dumbfounded is the only other fusion gene that is known to be restricted in expression to the founders and is absent in fusion competant myoblasts (FCM) (Ruiz-Gomez, 2000). In addition to the overlap in spatial expression, the temporal expression profiles of rols7 and duf in these tissues appear identical. These observations, the receptor-like nature of the molecule encoded by duf, and the loss of membrane-enriched Rols7 in the duf mutant led the authors to test whether Duf expression is sufficient to promote Rols7 membrane localization. This was done by overexpressing Rols7 early throughout the mesoderm and later in all muscles, either by itself or together with Duf, using the 24B-GAL4 driver. Under either of these conditions, no change was observed in the patterning of the somatic muscles. Rols7 localization was then examined late at stage 16, a time at which the expression of endogenous Rols7 and probably that of Duf is lost in wt muscles. When Rols7 is overexpressed alone, the protein appears as speckles throughout the cytosol of mature muscles. Despite its abundance, no membrane patches could be detected. In contrast, co-overexpression with Duf results in Rols7 becoming membrane enriched, with little or no protein remaining in the cytoplasm of muscles. In addition, higher levels of Rols7 are found along membranes that come in direct contact with other muscles. Muscles such as VL1, VO4, and VO6 that abut other muscles only on one side show significantly higher levels of Rols7 on the side in contact with its neighbor, whereas muscles such as VL2 and VO5, which lie between muscles, show equally high levels of Rols7 expression on either side of the membrane. This and observations in wt embryos, where Rols7 accumulates at discrete sites along the myotube membrane suggest that Duf and Rols7 may be present at specialized sites along the founder (or myotube) membrane that contact the FCM (Menon, 2001).
The aggregation of Rols in distinctive cytoplasmic locations in founder cells, and the presence of multiple protein-protein interaction motifs in the Rols protein prompted an examination of whether Rols plays a role during myoblast fusion by mediating interactions between molecules in the myoblast fusion pathway(s). To test whether Rols interacts with other fusion molecules, coimmunoprecipitation assays were performed in Drosophila S2 cells using MYC-tagged Rols and other fusion proteins, including Blow, Duf, Mbc, and Sns, tagged with the V5-epitope at their carboxyl termini. Rols interacts with the founder cell receptor Duf but not the fusion-competent cell receptor Sns, despite the high homology shared by Duf and Sns. This specific interaction between Rols and Duf is consistent with the founder cell-specific expression of Rols. A cleaved form of is generated when full-length Duf is expressed in S2 cells. This form migrates slightly slower than the Duf cytoplasmic domain alone, suggesting that it is likely to contain both the transmembrane and the cytoplasmic domains. Interestingly, this cleaved form also associates with Rols. However, when the Duf cytoplasmic domain alone was tested, no interaction was detected. These results suggest that the transmembrane domain of Duf is required for its interaction with Rols. In addition, protein-protein interaction was detected between an amino-terminal fragment of Mbc and Rols, while no interaction was detected between Blow and Rols. The interactions between full-length Mbc and Rols could not be tested, since the full-length Mbc was not expressed at a detectable level (Chen, 2001).
To locate the specific domain(s) of Rols that are required for its interaction with Duf, a carboxy-terminal deletion (Rols-DeltaC) was created that truncates the conserved region between Drosophila Rols and its mouse orthologs. This deletion construct was tested for its ability to associate with Duf in coimmunoprecipitation experiments. No interaction between the truncated Rols protein and Duf was detected, suggesting that the conserved region of Rols is required for its interaction with Duf. This conclusion is consistent with the genetic mutants, since the antsT321 allele produces carboxy-terminal-truncated protein that deletes the entire conserved region (Chen, 2001).
The interaction between Rols and Duf, together with the subcellular aggregation of the Rols protein, suggests that Rols is likely to colocalize with Duf during myoblast fusion. Because of the lack of Duf antibody, this hypothesis could not be tested directly. However, if Duf is involved in recruiting Rols to specific subcellular locations during fusion, one would expect a change in the pattern of Rols localization in duf mutant embryos. Examination of Rols protein in duf mutant embryos has shown this to be the case. Instead of localizing to discrete sites in the cytoplasm, Rols protein is distributed throughout the cytoplasm at the peripheral membrane region and appears as rings that outline the founder cells in the duf mutant embryo (Chen, 2001).
The protein encoded by the rols7 transcript has several distinct domains that can potentially participate in protein-protein interactions. At its N terminus, Rols7 carries a C3HC4 zinc finger, called the RING finger. While studies on several RING finger-containing proteins such as Cbl suggest that this domain is essential for E2-dependent ubiquitin protein ligase activity leading to protein destruction, the RING fingers in other proteins have been implicated in different modes of protein-protein interactions. Of note, the RING finger in the vertebrate muscle-specific proteins termed MURFs is essential to establish stable interaction between a specific MURF and Titin or the cytoskeletal network of microtubules (Centner, 2001; Spencer, 2000). At its C terminus, Rols7 encodes three different protein interaction motifs: a tandem array of nine ankyrin repeats followed closely by three TPR repeats and a coiled-coiled domain. Based on its overall structure, it is plausible that Rols7 could act as a focal point for the assembly of a multiprotein complex at the membrane where the Duf receptor is located, bringing the fusion machinery and directing changes in the cytoskeleton to sites where fusion would take place. In support of this, it has been shown that Rols7 is required for the enrichment of D-Titin to fusion sites in founders (or myotubes). However, this appears to be only one of the roles served by Rols7, since founders in the rols mutant either remain unfused or develop into small precursors, whereas fusion is arrested at a later stage in the D-Titin null allele (Menon, 2001).
The rolling pebbles gene of Drosophila encodes two proteins, one of which, Rols7, is essential for myoblast fusion. In addition, Rols 7 is expressed during myofibrillogenesis and in the mature muscles. Here it overlaps with alpha-Actinin (a-Actn) and the N-terminus of D-Titin/Kettin/Zormin in the Z-line of the sarcomeres. In the attachment sites of the somatic muscles, Rols7 and the immunoglobulin superfamily protein Dumbfounded/Kin of irreC (Duf/Kirre) colocalise. As Duf/Kirre is detectable only transiently, it may be involved in establishing the first contact of the outgrowing muscle fiber to the epidermal attachment site. It is proposed that Rols7 and Duf/Kirre link the terminal Z-disc to the cell membrane by direct interaction. This is supported by the fact that in yeast two hybrid assays the tetratricopeptide repeat E (TPR E) of Rols7 shows interaction with the intracellular domain of Duf/Kirre. The colocalisation of Rols7 with a-Actn and with D-Titin/Kettin/Zormin in the Z-dics is reflected in interactions with different domains of Rols7 in this assay. In summary, these data show that besides the role in myoblast fusion, Rols7 is a scaffold protein during myofibrillogenesis and in the Z-line of the sarcomere as well as in the terminal Z-disc linking the muscle to the epidermal attachment sites (Kreiskother, 2006).
