Gene name - arc
Cytological map position - 58D5-8
Function - Scaffolding protein
Symbol - a
FlyBase ID: FBgn0000008
Genetic map position - 2-99.2
Classification - PDZ domain protein
Cellular location - cytoplasmic
A P element (P lacZ) insert in line l(2)k11011 drives expression in multiple morphogenetically active epithelia during embryogenesis, e. g., in Malpighian tubules, hindgut, and foregut. Later, in third instar larvae, the reporter is expressed in imaginal discs: in the leg disc, in the prospective hinges and proximal wing blades of the wing imaginal disc, in the antennal disc, and in the morphogenetic furrow of the eye imaginal disc. Genetic evidence indicates that the P lacZ giving this expression pattern is inserted in the arc gene. Flies homozygous for this P element insert display downwardly bent wings, a phenotype of the original arc mutant described by Bridges and Morgan (1919). Both this P element insert and arc map to 58CD; complementation and reversion tests have shown that the P element insertion is allelic to arc. Therefore this P element insertion has been named aP. Additional arc alleles and deficiencies of the arc region have been generated by excision of the P element, as well as by X-ray and EMS mutagenesis. Intragenic complementation tests between different arc alleles based on wing phenotype (percentage of flies with the arc genotype displaying straight wings) reveal an allelic series with aP acting as a hypomorph (Liu, 2000).
Light microscopic analysis of sections of embryos stained with anti-Arc antibody reveals that the Arc protein is localized to the apical side of the expressing epithelia. Different PDZ domain proteins have been localized to either septate junctions/tight junctions (e.g., Drosophila Discs large and mammalian ZO-1 and ZO-2) or adherens junctions (e.g., Drosophila Polychaetoid (ZO-1), Canoe, and Discs lost: now redefined as Drosophila Patj ). Might Arc be localized to one or both of these cell-cell junctions? Using antibody staining and confocal microscopy, Arc was not found colocalized with Neurexin, a trans-membrane component of the septate junction. However, Arc is colocalized with Armadillo (the Drosophila beta-catenin homolog), which, together with DE-cadherin (Shotgun) and Dalpha-catenin, forms the cadherin-catenin complex in adherens junctions. Higher magnification shows that the Arc protein is almost entirely localized to an apical ring that largely overlaps with the junctionally localized Armadillo protein. Thus, while there may be some Arc in the cell apical membrane, its primary location is in the adherens junction. Interestingly, ARC mRNA is also found at the apical surface of the expressing cells, a characteristic that has been observed for mRNAs encoding some apical membrane localized proteins (e.g., Crumbs) and some secreted proteins (e.g., Wingless and Dpp) (Liu, 2000).
In addition to the wing phenotype, arc adult flies exhibit defects in the eye. arc allelic combinations can be arrayed in a series (as determined by quantitative RT-PCR) ranked by production of progressively less ARC mRNA. These allelic combinations produce reduction in eye size proportionate to their strengths. In the complete absence of arc activity, eye size is reduced to approximately half that of the wild type. Since the size of each individual ommatidium is almost the same in arc amorph and wild-type flies, the reduced eye size is primarily due to the presence of fewer ommatidia in the arc mutant eyes. Consistent with this interpretation, there are only 24 vertical columns, with 26 ommatidia in the tallest columns, in an arc amorph female eye, compared with 32 to 34 columns, and 32 ommatidia per tallest column, in a typical wild-type female eye (Liu, 2000).
Another feature of the arc amorph eyes is that the ommatidia, instead of being hexagonal with a bristle at every other vertex, are rhomboidal, with a bristle at each vertex. This phenotype is similar to that seen in nemo mutants, where a defect in ommatidial rotation leads to the formation of an abnormal pigment cell lattice, i. e., a disarrangement of bristle and secondary pigment cells. However, both photoreceptor arrangement within each ommatidium and ommatidial rotation appear normal in arc amorph eyes. Thus the apparent defect in the pigment cell lattice must arise from a different mechanism in arc versus nemo flies. Flies doubly homozygous for both arc and nemo do not show any defects beyond the sum of phenotypes of the singly homozygous mutants (Liu, 2000).
To obtain insight into the required function of arc in formation of the ommatidia, the expression of arc was examined in developing eye imaginal discs. Both ARC mRNA and Arc protein are expressed in the morphogenetic furrow in repeating clusters of cells along the dorsal/ventral axis. The size and distribution of the arc-expressing cell clusters are similar to those of the so-called 'arc forms' (no relation to the arc gene), clusters of ommatidial precursors that are just emerging from the morphogenetic furrow. The expression of arc in a pattern similar to that of the ommatidial precursor clusters, taken together with the reduced number of ommatidia in the absence of arc activity, suggests that arc may play a required role in the establishment of these clusters (Liu, 2000).
Proper expression of arc in the furrow depends on normal progession of the furrow. Thus, when hedgehog activity, which is required for normal furrow progression, is reduced, there is a dramatic reduction in arc expression (Liu, 2000).
How might the adult arc mutant phenotypes in wing and eye be related to the observed expression of arc in the wing and eye imaginal discs? The curved wing phenotype is quite subtle; while it could be due to a requirement for the expression of arc in the prospective hinge and adjacent blade regions of the disc, it might also be due to a required expression at a later time, such as during wing eversion. The phenotype that can most readily be attributed to a particular arc expression pattern is the reduced number of ommatidia in the arc null eye. The expression of arc in the lagging edge of the morphogenetic furrow could affect the number of ommatidia formed in at least two different ways: indirectly, by affecting the rate of furrow progression, or directly, by promoting the initiation of ommatidial clusters. A reduced rate of furrow progression in the absence of arc activity could result secondarily in a reduced rate of production of the ommatidial precursor clusters, the formations of which are initiated just behind the furrow. The idea that arc expression in the furrow is required for its normal progress is supported by the fact that expression of hh, which drives progression of the furrow, is required to drive a normal level arc expression in the furrow (Liu, 2000).
