InteractiveFly: GeneBrief

convoluted: Biological Overview | References

In metazoans, factors of the insulin family control growth, metabolism, longevity, and fertility in response to environmental cues. In Drosophila, a family of seven insulin-like peptides, called Dilps, activate a common insulin receptor. Some Dilp peptides carry both metabolic and growth functions, raising the possibility that various binding partners specify their functions. This study identifies ALS, the fly ortholog of the vertebrate insulin-like growth factor (IGF)-binding protein acid-labile subunit (ALS), as a Dilp partner that forms a circulating trimeric complex with one molecule of Dilp and one molecule of Imp-L2, an IgG-family molecule distantly related to mammalian IGF-binding proteins (IGFBPs). Drosophila ALS antagonizes Dilp function to control animal growth as well as carbohydrate and fat metabolism. These results lead to the proposal of an evolutionary perspective in which ALS function appeared prior to the separation between metabolic and growth effects that are associated with vertebrate insulin and IGFs (Arquier, 2008).

Members of the insulin-like peptide (ILP) family are found in a wide range of metazoans, where they control carbohydrate metabolism, tissue growth, reproduction, and longevity. The functional separation between insulin-like growth factor (IGF) and insulin signaling, as seen in mammals, dates back 600 million years, as the two types of molecules are already present in the lower metazoan tunicate phylum. Insulin and IGF-1 carry different biological functions, in part through their binding to closely related receptors, the insulin receptor (IR) and the IGF-1 receptor (IGF-1R), respectively. In contrast to insulin, which is produced and stored in specific endocrine tissues and is released by a highly regulated process, vertebrate IGF-1 accumulates in body fluids, where it associates with a large array of binding molecules globally referred to as IGF-binding proteins (IGFBPs). These play important though cryptic functions in controlling the biological activity of IGF-1 (Rosenfeld, 2000; Duan, 2005). Vertebrates have six bona fide IGFBPs, which directly bind IGF-1 and form stable binary complexes in the circulating blood. These binary complexes comprise 10% of the total plasma IGF-1. Most of the remaining plasma IGF-1 is bound up into ternary complexes comprising one molecule of IGFBP-3 or IGFBP-5 and one molecule of acid-labile subunit (ALS), leaving only 1% of plasma IGF-1 free (Boisclair, 2001). The binding of IGF partners greatly enhances the half-life of IGF-1 but also restrains its ability to interact with its receptor, IGF-1R, leading to the formation of a large reservoir of circulating IGF-1. The ALS partner is essential for the stabilization of circulating IGF-1; deficiencies in the ALS gene in both mouse and human lead to a drastic reduction in plasma levels of IGF-1 and IGFBP-3 (Domene, 2005; Domene, 2007). Ternary IGF-1 complexes have been proposed to contribute to the functional separation between IGF-1 and insulin by preventing illegitimate interaction between high blood concentrations of IGF-1 and the insulin receptor. While in vivo studies of IGFBP and ALS in mammals have revealed some key functions of these molecules in controlling the physiology of IGF-1, a full understanding of how these regulations take place in complex organisms is still lacking (Arquier, 2008).

Insects provide a simpler evolutionary alternative, possessing a single insulin-like system that represents a possible ancestor of the dual insulin/IGF system. In Drosophila, seven insulin-like peptide (Dilp)-encoding genes have been identified. These interact genetically with a unique insulin receptor called dInR. The dilp genes are expressed in different larval tissues, suggesting that they carry specific functions. At least three of them (dilp2, 3, and 5) are expressed in two symmetric clusters of seven neurosecretory cells called the insulin-producing cells (IPCs), located in each brain hemisphere. Ablation of these neurons leads to dramatically smaller body size with increased levels of circulating carbohydrate (trehalose in insects) in the hemolymph, indicating that the Dilp peptides produced in brain IPCs control both carbohydrate homeostasis and tissue growth. Interestingly, growth and trehalose level defects are rescued by ectopic expression of one Dilp, Dilp2, suggesting that at least this molecule carries both metabolic and growth functions. However, individual genetic analysis of dilp genes, which should yield important information toward the resolution of Dilp functions, has not yet been performed (Arquier, 2008).

