The Interactive Fly

Genes involved in tissue and organ development

Mesoderm and Muscle

Mesoderm migration in Drosophila is a multi-step process requiring FGF signaling and integrin activity
Gene expression during mesoderm development
A machine learning approach for identifying novel cell type-specific transcriptional regulators of myogenesis
Restricted gene expression patterns in somatic muscles
Notch and Ras signaling pathway effector genes expressed in fusion competent and founder cells during Drosophila myogenesis
Motoneurons regulate myoblast proliferation and patterning in Drosophila
Genetic control of the distinction between fat body and gonadal mesoderm
A nutrient sensor mechanism controls Drosophila growth
Role and regulation of starvation-induced autophagy in the Drosophila fat body
The two origins of hemocytes in Drosophila
The Drosophila lymph gland as a developmental model of hematopoiesis
Subdivision and developmental fate of the head mesoderm in Drosophila
Mononuclear muscle cells in Drosophila ovaries revealed by GFP protein traps
Variation in mesoderm specification across drosophilids is compensated by different rates of myoblast fusion during body wall musculature development
Founder cells regulate fiber number but not fiber formation during adult myogenesis in Drosophila
The bHLH transcription factor Hand is required for proper wing heart formation in Drosophila

Genes expressed in mesoderm and mesodermal derivatives including muscle

Please note: c, s, or v = expression in cardiac, somatic or visceral musculature; h = expression in head mesoderm, d = expression in dorsal medial cells, f = expression in fat body, h = expression in hemocytes

Mesoderm migration in Drosophila is a multi-step process requiring FGF signaling and integrin activity

Migration is a complex, dynamic process that has largely been studied using qualitative or static approaches. As technology has improved, it is now possible to take quantitative approaches towards understanding cell migration using in vivo imaging and tracking analyses. In this manner, a four-step model of mesoderm migration during Drosophila gastrulation was establised: (I) mesodermal tube formation, (II) collapse of the mesoderm, (III) dorsal migration and spreading and (IV) monolayer formation. The data provide evidence that these steps are temporally distinct and that each might require different chemical inputs. To support this, the role was analyzed of fibroblast growth factor (FGF) signaling, in particular the function of two Drosophila FGF ligands, Pyramus and Thisbe, during mesoderm migration. It was determined that FGF signaling through both ligands controls movements in the radial direction. Thisbe is required for the initial collapse of the mesoderm onto the ectoderm, whereas both Pyramus and Thisbe are required for monolayer formation. In addition, it was uncovered that the GTPase Rap1 regulates radial movement of cells and localization of the beta-integrin subunit, Myospheroid, which is also required for monolayer formation. These analyses suggest that distinct signals influence particular movements, since it was found that FGF signaling is involved in controlling collapse and monolayer formation but not dorsal movement, whereas integrins are required to support monolayer formation only and not earlier movements. This work demonstrates that complex cell migration is not necessarily a fluid process, but suggests instead that different types of movements are directed by distinct inputs in a stepwise manner (McMahon, 2010).

Mesoderm migration was found to be a combination of complex three-dimensional movements involving many molecular components. live imaging, coupled with quantitative analyses, is important for studying complex cell movements, as it allowed migration to be decomposed into different movement types and thus has allowed description of subtle phenotypes. First, analysis of the directional movements of mesoderm cells within wild-type embryos was extended, focusing on the temporal sequences of events. Cells were found follow a sequential and distinct set of trajectories: movement in the radial direction (tube collapse: -5 to 15 minutes, 0=onset of germband elongation), followed by movement in the angular direction (dorsal migration: 15 to 75 minutes) and ending with small intercalation movements in the radial direction (monolayer formation: 75 to 110 minutes). These movements appear temporally distinct (i.e. stepwise), and thus molecular signals controlling each process were sought (McMahon, 2010).

