InteractiveFly: GeneBrief

Myosin 31DF: Biological Overview | References

The internal organs of animals often have left-right asymmetry. Although the formation of the anterior-posterior and dorsal-ventral axes in Drosophila is well understood, left-right asymmetry has not been extensively studied. This study demonsrates that the handedness of the embryonic gut and the adult gut and testes is reversed (not randomized) in viable and fertile homozygous Myo31DF mutants. Myo31DF encodes an unconventional myosin, Drosophila MyoIA [also referred to as MyoID in mammals; Gillespie, 2001; Morgan, 1995), and is the first actin-based motor protein to be implicated in left-right patterning. Myo31DF is required in the hindgut epithelium for normal embryonic handedness. Disruption of actin filaments in the hindgut epithelium randomizes the handedness of the embryonic gut, suggesting that Myo31DF function requires the actin cytoskeleton. Consistent with this, it was found that Myo31DF colocalizes with the cytoskeleton. Overexpression of Myo61F (myosin-IB), another myosin I, reverses the handedness of the embryonic gut, and its knockdown also causes a left-right patterning defect. These two unconventional myosin I proteins may have antagonistic functions in left-right patterning. It is suggested that the actin cytoskeleton and myosin I proteins may be crucial for generating left-right asymmetry in invertebrates (Hozumi, 2006). An accompanying paper (Spéder, 2006, see below) demonstrates that Myo31DF regulates dextral looping of genitalia.

Mechanisms that create characteristic left-right asymmetry have been studied extensively in vertebrates. However, although the organs of many invertebrate species also have left-right asymmetry, the mechanisms by which this asymmetry arises are largely unknown. In Drosophila, several organs have left-right asymmetry, including the embryonic gut, the adult brain and the genitalia (Hozumi, 2006).

To identify genes involved in left-right asymmetry of the Drosophila embryonic gut, a genetic screen was performed using a collection of P-element lines. The embryonic gut is composed of three major parts, the foregut, midgut and hindgut, all of which have characteristic left-right asymmetry. 75.7% of homozygous Myo31DF^souther embryos show synchronous inversion of the midgut and hindgut. In these embryos, the hindgut and midgut are the mirror-image of those in wild-type embryos, rather than showing randomized patterning. In contrast, foregut handedness was normal in all cases examined, indicating that this phenotype was heterotaxic. Myo31DF^souther is a background mutation of the Gene Search Drosophila line GS14508. Deficiency mapping was used to map the cytological location of Myo31DF^souther to between 30D and 31F. Complementation tests were performed between Myo31DF^souther and lines bearing mutations that map to this region. Myo31DF^souther failed to complement Myo31DF^K1 (Spéder, 2006), an allele of Myo31DF encoding Drosophila MyoID (Hozumi, 2006).

The transposable element gypsy was inserted into the 5'-untranslated region of the Myo31DF gene in Myo31DF^souther. Myo31DF^L152 was one of five ethylmethanesulfonate (EMS)-induced Myo31DF alleles isolated in a large-scale EMS mutant screen. Myo31DF^L152 has a base substitution that introduces a premature stop codon at amino acid 331, resulting in a putative truncated product. Myo31DF overexpression from UAS-Myo31DF driven by byn-Gal4 in the hindgut and posterior midgut and their primordial counterparts rescued the left-right defects of Myo31DF^souther embryos, indicating that Myo31DF was responsible for the heterotaxia. In Myo31DF^souther embryos, NP2432-driven expression of Myo31DF in the hindgut epithelium, but not in other parts of the embryonic gut, such as the midgut and mesoderm, was sufficient to rescue this heterotaxia, suggesting that Myo31DF is required in the hindgut epithelium. Furthermore, the frequency of the handedness defect was similar in Myo31DF^L152 homozygous and Myo31DF^L152/Df(2L)J2 embryos. Thus, Myo31DF^L152 is probably a null mutant of Myo31DF. All homozygous Myo31DF mutants isolated in this study were viable and fertile, with normal hindgut tissue specification, suggesting that Myo31DF function is largely restricted to left-right patterning. No maternal phenotype or Myo31DF gene contribution was detected. Notably, the foregut became a mirror-image of its wild-type counterpart when Myo31DF was overexpressed in the entire embryo, but other parts of the gut were normal. Therefore, it is suggested that Myo31DF is not involved in the left-right asymmetrical development of the foregut in wild-type embryos, but can reverse foregut handedness (Hozumi, 2006).

