InteractiveFly: GeneBrief

wech: Biological Overview | References

Drosophila Dappled (DPLD, Wech) is a member of the RBCC/TRIM superfamily, a protein family involved in numerous diverse processes such as developmental timing and asymmetric cell divisions. DPLD belongs to the LIN-41 subclade, several members of which are micro RNA (miRNA) regulated. The LIN-41 subclade members was examined and their relation to other RBCC/TRIMs and dpld paralogs was examined, and a new Drosophila muscle specific RBCC/TRIM, Another B-Box Affiliate (ABBA) was identifed. In silico predictions of candidate miRNA regulators of dpld identified let-7 as the strongest candidate. Overexpression of dpld led to abnormal eye development, indicating that strict regulation of dpld mRNA levels is crucial for normal eye development. This phenotype was sensitive to let-7 dosage, suggesting let-7 regulation of dpld in the eye disc. A cell-based assay verified let-7 miRNA down-regulation of dpld expression by means of its 3'-untranslated region. Thus, dpld seems also to be miRNA regulated, suggesting that miRNAs represent an ancient mechanism of LIN-41 regulation (O'Farrell, 2008a).

The RING, B-box and coiled-coil/Tripartite Motif (RBCC/TRIM) proteins form a large protein superfamily with members spanning invertebrate to mammals. They have been shown to participate in numerous distinct processes, including roles in development, both normal and tumorous, asymmetric cell divisions and viral response, to name but a few. This protein superfamily is characterized by the presence of multiple protein-protein interaction domains, from which it has derived its name, that is, RING, B-box, and coiled-coil domains. In addition to these domains, members of this family often have distinct C-terminal domain(s), such as a series of NHL domains (first identified in NCL-1, HT2A and LIN-41 proteins; Slack (1998). For reviews on the RBCC/TRIM superfamily, see Torok (2001) and Meroni (2005) (O'Farrell, 2008a).

The dappled (dpld) gene, encodes a RBCC/TRIM protein highly homologous to C elegans lin-41 (Slack, 2000). Dappled protein (DPLD), however, lacks the N-terminal most RING domain found in many of the other members of this superfamily. Despite this, both DPLD and its paralog Brain tumour (BRAT) are considered to be RBCC/TRIM superfamily members due to the high sequence conservation of the remaining protein domains and the conservation of their order/positioning within the protein. dpld was first identified in a tumorigenesis screen (Rodriguez, 1996). The name itself relates to the tumorous phenotype, causing large melanotic spots within the animal, giving a dappled (an equestrian term) or spotted appearance. dpld mutations resulting from the screen were classified as likely causing a defect in cell growth or proliferation control (Rodriguez, 1996). Little else is known about either the function or regulation of dpld (O'Farrell, 2008a).

BRAT, on the other hand, has been the subject of intense study in recent times and as the name suggests, gives rise to brain tumours when mutated. Tumours arising in brat animals are highly proliferative, invasive, transplantable and lethal to the animal. Of interest, tumorigenic alleles of brat disrupt the NHL repeat region, implicating it as having a direct role in tumor suppression. Furthermore, the BRAT NHL domain has been shown to mediate protein-protein interactions of various kinds, including direct binding to and promoting asymmetric distribution of the cell fate determinant Prospero. Thus, both dpld and brat can cause tumors when mutated, suggesting at least some functional overlap (O'Farrell, 2008a and references therein).

The functionally characterized DPLD-like proteins to date, in addition to BRAT, include Drosophila MEI-P26, known to be involved in meiotic cell cycle control, and DPLD's orthologous protein from C. elegans, LIN-41. lin-41 is a heterochronic gene, regulating the timing of specific developmental events, demonstrated to govern the number of cell divisions during C. elegans larval hypodermal development. Mutants lack a terminal cell division and instead precociously differentiate to a terminal cell fate. The molecular mechanism by which this occurs is unknown. However, certain aspects of lin-41 regulation have been elucidated. Specifically, C. elegans lin-41 mRNA was one of the first transcripts experimentally demonstrated to be under micro RNA (miRNA) translational regulation, imposed in this case by the let-7 miRNA (Reinhart, 2000; Slack, 2000; O'Farrell, 2008a and references therein).

