of active Ras is dramatically enhanced in cno heterozygotes, even though cno mutation is completely recessive and flies heterozygous for cno exhibit no specific phenotype. Overexpression of cno+ in R7 equivalence group cells similarly enhances roughness of constitutively active Ras1, without increasing the number of R7 cells beyond the level found without cno+ overexpression. Since the cno mutations interfere with cone cell formation without affecting photoreceptor induction, it seems plausible that Cno interacts with Ras1 in the developing cone cells (Matsuo, 1997).

These observations raise a question: from what cells do the extra cone cells originate? Two possibilities are considered: (1) they originate from retinal precursors overproduced due to accelerated proliferation and/or (2) they are generated at the expense of pigment cells that differentiate after cone cells, and from cells that are fated to die by apoptosis. Since proliferative activity in third instar larval eye discs is indistinguishable in mutant discs, it is likely that supernumerary cone cells are recruited from a pool of retinal precursors that normally develop into pigment cells and/or from cells that would otherwise be eliminated by apoptosis. It is thought that Cno acts in oppositon to Ras1, but the exact mechanism is not known. Nevertheless, an attractive model is that direct binding of Cno to Ras1 occludes the effector domain of Ras, which otherwise provides the binding site for Raf and consequent transduction of the Ras signal. This hypothesis is based on the results of an in vitro experiment demonstrating that mammalian cRas-1 competes with AF-6 ( a mammalian homolog of Cno) in binding to the H-Ras effector domain (Kuriyama, 1996). The direct binding of Cno to Drosophila Ras1 has been confirmed to occur in yeast, by two-hybrid assay. However, the hypothesis cannot be used to satisfactorily explain why overexpression of cno+ results in enhancement, rather than inhibition, of constitutively active Ras action in cone cell formation (Matsuo, 1997).

An alternative to either of these two models is suggested. It is postulated that there are two counteracting signaling pathways in which Cno participates: one for promoting the induction of cone cells (via Ras) and another for repressing this induction. It is postulated that Cno acts as a negative regulator for both pathways. Cno predominately associates with the repressing pathway when Ras activation is minimal, where absence of Cno causes disinhibition of the repression of the cone cell formation. With maximal Ras activation, extra cone cells form because the extremely strong activation of Ras1 overrides the inhibition by Cno and the action of the repressing pathway. Under such conditions, reduction of the Cno levels results in disinhibition of the promoting pathway, leading to enhancement of extra cone cell formation (Matsuo, 1997).

It is worth noting that reduction in Notch activity results in either an increase or a decrease in the number of cone cells, depending on the timing of Notch inactivation. Thus the Notch action is likewise 'bidirectional' in cone cell formation, as found for Cno action. In this context, it is of interest to note that some components of the EGF receptor pathway display strong genetic interactions with the Notch group of genes in eye, wing and leg morphogenesis. Cno is a reasonable candidate for a molecule that mediates such crosstalk between the Notch pathway and the EGF receptor pathway (Matsuo, 1997).

The PDZ protein Canoe regulates the asymmetric division of Drosophila neuroblasts and muscle progenitors

Asymmetric cell division is a conserved mechanism to generate cellular diversity during animal development and a key process in cancer and stem cell biology. Despite the increasing number of proteins characterized, the complex network of proteins interactions established during asymmetric cell division is still poorly understood. This suggests that additional components must be contributing to orchestrate all the events underlying this tightly modulated process. The PDZ protein Canoe (Cno) and its mammalian counterparts AF-6 and Afadin are critical to regulate intracellular signaling and to organize cell junctions throughout development. Cno functions as a new effector of the apical proteins Inscuteable (Insc)-Partner of Inscuteable (Pins)-Gαi during the asymmetric division of Drosophila neuroblasts (NBs). Cno localizes apically in metaphase NBs and coimmnunoprecipitates with Pins in vivo. Furthermore, Cno functionally interacts with the apical proteins Insc, Gαi, and Mushroom body defect (Mud) to generate correct neuronal lineages. Failures in muscle and heart lineages are also detected in cno mutant embryos. These results strongly support a new function for Cno regulating key processes during asymmetric NB division: the localization of cell-fate determinants, the orientation of the mitotic spindle, and the generation of unequal-sized daughter cells (Speicher, 2008).