The adult hindgut and testes, which are regenerated during metamorphosis, were examined. These also showed inversed handedness in the Myo31DF homozygote. In most Myo31DF^L152 adults, the loop of the hindgut and spiral of the testes were reversed, although not always synchronously. The function of the Myo31DF gene was knocked-down using RNA interference (RNAi) in vivo. The expression of double-stranded Myo31DF RNA driven by byn-Gal4 caused inversion of the adult, but not the embryonic, hindgut. Thus, the left-right pattern involving Myo31DF is not transmitted during metamorphosis (Hozumi, 2006).

In situ hybridization revealed Myo31DF expression in the amnioproctodeal invagination at stage 6. At stages 12-14, Myo31DF messenger RNA was strongly detected in the primordial midgut and hindgut, and in the proventriculus, midgut and hindgut. A sense-strand probe of Myo31DF gave no detectable signal. Immunostaining of wild-type embryos with an anti-Myo31DF antibody (anti-Myo31DF-1P) also labelled the midgut and hindgut. These signals were absent in Myo31DF^souther and Myo31DF^L152 homozygotes, indicating that the staining was specific. Myo31DF mRNA and protein were detected in a symmetrical pattern before the development of left-right asymmetry. Myo31DF protein is expressed in the adult gut (Morgan, 1995). No Myo31DF expression was detected in the foregut, which may account for the absence of any laterality defect in the foregut of Myo31DF mutants (Hozumi, 2006).

Myo31DF protein binds to actin in an ATP-dependent manner (Morgan, 1995). The co-localization of Myo31DF was examined and the actin cytoskeleton in cultured Drosophila S2 cells. A green fluorescent protein (GFP)-tagged Myo31DF (Myo31DF-GFP) had wild-type function, given that its overexpression rescued Myo31DF^souther. Myo31DF-GFP co-localized with actin, mostly at cell protrusions. In epithelial cells of the hindgut, endogenous Myo31DF was detected as punctate staining, partly overlapping with cortical actin (Hozumi, 2006).

byn-Gal4-driven misexpression of GFP-tagged moesin, an actin-binding protein, in wild-type embryos caused a reduction in actin filaments in the apical region of the hindgut epithelium, where Myo31DF function is required. Notably, the midgut and hindgut always had the same handedness, but the handedness was random (not inversed), in wild-type and Myo31DF homozygous embryos. Embryo handedness was also affected by NP2432-driven GFP-moesin expressed in the hindgut epithelium only. However, GFP-moesin expression in the midgut only did not affect handedness (Hozumi, 2006).

To investigate further the functional link between Myo31DF and the actin cytoskeleton, the phenocritical period for inducing left-right defects was determined using Myo31DF RNAi and GFP-moesin misexpression, following the TARGET method. Both Myo31DF knockdown and GFP-moesin expression in the hindgut 0-24 h after pupation caused similar defects in adult gut handedness, suggesting that Myo31DF and the appropriate organization of actin filaments are required at the same time. GFP-moesin also affected handedness in the Myo31DF embryo, suggesting that the default handedness, which may be manifested in the Myo31DF homozygote, also depends on the actin cytoskeleton. The involvement of three Rho GTPase family proteins, Rho, Rac and Cdc42, which regulate the organization of the actin cytoskeleton, was examined. Expression of dominant-negative forms of these proteins, especially Rho, in the hindgut induced synchronous left-right defects in the embryonic midgut and hindgut. Together, these results suggest that Myo31DF depends on the actin cytoskeleton to generate left-right asymmetry (Hozumi, 2006).