miRNAs are short oligomers of RNA, 19-24 nucleotides long, encoded in the genome that are capable of recognizing and annealing to complementary sites within mRNA sequences. These are primarily located in the 3-untranslated region (3-UTR) of the target mRNAs. Formation of the double-stranded miRNA/mRNA duplex subsequently causes either a translational repression of the mRNA, or a targeted degradation of the RNA duplex. Regardless, the overall effect is a decrease in the level of translated protein from the targeted mRNA transcript. Recent work has shown that the avian and mammalian lin-41 orthologs are also likely candidates for this mode of regulation by means of conserved miRNA let-7 and miR-125 target sites, indicating possible conservation of this mode of regulation within the subclade (Lancman, 2005; Schulman, 2005; Kanamoto, 2006; O'Farrell, 2008a and references therein).

Based on these observations, this study set out to ascertain the phylogenetic positioning of DPLD within the LIN-41 subclade and RBCC/TRIM superfamily and to consider the possibility of its translational regulation by miRNA. Indeed, it could be demonstrated that dpld is targeted and repressed by let-7 in Drosophila, an miRNA that also specifically targets other members of the LIN-41 subclade. This finding suggests wide conservation of this regulatory mechanism between lin-41 family members. Furthermore, through phylogenetic reconstructions, a novel Drosophila RBCC/TRIM family member was identified orthologous to NHL-1 and provide a brief description (O'Farrell, 2008a).

To date, there are a total of nine annotated B-box proteins in the Drosophila melanogaster predicted proteome. Four of these additionally have NHL repeats in the C-terminal end. Two of the NHL containing proteins, Meiotic protein P26, MEI-P26, and CG15105 (which this study has named Another B-Box Affiliate, ABBA) are typical members of the RBCC/TRIM protein superfamily having RING, B-box, and coiled-coil domains. The others, Dappled, DPLD, and BRAT, while lacking the RING domain are also considered RBCC/TRIM family members due to the high sequence conservation of the remaining protein domains and the conservation of their positioning within the protein (Slack, 2000; Brody, 2002; Sempere, 2002; Meroni, 2005). Moreover, DPLD and LIN-41 belong to the same InParanoid metazoan ortholog pairs cluster. To provide an up-to-date understanding of the evolutionary relationships of the RBCC/TRIM proteins containing NHL domains in D. melanogaster, the RBCC/TRIM and B-box protein sequences were analyzed from several fully sequenced genomes (O'Farrell, 2008a).

Previously, all Drosophila RBCC/TRIM NHL proteins have been likened to the mammalian RBCC/TRIM proteins TRIM2 and 3. However, recent release of vertebrate sequence data brings new mammalian and vertebrate proteins to the LIN-41 subclade, showing greater sequence similarity between proteins within this subclade than to any other vertebrate RBCC/TRIM proteins. DPLD associates with the LIN-41 subclade with a very high bootstrap value of 999/1000 for this node, indicating a very high degree of reliability (O'Farrell, 2008a).

The majority of the conservation between fly, vertebrate, and worm LIN-41 proteins resides in the B-box domains and the C-terminal NHL-propeller region. Alignment of seven sequences, spanning from Homo sapiens to D. melanogaster, for these five NHL repeats reveals a high level of conservation, with 57% overall similarity and 41% identity. There is partial, weak conservation of a potential sixth C-terminal NHL domain in DPLD, which is considered a half-domain. Poor conservation of the sixth NHL domain seems common to those RBCC/TRIM proteins represented with the sixth NHL domain of C. elegans LIN-41, and D. melanogaster BRAT and MEI-P26 proteins having to be annotated manually. The three-dimensional structure of the NHL domain of BRAT has been determined and the sixth domain does form a portion the β-propeller structure (Edwards, 2003). As such, the lack of conservation in this portion of the propeller may represent functional divergence and, for example, alter protein interaction capabilities. Accordingly, despite the poor conservation in the DPLD C-terminal-most NHL domain, it likely represents a genuine NHL domain (O'Farrell, 2008a).