The current study tested the possibility that specific complexes between the Dilps and binding partners may restrict their interactions with dInR, thereby specifying their functions. Both a putative ALS ortholog (dALS; Colombani, 2003) and a possible candidate for an ILP-binding protein (encoded by the Imaginal morphogenesis protein-Late 2 [Imp-L2] gene; Garbe, 1993) have been identified in Drosophila. The Imp-L2 protein is a member of the immunoglobulin superfamily, which shares slight homology with the mammalian IGF-binding protein-related protein 1 (IGFBP-rP1/IGFBP-7; Yamanaka, 1997). It can bind human IGF-1 in vitro (Sloth Andersen, 2000) and functions as a growth inhibitor in Drosophila. This study presents evidence that Dilp2 forms a trimeric complex with the Drosophila ALS and Imp-L2 proteins. It was further demonstrated that dALS is required for the growth and metabolic functions of the Dilps. This indicates that the formation of a trimeric complex containing dALS and some Dilp peptides is a prerequisite for both metabolic and growth control in flies (Arquier, 2008).

CG8561 has been previously identified as a candidate gene encoding a putative Drosophila ortholog of the vertebrate ALS protein, which has been called dALS (Colombani, 2003). The dALS protein contains a series of 21 leucine-rich repeats (LRRs) that also form the core of the vertebrate ALS. Based on sequence similarity and the presence of LRRs, two additional related sequences were found in the Drosophila genome. The expression levels of all three genes was examined in larval tissues and in normally fed or starved animals. CG8561 is exclusively expressed in two larval tissues that play important roles in growth and metabolic regulation: the 14 IPCs in the brain, and the fat body (FB), a larval tissue that shares some functions with the vertebrate liver and fat (Colombani, 2003). Remarkably, dALS expression in the FB is suppressed under amino acid restriction, a finding reminiscent of the strong downregulation of the vertebrate ALS gene observed in the liver under starvation (Colombani, 2003). The two other related genes did not show clear expression in any of the larval tissues, nor did they show nutrition-regulated expression. Therefore the analysis focused on CG8561 (Arquier, 2008).

This work provides strong evidence for the formation of a trimeric complex involving Dilp2, ALS, and Imp-L2, a molecule with Dilp-binding protein function in Drosophila. No binding was observed between ALS and Dilp2 in the absence of Imp-L2, suggesting that, as with the trimeric IGF-1 complexes circulating in mammalian blood, the binding of ALS requires prior formation of a dimeric Dilp/Imp-L2 complex. Dilp5, another member of the ILP family in Drosophila, is also capable of forming a complex with ALS in cultured cells. Interestingly, the binding of Dilp5 and ALS is suppressed by excess Imp-L2, suggesting that one or more other Dilp-BPs produced in S2 cells compete with ALS binding for the formation of Dilp5 complexes. It is proposed that ALS may function as a common scaffold protein for different Dilp/Dilp-BP complexes in the hemolymph, with specific Dilp-BPs participating in the specialization of Dilp functions. At present, the technical difficulty of measuring the levels of endogenous Dilps in the hemolymph of Drosophila larvae precludes a detailed analysis of the types and amounts of circulating Dilp/Dilp-BP/ALS complexes (Arquier, 2008).

No abnormal phenotypes were observed upon ALS overexpression or silencing in the brain IPCs. This could be due to a lack of sensitivity in the method, as it was found that expressing ALS^M in the 14 IPCs leads to very low accumulation of ALS^M in the hemolymph as compared to its expression in the FB. Conversely, silencing ALS in the IPCs does not reduce global ALS transcript levels, possibly because an important ALS transcription from FB cells is masking this effect. It was also noticed that, when expressed in the IPCs, ALS^M is not present in the same vesicular structures as Dilp2, suggesting that the two molecules are not found in a preassembled complex before being released into the hemolymph. Determination of the function of IPC-produced ALS will require further examination (Arquier, 2008).