NBs delaminate from the neuroectoderm inheriting the apicobasal polarity of the neuroectodermal cells, in which the PDZ proteins Bazooka (Baz)/Par-3 and DmPar-6 and the kinase DaPKC localize apicolaterally. After delamination, NBs maintain the apical localization of Baz/DmPar-6/DaPKC. The cytoplasmic PDZ protein Cno localizes at the adherens junctions of some epithelial cells, and it was asked whether Cno was also present in the neuroectoderm and in the delaminated NBs. Double immunofluorescences with antibodies against Cno and Baz showed that these proteins colocalize both apicolaterally at the adherens junctions of neuroepithelial cells and apically in the delaminated metaphase NBs (mNBs). At later phases of the NB division, Cno was no longer detected (Speicher, 2008).

Apical proteins, such as Baz/Par-3, are critically involved in regulating cell-fate determinants localization and spindle orientation at metaphase. Given that Cno was detected in an apical crescent in mNBs, it was asked whether Cno was also required for modulating those events. In control embryos, the cell-fate determinant Numb was basally located in 95.4% of mNBs. In cno2 zygotic mutants, Numb was uniform or undetectable or was present in nonbasal crescents in 47.9% of the mNBs analyzed. cno2 has been defined as the strongest allele of cno, although the particular lesion associated is unknown. However, cno2 is probably a null allele because cno2 over the Df(3R)6-7 (covering the cno gene) showed a similar percent of Numb localization failures. Additionally, cno3, another strong allele of cno considered as a null displayed defects in Numb localization in comparable cases. The basal distribution of the scaffolding protein Miranda (Mira) was also altered in 16.9% of mNBs of cno2 mutants. Indeed, the localization of two Mira cargo proteins, the cell-fate determinants Prospero (Pros) and Brain Tumor (Brat), was affected in mNBs. The variable penetrance of the cno2 mutant phenotype observed for the different proteins analyzed may reflect, at least in part, the different sensitivity of the antibodies used (Speicher, 2008).

Intriguingly, the orientation of the mitotic spindle in mNBs of cno2 mutants was randomized in 18.3% of the cases. In control embryos, the spindle is tightly aligned with the center of Numb crescents in mNBs. In cno2 mutants, the spindle was uncoupled with the Numb crescent in 7.7% of the mNBs that showed these crescents (either basal or at other incorrect localizations). The maternal contribution of cno might reduce the penetrance of these phenotypes (Speicher, 2008).

The overexpression of Cno also caused Numb localization failures (45.8%) and aberrant spindle orientations (39%) in mNBs. Hence, the results showed that Cno regulates essential processes during asymmetric NB division: the basal localization of cell-fate determinants and the proper orientation of the mitotic spindle (Speicher, 2008).

Another characteristic feature of asymmetric NB division is the different cell size of the progeny. Hence, whether Cno was also regulating this process was analyzed. Control telophase NBs (tNBs) showed unequal-sized daughter cells in 100% of the cases analyzed. In cno2 mutants, equal-size divisions were observed in 21.3% of tNBs. Two redundant pathways, Baz/DaPKC/Insc and Pins-Gαi, regulate cell size and mitotic-spindle asymmetry at the NB apical pole . Only when both pathways are compromised is the different size of the daughter cells affected. The data suggested that Cno functions downstream of Gαi. Thus, Cno might belong to the Pins-Gαi pathway. Indeed, when both insc and cno were eliminated, 85.2% of tNBs showed equal-sized daughter cells, a much more penetrant phenotype than those displayed by each single mutant. Moreover, ΔGαi, cno2 double mutants showed a much lower percentage of equal-sized divisions (30.4%) than the inscP49; cno2 double mutants. Hence, these results strongly suggest that Cno participates within the Pins-Gαi pathway to regulate NB progeny size (Speicher, 2008).