Cell division and cell death do not occur during left-right asymmetric development of the hindgut. It is therefore speculated that the rearrangement of hindgut epithelial cells may be part of this process. To test this possibility, a time-lapse analysis was performed. The position of each cell was visualized by labelling the nucleus with GFP. Cell rearrangement, which coincided with the left-right bias associated with left-handed rotation of the hindgut, was suggested by the significant intercalation of some cells (Hozumi, 2006).

Another myosin I protein, Myo61F (Drosophila MyoIB, also referred to as MyoIC in mammals), has been reported in Drosophila. Myo61F protein is detected in the embryo and adult gut. To test whether Myo61F is also involved in left-right asymmetry, Myo61F was overexpressed using UAS-Myo61F or GS9889 driven by byn-Gal4. Unexpectedly, Myo61F overexpression resulted in inversion of the midgut and hindgut in both cases. In contrast, Myo31DF overexpression did not affect the handedness of these organs. These results suggest that Myo31DF and Myo61F have antagonistic functions in creating the left-right asymmetry of these organs. The involvement of Myo61F in left-right asymmetry is also supported by the finding that its knockdown by RNAi results in the left-right defect in the embryonic midgut (Hozumi, 2006).

Homozygous Myo31DF embryos show reversed handedness of embryonic and adult visceral organs, which may represent the default state of left-right asymmetry in Drosophila. This situation is similar to the function of the sinistral gene in the freshwater snail, Limnea (although the sinistral gene is required maternally) (Wood, 1997; Shibazaki, 2004). Normal handedness is still seen in 25% of Myo31DF homozygotes. It is speculate that some other myosin gene(s) has a redundant function in left-right patterning. Inversion of the anteroposterior axis does not affect laterality, suggesting that left-right pattering occurs zygotically; this is consistent with the zygotic function of Myo31DF. These results also suggest that an actin-based mechanism, which can align itself to either an anteroposterior-dorsoventral reference or the pre-existing sinistral handedness, exists to direct the rotation of the hindgut epithelium. Since myosin I proteins are involved in vesicular transport (Huber, 2000), it is proposed that Myo31DF and Myo61F, which on the basis of their structures are believed to move to the plus ends of actin filaments, carry left-right determinants with opposite activities. Thus, both left-right determinants would be concentrated in the plus ends of actin filaments that have a hypothetical planar polarity. In the Myo31DF mutant, only the opposing determinant is concentrated here, which reverses the handedness. According to this model, disruption in actin organization would result in left-right randomization, as was indeed observed experimentally (Hozumi, 2006).

Type ID unconventional myosin controls left-right asymmetry in Drosophila

Breaking left-right symmetry in Bilateria embryos is a major event in body plan organization that leads to polarized adult morphology, directional organ looping, and heart and brain function. However, the molecular nature of the determinant(s) responsible for the invariant orientation of the left-right axis (situs choice) remains largely unknown. Mutations producing a complete reversal of left-right asymmetry (situs inversus) are instrumental for identifying mechanisms controlling handedness, yet only one such mutation has been found in mice (inversin; Morgan, 1998) and snails (Ueshima, 2003; Freeman, 1982). The conserved type ID unconventional myosin 31DF gene (Myo31DF) has been identified as a unique situs inversus locus in Drosophila. Myo31DF mutations reverse the dextral looping of genitalia, a prominent left-right marker in adult flies. Genetic mosaic analysis pinpoints the A8 segment of the genital disc as a left-right organizer and reveals an anterior-posterior compartmentalization of Myo31DF function that directs dextral development and represses a sinistral default state. As expected of a determinant, Myo31DF has a trigger-like function and is expressed symmetrically in the organizer, and its symmetrical overexpression does not impair left-right asymmetry. Thus Myo31DF is a dextral gene with actin-based motor activity controlling situs choice. Like mouse inversin (Nurnberger, 2002), Myo31DF interacts and colocalizes with β-catenin, suggesting that situs inversus genes can direct left-right development through the adherens junction (Spéder, 2006).