There are several interesting observations associated with this cladogram. First, the LIN-41 subclade of the RBCC/TRIM family contains members spanning from Drosophila/Caenorhabditis to mammals and as such highlights the potential for RBCC/TRIM protein and regulatory studies using the model organisms within this subclade. However, and a second point of interest, all the vertebrate members of the LIN-41 subclade possess N-terminal RING domains, based on both protein predictions and experimental evidence as does C. elegans LIN-41. This finding is in contrast to the dipteran insect members of the subclade, which lack the RING domain, illustrating a potential limit to functional comparisons. From their relative positions in the cladogram, the lack of a RING domain in the insect proteins DPLD and BRAT is interpreted to represent two independent events, whereby the RING domains of each protein were lost. Despite the presence of a RING domain in LIN-41 vertebrate proteins, these are more closely related to the RING-less insect members than to any vertebrate protein relatives (O'Farrell, 2008a).

Manually searching up to 300 kb of ungapped sequence upstream to the dpld ATG, searching for cysteine and histidine residues appropriately spaced to fulfill the RING consensus sequence did not reveal any canonical RING domain to which DPLD could be associated. This observation, together with the presence of numerous dpld cDNAs (over 60), which do not encode RING domains, leads to the conclusion that DPLD is RING-less. As mentioned, no vertebrate members of the MEI-P26 and BRAT subclades have been described nor has any evidence been found suggesting that vertebrate members exist, indicating this branch to be protostome-specific; however, with constantly increasing genome sequence coverage, this may change (O'Farrell, 2008a).

A notable addition to the cladogram is the protein CG15105, ABBA, which is most similar to NHL-1 in C. elegans, a known ubiquitin ligase (Gudgen, 2004). ABBA is a RBCC/TRIM family member possessing the N-terminal RING, suggesting a possible ubiquitinating ability of this protein (TRIM/RBCC RING-dependent ubiquitinase activities reviewed by Meroni, 2005). The data shows that ABBA is more closely related to the TRIM2 and 3 subclades of the RBCC/TRIM superfamily than DPLD or other Drosophila B-box proteins are. ABBA has an overall identity of 19% with 26% similarity to human TRIM2, the majority of the conserved amino acids lying in the RING and NHL domains (38% identical and 57% similar in a 272 amino acid stretch of the NHL repeat region) (O'Farrell, 2008a).

The expression patterns of all four Drosophila RBCC/TRIMs (DPLD, BRAT, MEI-26, and ABBA) have been partially characterized during embryogenesis, and the expression pattern of brat has been extensively documented in developing adult tissues. The four RBCC/TRIM paralogs have very different zygotic transcription patterns, the only exception being an overlap between brat and dpld, which are coexpressed in the developing embryonic central nervous system. To gain a better understanding of the potential functions of those less-characterized Drosophila family members, abba and dpld, in situ hybridization (ISH) was performed on a variety of tissues at various developmental stages. This revealed that, in contrast to the other three RBCC/TRIMs which all are expressed maternally, there was no maternal contribution of abba. Rather, expression of abba was detectable only from mid-embryogenesis onward, specifically in developing muscles. This expression pattern was maintained in the differentiated larval muscles while absent in other tissues. The C. elegans ortholog of ABBA, NHL-1 is also predominantly expressed in muscle tissue (O'Farrell, 2008a).

ISH experiments in Drosophila embryos revealed dpld mRNA to be present ubiquitously in the early embryo, thus maternally contributed. Later during embryogenesis, dpld mRNA was expressed in a spatially restricted but dynamic fashion during both CNS and peripheral nervous system (PNS) development, gradually assuming a more defined and stable expression in neurons of the CNS and PNS. Double labeling with antibodies directed against the neuron (anti-ELAV) and sheath (anti-PROS) cells of the PNS in conjunction with ISH, demonstrated that the neurons of the PNS (both chordotonal and multidendritic neurons) expressed dpld at the end of PNS development. Further examination of dpld mRNA distribution in the imaginal discs of late third instar larvae, focusing on the larger wing and eye-antennal discs, revealed dpld mRNA to be expressed in specific cells along the future anterior wing margin of the wing. These cells are likely precursors of the innervated mechano- and/or chemosensory bristles that line the wing margin, in analogy with dpld expression in other neuronal cell types. Some background staining is visible in the surrounding wing pouch; this is, however, unspecific signal, which unlike that of the wing margin staining was not reproducible. In addition, dpld was expressed in a particular pattern during eye development in the eye-antennal imaginal disc (O'Farrell, 2008a).