The results point to a dual effect of ALS in the control of IIS that depends on nutritional status. This dual effect is interpreted in light of the complex functions of IGFBPs and ALS in mammals. Under optimal nutritional conditions, Dilps are not limiting, and overexpression of ALS can induce the recruitment of more Dilps into stable but inactive trimeric complexes. If the release of active Dilp molecules is limited by the amounts of the various proteases that break apart the trimeric complexes, the net effect of ALS overexpression will be growth inhibition, as observed in vivo. In contrast, fasting leads to a general inhibition of IIS that may reveal a positive function for ALS: Dilp molecules becoming limiting, and ALS overexpression may increase the half-life of circulating Dilps and thereby enhance Dilp signaling (as long as the proteases are not limiting). Along these lines, the severe downregulation of ALS transcription observed under limited nutrient conditions (Colombani, 2003) suggests that ALS participates in the adaptation of IIS to limited nutrition and the necessity of slowing down growth rate as well as carbohydrate and fat metabolism. Alternatively, the opposing results observed in starved versus fed conditions could be explained by the differential regulation of Dilp/ALS complexes involved in distinct regulations of IIS in response to nutritional conditions (Arquier, 2008).

It has been proposed that in vertebrates, the formation of trimeric IGF/IGFBP/ALS complexes contributes to the functional separation between insulin and IGFs. This study has provided evidence that such complexes are required for both the growth and metabolic functions carried out by the Dilps in Drosophila. The work suggests an alternative scenario in which ALS, Imp-L2, and possibly additional Dilp-BPs participate in an ancestral function used for both metabolism and growth control (Arquier, 2008).

Drosophila convoluted/dALS is an essential gene required for tracheal tube morphogenesis and apical matrix organization

Insulin-like growth factors (IGFs) control cell and organism growth through evolutionarily conserved signaling pathways. The mammalian acid-labile subunit (ALS) is a secreted protein that complexes with IGFs to modulate their activity. Recent work has shown that a Drosophila homolog of ALS, dALS, can also complex with and modulate the activity of a Drosophila IGF. This study reports the first mutations in the gene encoding dALS. Unexpectedly, it was found that these mutations are allelic to a previously described mutation in convoluted (conv), a gene required for epithelial morphogenesis. In conv mutants, the tubes of the Drosophila tracheal system become abnormally elongated without altering tracheal cell number. conv null mutations cause larval lethality, but do not disrupt several processes required for tracheal tube size control, including septate junction formation, deposition of a lumenal/apical extracellular matrix, and lumenal secretion of Vermiform and Serpentine, two putative matrix-modifying proteins. Clearance of lumenal matrix and subcellular localization of clathrin also appear normal in conv mutants. However, Conv/dALS is required for the dynamic organization of the transient lumenal matrix and normal structure of the cuticle that lines the tracheal lumen. These and other data suggest that the Conv/dALS-dependent tube size control mechanism is distinct from other known processes involved in tracheal tube size regulation. Moreover, evidence is presented indicating that Conv/dALS has a novel, IGF-signaling independent function in tracheal morphogenesis (Swanson, 2009).

Insulin and insulin-like growth factors (IGFs) control energy homeostasis and growth through evolutionarily conserved signaling pathways. Key regulators of these pathways are IGF binding proteins (IGFBPs) that modulate IGF activity, transport, and stability. The mammalian acid-labile subunit (ALS) forms ternary complexes with IGFs and IGFBP-3 or IGFBP-5. It has recently been shown that a Drosophila homolog of ALS, dALS, forms a ternary complex with Drosophila insulin-like peptide-2 (Dilp2) and the binding protein IMP-L2 (Arquier, 2008). Surprisingly, this study found that the gene encoding dALS is allelic to convoluted (conv), a gene previously shown (Beitel, 2000) to be required for regulating the length of epithelial tubes in the Drosophila tracheal system (Swanson, 2009).