In wild-type males, the genital plate, to which the spermiduct is attached, undergoes a 360° clockwise (dextral) rotation when viewed from the posterior pole. This directional looping is reminiscent of other coiling processes such as mammalian heart tube looping and snail spiral development. As in other species, one direction is dominant among the Drosophilidae, the dextral rotation, with no sinistral species reported to date. This study identified an insertional mutation, KG02246, which shows a striking inverted phenotype when combined with deficiencies covering the 31DF genomic region. In KG02246/Df(2L)Exel7048 or KG02246/Df(2L)J3 males, genitalia rotation is variable, with ~60% of individuals showing sinistral rotation. Imprecise excision of KG02246 generated two genomic deletions, KG02246¹ and KG02246², both of which presented a stronger, 100% inverted (sinistral) genitalia rotation phenotype. Thus, KG02246 alleles identify the first situs inversus mutations in Drosophila and provide genetic evidence for the existence of a left-right axis in this organism (Spéder, 2006).

Drosophila Myo31DF is a conserved myosin belonging to the Myo1D family. It is known to interact with F-actin (Morgan, 1995) and it colocalizes with actin-rich structures in different tissues. After mouse inversin, Drosophila Myo31DF represents the second situs inversus gene to be molecularly identified. The mouse inversin gene (Invs) encodes a protein with ankyrin repeats and two IQ domains (Morgan, 2002; Eley, 2004) that bind calmodulin, a Ca²⁺ -dependent regulatory protein (Bahler, 2002). Like inversin, Myo31DF contains two IQ domains essential for its function. Indeed, a Myo31DF form lacking IQ domains (Myo31DFDeltaIQ) is unable to rescue Myo31DF mutations (Spéder, 2006).

To determine the function of Myo31DF in genitalia rotation, its expression was examined in the genital disc, the precursor of adult genitalia. The genital disc is composed of segments A8, A9 and A10, each with an anterior and a posterior. Immunostaining of wild-type flies with two polyclonal antibodies directed against overlapping regions of the Myo31DF tail domain, anti-Myo31DF-1P and anti-Myo31DF-3P, or using a Myo31DF-Gal4 enhancer-trap line (NP1548) revealed symmetrical expression of Myo31DF in a double chevron-like pattern restricted to the ventral domain of the male genital disc. This expression, starting in third instar larvae and remaining unchanged during this stage, was absent in Myo31DF^K2 mutant discs. Double staining with an A8-specific marker in the genital disc (tsh-Gal4 > myr-RFP) indicated that Myo31DF is expressed in the A8 segment, with one chevron in the posterior compartment and the other in the anterior compartment. Consistently, expression of two copies of an inhibitory RNA gene (2 x Myo31DF^RNAi) specifically in the A8 segment led to loss of Myo31DF expression and to inverted phenotypes that mimicked Myo31DF mutations. Together, these data identify the A8 segment as a left-right organizer that is required for situs choice in Drosophila (Spéder, 2006).