Which mesoderm movements were FGF-dependent were investigated and, in particular, either Ths- or Pyr-dependent. The interaction between Htl and its two ligands provides a simpler system relative to vertebrates (which exhibit over 120 receptor-ligand interactions) in which to study how and why multiple FGF ligands interact with the same receptor. Previously, it was found that FGF signaling via the Htl FGFR controls collapse of the mesodermal tube but not dorsal-directed spreading (McMahon, 2008). This study demonstrated that FGF signaling is also required for monolayer formation. In addition, distinct, non-redundant roles were defined for the FGF ligands: Ths (but not Pyr) is required for collapse of the mesodermal tube, whereas both Pyr and Ths are required for proper intercalation of mesoderm cells after dorsal spreading (McMahon, 2010).

This analysis raises questions about ligand choice during collapse and monolayer formation. Within the mesodermal tube, cells at the top require a long-range signal in order to orient towards the ectoderm during tube collapse, whereas the signals controlling intercalation during monolayer formation can be of shorter range. It is suggested that the ligands have different activities that are appropriately tuned for these processes. In fact, recent studies of the functional domains of these proteins suggest that Ths has a longer range of action than Pyr, in agreement with the analysis that Pyr does not support tube collapse, but does have a hand in monolayer formation (McMahon, 2010).

This study has demonstrated that Rap1 mutants have a similar mesoderm phenotype to the FGFR htl mutant, with defects in collapse and monolayer formation. It was not possible to establish whether Rap1 acts downstream of FGF signaling, as the complete loss of Mys in Rap1 mutants is more severe than the patchy expression of Mys seen in htl mutants. Therefore, Rap1 could be working in parallel to or downstream of FGF signaling during mesoderm migration. Rap1 has been implicated in several morphogenetic events during Drosophila gastrulation and probably interacts with many different signaling pathways. Further study of Rap1, along with other GTPases, will shed light onto their role during mesoderm migration, how they interact with one another and what signaling pathways control them (McMahon, 2010).

Focus was placed on the more specific phenotype of mys mutants, as its localization is affected in htl mutants and it exhibits a monolayer defect that is similar to pyr and ths mutants. Integrins are important for cell adhesion, so it is not surprising that cells fail to make stable contact with the ectoderm through intercalation in mys mutants. However, some cells do contribute to monolayer formation in the absence of Mys, implying that other adhesion molecules are involved in maintaining contact between the mesoderm and ectoderm. These other adhesion molecules might be activated downstream of FGF signaling as the htl mutant monolayer phenotype is more severe than the mys mutant. Discovering the downstream targets of Htl, which might regulate cell adhesion properties, will help to shed light on the mechanisms supporting collapse of the mesodermal tube (which is not dependent on Mys) and monolayer formation (which is Mys-dependent) (McMahon, 2010).

Cell protrusions, such as filopodia, are important for sensing chemoattractants and polarizing movement during migration. Previous studies have focused on protrusive activity at the leading edge during mesoderm migration in Drosophila and shown that these protrusions are FGF-dependent. In this study, it was found that protrusions exist in all mesoderm cells, not just the leading edge, and that these protrusions also extend into the ectoderm (McMahon, 2010).

The study demonstrates that FGF signaling, as well as integrin activity, is required to support protrusive activity into the ectoderm; this is a potential mechanism by which FGF signaling and Mys could control movement toward the ectoderm during monolayer formation. The function of protrusions at the leading edge remains unclear, as they appear to be reduced in pyr and mys mutants, but migration in the dorsal direction still occurs in both mutant backgrounds. One interpretation is that FGF and Mys are important for generalized protrusive activity and that extensive protrusions are required for intercalation but not dorsal migration (McMahon, 2010).

Based on this study, it is proposed that mesoderm migration is a stepwise process, with each event requiring different molecular cues to achieve collective migration. Invagination of the mesoderm is the first step in this process and is dependent on Snail, Twist, Concertina, Fog and several other genes. Next, collapse of the mesoderm tube onto the ectoderm requires Htl activation via Ths. Rap1 might be involved in this process as well but the phenotype of Rap1 mutants is complex and it is unclear which phenotypes are primary defects (McMahon, 2010).

Following collapse, mesoderm cells spread dorsally by an unknown mechanism. Dorsal migration is unaffected in pyr and ths mutants and occurs in all cells that contact the ectoderm in htl mutants, implying that FGF signaling is, at most, indirectly involved in this step owing to the earlier tube collapse defect (McMahon, 2008). Whether dorsal migration requires chemoattractive signals or whether the cells simply move in this direction because it is the area of least resistance remains unclear (McMahon, 2010).