The Drosophila tracheal system is a ramifying network of epithelial tubes that delivers oxygen directly to target tissues. The tracheal system is among the best characterized systems for investigating branching morphogenesis and for control of epithelial tube size, an essential feature of many vital organs such as lung and kidney. The dimensions of tracheal tubes are regulated by at least two distinct mechanisms, one of which involves a transient lumenal extracellular matrix (ECM) and the putative matrix-modifying proteins Vermiform (Verm) and Serpentine (Serp). Importantly, lumenal secretion of Verm requires the septate junction (SJ), which restricts paracellular diffusion similar to the vertebrate tight junction, but is located in the basolateral membrane and contains polarity proteins that promote basal membrane identity. Mutations that disrupt the matrix or Verm secretion cause individual tracheal cells, and thus the overall tubes, to become too long. It has not been established whether the ECM provides a signal to the epithelial cells that causes them to adjust their dimensions or whether the ECM serves as a mandrel that shapes the tubes by physical forces. However, in addition to the matrix-based size control mechanism, there is evidence that growth factor signaling constitutes a second mechanism that controls tracheal tube size because mutation of chico, which encodes the Drosophila homolog of the mammalian insulin receptor substrates 1-4, shortens tracheal tube length to match the reduced body size of the chico mutant larva (Swanson, 2009).

Analysis of the conv/dALS locus reveals that Conv activity defines a new step in the matrix-based size control process. Further, although Conv/dALS could potentially act through the insulin growth factor pathway to regulate tube size, the results suggest that the tracheal matrix organization function of Conv/dALS represents a distinct function from the IGF pathway function (Swanson, 2009).

conv has very low embryonic expression levels, and no conv cDNAs were present in the publicly available cDNA libraries. RT-PCR was used to assemble a complete cDNA and to confirm the predicted gene structure. The full-length cDNA is 3.3 kb long and has an open reading frame (ORF) of 1092 amino acids (aa) preceded by in-frame stop codons. No evidence for alternative splicing was observed (Swanson, 2009).

The Conv protein has a strongly predicted signal sequence at the N terminus and multiple leucine-rich repeats (LRR) that are commonly involved in protein-protein interactions. A BLASTP search revealed that the closest human homolog of Conv is ALS of the insulin growth factor binding complex (Boisclair, 2001). In strong support of Conv having functional as well as sequence similarity to human ALS, recent work by Arquier (2008) has demonstrated that Conv/dALS can bind and antagonize Dilp2 and that altering Conv/dALS levels can affect metabolism. Since larval tracheal length is reduced in chico mutants, which have reduced insulin-like growth factor signaling, the increased length of trachea in conv mutants is consistent with conv functioning to regulate tracheal tube length through insulin-like peptide signaling. Furthermore, the R278 mutation is located in a region of significant similarity between ALS and Conv/dALS and could potentially disrupt Conv/dALS insulin binding functions (Swanson, 2009).

Alternatively, although ALS is the closest human homolog of Conv/dALS, there are substantial differences between the two proteins that have not previously been noted. ALS is 605 aa long, has 19 tightly packed typical LRR domains, and has both N- and C-terminal class LRR domains. In contrast, Conv is 1092 aa long, has 23 somewhat dispersed LRR domains, and has a C- but not N-terminal class LRR domain. BLASTP searches of the Drosophila proteome using Conv/dALS reveal that Conv/dALS and human ALS (BLASTP score 181) have similarity that is less than or comparable to Conv/dALS and other Drosophila LRR proteins, including the cell adhesion protein Chaoptin (score 199), the Toll-7 and Toll-6 receptors (scores 181 and 179), and the 18-wheeler transmembrane protein (score 171) that controls salivary gland cell shape. Similarly, ClustalW alignment and Bootstrap tree analysis indicates that Conv/dALS is as related to Chaoptin and other LRR proteins as it is to ALS. Thus, while strong evidence supports Conv/dALS being the functional homolog of human ALS in IGF signaling, the divergence of human ALS and Conv/dALS is also consistent with Conv/dALS having an unanticipated IGF-independent function in tracheal morphogenesis (Swanson, 2009).