Drosophila neuronal photoreceptor cells of the adult eye are known to be selected from a monolayer of undifferentiated, proliferating cells during larval third instar. During the process of eye cell specification and differentiation, a physical indent known as the morphogenetic furrow sweeps across the disc. Within this furrow, the first photoreceptor cells, the R8, are born at regular intervals. These then progressively recruit additional photoreceptor and accessory cells necessary to construct a complete ommatidium. dpld was found to be broadly expressed both immediately anterior and posterior to the morphogenetic furrow. Interestingly, a more discrete expression was observed at some distance posterior to the furrow in distinct regularly ordered cells, this in contrast to the broad band(s) of expression around the furrow. In an effort to identify the nature of this cell type, double-labeling experiments using ISH and anti-ELAV antibodies were performed, demonstrating dpld to be expressed in neuronal photoreceptor cells. dpld is reiteratively expressed in the neuron both before and after its specification, suggesting that DPLD may fulfill different roles in proliferating versus differentiated cells. This finding is similar to abba, which is expressed both during and after muscle cell differentiation (O'Farrell, 2008a).

From the eye disc, the developing photoreceptor neurons send axons to synapse in the optic lobes of the larval CNS. Two distinct concentric arches of dpld expression were clearly visible in both lateral and dorsal views of the optic lobe, whereas another less distinct arch encompassing the inner two was also observed. The outermost arch was narrower and labeled individual, regularly spaced cells. From a dorsal view the expression pattern had the distinctive heart-shaped pattern particular to the proliferative zones of the optic lobe. Vertebrate lin-41 genes are also expressed during embryogenesis in, among other places, developing nervous tissue (Lancman, 2005; Schulman, 2005; Kanamoto, 2006). One interesting conclusion from the ISH experiments is that the zone of dpld expression in the larval optic lobes is concurrent with dpld expression in neurons of the eye, suggesting a possible function for DPLD in both the photoreceptor neurons and their target brain region (O'Farrell, 2008a).

Multiple computational approaches have indicated dpld as a possible miRNA target on the basis of one or more let-7 (in addition to other miRNA) regulatory sites present in the 3' region of dpld mRNA transcripts. Furthermore, recent studies demonstrated miRNA regulation of dpld homologues (lin-41) in other species (Lancman, 2005; Schulman, 2005; Kanamoto, 2006). However, miRNA regulation of dpld in Drosophila has yet to be demonstrated. Toward this end, a combined in silico approach was used in which all possible miRNA sites in the dpld 3'-UTR were examined using miRNA prediction software, whereas independently the 3' region of a large number of insect dpld genes were compared and analyzed using standard multiple alignment techniques. Subsequently, target sites were sought within the highly conserved region of the 3'-UTR, in an effort to pinpoint potential miRNA target site(s). In this manner, the 3' region of dpld orthologs in several species were examined and several candidate miRNA sites were identified. The best candidate miRNA was let-7, which had three well-conserved sites (one of which was perfectly conserved within Drosophilidae) with favorable G folding values, followed by miR-31a and miR-31b. The miR-31a site M31a 1 was not perfectly conserved between all 11 species. In conclusion, the in silico approach enabled prediction of several candidate miRNA regulators of dpld mRNA (O'Farrell, 2008a).

Next, the relevance of these potential dpld regulators was assessed in the fly. To do this, dpld was expressed in transgenic animals using the UAS/GAL4 system. dpld normally is expressed in a spatially and temporally restricted pattern in the developing eye field. Ectopic expression of dpld by the GMR-GAL4 driver, which drives expression of GAL4 in all cells behind the morphogenetic furrow caused a severe eye phenotype as revealed both by anti-ELAV staining of third instar larval eye discs and scanning electron images of the surface of the eye. The pattern of ELAV expression was altered in response to dpld ectopic expression, ommatidial units appeared enlarged as did the individual neurons and in places seemed to be shared between or overlap other ommatidial units. These ommatidial units, which normally arrange themselves into a highly organized crystal lattice-like array, were additionally disturbed on the surface of the eye in GMR>dpld flies, with many of the ommatidial units appearing fused together. At higher magnifications, the lens material, which is secreted by the underlying cone cells, appeared to have spread and to be shared between ommatidial units. This finding, together with the irregular spacing of the interommatidial bristles throughout the eye, accounts for most of the externally visible defects. In addition, a clear reduction in pigment was observed under the light microscope, indicating a lack of pigment cells or defects therein leading to lowered pigment production. Cone and pigment cells, are specified from cells born in the second mitotic wave of the eye disk, a time point and position where ISH experiments have shown dpld to normally be expressed in a spatially restricted manner. Hence, ectopic expression of dpld post furrow interfered with the growth/cell size of many cell types that constitute the adult eye, including neurons, and possibly the subsequent recruitment of cone and pigment cells. Studies directed toward elucidating the underlying mechanism by which dpld acts in the eye are currently under way. The phenotype resulting from ectopically expressed dpld could be modulated by either varying the GAL4 efficiency (by culturing GMR>dpld flies at 29°C or altering gene dosage). It is therefore concluded that the eye phenotype resulting from ectopic expression of dpld was sensitive to levels of DPLD, and an enhancer/suppressor screen for genetic modifiers was consequently performed (O'Farrell, 2008a).