Finally, after dorsal spreading is complete, any remaining cells not contacting the ectoderm intercalate to form a monolayer. This process is controlled by a combination of both Pyr and Ths interacting through Htl and also by Rap1 and Mys. In other systems, intercalation can lead to changes in the properties of the cell collective, for instance, lengthening of a body plan. However, this study has shown that dorsal migration and spreading are not a result of intercalation, as intercalation occurs after spreading has finished (McMahon, 2010).

Coordination of these signals to control collective migration enables the mesoderm to form a symmetrical structure, which is essential for embryo survival. This model begins to address the question of how hundreds of cells move in concerted fashion and is relevant for a generalized understanding of embryogenesis and organogenesis. It was found that mesoderm migration is accomplished through sequential movements in different directions, implying that collective migration might be best achieved by distinct phases of movement (McMahon, 2010).

Gene expression during mesoderm development

The transcription factor Twist initiates Drosophila mesoderm development, resulting in the formation of heart, somatic muscle, and other cell types. Using a Drosophila embryo sorter, enough homozygous twist mutant embryos were isolated to perform DNA microarray experiments. Transcription profiles of twist loss-of-function embryos, embryos with ubiquitous twist expression, and wild-type embryos were compared at different developmental stages. The results implicate hundreds of genes, many with vertebrate homologs, in stage-specific processes in mesoderm development. One such gene, gleeful/lame duck, related to the vertebrate Gli genes, is essential for somatic muscle development and is sufficient to cause neural cells to express a muscle marker (Furlong, 2001).

Formation of muscles during embryonic development is a complex process that requires coordinate actions of many genes. Somatic, visceral, and heart muscle are all derived from mesoderm progenitor cells. The Drosophila twist gene, which encodes a bHLH transcription factor, is essential for multiple steps of mesoderm development: invagination of mesoderm precursors during gastrulation, segmentation, and specification of muscle types. The role of twist in mesoderm development has been conserved during evolution, perhaps because it controls conserved regulatory mesoderm genes. For example, tinman and dMef2 are regulated by Twist in flies and are highly conserved in sequence and function in vertebrates (Furlong, 2001).

In Drosophila, somatic muscle forms from progenitor cells that divide to become muscle founder cells. Founder cells acquire unique identities controlled by transcription factors including Krüppel, S59, vestigial, and apterous. Each of the 30 body wall muscles in an abdominal hemisegment is initiated by a single founder cell and has unique attachments and innervations. To further clarify mechanisms underlying founder cell specification, myoblast fusion, and muscle patterning, Drosophila mutants together with microarrays of cDNA clones were used (Furlong, 2001).

twist mutant embryos develop no mesoderm. A population of mRNA species isolated from twist homozygous embryos was compared with that of stage-matched wild-type embryos. Drosophila lethal mutations are maintained as heterozygotes, in trans to balancer chromosomes. A twist mutation was established in trans to a balancer chromosome carrying a transgene encoding green fluorescent protein (GFP). Embryos were collected from wild-type and twist/GFP-balancer fly stocks at specific developmental stages. The twist/GFP-balancer collections contain a mixed population of embryos: one-quarter twist homozygotes lacking GFP, half heterozygotes with one copy of GFP, and one-quarter homozygous for the balancer chromosome with two copies of GFP. Homozygous twist mutant embryos were separated from their siblings using an embryo sorter. Putative homozygous twist embryos were assessed by immunostaining with an antibody to dMef2. More than 99% of the selected embryos had the twist phenotype (Furlong, 2001).

Three different periods of mesoderm development were analyzed: stages 9-10, 11, and 11-12 . During stage 9-10, the earliest time GFP is detectable in the balancer embryos, mesoderm cells migrate dorsally and become specified as somatic, visceral, cardiac, and fat body mesoderm. twist and its direct targets tinman and dMef2 are expressed throughout stage 9-10 mesoderm. The middle period contains stage 11 embryos and is a transition between the first period (stage 9-10) and the third period (late stage 11-12). During late stage 11 and stage 12, myoblast fusion begins and twist expression remains prominent in only a subset of the somatic muscle cells (Furlong, 2001).