A number of lines of evidence, however, suggest that arc affects the number of ommatidia by other mechanisms, in addition to any possible effect arc has on furrow progression. (1) arcP;hh1 eyes are smaller in the dorsal-ventral axis (which is perpendicular to the anterior-posterior axis of furrow progression) than hh1 eyes. (2) In the absence of arc activity, there are fewer ommatidia in both the rows and the columns of the eye (if arc affected only furrow progression, one would expect to see the same number of rows as in the wild type, but fewer columns). The expression of arc in the furrow, in groups of cells of roughly the same size and spacing as the forming ommatidial precursor clusters, suggests that arc function might contribute directly to the initiation of the ommatidial clusters. This is supported by the observation that ectopic expression of arc throughout the eye imaginal disc (by Actin 5C-GAL4), or in all cells behind the morphogenetic furrow (by GMR-GAL4), results in larger eyes with fused ommatidia (i.e., more ommatidia overall). Given its localization to the adherens junction, Arc might promote the association of cells into ommatidial clusters by modulating adherens junctions in cells just behind the furrow (Liu, 2000).
While the organization of the photoreceptors in arc null eyes appears normal, there is a distortion of the overall shape of the ommatidium: rhomboidal, rather than hexagonal. In contrast to what has been described for nemo mutants, this shape distortion does not arise by the failure of the ommatidia to rotate. The defect in the pigment cell lattice, the apparent basis of the rhomboidal morphology, must then arise by some other mechanism. Since the cells of this lattice are recruited into the ommatidial cluster during the first third of pupation, well after arc expression in the morphogenetic furrow, there may be a later expression of arc during pupation that is required for the proper association of pigment cells with the ommatidium. As was proposed for formation of the ommatidial clusters, Arc might also promote this second association by modulating the adherens junctions of the interacting cells (Liu, 2000).
Of the 17 genes encoding PDZ domain proteins that have been identified in Drosophila, bazooka, discs large, canoe and polychaetoid have been characterized in the most detail; these are expressed in epithelia throughout embryogenesis. Relative to these rather ubiquitously expressed genes, arc is unusual in being expressed only in epithelia as they are undergoing morphogenesis. Thus Arc is present in the amnioproctodeal invagination during gastrulation, but disappears from this epithelium by stage 10. Arc appears in the Malpighian tubule primordia just as they are being specified during stage 10 and becomes very strongly expressed during bud evagination and tubule elongation. Arc is also expressed in other tubular structures that are elongating, such as hindgut, foregut, salivary glands, and tracheae. arc is transcriptionally activated as a common target of different patterning systems that regulate development of various epithelia undergoing morphogenesis. In spite of the exquisitely regulated expression pattern of arc expression in developing embryonic tissues, the complete lack of arc activity does not detectably affect developing embryonic morphology and is not required for viability. This is not due to a rescuing effect of maternal ARC mRNA, since mothers lacking arc activity produce normal appearing embryos even when crossed with arc heterozygous fathers. Another possible explanation for the viability of arc null flies would be that there are other genes that play a role redundant with that of arc (Liu, 2000).
The most likely candidates for such genes are those with high sequence similarity to arc. Although a search of the largely completed Drosophila genome sequence identifies genes with some similarity to arc (baz, inaD, cno, dlg, and dlt, as well as three previously undescribed PDZ domain encoding sequences); none of these has high similarity to arc, and none display the arrangement of a highly variant PDZ domain followed by a well-conserved PDZ domain that is found in Arc. Furthermore, genetic studies with cno, pyd, and dlt, which encode adherens junctions associated PDZ domain proteins, do not reveal any genetic interaction with arc. Thus there does not appear to be a particular protein highly similar to Arc that might provide redundant function. It is of course possible either that multiple PDZ proteins acting together, or a protein with entirely different sequence motifs, might substitute for arc function in developing embryonic epithelia (Liu, 2000).
Since no protein likely to play a redundant function with Arc has been identified, the possibility must be considered that, in developing embryonic epithelia, Arc plays a specialized, subtle role that is not detectable by the criteria employed to date and that is not required for viability under laboratory conditions (Liu, 2000).
There are three different arc transcripts, which differ only slightly at their 5' ends, due to differences in initiation site and splicing of the 5' UTR. Transcripts I and II, 4644 and 4826 nucleotides in length, respectively, are initiated at the same position (designated as +1), but have different splice donor sites: exon 1a of transcript I (+38) vs. exon 1b of transcript II (1220). Transcript III, 5173 nucleotides in length, is initiated at a different position (+445, exon 2a). Starting with exon 2b (+1012), all three transcripts share the same intron/exon structure, including a single long open reading frame (ORF). RT-PCR analysis suggests that transcript III is more abundant than the other two transcripts. Consistent with this, Northern blot analysis of embryonic mRNA with a probe derived from the common 3' region of the arc cDNA reveals only one band migrating as 5.5 kb; most likely this corresponds to transcript III (Liu, 2000).
The arc protein contains two PDZ protein-protein interaction domains. While domain 1 is not as similar to the consensus PDZ domain as is domain 2, it is still highly significant by the statistical criteria of Pfam analysis, in fact more significant by these criteria than the well-characterized PDZ domain 4 of the Drosophila InaD protein. Outside of the PDZ domains, arc shows no close structural similarity to other PDZ domain proteins and does not contain other known structural motifs (Liu, 2000).
date revised: 10 June 2000
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.