Given the defects observed in cno2 mutant embryos during NBs division, it was asked whether neuronal lineages were altered in cno2 mutants. The lineage of the ganglion mother cell (GMC) 4-2a has been extensively studied. This GMC expresses the transcription factor Even-Skipped (Eve) and divides asymmetrically to give rise to two different neurons called RP2 and RP2 sibling. Both maintain the expression of Eve initially; however, at later stages of embryogenesis, only the RP2 neuron keeps expressing Eve . In control embryos, 0.9% of the segments analyzed showed defects in the number of RP2s. In cno2 mutants, two or no RP2s were detected per hemisegment in 5.7% of the segments analyzed. Such a result suggested failures in the GMC 4-2a asymmetric division. This phenotype was also observed in cnomis1 hypomorph mutants (4.6%) as well as in mutants for genes that are critical during asymmetric cell division. For example, homozygotes for DaPKCk06403, inscP49, ΔGαi, and mud4 (zygotic null mutant embryos) showed defects in the GMC 4-2a lineage in 6.4%, 13.8%, 2.5%, and 8.3% of the segments analyzed, respectively. Hence, it was next investigated whether Cno was interacting with these proteins to properly generate the GMC 4-2a neuronal lineage. Double heterozygotes DaPKCk06403/+; cno2/+ showed defective RP2 number in 0.8% of segments. This result is consistent with a lack of functional interactions between DaPKC and Cno. However, double heterozygotes inscP49/+; cno2/+ and ΔGαi, +/+, cno2 showed an altered RP2 lineage in 14.4% and 7.6% of the segments analyzed. In addition to the analysis of double heterozygotes, it was found that the cnomis1 phenotype was significantly enhanced in a mud4 zygotic null mutant background. Altogether, these results indicated that Cno functionally interacts with the apical proteins Insc, Gαi, and Mud during the asymmetric cell divisions that generate specific neuronal lineages in the CNS (Speicher, 2008).

Since Cno functionally interacts with Insc, Gαi, and Mud, the epistatic relationships between them were analyzed. To investigate whether Cno was acting upstream of the apical proteins, the localization of Baz, Insc, and Gαi was examined in cno2 mutant embryos. The distribution of all these proteins was normal in cno2 mutants. This result suggested that Cno acts either downstream or in parallel to Baz, Insc, and Gαi. To clarify this point, the distribution of Cno was analyzed in loss- and gain-of-function (lof and gof) mutants for several apical proteins. In inscP49 lof mutants, Cno was untraceable or showed a wrong orientation in 78.8% of the mNBs analyzed. Insc overexpression also caused failures in Cno localization (76%); Cno was either undetectable (13/21) or present in not-apical crescents (3/21). Likewise, in Gβ13F maternal and zygotic null mutant embryos, in which Gαi is lost, Cno was mislocalized or undetectable in 94% of the mNBs. Moreover, the overexpression of Gαi caused a striking mislocalization of Cno in 100% of the mNBs analyzed. The NuMA-related protein Mud binds the apical protein Pins and functions downstream of Pins-Gαi to regulate spindle orientation. In mud mutant NBs, the spindle fails to tightly align with the basal crescent, and this failure is also shown by cno2 mutant NBs. Additionally, Cno and Mud interacted genetically. Hence, it was asked whether Cno functions along with Mud to regulate spindle orientation. In control embryos, Mud localized at the apical cortex of mNBs (97%) and at the two centrosomal regions (100%). In cno2 lof mutants, Mud failed to accumulate apically in 49% of mNBs, and 15% of NBs showed Mud localization in one or none of the two centrosomes. cno gof also caused failures in Mud localization (38%). Altogether, these results strongly support a function of Cno downstream of Insc and Pins-Gαi and upstream of Mud during asymmetric NB division (Speicher, 2008).

Given the functional relationships found between Cno and apical proteins during asymmetric NB division, it was asked whether Cno was physically interacting with some of these proteins. Coimmunoprecipitation experiments from Drosophila embryo extracts showed that Cno is forming a complex with Pins. Cno did not physically interact with DmPar6, Baz, DaPKC, or other apical proteins tested such as Insc, Gαi, and Mud (Speicher, 2008).

Pins also forms a complex in the delaminated metaphase NBs with the tumor-suppressor protein Discs Large (Dlg) and the kinesin Khc-73, an astral microtubule-binding protein. First, at prophase, the DmPar6/Insc pathway is required to polarize Pins/Gαi at the apical pole of the NB. Then, at metaphase, the Pins/Gαi/Dlg/Khc-73 complex forms, and it is key for tightly coupling cortical polarity with spindle orientation. Hence, it was asked whether Cno was also forming part of this complex. Experiments showed that neither Dlg nor Khc-73 coimmunoprecipitate Cno in embryo extracts. This result indicated that Cno is not forming part of the Dlg/Khc-73 complex (Speicher, 2008).