For each developmental period, three independent embryo collections, embryo sortings, and microarray hybridizations were conducted. The microarrays used for the analysis contained over 8500 cDNAs corresponding to 5081 unique genes plus a variety of controls. Each embryonic RNA sample was compared with a reference sample, containing RNA made from all stages of the Drosophila life cycle to allow direct comparisons among all the experiments (Furlong, 2001).

To determine how transcription was affected by the twist mutation, SAM (significance analysis of microarrays) analysis was used. Genes that are normally highly expressed in mesoderm should have lower transcript levels in twist homozygotes. Genes in other tissues whose expression depends on signals from the mesoderm might also have reduced expression. Transcripts of 130 genes, the 'Twist-low' group, were significantly lower in twist mutants than in wild type. Conversely, cells that would have formed mesoderm may take on other fates in the absence of twist, such as neuroectoderm; therefore, many transcript levels could increase in twist mutants. Genes whose transcription is repressed by signals from the mesoderm would also be enriched in twist mutants. One hundred fifty genes, called the 'Twist-high' group, have increased levels of RNA in twist mutant embryos (Furlong, 2001).

In total, 280 of ~5000 genes had significant changes in transcript levels, with 10 false positives. The genes on the array include 15 previously characterized mesoderm-specific genes, all of which are significantly reduced in twist mutant embryos. The arrays also contain genes known to be transcribed in both mesoderm and other cell types. Significant changes in expression were detected for many of these genes (Furlong, 2001).

The 130 Twist-low genes were divided into three groups with similar trends of expression by a self-organizing map (SOM) clustering program. The 24 group A genes, which include tinman, dMef2, and bagpipe, have reduced transcript levels in twist mutants at all developmental stages assayed. Most of the Twist-low genes fall into the B and C groups. The 62 group B 'early genes' encode transcripts with reduced levels of expression in twist mutants only during stages 9-10, not later. One member of group B, stumps (dof/hbr) is essential for mesoderm cell migration. stumps RNA is abundant in the mesoderm at stages 9-10 and is strongly reduced by stage 11. At stage 11, stumps RNA accumulates in trachea, which is largely unaffected in twist mutants (Furlong, 2001).

The 44 group C genes have reduced transcript levels in twist mutant embryos only during late stage 11 and stage 12. These 'lategenes' include blown fuse, a gene essential for myoblast fusion; delilah, a gene required for somatic muscle attachment, and genes such as kettin, which is required to form contractile muscle. Given the predominantly early expression of twist, the early genes in groups A and B are the best candidates for direct transcription targets of Twist, though some indirectly activated genes may be present within these groups. Group C late genes are likely to be regulated by products of genes that are activated by Twist (Furlong, 2001).

In situ hybridizations were done using a previously uncharacterized representative of each Twist-low group. In each case, the hybridization pattern was consistent with the predicted time of transcription. A group A gene, CG15015 (GH16741), is transcribed in somatic muscle throughout stages 9-12. A group B gene, CG12177 (GH22706), is transcribed during early mesoderm development, but not later. CG14848 (GH21860), a group C gene, is expressed in the stomodeum but not the mesoderm during stages 9-10. Its mesoderm expression initiates during stage 11, the latest period of the twist experiment. Thus, combining loss-of-function mutant embryo analysis with staged embryo collections provides gene expression information for both tissue specificity and temporal expression (Furlong, 2001).

The mis-expression of twist in the ectoderm is sufficient to convert both neuronal and epidermal tissues to a myogenic cell fate. RNA from embryos with ubiquitous twist expression was used to evaluate the ability of Twist to initiate mesoderm-like gene expression in cells that would normally form other tissue types. Genes whose transcript levels decrease in twist loss-of-function embryos and increase when twist is ubiquitous are excellent candidates for regulators of mesoderm development or differentiation (Furlong, 2001).