The scaffold protein Rols7 has been shown to be essential for myoblast fusion in the somatic mesoderm during Drosophila embryogenesis where it might interact with several components of the fusion machinery. Evidence is presented that Rols7 has an additional function in the establishment of the muscle attachment and the formation of the Z-discs, as well as in the Z-discs of the mature muscles (Kreiskother, 2006).
During myoblast fusion, Rols7 mRNA decays at stage 15. Antibody staining of stage 17 embryos reveal a concentration of Rols7 at the muscle ends next to the epidermal attachment sites, which are caused by new transcription in RT-PCR experiments. Later on in the mature larval muscles Rols7 is detected in the sarcomeric Z-discs (Kreiskother, 2006).
During the early stages of myogenesis, the interaction of the founder cell specific protein Duf/Kirre and the fusion competent myoblasts (fcm) specific Sns leads to the adhesion of the two cell types, which is a prerequisite for further steps of the fusion process. Besides this, Duf/Kirre transiently are concentrated at the end of the developing muscles at stage 15 and 16, while it disappears again at stage 17. This led to a hypothesis that Duf/Kirre might participate in the first contact of the outgrowing muscle to the attachment site, as does Vein. This would require an interaction partner in the extracellular matrix or at the epidermal site. Since the sns transcript is present in the muscle attachment sites at a low level at stage 17, antibody staining for Sns was performed, but a distinct signal in the attachment sites could not be detected. As well as the transcript of sns, its paralog, Hibris (Hbs), is also found in the muscle attachment sites, and, more exactly, localised to the contact site between the cells at the epidermal attachments. Thus, it could function as an interaction partner for Duf/Kirre (Kreiskother, 2006).
As a further possible interaction partner Rst/IrreC was considered, since Rst/IrreC, the paralogue of Duf/Kirre, shows expression in the epidermal tendon cells during embryonic stages. Due to the fact that Duf/Kirre and Rst/IrreC are indeed able to undergo heterophilic interaction in cell culture experiments, the conclusion is drawn that Rst/IrreC might be the candidate for an interaction partner of Duf/Kirre on the epidermal site, thus enabling an early contact of the muscle to the epidermal attachment site (Kreiskother, 2006).
Rols7, which interacts with the intracellular domain of Duf/Kirre, is also localised at the muscle ends from late stage 16 onwards shortly before Duf/Kirre disappears. It is speculated that Rols7 is brought to the membrane where it interacts with Duf/Kirre (Kreiskother, 2006).
alpha actinin (α-Actn) and D-Titin/Kettin, both found at the muscle attachment site in a similar pattern, also interact with Rols7 (at least in the yeast two hybrid assay) and participate in the establishment of the terminal Z-disc. For the flight muscle it was shown that α-Actn is essential for the formation of this structure and for obtaining a correct insertion of the myofibril to the epidermal tendon cell. Furthermore the yeast assay showed an interaction of α-Actn with Duf/Kirreintra (Kreiskother, 2006).
kettin mutants show strong defects in terminal Z-disc function. This study proposes that, in addition to Kettin, Rols7 and α-Actn are important for the formation of this structure. The process might be connected to Muscleblind (Mbl), since in mutants for mbl, Z-discs are not assembled correctly. Unfortunately, a mutant analysis of Rols7 function in terminal Z-disc formation is difficult due to its essential function during myoblast fusion (Kreiskother, 2006).
Apart from the myoblasts and attachment sites, Rols7 is expressed in the developing sarcomeres of larval and adult muscles and localises to the Z-discs, as was shown using antibodies for α-Actn and D-Titin/Kettin as markers. Yeast interaction assays revealed that Rols7 might directly interact with α-Actn and Zormin, which, like Kettin, is an isoform derived from the sallimus (sls) gene and also localises to the Z-discs. Therefore, it is postulated that Rols7 serves as a scaffold protein that links α-Actn and Zormin in the Z-disc. Furthermore, the analyses of alpha actinin mutants showed that the presence of α-Actn is not necessary for Rols7 localisation to the Z-discs. In addition, Rols7, as well as α-Actn and D-Titin/Kettin, is present during the assembly of the sarcomere. In vertebrates it has been shown that in spreading edges of rat cardiomyocytes, dense bodies that contain Z-disc proteins assemble at the spreading membrane and align to premyofibrils in cooperation with newly formed actin filaments and small myosin filaments (Kreiskother, 2006).
Antibody staining showed protein aggregates that aligned to form kinds of premyofibrils and demonstrated that in Drosophila, the assembly of the Z-discs seems to be similar to that of vertebrates. So, Rols7 is the first protein that is essential for myoblast fusion and plays an additional role in the sarcomere assembly as well as in the Z-discs of mature muscles, where it is proposed that it links α-Actn and D-Titin/Kettin/Zormin. Δ-titin/kettin-mutants have a weaker fusion phenotype than rols7-mutants, however, D-Titin/Kettin is clearly expressed during myoblast fusion as a component of the adhesion complex between founder cell and fcm. Individual Rols7 domains serve different function in distinct processes of myogenesis (Kreiskother, 2006).
From coimmunoprecipitation experiments it was already supposed that the intracellular domain of Duf interacts with Rols. Furthermore, cell culture cotransfection assays showed colocalisation of Duf, Rols and D-Titin. In yeast interaction assays the individual domains of Rols7 were tested for interaction with potential partners that included components of the fusion machinery which might be relevant for muscle attachment or sarcomere assembly as well. Indeed, the different domains interact with different partners in different developmental contexts, and it is concluded that Rols7 is a multifunctional protein (Kreiskother, 2006).