Because Conv/dALS might be a multifunctional protein and it was unclear whether conv^K6507b and conv^R278 were null alleles, a conv null allele was created with which to definitively characterize Conv/dALS functions. Imprecise excision of the P-element SelD^SH1599 created the small deficiency Df(2R)conv^Y58 that deletes the entire intergenic region 5' of conv as well as the first exon and a half of conv that include the transcriptional start site. No conv transcript is detected by RT-PCR in conv^Y58 homozygotes. Although conv^Y58 also disrupts the adjacent gene SelD, SelD does not have a role in tracheal development because no tracheal defects are apparent in SelD^K11320, SelD^SH1599, or nine new excision alleles that disrupt SelD but complement conv^R278. More definitively, the 6-kb conv genomic fragment that does not include any SelD ORF completely rescues the tracheal defects of conv^Y58 and Df(2R)7131 homozygous embryos . Therefore conv^Y58 is considered a molecular null allele of conv (Swanson, 2009).

Embryonic morphogenesis appears normal in both conv^R278 and conv^Y58, with the notable exception of tracheal tube size-control defects. Conv/dALS function is largely or entirely dispensable for neural and muscle function as larvae homozygous for conv^R278 or conv^Y58 hatch and begin crawling. However, neither these homozygous larvae nor larvae trans-heterozygous for conv^R278/conv^Y58 survive past the second larval stage, which is presumably due to the 100% penetrant failure of conv mutant trachea to fill with air and thereby provide oxygen to target tissues. Thus conv/dALS is an essential gene under normal conditions, which is in contrast to the viable null phenotype of Imp-L2, the insulin-binding protein required for Conv/dALS to bind Dilp2-containing complexes, and in contrast to the viable null phenotype of chico, the fly homolog of human insulin receptor substrates 1-4 (Swanson, 2009).

The trachea of embryos homozygous or trans-heterozyogous for conv mutations becomes abnormally long during stage 16. The onset and severity of the phenotypes caused by conv^R278, conv^Y58, Df(2R)7131, or conv^R278 in trans to conv^Y58 or Df(2R)7131 are indistinguishable, suggesting that conv^R278 is null for the tracheal functions of conv (Swanson, 2009).

To understand the role of Conv/dALS in tracheal tube-size control, it was asked if conv mutations affect known mechanisms of tube-size control. Localization of apical polarity and SJ markers appears normal in conv null mutants, as is the ultrastructure of intercellular septa that form paracellular diffusion barriers. Consistent with this, conv mutants have no paracellular barrier defects as evidenced by wild-type-like impermeability of trachea and salivary glands in conv^Y58/Df(2R)7131 embryos to a 10-kDa dextran dye. Chitin biosynthesis is not affected, as chitin accumulates normally in the tracheal lumen, as do the putative chitin deacetylases Verm and Serp. Together, these results indicate that Conv/dALS is not required for the known tracheal tube-size control mechanisms that operate before stage 16 (Swanson, 2009).

The defect common to most currently identified mutations affecting tracheal tube-size control is that they disrupt organization of the tracheal apical and/or lumenal extracellular matrix. In conv mutants, EMs show that at stage 16, the lumenal matrix appears to be less dense and somewhat grainier than in WT. By stage 17 the lumenal matrix in conv mutants appears even more sparse, and while WT embryos create a gap between the lumenal matrix and the tracheal tube surface, in conv mutants the matrix still extends to the tracheal surface. The failure of conv mutants to create a gap is also detectable by immunohistochemical staining of Verm. In stage 17 WT embryos, gaps are visible between the apical surface and the Verm-stained lumenal matrix, spaces adjacent to arrowheads that mark the apical surface of the tracheal cell), while in conv mutants Verm continues to occupy the entire lumenal space and is not organized into fibrils (Swanson, 2009).

In addition to lumenal matrix defects, conv mutants have grossly abnormal morphogenesis of the apical matrix (cuticle) that lines the tracheal surface. In conv mutants, the taenidia, which normally are stereotyped periodic ridges in the highly organized apical matrix, are frequently misshapen and flattened. Thus, Conv/dALS is required for normal organization and modulation of apical and lumenal matrices that control tracheal tube size. This result is unexpected because neither human ALS nor components of the Drosophila IGF pathway have been observed to be required for apical extracellular matrix organization and no aberrant embryonic tracheal morphologies have been observed in either chico or Imp-L2 mutants (Swanson, 2009).