To ectopically express twist, a dominant gain-of-function mutation of the maternal gene Toll (Toll10B) was used. Activated Toll induces the expression of twist and snail in early embryos and of immune response genes in older embryos. Thus, the effects of Toll10B on gene expression reflect the activities of Twist as well as Snail and Dorsal, or their combined actions. Toll10B embryos are essentially bags of mesoderm; epidermal structures are absent or greatly reduced, and they have been used successfully in subtractive hybridization screens to identify mesoderm genes. The gene transcription profile of Toll10B embryos was compared with that of wild-type embryos during four periods of development, using the reference sample to normalize experiments. The earliest period, stage 5, is when twist is initially expressed in presumptive mesoderm. The other three periods analyzed were those used in the twist mutant analysis: stages 9-10, 11, and 11-12 (Furlong, 2001).

In Toll10B embryos, 447 genes had significant changes in RNA levels compared with stage-matched wild-type embryos, 16 of which are predicted to be false positives. Transcripts from 166 genes were reduced in Toll10B embryos compared with wild type. These genes may be involved in neuroectoderm events that are blocked when cells are turned into mesoderm. Transcripts of 281 'Toll-high' genes were increased in Toll10B embryos. Of the 21 previously characterized mesoderm-specific genes on the arrays, 18 have significantly increased transcript levels in Toll10B embryos. The remainder may require activators other than Toll, such as signals from the severely altered ectoderm (Furlong, 2001).

Genes with altered transcription in twist and Toll10B mutants were analyzed with a hierarchical clustering program to identify similar transcription profiles. The genes were divided into putative 'mesoderm' and 'non-mesoderm' groups. Non-mesoderm genes were defined as having increased transcript levels in mesoderm-deficient (twist) embryos and/or decreased expression in mesoderm-enriched (Toll10B) embryos. Mesoderm genes were defined as having decreased transcript levels in twist mutants and/or increased transcripts in Toll10B mutants. The mesoderm genes group would also contain genes expressed in other tissues in a mesoderm-dependent manner (Furlong, 2001).

Non-mesoderm genes in clusters B and C are repressed in Toll10B mutants. Cluster B genes have increased RNA levels in twist mutant embryos, whereas cluster C genes do not change significantly. The overexpression of twist in the presumptive ectoderm in Toll10B embryos results in a conversion of ectodermal cell fate into mesoderm. snail and dorsal are ectopically expressed in Toll10B embryos and transcriptionally repress the expression of neuroectoderm and ectoderm genes. The conversion of ectoderm to mesoderm due to twist misexpression, and the ability of Snail and Dorsal to repress ectoderm genes, suggests that the B and C clusters should contain primarily neuroectodermal genes. Indeed, the non-mesoderm genes include 31 previously characterized neuroectodermal genes. One previously unknown cluster B gene that encodes a putative cell adhesion protein is transcribed in the ventral nerve cord. Another previously unknown gene within cluster C is transcribed within the developing brain (Furlong, 2001).

The Twist-low and Toll-high genes have in common 51 genes that are highly likely to be involved mesoderm development. For example, transcription of the genes in cluster D is reduced in twist mutants and increased in Toll10B mutants during most or all time periods (Furlong, 2001).

A complete overlap between Twist-low and Toll-high gene sets is not expected for three reasons: (1) development of dorsal mesoderm, and of muscle founder cells marked by apterous and connectin, requires the Decapentaplegic signal and perhaps others. Changed characteristics of cells that form ectoderm in Toll10B embryos interfere with these signaling events. A significant reduction is observed in dpp RNA levels in Toll10B embryos. (2) During midgut development, endoderm cells migrate along the mesoderm. Midgut endoderm development is affected in twist mutants. Some Twist-low genes with unchanged expression in Toll10B embryos are transcribed in the midgut. (3) Ectopic Twist inhibits visceral mesoderm and heart development and promotes excess somatic muscle development. Toll10B embryos produce high levels of Twist throughout the embryo, so genes that have reduced RNA levels in both twist and Toll10B mutant embryos are likely to be visceral muscle and heart genes. Indeed, bagpipe and connectin, genes expressed in visceral mesoderm, are among the 79 Twist-low genes not induced by ectopic Twist (Furlong, 2001).