The interaction of Rols7 with the intracellular domain of Duf/Kirre was confirmed and it was found that the interaction probably is mediated by the TPR repeats of Rols7, respectively, by the most C-terminal TPR E repeat and the R1 fragment that contains the RING finger and an additional part of 321 amino acids downstream. In contrast α-Actn interacts with the R1 domain and with both TPR repeats, the TPR E and the TPR X, whereas the N-terminal part of Zormin interacts only with the R1 domain in this assay. No interaction was detected for the N-terminal part of Kettin and the Rols7 domains. These results, together with the rescue capability of truncated Rols7 versions, led to the proposal of certain functions to individual Rols7 domains. Either the RING finger domain, the TPR repeats or the ankyrin repeats and the TPR repeats have been deleated and the remaining parts of Rols7 were examined for their competence to rescue the rols fusion defect. A deletion of the RING finger domain does not affect the rescue of the rols fusion phenotype, whereas a deletion of the TPR repeats leads to a partial rescue and a deletion of ankyrin repeats and TPR repeats together does not rescue fusion at all. The Rols7 version without the RING finger rescues the fusion phenotype. This RING finger is included in the R1 fragment which interacts with Duf/Kirreintra, Blow, Zormin and α-Actn. Thus, it is proposed that the R1 domain is a candidate to mediate the transient interaction of Duf/Kirreinttra at the muscle attachment sites. The R1 domain of Rols7 could then mediate the interaction with Zormin in the Z-discs in all larval muscles. R1 is the only Rols7 fragment that interacts with Zormin. R1 and TPR E as well as TPR X have the capability to interact with α-Actn. It cannot be decide whether both domains of Rols7 interact with α-Actn in the Z-discs. The interaction of Duf/Kirreintra with the TPR E repeat indicates a function of the TPR E repeat during myoblast fusion, since its deletion only leads to a partial rescue of the rols fusion phenotype. The ankyrin repeats did not interact with any of the proteins which have been tested in the yeast assay and which are characteristic for sarcomere assembly and muscle attachment. Taking this together with the fact that a deletion of this domain, in addition to a deletion of the TPR repeats, prevents the rescue of the fusion defect, indicates that the ankyrin repeats predominantly function during myoblast fusion (Kreiskother, 2006).
Rols7 is a scaffold protein which contains distinct domains characteristic of protein-protein interaction. It is proposed that the interaction of the appropriate domain with certain proteins is specific for the process of myogenesis, myoblast fusion, muscle attachment or sarcomere assembly (Kreiskother, 2006).
Drosophila body wall muscles are multinucleated syncytia formed by successive fusions between a founder myoblast and several fusion competent myoblasts. Initial fusion gives rise to a bi/trinucleate precursor followed by more fusion cycles forming a mature muscle. This process requires the functions of various molecules including the transmembrane myoblast attractants Dumbfounded (Duf) and its paralogue Roughest (Rst), a scaffold protein Rolling pebbles (Rols) and a guanine nucleotide exchange factor Loner (Schizo). Fusion completely fails in a duf, rst mutant, and is blocked at the bi/trinucleate stage in rols and loner single mutants. This study analysed the transmembrane and intracellular domains of Duf, by mutating conserved putative signaling sites and serially deleting the intracellular domain. These were tested for their ability to translocate and interact with Rols and Loner and to rescue the fusion defect in duf, rst mutant embryos. Studying combinations of double mutants, further tested the function of Rols, Loner and other fusion molecules. This study shows that serial truncations of the Duf intracellular domain successively compromise its function to translocate and interact with Rols and Loner in addition to affecting myoblast fusion efficiency in embryos. Putative phosphorylation sites function additively while the extreme C terminus including a PDZ binding domain is dispensable for its function. It was also shown that fusion is completely blocked in a rols, loner double mutant and is compromised in other double mutants. These results suggest an additive function of the intracellular domain of Duf and an early function of Rols and Loner which is independent of Duf (Bulchand, 2010).
This study has shown that in order to ensure successful fusion a large part of the intracellular region of Duf is required for its function. Serial truncations of the intracellular domain reveal that the efficiency of fusion is decreased as larger regions are removed. Also, conserved putative phosphorylation signalling sites function additively resulting in efficient myoblast fusion and the formation of a mature myotube. Several parallels can be drawn from this data and that published by Kocherlakota (2008), on the intracellular domain of the Duf ligand SNS. Similar to what has been found for SNS, the PDZ binding domain is not required for the function of Duf during myoblast fusion. This is contrary to the role of this domain in the function of Rst in the developing eye. While the intracellular domain of SNS is important for its function, the C terminal end of SNS is dispensable similar to that of Duf as shown by Duf ΔCT1-flag in the Rols/Loner translocation assay in S2 cells and rescue of the fusion defect in duf, rst embryos. The membrane proximal intracellular regions of SNS and Duf are more important for their functions. While SNS is phosphorylated on tyrosine residues, the ability of Duf 4 phos-flag to only partially rescue the duf, rst mutant, implies that phosphorylation of these sites also contributes to Duf function (Bulchand, 2010).
Membrane anchored forms of Duf irrespective of the sequence of the transmembrane domain, appear to be sufficient for successful fusion. This suggests that the transmembrane domain of Duf does not perform any essential role or contribute to downstream signalling activity and only serves to anchor Duf to the plasma membrane. The PADVI motif, though not essential for myoblast fusion, might have a function in the context of a different tissue type that has not been tested so far. That the functions of Duf cannot be attributed to particular motifs might be a strategy utilised to ensure that normal myotube development occurs in a robust manner and compromising the function of any of these motifs singly, does not drastically affect the overall process. As has been suggested for the downstream functions of SNS, Duf too might transduce signals to cytoskeletal elements via its intracellular domain, to ensure successful myoblast fusion (Bulchand, 2010).
Previous studies proposed that myoblast fusion molecules can be categorised into those that participate in the early versus later phases of fusion. More recently it has been proposed that all fusion molecules are required in all fusion events. Molecules like Rols and Loner have been individually shown to function in the second phase of fusion after the formation of the bi/trinucelate precursor. This study has shown that removal of both rols and loner completely blocks fusion similar to the duf, rst mutant. Analyses of other similar double mutants demonstrate that genes involved in myoblast fusion might interact with each other to affect fusion efficiency. It is possible that what this study has shown with a few myoblast genes is true for other genes that have thus far been characterised for their role in the later stages of fusion. Such interactions have been shown for Kette/Hem/Nap1/GEX-3 and Blow (Bulchand, 2010).