Typically, in enhancer/suppressor screens, altering levels of various factors involved in regulation of the level of active of protein, including in this context miRNA translational regulation, would affect the severity of the observable phenotype, in this case the rough eye resulting from GMR>dpld. Almost the entire Drosophila genome has chromosomal deletion coverage, publicly available as individual fly stocks collectively termed the Drosophila deletion kit. Included in this deletion coverage are most miRNA transcriptional units, facilitating a rapid screen through regions of the genome for genetic interactions. In this manner one can appraise the impact of lowering the gene dosage of candidate interactors in a dpld sensitized background. Importantly, specific mutations in the various miRNA are not available in flies, in fact, due to the small size of each miRNA locus, miRNA represent a very difficult target for traditional mutagenesis. Based on the results from the in silico approach predicting potential miRNA regulators of dpld, deletions and duplications including miRNA transcriptional units were selected for further analysis (O'Farrell, 2008a).

Of interest, a duplication that included the let-7 locus, mildly suppressed the GMR>dpld phenotype, whereas a deletion uncovering let-7 enhanced the phenotype, consistent with a role for let-7 in dpld regulation in vivo. This was further examined using a range of independent deletions in the let-7 area, which either do or do not remove the let-7 locus. By means of the use of overlapping deletions, the chromosomal region genetically interacting with dpld was narrowed downto 23 loci. The enhancement of the GMR>dpld phenotype in a let-7 background most likely reflects an increase in levels of DPLD and/or an expansion of DPLD activity within the eye field (O'Farrell, 2008a).

The use of independent deletions, uncovering miR-31a, miR-31b, and miR-210, produced milder but reproducible enhancements of the GMR>dpld eye phenotype. ISH was performed using antisense locked nucleic acid analogue (LNA) probes directed against miR-31a and let-7 (it having been shown to be present in pupal imaginal discs and CNS by means of Northern techniques only. It was found that let-7 was expressed in the two larger discs of wild-type third instar larvae, the wing and eye-antennal discs. Within the wing disc, let-7 expression was detected in the wing pouch, whereas in the eye, let-7 was found to be expressed in a specific pattern at a distance post-furrow in the portion of the disc where the assembling ommatidial units are located. Thus far the specific cell type(s) expressing let-7 have not been identified. miR-31a was expressed in the optic lobe region of the CNS. As neither the miR-31a or miR-210 potential targeting of the dpld 3'-UTR is conserved to other species, these findings were not investigated further. Similarly, the lack of potential conserved miR-125 target sites within the dpld 3'-UTR (being thought to regulate other lin-41 homologues) likely reflects the lack of conservation of this regulatory mechanism between vertebrate lin-41 and Drosophila dpld (O'Farrell, 2008a).

An advantage of the enhancer/suppressor approach described is the availability of deletions and the relative ease and speed of screening. A limit to the system, however, is the large size of the genomic deletions. Although the approach used in this study facilitated rapid screening, from which non-interacting deletions containing potential miRNA regulators of dpld could be discounted, subsequent detailed verification of the putative miRNA is required. The strongest interactor, let-7, which additionally represents a potentially conserved mechanism of miRNA regulation of lin-41 transcripts, was chosen for further molecular study (O'Farrell, 2008a).