Of the 281 Toll-high genes, 230 are unaffected in twist mutants. Some of the 230 are normally expressed late in embryogenesis in wild-type embryos but are expressed prematurely in Toll10B embryos due to ectopic Twist. These include Myo61F, MSP-300, and Paramyosin, genes normally active in terminally differentiated muscle (stage 16). Ectopic Snail and Dorsal in Toll10B embryos may activate genes that are unaffected in twist mutants. Snail can repress neuroectodermal genes and may also activate mesoderm genes. Dorsal activates immune response genes later in development. relish, drosomycin, and metchnikowin genes -- all immuneresponse genes -- have higher transcript levels in Toll10B embryos (Furlong, 2001).

Data from loss- and gain-of-function experiments, combined with careful staging, yield a useful picture of genes that are likely to be required for mesoderm specification and muscle differentiation. Of 360 identified mesoderm genes, 273 have not been the focus of developmental studies. The predicted proteins encode transcription factors, signal transduction molecules, kinases, and pioneer proteins. The stage at which each gene is active is one criterion for assigning possible functions. Another key criterion will be finding a mutant phenotype. As a pilot, this additional step was undertaken for the gene CG4677 (LD47926). Changes in CG4677 transcript levels were also observed in a Toll10B subtractive hybridization screen (Furlong, 2001).

CG4677 is transcribed in the visceral mesoderm at stages 10-13 and the somatic mesoderm during stages 11-13. This gene encodes a C2H2 zinc finger transcription factor with high sequence similarity to vertebrate Gli proteins: the gene has been named gleeful (gfl). Mammalian Gli proteins act downstream of Hedgehog signaling proteins to control target gene transcription (Furlong, 2001).

The role of gfl in mesoderm development was assessed by disrupting its function using RNA interference. Injection of a double-stranded RNA (dsRNA) control sequence had no effect on mesoderm development. In contrast, gfl dsRNA injection causes severe loss and disorganization of somatic muscle cells, whereas heart and visceral muscle are unaffected. A similar phenotype is seen in Df(3R)hh homozygous embryos: the deficiency removes gfl but not the nearby hedgehog gene (Furlong, 2001).

To determine whether gfl can induce muscle cell development, a UAS-gfl transgenic fly strain was generated. Ectopic expression of gfl using an en-GAL4 driver results in lethality and induction of ectopic dMEF2 expression in the ventral nerve cord. Remarkably, Gfl is sufficient to induce expression of a muscle gene in neuronal cells. Previous studies have shown an essential role for Sonic hedgehog signaling in the formation of slow muscle in avian and zebrafish embryos. gfl may be performing a similar role in Drosophila somatic muscle development (Furlong, 2001).

A machine learning approach for identifying novel cell type-specific transcriptional regulators of myogenesis

Transcriptional enhancers integrate the contributions of multiple classes of transcription factors (TFs) to orchestrate the myriad spatio-temporal gene expression programs that occur during development. A molecular understanding of enhancers with similar activities requires the identification of both their unique and their shared sequence features. To address this problem, phylogenetic profiling was combined with a DNA-based enhancer sequence classifier that analyzes the TF binding sites (TFBSs) governing the transcription of a co-expressed gene set. A small number of enhancers were assembled that are active in Drosophila melanogaster muscle founder cells (FCs) and other mesodermal cell types. Using phylogenetic profiling, the number of enhancers was increased by incorporating orthologous but divergent sequences from other Drosophila species. Functional assays revealed that the diverged enhancer orthologs were active in largely similar patterns as their D. melanogaster counterparts, although there was extensive evolutionary shuffling of known TFBSs. A classifier using this enhancer set was then built and trained, and additional related enhancers were identified based on the presence or absence of known and putative TFBSs. Predicted FC enhancers were over-represented in proximity to known FC genes; and many of the TFBSs learned by the classifier were found to be critical for enhancer activity, including POU homeodomain, Myb, Ets, Forkhead, and T-box motifs. Empirical testing also revealed that the T-box TF encoded by org-1 is a previously uncharacterized regulator of muscle cell identity. Finally, extensive diversity was found in the composition of TFBSs within known FC enhancers, suggesting that motif combinatorics plays an essential role in the cellular specificity exhibited by such enhancers. In summary, machine learning combined with evolutionary sequence analysis is useful for recognizing novel TFBSs and for facilitating the identification of cognate TFs that coordinate cell type-specific developmental gene expression patterns (Busser, 2012).