Altogether, a working network of protein interactions is proposed. Analysis of epistatic relationships between apical proteins and Cno showed that Cno is acting downstream of Insc-Pins-Gαi and upstream of Mud. Indeed, genetic analysis suggests that, at least for the control of daughter cells size asymmetry, Cno functions within the Pins-Gαi pathway, in parallel to the DaPKC-Baz-Insc pathway. Accordingly, Cno was found to form a complex with Pins in vivo. Cno did not coimmunoprecipitate with Gαi, though. One possibility is that Cno and Gαi are mutually exclusive in the complex that each of them forms with Pins. Additionally, transient or labile interactions between Cno and Gαi may occur that were not possible to detect. Another Pins interacting partner, the microtubule-binding protein Mud contributes to coordinate spindle orientation with cortical polarity. Given the functional relationships that were found between Cno and Mud, Cno could act in a complex with Pins to modulate Mud localization and, consequently, spindle orientation (Speicher, 2008).

Finally, it was asked whether the function of Cno during asymmetric cell division was conserved in different tissues. Since the NBs of the CNS, the Drosophila somatic muscle and heart progenitors divide asymmetrically to give rise to two different founder cells. Cno is present in the somatic mesoderm and is required for muscle and heart progenitor specification. Hence, it was aked whether Cno was also functioning during the asymmetric division of muscle and heart progenitors. For this analysis, focus was placed on two dorsal progenitors called P2 and P15 that express the transcription factor Eve and whose lineages have been characterized in detail. In this study, it was found that the transcription factor Seven-up (Svp), a characteristic marker of a subset of cardial cells, was expressed in a dorsal founder cell of unknown identity until now, which is here named founder of Svp cardial cells (FSvpCs). With all these markers, specific for individual derivatives, whether dorsal muscle and cardial lineages were altered in cno2 mutants was analyzed. It was found that at late stages (stage 14), 3.1% of hemisegments (n = 96) showed simultaneously either loss of EPCs and gain of DO2 muscle or gain of EPCs and loss of the DO2 muscle (P2 lineage). In control embryos, this phenotype was not observed in any of the hemisegments analyzed. Indeed, Numb localization, which was basal in 100% of the metaphase P2s analyzed in control embryos, was altered in 93% of metaphase P2s in cno2 mutants. Hemisegments showing duplication of DA1 muscle and loss of SvpCs or DA1 muscle loss and gain of SvpCs (P15 lineage) were also detected in cno2 mutants. Hence, Cno was required for the asymmetric division of progenitor cells both in the CNS and in the mesoderm (Speicher, 2008).

In conclusion, the discovery of new modulators of asymmetric cell division, as described in this study, for the PDZ protein Cno, is key to complete understanding of this intricate process. Especially challenging in the future will be unraveling the complete network of connections between all the players required for an accurate asymmetric cell division (Speicher, 2008).


Genome length - 70 kb

cDNA length - 7296

Transcript lengths - 7.0, 7.5 and 8.2 kb

Bases in 5' UTR - 484

Exons - 15

Bases in 3' UTR - 1133


Amino Acids - 1893

Structural Domains

The sequence of Cno cDNA clones isolated from an embryonic cDNA library reveal a long open reading frame that encodes protein with the GLGF/DHR motif, a conserved sequence in Discs large, Dishevelled, and some other proteins associated with cellular junctions (Miyamoto, 1995). The molecular structure of Cno suggests its direct association with Ras. It has significant homology with a mammalian Ras-binding protein AF-6 (Kuriyama, 1996), which as been cloned as a fusion partner of All-1, a protein involved in acute myeloid leukemias in humans and a homolog of Drosophila Trithorax. Cno and AF-6 share two putative Ras-binding domains (RA1 and RAs), located at the N-terminus, as well as a kinesin-like and a myosin-V-like domain, and a Discs large homologous region, (DHR, also known as the GLGF or PDA motif) (Matsuo, 1997 and Kuriyama, 1996). The direct interaction of Raf, phosphatidylinositol-3-OH kinase, Ral GDS and Rin1 with activated Ras has been demonstrated. There is no obvious homology among Ras-interacting interfaces of these proteins with either AF-6 or Canoe, indicating that activated Ras can recognize a variety of target interfaces (Kuriyama, 1996 and references)

canoe: Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

date revised: 18  January 98

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.