This study has shown that membrane anchored Duf without its intracellular domain and without any interaction with Rols and Loner, is sufficient to initiate fusion. It is possible that even in the absence of robust Duf dependent signal transduction, requirements for the formation of a bi/trinucleate precursor are met. It was also shown that Rols and Loner are required, albeit redundantly, for precursor formation or the initial phase of fusion suggesting that this 'early function' of these molecules appears to be independent of Duf. This fusion defect was observed in late stage 15-early stage 16 embryos to ensure that the observations and interpretation thereof are not due to a delay in fusion. Rols and Loner may perform different roles early versus later on during myoblast fusion. In the later phase of fusion, Rols and Loner appear to sustain fusion by interacting with and translocating Duf to the surface of the myotube. As has been suggested in the case of Rols, Loner too might serve to regulate Duf at the surface of the myotube through as yet unknown mechanisms. It is possible that these supposed distinct early versus late mechanisms are used in mutant conditions in an effort to overcome fusion blocks, thus leading to delayed fusion events (Bulchand, 2010).
The embryonic expression of the rols transcripts was analyzed using antisense RNA directed against the N-terminal exons unique to either rols7 or rols6, the two identified transcripts (see gene structure section). Transcripts corresponding specifically to rols7 were first observed at about stage 11 in clusters of mesodermal cells, and this pattern quickly resolves into expression in single cells. By virtue of their position within the mesoderm, it is concluded that these cells are the progenitors of the somatic, pharyngeal, and visceral musculature, with the smaller cells in the somatic mesoderm, often seen in pairs, being sibling founders arising from the division of progenitors. The expression of rols7 in the somatic and pharyngeal mesoderm closely follows the fusion profile, appearing first in the presumptive progenitors and founders and then in fusing myotubes. In very young myotubes, rols7 RNA can be seen at higher levels in one nucleus, likely to be that from the founder cell. Expression rapidly declines by about stage 15 as fusion comes to completion, and by stage 16, rols7 expression is completely lost. In the visceral mesoderm, rols7 expression is maintained until about stage 12 and then disappears (Menon, 2001).
In contrast, rols6 transcripts are never observed in the somatic or visceral mesoderm. Expression is initiated relatively later at stage 13 in epithelial and endodermal tissues and the pharynx, and persists until the end of embryogenesis. Hence, the two rols transcripts show distinct temporal and spatial expression patterns (Menon, 2001).
To analyze the distribution of the rols7 protein, antibodies were raised against a region within its unique N terminus and expression was studied in an rP298-lacZ enhancer trap line. As anticipated from analyses of rols7 RNA, the protein is coexpressed with ß-galactosidase in somatic muscle progenitors and founders, with a higher level of expression consistently detected in the founders. Expression persists in myotubes, but levels decline as fusion comes to an end. In a loss-of-function allele of Notch (N), where the fusion competent myoblasts (FCM) take on the founder cell fate and fusion is abrogated, a corresponding expansion of the rols7 domain is observed both at the RNA and protein levels. Rols7 is also present in the pharyngeal and visceral mesoderm. No staining is observed with this antibody in Df(3L)BK9 embryos (Menon, 2001).
At the subcellular level, Rols7 appears to be cytoplasmic in muscle founders, whereas it is seen enriched along the membrane of nascent myotubes in punctate domains. As the myotube matures, more protein is detected along the membrane with less in the cytoplasm. In fusion mutants such as sns, D-mef2, mbc, and blow, Rols7 is found in the cytoplasm of founders but with time it becomes increasingly enriched along the cell membrane of unfused founders. This suggests that during normal development, fusion begins soon after Rols7 becomes membrane enriched in muscle founders, and this normally transient state is made apparent when fusion is stalled as a result of various mutations (Menon, 2001).
Attempts were made to place rols7 within the hierarchy of other genes essential for myoblast fusion by examining Rols7 expression in various fusion mutants. In embryos carrying a null allele of sns, Rols7 is seen in muscle progenitors and founders at stage 11, just as in wt embryos, and remains detectable in unfused founders until stage 15. In addition, the protein continues to become localized to discrete sites along the founder cell membrane, with more protein being detected along the membrane and less in the cytoplasm with time. From this, it appears that Sns is not required for the initiation, maintenance, or localization of Rols7. Similar analyses were performed for null alleles of D-mef2, mbc, blow, and duf. Strikingly, all mutants exhibit Rols7 membrane enrichment in unfused founders with the exception of duf. In duf mutants, Rols7 appears to form speckles throughout the cytoplasm of unfused founders, that later turn into larger aggregates (Menon, 2001).
Expression of the rols7 starts at the extended germ band stage in progenitor/founder cells and persists during fusion to myofibers of the body wall and pharyngeal muscles. rols7 expression is first observed at the extended germ band stage in a few mesodermal cells per segment and in the head region where the precursors of the pharyngeal muscles arise. Transient expression of rols7 is observed in the visceral mesoderm during germband retraction and then it vanishes. The functional significance of rols7 expression in the visceral mesoderm remains to be clarified (Rau, 2001).
During germband retraction, myogenesis proceeds and several cells express rols7 in every segment. The precursors formed at stage 13 accumulate rols7 mainly around one nucleus. The mRNA is of very low abundance so that colocalization experiments with other markers were not of sufficient quality. Therefore, the expression pattern of rols7 was compared to that of Mef2 which is expressed throughout embryogenesis and with rP298 (an enhancer trap pattern corresponding to dumbfounded). In these experiments rP298 was used to identify muscle founder cells. Only a few myoblasts are rP298 positive when compared to Mef2 in all nuclei of myoblasts. By contrast, the rols7 expression is very similar to the expression pattern of rP298 in the somatic mesoderm, indicating that rols7 is expressed in progenitor/founder cells up to the muscle precursor cells (Rau, 2001).