To directly address the potential regulatory role of let-7 on dpld mRNA, a luciferase reporter assay was implemented in Drosophila S2 cell lines. Importantly, these cell lines have been previously shown not to express let-7 under normal circumstances, although they can be induced to upon exposure to ecdysone, the insect moulting hormone. Multiple versions of the dpld 3-UTR were created stemming from two publicly available cDNAs that differ in their 3'-UTR length, LD39167, which has a 3'-UTR 1527 nucleotides in length (luciferase constructs tagged with this 3'-UTR are referred to as dpld-L) and LD02463, which has a 3'-UTR 242 nucleotides in length (luciferase constructs tagged with this 3'-UTR are referred to as dpld-S). The entire sequence of the dpld-S 3'-UTR is in common to both UTRs. Within this region lies the first predicted let-7 target site (LCS1). The longer 3'-UTR variant, dpld-L, contains an additional two potential let-7 target sites, which lie in close proximity to each other. Mutated versions of both 3'-UTRs were generated in which specifically the let-7 target sites were removed, referred to as dpld-LD and dpld-S. The sensitivity of both the wild-type UTRs (dpld-L and dpld-S) and mutated versions (dpld-LD and dpld-S) was tested to determine whether or not these UTRs were, first, sensitive to let-7, and, second, whether the let-7 target sites were responsible for conferring this sensitivity and finally to compare the different versions of the 3'-UTRs to assess whether the naturally occurring length variation held any significance in terms of let-7 regulation. Both the wild-type and mutated dpld-L constructs promoted equal levels of luciferase reporter gene activity in the absence of let-7. Upon coexpression of a let-7 expression plasmid with the dpld-L 3'-UTR construct strong suppression of reporter gene activity was observed. In contrast, let-7 did not suppress the expression from the mutant dpld-LD construct. Furthermore, coexpression of either 3'-UTR reporter construct with the miRNA miR-92b (not predicted to target the dpld 3'-UTR) had no suppressing effect, indicating the let-7 miRNA effect to be specific. From this two conclusions are drawn: the dpld 3'-UTR is sensitive to let-7 expression and indeed the predicted, conserved let-7 target sites are responsible for this effect. Similarly, both the wild-type and mutated dpld-S constructs promoted roughly equal levels of luciferase reporter gene activity in the absence of let-7; however, in the presence of let-7, the mutated dpld-S, lacking the single predicted let-7 target site, was expressed at levels approximately threefold that of the intact version. It is concluded that the short version was also sensitive to let-7 and that again the predicted target site (LCS1) conferred this sensitivity. Moreover, it is concluded that the difference in length of the UTRs did not represent a plausible mechanism by which dpld could overcome let-7 repression. Furthermore, an additional construct was tested, dpld-L, minus only the let-7 site in common to both UTRs. This construct behaved in a similar fashion to dpld-LD in the presence of let-7, although expression levels were slightly lower, indicating a minor contribution of these two let-7 target sites to let-7 repression of the dpld-L construct. This finding suggested that the let-7 target site (LCS1) in common to both UTR variants, and the most widely conserved LCS within Drosophilidae conferred the majority of the dpld 3'-UTRs sensitivity to let-7 (O'Farrell, 2008a).

A mis-expression study of factors affecting Drosophila PNS cell identity

Drosophila PNS sense organs arise from single sensory organ precursor (SOP) cells through a series of asymmetric divisions. In a mis-expression screen for factors affecting PNS development, string and dappled were identified as being important for the proper formation of adult external sensory (ES) organs. string is a G2 regulator. dappled has no described function but is implicated in tumorigenesis. The mis-expression effect from string was analysed using timed over expression during adult ES-organ and, for comparison, embryonic Chordotonal (Ch) organ formation. Surprisingly, string mis-expression prior to SOP division gave the greatest effect in both systems. In adult ES-organs, this lead to cell fate transformations producing structural cells, while in the embryo organs were lost, hence differences within the lineages exist. Mis-expression of dappled, lead to loss and duplications of entire organs in both systems, potentially affecting SOP specification, in addition to affecting neuronal guidance (O'Farrell, 2008b).