To investigate the relationship between the anterior-posterior and left-right axes, the putative domain(s) of Myo31DF function was mapped by selectively silencing the gene in different compartments, using specific Gal4 lines driving 2 x Myo31DF^RNAi. The combinatorial removal of Myo31DF function in the anterior and/or posterior domains led to distinct phenotypes, indicating a dual function for the Myo31DF protein in A8. First, blocking Myo31DF function posteriorly in A8 (using hh-Gal4 or en-Gal4) resulted in a striking non-rotated genitalia phenotype. This finding was confirmed in a complementary experiment using dpp-Gal4 to rescue Myo31DF^K1 in the anterior compartment. The absence of situs choice observed in these experiments indicates that posterior Myo31DF has an instructive role in dextral looping. Second, blocking Myo31DF function solely in the anterior compartment (in dpp-Gal4 > 2 x Myo31DF^RNAi or in Myo31DF^K1/Myo31DF^K1; hh-Gal4 > Myo31DF rescued males) led to partial dextral rotation, suggesting a permissive role for Myo31DF in this compartment. Comparing this outcome to the effect of concomitant removal of both anterior and posterior functions (the only context that led to complete reversal) indicates that the function of Myo31DF in the anterior compartment is to repress sinistral looping. These experiments demonstrate that Myo31DF is essential both anteriorly and posteriorly. A model illustrating the dual function of Myo31DF in regulating dextral development within the A8 segment is presented. In this model, A8 contains information to specify both sinistral and dextral rotation, with sinistral information being anterior and dextral information posterior. Dextral information is dominant over sinistral information, and Myo31DF function is required both posteriorly, to induce dextral development, and anteriorly, to repress sinistral development. Only in the absence of Myo31DF can sinistral development occur as a default state (Spéder, 2006).

As expected of a left-right determinant with a function that precedes asymmetry, Myo31DF is expressed symmetrically in A8. In addition, as with mouse inversin (Watanabe, 2003), symmetrical overexpression does not lead to left-right defects (AbdB-Gal4, tsh-Gal4 and ptc-Gal4 lines). Another predicted feature of left-right determinants is their temporally restricted, trigger-like function, which is later relayed by mechanisms acting to maintain the initial symmetry-breaking event. To test this, temperature-shift experiments were carried out using a genetically engineered temperature-sensitive Myo31DF allele. Single temperature-shift experiments with 24-h or 6-h resolution indicated that Myo31DF function is required at day 6 of development, between 126 and 132 h. Double temperature-shift experiments allowed to determine that Myo31DF function is required for as little as 3 h within this period. These data provide high temporal resolution, allowing the temporal mapping of a left-right symmetry-breaking event and demonstrating that situs choice depends on a peak of Myo31DF function in the left-right organizer (Spéder, 2006).

Cilia have emerged as important cellular structures for generating left-right asymmetry in vertebrate embryos. To address their possible contribution to invertebrate left-right determination, genital discs were stained with GT335, an antibody that labels glutamylated tubulin in cilia across species, including Drosophila. GT335 did not detect any cilia in genital discs. Additionally, mutations in the conserved Rfx gene, which controls the formation of ciliated neurons in Drosophila, did not affect genitalia rotation. Together, these data suggest that cilia are not involved in dextral development in Drosophila. It is proposed that Drosophila uses primarily the actin cytoskeleton to determine left-right asymmetry. Consistently, the small GTPase Drac1 and the JNK pathway, known regulators of the actin cytoskeleton, showed specific genetic interactions with Myo31DF. Notably, actin is also important for situs choice in snails, suggesting that in invertebrates the actin cytoskeleton has a central role in left-right determination (Spéder, 2006).

To start investigating how Myo31DF might act to determine left-right asymmetry, two-hybrid screening was used to identify Myo31DF interactors or cargo(es). Using the Myo31DF tail domain (amino acids 737-1011) as bait, several positive clones were found encoding a carboxy-terminal fragment containing ARM repeats 6-12 of the Armadillo/β-catenin protein. This interaction was direct, as shown by glutathione S-transferase (GST)-pulldown experiments using purified proteins or S2 cell extracts. Furthermore, endogenous Myo31DF and a functional Myo31DF-GFP (green fluorescent protein) fusion protein colocalized with Armadillo at the adherens junctions in A8, suggesting that the two proteins can interact in vivo. Notably, inversin has been shown to colocalize and interact with β-catenin in vertebrate epithelial cells, indicating that an interaction with β-catenin is a common feature of both known situs inversus genes. These results are consistent with a demonstrated role of N-cadherin in left-right asymmetry (Spéder, 2006).