The pharyngeal musculature is also syncytial and the development of these muscles is also dependent of rols. sticks and stones mutants show fusion defects in the development of the pharyngeal musculature. Indeed, expression analysis revealed two distinct cell populations in the clypeolabrum at late stage 12 that later invaginate to form the pharyngeal muscles. rols7 and rP298 show overlapping expression patterns in the ventrally localized mesodermal cells of the clypeolabrum, while sns transcripts, which mark fusion competent myoblasts, are localized in the dorsally adjacent mesodermal cells. Since the rols7 pattern overlaps with the rP298 pattern (reflecting duf expression), it is suggested that the pharyngeal musculature also consists of founder cells and fusion competent cells (Rau, 2001).
rols6 exhibits a distinct transcript profile. At the extended germ band stage, rols6 is expressed strongly in the invaginating endoderm. At stage 14 and later, rols6 is expressed in ectodermal cells of the head region and the developing malpighian tubules are stained weakly . Furthermore, a stripe-like expression pattern is observed in the ectoderm at stage 16. The expression pattern of rols6 is thus very similar to the enhancer trap pattern observed in rolsP1729 in the apodemes and in rolsP1027 in the endoderm. This distinct distribution shows that rols7 is transcribed during development of the Drosophila pharyngeal and body wall musculature. Because only rols7 is detected in myoblasts during fusion and ceases beyond stage 14, it is proposed that this is the relevant transcript. Comparison with the rP298 staining indicates that rols7 is expressed starting with progenitors/founders up to muscle precursor cells during myoblast fusion (Rau, 2001).
The scaffold-like protein D-Titin (Sallimus) is expressed in prefusion myoblasts: during fusion, it accumulates along the membrane at sites of myoblast-myotube contact. Although fusion is initiated in the absence of D-Titin, it does not go to completion, and results in the formation of significantly smaller muscles that show aberrant myotube morphology with many unfused myoblasts. This suggests that proper myotube formation is achieved through the synchronization of fusion events occurring at the cell (or myotube) surface and changes in cytoskeletal architecture that take place within the cytoplasm. Whether Rols7 might serve to coordinate these two processes in the founder (or myotube) by linking Duf function to D-Titin expression and/or localization was tested (Menon, 2001).
In wt embryos, D-Titin is strongly expressed along the periphery of the FCM, whereas its expression around the founder cell periphery is relatively weak. Upon fusion, the myotube expresses D-Titin along its periphery, with enrichment at sites of myoblast-myotube contact that coincide with sites where Rols7 appears to be membrane associated. In D-mef2, sns, duf, and Df(3L)BK9 embryos, D-Titin expression in unfused founders is initially weak but increases with time. However, while the founders in D-mef2 and sns embryos continue to show peripheral D-Titin expression and enrichment at discrete sites that colocalize with membrane-associated Rols7, its localization in duf and Df(3L)BK9 founders is different. In duf, D-Titin no longer shows membrane enrichment and the protein appears cytosolic. In Df(3L)BK9 embryos, D-Titin can sometimes be seen around the membrane of some precursors without showing enrichment. More often, membrane-associated D-Titin is not detectable and the protein appears cytosolic. However, by reintroducing Rols7 into the founders of Df(3L)BK9 embryos using rP298-GAL4/UAS-rols7, D-Titin is seen to regain peripheral localization and becomes clearly enriched at discrete sites along the myotube membrane that colocalizes with Rols7 and contacts other myoblasts. Since Rols7 expression and membrane localization remain unaffected in the D-Titin mutant, Rols7 appears to function upstream of D-Titin. From these results, it is concluded that the enrichment of D-Titin at fusion sites within the founder or precursor requires Rols7 (Menon, 2001).
The expression profile of rols7 suggests that it may act at the level of muscle founder or precursor cells. The analysis of the mutant phenotype reveals the appearance of elongated mini-muscles containing more than one nucleus in several cases. In order to test the hypothesis that Rols acts on the precursor level, the fusion competence of rols mutant cells were examined in a wild-type background with a cell transplantation strategy. The wild-type host embryos contain a daGAL4 construct, which is expressed in all cells, while the transplanted cells of the rols mutant contained a UAS-lacZ gene. Thus rols mutant cells in a wild-type background express only the reporter ß-galactosidase after successful cell fusion with the wild-type host cells. One might expect that successful fusion is dependent on the transplanted cell type. Since ventral mesodermal cells were transplated at the cellular blastoderm stage, they develop either to founders or to fusion-competent cells, the latter being the far more abundant cell type. If Rols is indeed required in muscle precursor cells -- as suggested by its expression and mutant phenotype -- but not in FCMs, wild-type precursor cells of the host embryo should be able to recruit FCMs from rols mutants to form myofibers that express ß-galactosidase. However, transplanted donor cells that develop into precursors should not be able to recruit further host myoblasts for fusion (Rau, 2001).
The deficiency Df(3L)BK9 was chosen as a null allele for rols. Since mutant mesodermal cells were transplanted into wild-type embryos, the loss of rols6 in the endoderm and semaphorin 5c in the ectoderm as well as the loss of other genes localized in the deficiency should not influence this assay. After transplantation the recipient embryos were allowed to develop until 3rd instar larval stage and the muscle pattern was examined with respect to myotube development (Rau, 2001).
From a total of 191 transplantations, 119 embryos (62%) reached the third larval instar. Of these larvae, 53 showed a clone derived from the transplanted cell descendants, which had fused to host cells. In 11 cases, the clones derived from homozygous rols donors and were found in the musculature. These clones were compared at the morphological level with 42 clones derived from control donors with regard to size, shape and correct attachment. The 11 clones derived from homozygous rols embryos were found in the ventral, lateral and dorsal body wall muscles of third instar larvae, and they demonstrate that at least a population of rols mutant cells is able to fuse with host cells to multinucleate myotubes. These data show that rols mutant cells can participate in fusion. Since these large muscle clones are abundant, it is suggested that in these clones rols mutant FCMs fuse with wild-type founders. Moreover the clones were found at the same positions in the thoracic and the abdominal segments and approximately in the same size as the control clones. In the case of transplanted rols mutant cells, small muscle-like structures referred to as mini-muscles or compact, not elongated muscle-like cells were found in five larvae. These mini-muscles represent only a small part of the observed clones and contain two to five nuclei each. These mini-muscles are considered as precursor cells formed by a rols mutant founder and fusion competent cells of the donor. In one case, two of these mini-muscles were observed close to each other. The appearance of duplicated mini-muscles is interpreted as evidence for two precursor cells derived from asymmetric cell division of a progenitor. On the basis of these observations, it is concluded that the mini-muscles can be hybrid precursor cells derived from a rols mutant founder (with UAS-lacZ) and one to four cells of the wild-type host (carrying daGAL4). Therefore it is suggested that Rols acts in muscle precursors to recruit further FCMs. With rols-deficient cells, nearly 50% of the host embryos were found to contain mini-muscles. In four cases of the 43 control transplantations (in about 10% of the embryos) with hetero- or homo-zygous balancer embryos, similar defects were detected and it is assumed that this is probably a result of a dosage effect or may be due to incompatibility of homozygous balancer cells in the mosaic clones (Rau, 2001).