As mutations in genes encoding SJ proteins and conv both cause lumenal matrix defects, additional genetic tests were performed to determine whether there were differences between the effects of conv and SJ mutations on the extracellular matrix. The genetic interactions of varicose (vari), which encodes an adaptor protein critical for SJ formation, and of conv with piopio (pio), and dumpy (dp). Pio and Dp are extracellular matrix proteins deposited in the tracheal lumen that contain ZP domains. ZP domains are named after proteins that form the zona pelucida, a gel-like substance surrounding mammalian oocytes. Intriguingly, ZP domains can polymerize to form strands, which in the trachea could potentially play a role parallel to that of the chitin-based fibrillar matrix. Alternatively, Pio and Dp are transmembrane proteins and thus could act as mediators of chitin-fibril-based signaling or scaffolding. Pio and Dp are required for the cell intercalation that produces unicellular tracheal branches (Jazwinska, 2003; Ribeiro, 2004). In strong pio and dp mutants intercalation fails to stop and unicellular tracheal branches become disconnected. As the multicellular tracheal tubes in pio and dp mutants have normal length and diameter, these ZP proteins were not thought to have a role in tracheal tube-size control. However, both a strong mutation in pio, pio^2R-20, and a viable mutation in dp that does not cause branch breaks, dp^ov1, significantly enhanced the tracheal tube elongation defects of both weak and strong mutations in vari. This enhancement was specific because a lethal mutation affecting DE-cadherin, shg^G119, that reduces DE-cadherin levels by >50% and causes sporadic tracheal branch breaks, did not enhance the vari, dp, or pio mutant phenotypes (Swanson, 2009).

In contrast to the enhancement of the vari length defects by dp and pio, double-mutant combinations of a strong conv allele and dp or pio did not have increased tracheal length, even when the dp conv double mutant was constructed with a stronger allele of dp, dp^OVLR, that causes a fully penetrant branch-break phenotype. Thus, although the exact role of pio and dp in tracheal length control remains to be determined, these results provide genetic evidence that conv has distinct effects from a SJ mutant on lumen matrix and size control. This possibility is further supported by double-mutant combinations of conv and vari and of conv and coracle showing enhanced tracheal length defects. Thus, the interactions of conv mutations with mutations in ZP and SJ genes indicate that Conv has a distinct role from SJ proteins controlling tracheal tube size (Swanson, 2009).

Mutations in conv cause tracheal tube-length and matrix defects similar to those caused by mutations in the wurst locus, which encodes a J-domain transmembrane protein that is required for clathrin-mediated endocytosis of lumenal material (Behr, 2007). However, in contrast to wurst mutants, lumenal clearance of the 2A12 marker and Verm in the conv mutant was the same as in wild type. Similarly, in conv mutants, clathrin had a dispersed cytoplasmic localization in epidermal and tracheal cells that was indistinguishable from that of wild-type embryos and markedly different from the striking membrane localization observed in wurst mutant epidermis (Behr, 2007). Therefore, Conv/dALS is not required for lumenal clearance and does not appear to regulate endocytosis. Taken together, these data suggest that in the temporal sequence of events, Conv/dALS acts after SJ proteins but before Wurst in controlling tracheal tube size and defines a new step in this process (Swanson, 2009).

If Conv/dALS functions as a matrix-organizing or cell-adhesion protein, one would expect it to be localized to the tracheal apical cell surface or lumen. Unfortunately, attempts to raise antibodies to Conv/dALS protein have been unsuccessful. Therefore attempts were made to determine the subcellular localization of Conv/dALS by expressing a YFP-tagged protein in the tracheal system using the ubiquitous da-Gal4 that efficiently expresses and rescues many tracheal genes. When expressed with da-Gal4, Conv::YFP showed little accumulation in the tracheal system and its subcellular localization could not be reliably determined in tracheal cells by YFP fluorescence or by anti-YFP immunofluorescent staining. Interestingly, although little Conv::YFP was evident in the trachea, da-Gal4 driving Conv::YFP almost completely rescued embryonic tracheal morphological defects, suggesting either that very little Conv/dALS is required in the embryonic tracheal system or that Conv/dALS does not act in the embryonic tracheal system (Swanson, 2009).