DPLD is a member of the Tripartite Motif (TRIM) superfamily of proteins originally isolated in a screen for tumour causing genes. Its molecular function remains unknown. Plasmid rescue experiments placed the P{GS}dpld^GS8193 insertion site 0.45 kb upstream of the dappled translational start site. sca-dpld^GS8193 had strongest effects on macrochaetae development. General mis-spacing of organs over the thorax was noted with some individuals lacking macrochaetae while other had excess. Occasionally, duplications of bristles (twinning) from individual organs was observed. The ectopic and mis-positioned organs suggest a role for dappled in SOP cell selection. However, a caveat to mis-expression screening is the relevance of the phenotype, i.e., whether the gene is normally expressed within the affected tissue. In the case of dappled, no expression in the thorax has been documented. Expression of dappled within the developing embryonic PNS has been document. This, and the fact that embryo possesses adult ES analogous ES- and Ch-organs, which are easily studied, prompted continued study of dappled within these PNS lineages. sca-dpld^GS8193 expression also gave rise to embryonic PNS phenotypes. Frequently, PNS axonal patterning was non-wildtype, with lateral sense organs, Lch, Les and Lda aberrantly positioned, including Lch5 organs that occasionally appeared fused to Lch5s of adjacent segments. In all cases, the fused appearance paralleled a loss of organ in the adjacent segment, discarding a simple duplication explanation. Instead, the positional defects are interpreted to arise potentially as a result of axonal guidance problems. Lch5 axons follow the tracheal spiracular branch (TSB) to reach the intersegmental nerve leading to the CNS. The LdaA sense organ marks the first turning point for these axons to seek contact with the TSB. As the PNS pattern deviates from wildtype in sca-dpld embryos, it is possible that missing guidance cues resulted in misrouting of Lch5 axons. Also, post-mitotic PNS neurons normally express dappled during the period of axonal outgrowth, indicating that the pathfinding defect may constitute a relevant cell autonomous effect, where misrouted axons subsequently pulled their corresponding organs into the neighbouring segment (O'Farrell, 2008b).

Lch1 entire organ duplications were observed. This likely resulted from an ectopic Lch1 SOP as two complete organs formed. This, similarly to observations made in the adult ES-organ, suggests a role, or interference, of dappled in the lateral inhibition process. It should be noted that Lch1 is peculiar in that it lacks a ligament cell, as assessed by the absence of normal ligament cell markers (anti-β-tubulin 85E and anti-reversed polarity). It is however difficult to speculate whether this difference to the lineage renders the Lch1 organ more sensitive to duplications. In line with an effect of dappled upon SOP selection incomplete Lch5 organs were observed specifically lacking complete Ch-organs. Within the Lch5 and other lineages a wildtype ratio of sheath to neuronal cells was observed indicating normal pIIIb divisional output. Likely, mistimed onset and/or elevated activity of dappled in cells downstream of the SOP is tolerated by the programmes generating the proper cell types of sense organs (O'Farrell, 2008b).

Identification of immune system and response genes, and novel mutations causing melanotic tumor formation in Drosophila melanogaster

Two types of screens in Drosophila were conducted using enhancer detector strains to find genes involved in immunity and tumor formation; genes expressed in the immune system (type A; hemocytes, lymph glands and fat body) and genes increased in expression by bacterial infection (type B). For type A, tissue-specific reporter gene activity was determined. For type B, a variation of enhancer detection was devised in which beta-galactosidase is assayed spectrophotometrically with and without bacterial infection. Because of immune system involvement in melanotic tumor formation, a third type was hypothesized to be found among types A and B; genes that, when mutated, have a melanotic tumor phenotype. Enhancer detector strains (2800) were screened for type A, 900 for B, and 11 retained for further analysis. Complementation tests, cytological mapping, P-element mobilization, and determination of lethal phase and mutant phenotype have identified six novel genes, Dorothy, wizard, toto, viking, Thor and dappled, and one previously identified gene, Collagen IV. All are associated with reporter gene expression in at least one immune system tissue. Thor has increased expression upon infection. Mutations of wizard and dappled have a melanotic tumor phenotype (Rodriguez, 1996).