During myoblast fusion, cell-cell recognition along with cell migration and adhesion are essential biological processes. The factors involved in these processes include members of the immunoglobulin superfamily like Sticks and stones (Sns), Dumbfounded (Duf) and Hibris (Hbs), SH3 domain-containing adaptor molecules like Myoblast city (Mbc) and multidomain proteins like Rolling pebbles (Rols). For rolling pebbles, two differentially expressed transcripts have been defined (rols7 and rols6). However, to date, only a muscle fusion phenotype has been described and assigned to the lack of the mesoderm-specific expressed rols7 transcript. This study shows that a loss of the second rolling pebbles transcript, rols6, which is expressed from the early bud to later embryonic stages during Malpighian tubule (MpT) development, leads to an abnormal MpT morphology that is not due to defects in cell determination or proliferation but to aberrant morphogenesis. In addition, when Myoblast city or Rac are knocked out, a similar phenotype is observed. Myoblast city and Rac are essentially involved in the development of the somatic muscles and are proposed to be interaction partners of Rols7. Because of the predicted structural similarities of the Rols7 and Rols6 proteins, it is argued that genetic interaction of rols6, mbc and rac might lead to proper MpT morphology. It is also proposed that these interactions result in stable cell connections due to rearrangement of the cytoskeleton (Putz, 2005).
The Malpighian tubules (MpTs) of Drosophila arise as four buds from the hindgut anlage close to its boundary with the posterior midgut primordium. The cells of the four buds are characterised by the expression of the transcription factor Cut (Ct) at stage 10 of embryogenesis. During germ band extension at stage 11, the cells of the four tubule primordia undergo cell proliferation, and the tubules begin to bud out. By stage 13, proliferation is complete and short tubules have formed. From stage 13 onwards, cells from the caudal mesoderm join the MpT primordia and later the stellate cells (SCs). From the end of germ band retraction, the tubules begin to elongate due to cell rearrangement. In stage 15 and 16 embryos, the characteristic stereotypic course of the four renal tubules through the embryonic body is clearly visible. The paired posterior tubules span the posterior abdominal and terminal segments of the embryo. The anterior tubules extend forwards into abdominal segments 2/3 where the tubule loops back on itself so that the tips of both anterior tubules lie more posteriorly within the abdomen (Putz, 2005).
Since Rols6 is expressed in the Malpighian tubules (MpTs) throughout their development, the role of Rols6 in the generation of this tissue was investigated. For this purpose, a rols6-specific mutant was generated, in which the majority of the putative promoter region of rols6 was deleted, and thereby rols6 transcription was knocked out, while rols7 expression persisted as in the wild type. In this rols6-specific mutant, the early phase of organogenesis is the same as in wild type, i.e. the MpTs consist of two cell types, the principal cells (PCs) and the SCs. As the SCs originate from the mesoderm, one might expect that they would be affected in rols mutants. However, in the specific rols6 mutation generated, the SCs are able to migrate and integrate between the PCs as observed in the wild type. However, the PCs and SCs do not arrange correctly, and therefore, the typical MpT arrangement as found in wild-type embryo is not observed for stage 15 embryos onwards. The anterior tubules often show abnormal curves and lasso-like structures and fail to extend through the abdominal cavity. These navigation defects might well result from incorrect cell rearrangements, indicated by thickened regions of the tubules, whereas other parts seem to have a typical wild-type organisation (Putz, 2005).
Evidence is presented that correct cell rearrangement is dependent on Rols6 and proteins such as Mbc and Rac. These factors have been proposed to act with Rols7 in a common signalling cascade during myoblast fusion. An additional defect is the disorientation of MpTs in the body cavity, which again is characteristic for rols6, mbc and rac mutants (Putz, 2005).
Homozygous rols6 mutants are viable, indicating that the physiological functions of principal cells and stellate cells are largely unaffected. Loss of rols6 expression only moderately affects embryonic viability. Furthermore, homozygous EP(3)3330*5a flies do not die prematurely in contrast to those lacking another gene essential for MpT formation, hibris. The strongest hibris allelic combination die early as adults. Also, in contrast to rols6 mutants, in hibris mutants, the number of SCs is strongly reduced. This might cause defects in excretory function of the tubules, and thus leads to the observed lethality (Putz, 2005).
rols6-specific mutants show no distortion in rols7 transcription and in muscle development indicating that rols6 is specific for MpT development, while rols7 is essential for myogenesis. This is consistent with the observatio that Rols6 is not able to rescue the myogenic defect in rols mutants (Putz, 2005).
myoblast city mutants, rolling pebbles mutants and rac1/rac2 double mutants show late defects in Malpighian tubule differentiation. mbc mutants exhibit a MpT phenotype and it is proposed that this might be due to a failure to complete cell rearrangement; a phenomenon which is more apparent in mbc mutants than for rolling pebbles ones. Mbc, the homologue of vertebrate DOCK180 in Drosophila, associates with the adapter protein Crk. This interaction regulates cell migration and cytoskeleton organisation in a Rac-dependent manner. This agrees with the finding that rac1/rac2 double mutants exhibit the characteristic MpT defects as rols6 and mbc mutants do. Rols7 and Duf have been shown to interact in myogenesis. The strong similarity between the Rols proteins and their proposed functions leads to the hypothesis that Rols6 interacts with a so far unknown partner in the PCs. It is proposed that Rols6 initiates a signalling cascade via Mbc and Rac that leads to the correct rearrangement of cells, presumedly by rearranging the cytoskeleton, as has been proposed for Rols7 in the myogenic precursor cells. In the development of the somatic musculature, rearrangement of cytoskeleton is mediated by Blown fuse (Blow) and Kette in the second fusion wave (Putz, 2005).