To more directly address whether Conv/dALS acts in the tracheal system, Conv/dALS and Conv::YFP were expressed using the btl-Gal4 driver that expresses only in the tracheal system and in some glia in the central nervous system. With this driver, Conv::YFP localized to the tracheal cytoplasm, suggesting that it was inefficiently trafficked to the cell surface. Similar results were obtained by Arquier (2008) with a myc::dALS construct expressed in the fat body and in cultured S2 cells. For the Conv::YFP fusion, the cytoplasmic localization was not an artifact of cleavage of the C-terminal YFP tag; Western blots showed that almost all detectable GFP immunoreactivity was in a high molecular weight band that corresponds to the correct size of the Conv::YFP fusion protein. There was no obvious accumulation of tagged protein in the tracheal lumen, apical surfaces, or basal surfaces (Swanson, 2009).

Despite poor trafficking of Conv::YFP in the tracheal system, both tagged- and untagged-expression constructs fairly efficiently rescued embryonic tracheal defects in conv mutants, but did not rescue the viability defects. As this rescue was achieved using the btl-Gal4 driver, this result is consistent with Conv/dALS acting as a cell-adhesion or matrix-organizing factor in the tracheal lumen. However, this result is also consistent with local rescue of IGF pathway function since Honegger (2008) has shown that ectopic expression of Imp-L2 in imaginal eye cell clones can locally reduce ommatidia size. Therefore Conv/dALS was expressed in the fat body using the ppl-Gal4 driver. Although this driver was successfully used in combination with Conv/dALS constructs by Arquier (2008) to alter IGF signaling pathway functions, no rescue of the tracheal defects was observed using either the untagged or the tagged constructs. A similar lack of rescue was observed using the cg-Gal4 driver, which expresses in the fat body as well as in hemocytes. These results suggest that Conv/dALS acts autonomously in the tracheal system (Swanson, 2009).

In conclusion, Conv/dALS, an important player in the IGF signaling pathway where it forms a ternary complex containing Imp-L2 (Arquier, 2008), is an essential gene. Surprisingly, Conv/dALS has an important role in tracheal epithelial morphogenesis, where it limits tube elongation. Although these results raise the fascinating possibility that Conv/dALS could act by dampening IGF signaling to prevent abnormal tracheal growth, the observations that Imp-L2 is not required for tracheal morphogenesis and that Conv/dALS appears to act autonomously in the tracheal system suggest that Conv/dALS has a tracheal-matrix organizing function that is distinct from its IGF-binding function. The exact role of Conv/dALS in matrix organization is unclear, but the apparently low level of embryonic Conv expression suggests that Conv/dALS acts as an important regulator rather than a structural component of the lumenal extracellular matrix (Swanson, 2009).

A putative Dilp cofactor is expressed in the FB

dilp expression results suggest that the general growth defects observed in fat body (FB) starved animals might be mediated through other diffusible factors linked to insulin/IGF signaling. In mammals, most circulating IGF-I is stoichiometrically associated in a ternary complex with IGF-BPs and a third partner called acid labile subunit (ALS) (Boisclair, 2001; Duan 2002). This ternary complex is known to regulate most of IGF-I biological functions and plays an important role in the stabilization of circulating IGF-I. ALS is a liver-secreted glycoprotein whose concentration in the serum varies with nutritional conditions. A putative Drosophila ALS ortholog gene (dALS) encodes a protein presenting 46% homology with human ALS in a 444 amino acid central region. Strikingly, the gene is expressed in the same seven dilp-expressing median neurosecretory cells in each larval brain lobe, reinforcing the notion of a functional link with Dilps. In contrast to the dilp genes, dALS is also strongly expressed in the larval FB, but not in any other larval tissue. In response to different starvation conditions, dALS is severely downregulated in the FB and in the median neurosecretory cells. Interestingly, dALS expression in the FB is strongly suppressed when amino acid restriction is induced in this tissue, suggesting that it is a direct target of the FB sensor mechanism (Colombani, 2003).