How could anterior-posterior organization of Myo31DF and interaction with Armadillo account for left-right asymmetry? The actin cable network might serve as a track for Myo31DF to deliver specific cargoes or vesicles at the adherens junction. Indeed, it was found that Myo31DF interacts with dynamin, consistent with the role of rat MyoID in vesicular transport (Huber, 2000). The anterior-posterior boundary itself creates an asymmetric junction that could serve as a scaffold for Myo31DF to assemble a dextral-specific complex on the anterior side of posterior Myo31DF-expressing cells. Anterior-posterior asymmetry of dextral information could later be translated into left-right asymmetry through the remodelling of cell contacts (cell intercalation or rotation), as seen in other epithelia. For example, 90° rotation of epithelial cells relative to the main body axis is observed in the Drosophila eye imaginal disc. A similar, 90° planar rotation of Myo31DF-expressing cells would result in left-right orientation of the dextral junctional complex in the tissue. In this working model, a short pulse of Myo31DF activity would be essential for spatially restricting the dextral junction, as has been observed (Spéder, 2006).

Search PubMed for articles about Drosophila Myosin 31DF

Bahler, M. and Rhoads, A. (2002) Calmodulin signaling via the IQ motif. FEBS Lett. 513: 107-113. PubMed citation: 11911888

Eley, L., et al. (2004). A perspective on inversin. Cell Biol. Int. 28: 119-124. PubMed citation: 14984757

Freeman, G. and Lundelius, J. (1982). The developmental genetics of dextrality and sinistrality in the gastropod Lymnaea peregra. Wilhelm Roux's Archives 191: 69-83

Gillespie, P. G., et al. (2001). Myosin-I nomenclature. J. Cell Biol. 155: 703-704. PubMed citation: 11724811

Hozumi, S., et al. (2006). An unconventional myosin in Drosophila reverses the default handedness in visceral organs. Nature 440(7085): 798-802. PubMed citation: 16598258

Huber, L. A., et al. (2000). Both calmodulin and unconventional myosin Myr4 regulate membrane trafficking along the recycling pathway of MDCK cells. Traffic 1: 494-503. PubMed citation: 11208135

Morgan, D., et al. (1998). Inversin, a novel gene in the vertebrate left-right axis pathway, is partially deleted in the inv mouse. Nature Genet. 20: 149-156. PubMed citation: 9771707

Morgan, D., et al. (2002). The left-right determinant inversin has highly conserved ankyrin repeat and IQ domains and interacts with calmodulin. Hum. Genet. 110: 377-384. PubMed citation: 11941489

Morgan, N. S., Heintzelman, M. B. and Mooseker, M. S. (1995). Characterization of Myosin-IA and Myosin-IB, two unconventional myosins associated with the Drosophila brush border cytoskeleton. Dev. Biol. 172: 51-71. PubMed citation: 7589814

Nurnberger, J., Bacallao, R. L. and Phillips, C. L. (2002). Inversin forms a complex with catenins and N-cadherin in polarized epithelial cells. Mol. Biol. Cell 13: 3096-3106. PubMed citation: 12221118

Shibazaki, Y., Shimizu, M. and Kuroda, R. (2004) Body handedness is directed by genetically determined cytoskeletal dynamics in the early embryo. Curr. Biol. 14: 1462-1467. PubMed citation: 15324662

Spéder P, Adám G, Noselli S. (2006). Type ID unconventional myosin controls left-right asymmetry in Drosophila. Nature 440(7085): 803-7. PubMed citation: 16598259

Ueshima, R. and Asami, T. (2003). Evolution: single-gene speciation by left-right reversal. Nature 425: 679. PubMed citation: 14562091

Watanabe, D., et al. (2003). The left-right determinant inversin is a component of node monocilia and other 9 + 0 cilia. Development 130: 1725-1734. PubMed citation: 12642479

Wood, W. B. (1997). Left-right asymmetry in animal development. Annu. Rev. Cell Dev. Biol. 13: 53-82. PubMed citation: 9442868

date revised: 20 March 2008

Home page: The Interactive Fly © 2008 Thomas Brody, Ph.D.