Individual factors and protein complexes involved in cell migration and cytoskeleton arrangement have been described from many model organisms as well as from cell culture experiments. DOCK180/CED-5, the homologues of Drosophila Myoblast city (Mbc) in vertebrates and in C. elegans, form a complex with ELMO1/CED-12 that functions as a guanine nucleotide exchange factor (GEF). This functional GEF promotes Rac activation, and thus facilitates cell migration and rearrangement of the cytoskeleton. In vertebrates, additional protein complexes are built via DOCK180/p130Cas/Crk interaction and regulate cell migration and cytoskeletal organisation in a Rac-dependent manner. From kidney cells of human and mouse, the signalling molecule NEPHRIN is known to be of major importance in the podocyte for slit-diaphragm formation. Mutations in the nephrin gene are the major cause of congenital nephrotic syndrome in humans. In Drosophila, the homologue of vertebrate Nephrin, Hibris (hbs), is expressed during MpT development specifically in SCs. Therefore, it is likely that during MpT differentiation, Hibris mediates cell adhesion and arrangement between the PCs and the SCs, a mechanism comparable to myogenesis. In vertebrates, CMS/CD2AP has been identified as an interaction partner for Nephrin. The CMS/CD2AP homologue in Drosophila can be detected in silico as CG11316. CD2AP knock-out mice die due to kidney failure. Moreover, the Nephrin/CD2AP complex is able to bind to actin and to p130Cas (corresponding to CG1212). In Drosophila, homologues have been identified for all the above-mentioned factors involved in these protein complexes. However, little is known about their role in the developmental processes taking place during MpT development (Putz, 2005).
In Drosophila, a group of immunoglobulin-like proteins act in cell-cell recognition and attraction during myogenesis. These processes are also of importance in MpT development. Rolling pebbles is a multidomain and adapter-like protein. It is proposed that Rols6 interacts in Malpighian tubule development with proteins also involved in myogenesis such as Mbc and Rac. It is assumed that Rolling pebbles interacts with Mbc, and thus activates Rac. This hypothesis is supported by the observations that mbc and rac mutants exhibit defects in MpT development which might be linked to cell organisation in this tissue (Putz, 2005).
The mechanisms underlying the stereotypic course of the MpTs through the body cavity are still unclear. However, studies of phenotypes of early determination mutants like numb show that the tip cell and its sibling might both play a critical role in controlling the spatial arrangement of the growing tubules. This is indicated by the MpT phenotypes of numb mutants and UAS-numb embryos, where numb is overexpressed. These embryos lack either the tip cell or the sibling cell but form elongated MpTs with normally rearranged PCs. Although the PCs rearrange normally in these numb alleles, the MpTs are misrouted through the body cavity, as has been observed for rols and rac mutants. This raises the question whether determination of the tip cells is affected in rols mutants (Putz, 2005).
Essential transcription factors for Tip cell determination and PC cell proliferation are the A-SC, Krüppel and Seven up. These factors could be required for rols6 expression in the MpTs. However, since rols6 is expressed in the rudimentary primordia of MpTs in Krüppel mutants and in seven up mutants, this is unlikely. It is assumed, therefore, that Rolling pebbles is not a signalling molecule involved in cell specification through direct regulation of early genes, but rather that it plays a role as an adapter molecule in a protein complex connecting the cells in the tissue as Rols7 does in myogenesis. Since rols6, mbc and rac mutant embryos exhibit the described MpT phenotype, it is likely that they belong to a group of genes that can be helpful in discovering the mechanisms in MpT development that lead to the typical thin tubule morphology through cell rearrangement (Putz, 2005).
Rhabdomyosarcoma (RMS - see Drosophila as a Model for Human Diseases: Rhabdomyosarcoma) is a malignancy of muscle myoblasts, which fail to exit the cell cycle, resist terminal differentiation, and are blocked from fusing into syncytial skeletal muscle. In some patients, RMS is caused by a translocation that generates the fusion oncoprotein PAX-FOXO1, but the underlying RMS pathogenetic mechanisms that impede differentiation and promote neoplastic transformation remain unclear. Using a Drosophila model of PAX-FOXO1–mediated transformation, this study shows that mutation in the myoblast fusion gene rolling pebbles (rols) dominantly suppresses PAX-FOXO1 lethality. Further analysis indicates that PAX-FOXO1 expression causes upregulation of rols, which suggests that Rols acts downstream of PAX-FOXO1. In mammalian myoblasts, gene silencing of Tanc1, an ortholog of rols, reveals that it is essential for myoblast fusion, but is dispensable for terminal differentiation. Misexpression of PAX-FOXO1 in myoblasts upregulates Tanc1 and blocks differentiation, whereas subsequent reduction of Tanc1 expression to native levels by RNAi restorrs both fusion and differentiation. Furthermore, decreasing human TANC1 gene expression causes RMS cancer cells to lose their neoplastic state, undergo fusion, and form differentiated syncytial muscle. Taken together, these findings identify misregulated myoblast fusion caused by ectopic TANC1 expression as a RMS neoplasia mechanism and suggest fusion molecules as candidates for targeted RMS therapy (Avirneni-Vadlamudi, 2012).
This study uses a Drosophila chromosomal deletion, Df(3L)vin5, that dominantly suppresses PAX7-FOXO1–induced lethality. Human PAX7 demonstrates slightly higher sequence identity to Drosophila PAX3/7 than does human PAX3 and is therefore used in flies in this study. Df(3L)vin5 deletes segments 68A2–69A1 on chromosome 3, which includes the muscle-patterning gene rols, located at 68F1. rols encodes an essential adaptor molecule that links the Kirre transmembrane receptor with the machinery that drives myoblast cell-cell fusion and syncytial muscle formation; therefore, rols expression in the somatic mesoderm temporally coincides precisely with embryonic myoblast fusion. However, it was found by mRNA expression profiling that rols is misexpressed in PAX7-FOXO1 larval muscle. Thus, the study hypothesizes that heterozygous deletion of the rols locus might account for Df(3L)vin5-mediated PAX7-FOXO1 suppression and that rols might act as a PAX7-FOXO1 target gene (Avirneni-Vadlamudi, 2012).
Of the 2 alternative transcripts expressed from the rols locus, only one of which is expressed in myoblasts; expression of the second is restricted to endodermal/ectodermal precursors. In this study, 2 rols homozygous-lethal, P-element insertion loss-of-function alleles, P1027 and P1729, were tested for suppression of PAX7-FOXO1. Of these 2 alleles, only the P1729 insertion disrupts expression of the myoblast rols transcript (myoblast expression of rols is unperturbed in P1027); accordingly, only the rolsP1729 allele suppresses PAX7-FOXO1–induced lethality and muscle pathogenicity (Avirneni-Vadlamudi, 2012).