The novel approach of identifylng mutations with a melanotic tumor phenotype based only on selection of tissue specificity and/or infection inducibility of the reporter gene was successful for tissue specific selection and has identified the genes wizard and dappled. The lethality of both wizard and dappled^Ef1 is early in larval development, and thus would have gone undetected in previous screens, which have selected directly for a tumor formation phenotype in late larvae and pupae. The viability of dappled^MLB is not unusual for a melanotic tumor mutation, but the combination with 100% tumor formation rather than the typical variable and lower levels makes this mutation a unique practical tool for isolation of new genes that suppress or enhance melanotic tumors. In general, melanotic tumors are postulated to be a reaction to abnormal development, and this has been extended to the proposal that all melanotic tumor mutants can be categorized as belonging to one of two classes: class 1 - melanotic tumors associated with apparently normal immune systems that are responding to abnormal tissues, and class 2 - melanotic tumors associated with obvious defects of the immune system's lymph glands and hemocytes. Since the lymph glands are normal in dappled^MLB and dappled^Ef1, these are class 1 mutations. wizard has an expression pattern in lymph glands, oenocytes and head. wizard may thus be a class 1 or 2 mutation and requires further analysis to distinguish between the two possibilities. Some class 2 tumor forming genes have been molecularly characterized. Molecular characterization of class 1 tumor formation genes is lacking, and dappled is being cloned also to provide molecular information about this category (Rodriguez, 1996).

Search PubMed for articles about Drosophila Dappled

Brody, T., Stivers, C., Nagle, J. and Odenwald, W. F. (2002). Identification of novel Drosophila neural precursor genes using a differential embryonic head cDNA screen. Mech. Dev. 113: 41-59. PubMed ID: 11900973

Edwards, T. A., Wilkinson, B. D., Wharton, R. P. and Aggarwal, A. K. (2003). Model of the brain tumor-Pumilio translation repressor complex. Genes Dev. 17: 2508-2513. PubMed ID: 14561773

Gudgen, M., Chandrasekaran, A., Frazier, T. and Boyd, L. (2004). Interactions within the ubiquitin pathway of Caenorhabditis elegans. Biochem. Biophys. Res. Commun. 325: 479-486. PubMed ID: 15530417

Kanamoto, T., Terada, K., Yoshikawa, H. and Furukawa, T. (2006). Cloning and regulation of the vertebrate homologue of lin-41 that functions as a heterochronic gene in Caenorhabditis elegans. Dev. Dyn. 235: 1142-1149. PubMed ID: 16477647

Lancman, J. J., et al. (2005). Analysis of the regulation of lin-41 during chick and mouse limb development. Dev. Dyn. 234: 948-960. PubMed ID: 16245339

Meroni, G. and Diez-Roux, G. (2005). TRIM/RBCC, a novel class of single protein RING finger E3 ubiquitin ligases. Bioessays 27: 1147-1157. PubMed ID: 16237670

O'Farrell, F., Esfahani, S. S., Engström, Y. and Kylsten, P. (2008a). Regulation of the Drosophila lin-41 homologue dappled by let-7 reveals conservation of a regulatory mechanism within the LIN-41 subclade. Dev. Dyn. 237(1): 196-208. PubMed ID: 18069688

O'Farrell, F. and Kylsten, P. (2008b). A mis-expression study of factors affecting Drosophila PNS cell identity. Biochem. Biophys. Res. Commun. 370(4): 657-62. PubMed ID: 18420029

Reinhart, B. J., et al. (2000). The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans. Nature 403: 901-906. PubMed ID: 10706289

Rodriguez, A., et al. (1996). Identification of immune system and response genes, and novel mutations causing melanotic tumor formation in Drosophila melanogaster. Genetics 143(2): 929-40. PubMed ID: 8725239

Schulman, B. R., Esquela-Kerscher, A. and Slack, F. J. (2005). Reciprocal expression of lin-41 and the microRNAs let-7 and mir-125 during mouse embryogenesis. Dev. Dyn. 234: 1046-1054. PubMed ID: 16247770

Sempere, L. F., et al. (2002). The expression of the let-7 small regulatory RNA is controlled by ecdysone during metamorphosis in Drosophila melanogaster. Dev. Biol. 244: 170-179. PubMed ID: 11900466

Slack, F. J. and Ruvkun, G. (1998). A novel repeat domain that is often associated with RING finger and B-box motifs. Trends Biochem. Sci. 23: 474-475. PubMed ID: 9868369

Slack, F. J., et al. (2000). The lin-41 RBCC gene acts in the C. elegans heterochronic pathway between the let-7 regulatory RNA and the LIN-29 transcription factor. Mol. Cell 5: 659-669. PubMed ID: 10882102

Torok, M. and Etkin, L. D. (2001). Two B or not two B? Overview of the rapidly expanding B-box family of proteins. Differentiation 67: 63-71. PubMed ID: 11428128

date revised: 10 July 2009

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

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