Search PubMed for articles about Drosophila

Arquier, N., et al. (2008). Drosophila ALS regulates growth and metabolism through functional interaction with insulin-like peptides. Cell Metab. 7(4): 333-8. PubMed ID: 18396139

Behr, M., et al. (2007). Wurst is essential for airway clearance and respiratory-tube size control. Nat. Cell Biol. 9: 847-853. PubMed ID: 17558392

Beitel, G. J. and Krasnow, M. A. (2000). Genetic control of epithelial tube size in the Drosophila tracheal system. Development 127: 3271-3282. PubMed ID: 10887083

Boisclair, Y. R., et al. (2001). The acid-labile subunit (ALS) of the 150 kDa IGF-binding protein complex: an important but forgotten component of the circulating IGF system. J. Endocrinol. 170: 63-70. PubMed ID: 11431138

Colombani, J., Raisin, S., Pantalacci, S., Radimerski, T., Montagne, J., Leopold, P. (2003). A nutrient sensor mechanism controls Drosophila growth. Cell 114(6): 739-749. PubMed ID: 14505573

Domene, H. M. et al. (2005). Acid-labile subunit deficiency: phenotypic similarities and differences between human and mouse. J. Endocrinol. Invest. 28: 43-46. PubMed ID: 16114275

Domene, H. M., et al. (2007). Phenotypic effects of null and haploinsufficiency of acid-labile subunit in a family with two novel IGFALS gene mutations. J. Clin. Endocrinol. Metab. 92: 4444-4450. PubMed ID: 17726072

Duan, C. (2002). Specifying the cellular responses to IGF signals: roles of IGF-binding proteins. J. Endocrinol. 175: 41-54. PubMed ID: 12379489

Duan, C. and Xu, Q. (2005). Roles of insulin-like growth factor (IGF) binding proteins in regulating IGF actions. Gen. Comp. Endocrinol. 142: 44-52. PubMed ID: 15862547

Garbe, J. C., Yang, E. and Fristrom, J. W. (1993). IMP-L2: an essential secreted immunoglobulin family member implicated in neural and ectodermal development in Drosophila. Development 119: 1237-1250. PubMed ID: 8306886

Honegger, B., Galic, M., Köhler, K., Wittwer, F., Brogiolo, W., Hafen, E. and Stocker, H. (2008). Imp-L2, a putative homolog of vertebrate IGF-binding protein 7, counteracts insulin signaling in Drosophila and is essential for starvation resistance. J. Biol. 7(3): 10. PubMed ID: 18412985

Jazwinska, A., Ribeiro, C. and Affolter, M. (2003). Epithelial tube morphogenesis during Drosophila tracheal development requires Piopio, a luminal ZP protein. Nat. Cell Biol. 5: 895-901. PubMed ID: 12973360

Ribeiro, C., Neumann, M. and Affolter, M. (2004). Genetic control of cell intercalation during tracheal morphogenesis in Drosophila. Curr. Biol. 14: 2197-2207. PubMed ID: 15620646

Rosenfeld, R. G., et al. (2000). The insulin-like growth factor-binding protein superfamily. Growth Horm. IGF Res. 10 (Suppl A): S16-S17. PubMed ID: 10984276

Sloth Andersen, A., Hertz Hansen, P., Schaffer, L. and Kristensen, C. (2000). A new secreted insect protein belonging to the immunoglobulin superfamily binds insulin and related peptides and inhibits their activities, J. Biol. Chem. 275: 16948-16953. PubMed ID: 10748036

Swanson, L. E., Yu, M., Nelson, K. S., Laprise, P., Tepass, U. and Beitel, G. J. (2009). Drosophila convoluted/dALS is an essential gene required for tracheal tube morphogenesis and apical matrix organization. Genetics 181(4):1281-90. PubMed ID: 19171940

Yamanaka, Y., et al. (1997). Inhibition of insulin receptor activation by insulin-like growth factor binding proteins. J. Biol. Chem. 272: 30729-30734. PubMed ID: 9388210

date revised: 30 October 2009

Home page: The Interactive Fly © 2009 Thomas Brody, Ph.D.

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