To investigate whether rols acts as a downstream PAX-FOXO1 target, the daughterless-Gal4 transgene was used to drive ubiquitous embryonic expression of UAS-PAX7-FOXO1 and probed for Rols misexpression. Since native Rols expression initiates at embryonic stage 11, the study focused only on embryos stage 10 or earlier. Diffuse expression of PAX7-FOXO1 and Rols is observed in blastoderm (stage 4–5) embryos, which consist of uncommitted precursor cells, and expression persists in all examined cells — including nonmyogenic ectodermal and endodermal cells — of gastrulated (stage 9–10) embryos. Taken together, these Drosophila studies reveal that rols acts as a PAX7-FOXO1 downstream target gene, direct or indirect, and as a bona fide genetic effector (Avirneni-Vadlamudi, 2012).
RMS model systems conveniently promote insights into not only neoplasia, but also muscle development. Although ultrastructural studies suggest that myoblast fusion biology is conserved, few of the Drosophila fusigenic genes have been identified as essential in mammals, and none of these are from the founder subfamily. As the name implies, founder myoblasts are seminal to Drosophila myogenesis, uniquely dictating the location and physiology of each individual muscle. With rols and Tanc1, it was shown that founder gene function is conserved in mammals and, furthermore, participates in human disease. How founder gene activity influences other forms of neuromuscular disease now becomes an intriguing issue (Avirneni-Vadlamudi, 2012).
Genetic screening in a Drosophila model and loss-of-function/gain-of-function studies in mammalian platforms have collaboratively uncovered a PAX-FOXO1-to-TANC1 neoplasia axis, a finding that the study suggests to be novel. Results from this study also argue that the relationship between myogenesis transcription factor (e.g., MyoD) signaling and myoblast fusion genes is intricate. In the presence of altered fusion potential, both Drosophila and mammalian myoblasts transition to differentiated myocytes, which suggests that later aspects of myogenesis signaling must uncouple from the TANC1 fusigenic pathway. Yet correcting PAX-FOXO1–mediated overexpression of rols/TANC1 rescues PAX-FOXO1–induced differentiation and arrest. These results intimate that correction of the TANC1 fusigenic axis feeds back to and rescues PAX-FOXO1–mediated misregulation of myogenic signaling, raising fascinating questions regarding the mechanisms by which this occurs. The observation in this study that PAX3-FOXO1 protein levels remain unchanged in TANC1-silenced cells argues that rescue does not originate from decreased expression of PAX3-FOXO1 from the PAX3 promoter. Thus, the study speculates that rescue occurs epistatically downstream of PAX3-FOXO1 (Avirneni-Vadlamudi, 2012).
Avirneni-Vadlamudi, U., Galindo, K.A., Endicott, T.R., Paulson, V., Cameron, S. and Galindo, R.L. (2012). Drosophila and mammalian models uncover a role for the myoblast fusion gene TANC1 in rhabdomyosarcoma. J Clin Invest 122: 403-407. PubMed
Bour, B. A., Chakravarti, M., West, J. M. and Abmayr, S. M. (2000). Drosophila Sns, a member of the immunoglobulin superfamily that is essential for myoblast fusion. Genes Dev. 14: 1498-1511. 10859168
Bulchand, S., Menon, S. D., George, S. E. and Chia, W. (2010). The intracellular domain of Dumbfounded affects myoblast fusion efficiency and interacts with Rolling pebbles and Loner. PLoS One 5(2): e9374. PubMed Citation: 20186342
Centner, T., et al. (2001). Identification of muscle specific ring finger proteins as potential regulators of the titin kinase domain. J. Mol. Biol. 306: 717-726. 11243782
Chen, E. H. and Olson, E. N. (2001). Antisocial, an intracellular adaptor protein, is required for myoblast fusion in Drosophila. Dev. Cell 1: 705-715. 11709190
Doberstein, S. K., Fetter, R. D., Mehta, A. Y. and Goodman, C. S. (1997). Genetic analysis of myoblast fusion: blown fuse is required for progression beyond the prefusion complex. J. Cell Biol. 136: 1249-1261. 9087441
Frasch, M. and Leptin, M. (2000). Mergers and acquisitions: unequal partnerships in Drosophila myoblast fusion. Cell 102: 127-129. 10943831
Kocherlakota, K. S., et al. (2008), Analysis of the cell adhesion molecule sticks-and-stones reveals multiple redundant functional domains, protein-interaction motifs and phosphorylated tyrosines that direct myoblast fusion in Drosophila melanogaster. Genetics 178: 1371-1383. PubMed Citation: 18245830
Kreiskother, N., Reichert, N., Buttgereit, D., Hertenstein, A., Fischbach, K. F. and Renkawitz-Pohl, R. (2006). Drosophila rolling pebbles colocalises and putatively interacts with alpha-Actinin and the Sls isoform Zormin in the Z-discs of the sarcomere and with Dumbfounded/Kirre, alpha-Actinin and Zormin in the terminal Z-discs. J. Muscle Res. Cell Motil. 27(1): 93-106. 16699917
Menon, S. D. and Chia, W. (2001). Drosophila Rolling pebbles: A multidomain protein required for myoblast fusion that recruits D-Titin in response to the myoblast attractant Dumbfounded. Dev. Cell 1: 691-703. 11709189
Putz, M., et al. (2005). In Drosophila melanogaster, the Rolling pebbles isoform 6 (Rols6) is essential for proper Malpighian tubule morphology. Mech. Dev. 122(11): 1206-17. 16169193
Rau, A., et al. (2001). rolling pebbles (rols) is required in Drosophila muscle precursors for recruitment of myoblasts for fusion. Development 128: 5061-5073. 11748142
Rúiz-Gomez, M., Coutts, N., Price, A., Taylor, M. V. and Bate, M. (2000) Drosophila dumbfounded: a myoblast attractant essential for fusion. Cell 102: 189-198. 10943839
Spencer, J. A., Eliazer, S., Ilaria, R. L., Richardson, J. A. and Olson E. N. (2000) Regulation of microtubule dynamics and myogenic differentiation by MURF, a striated muscle RING-finger protein. J. Cell Biol. 150: 771-784. 10953002
Suzuki, T., et al. (2004). A novel scaffold protein, TANC, possibly a rat homolog of Drosophila Rolling pebbles (Rols), forms a multiprotein complex with various postsynaptic density proteins. Eur. J. Neurosci. 21(2): 339-50. 15673434
date revised: 20 July 2012
Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.