There are three main approaches for the prediction of tissue-specific regulatory elements that are based on high-throughput sequencing coupled with chromatin immunoprecipitation (ChIP-Seq), DNA sequence pattern analysis, or hybrid methods that combine both of these strategies. ChIP-Seq for p300 using mouse embryonic tissue has proven to be an accurate means for identifying enhancers and their associated activities, with in vivo validation rates varying from 62% to 88%. Computational analysis of whole-genome histone modification profiles using hidden Markov models and machine learning techniques has also been highly successful at linking chromatin signatures with regulatory elements. Finally, computational models that identify tissue-specific enhancers relying on sequence motifs and linear regression and support vector machines have been similarly effective, with in vivo validation rates of de novo predictions ranging from 62% for heart enhancers to 91% for brain enhancers. Although experimental techniques are often preferred for identifying enhancers on a genome-wide scale, ChIP-Seq has several limitations. For example, ChIP-Seq experiments are typically carried out in only one species and for individual cell types, and are currently not sufficiently precise for low-quality genome sequences. Thus, de novo prediction of regulatory elements based on ChIP-Seq data critically depends on the availability of relevant data for the species, cell type and genomic regions of interest. Currently, computational analysis of DNA sequence patterns shared by a set of regulatory elements with the same or similar biological activity remains a highly effective method for the de novo discovery of tissue-specific enhancers, and the simultaneous elucidation of cell type-specific regulatory codes. The method presented in this study further extends the usefulness of computational sequence analysis by exploring phylogenetic information that can be used to improve the classification accuracy, a strategy that promises to be advantageous in the large number of cases where comparative genomics data are available (Busser, 2012).

Computational approaches for predicting cis-regulatory modules are commonly based on machine learning of arrangements of TFBSs in enhancers that have common functions. These methods rely heavily on a training set of related enhancers to detect over-represented TFBS combinations. Unfortunately, in the vast majority of cases -- including the present study of Drosophila muscle FC enhancers -- the size of the training set is limited by the lack of experimentally validated tissue- and cell type-specific enhancers, which results in overfitting of computational models and poor accuracy of predictions. To overcome this problem, and to provide a generalizable approach for increasing the size of the training set, a phylogenetic profiling strategy was developed based on a search for diverged orthologous counterparts of available enhancers from distantly related species. Twenty-four Drosophila orthologs were identified using this approach, which more than doubled the size of the training set. The ability to accurately distinguish FC enhancers was assessed in a cross-validation framework using the extended training set, and it was determined that the classifier accuracy is 89% as assessed by the AUC approach. This classifier was then developed to scan the entire genome of D. melanogaster for novel FC enhancers, retrieving 5,500 high-scoring predictions at a FPR of 5%. These predictions were significantly associated with genes expressed in FCs, demonstrating that the model was able to capture essential features of FC gene co-regulation. A similar machine learning approach could be applied to a diverse array of datasets, including experimentally-verified regulatory elements from co-expressed targets at either a germ layer, organ, tissue or cellular level from invertebrate and vertebrate databases. Alternatively, a similar approach could be coupled to a training set of predicted regulatory elements derived from genome-wide analyses of chromatin marks or DNAse hypersensitive sites in active enhancers associated with a co-expressed gene set (Busser, 2012).

Evolutionary constraint of functional sequences is routinely employed as an effective filter to improve the prediction of regulatory elements. Furthermore, cross-species comparisons have been successfully exploited to obtain evidence for functional TFBSs. For example, Rouault (2010) used twelve Drosophila species to identify over-represented motifs in the regulatory elements of genes expressed in neural progenitor cells, with sequence orthologs used to enrich the training set and to give prominence to conserved motifs. However, this method extends this approach by including suitably diverged orthologous enhancers from other Drosophila species in the dataset used to train the classifier. The purpose in designing this strategy was two-fold. First, it was desired to enrich for relevant sequence motifs in the training data, allowing for a level of variation that would improve the generalization of the model. Second, a potentially wider variety of TFBS arrangements would be provided that characterize the architecture of authentic FC enhancers. In essence, the addition of orthologous sequences boosts the statistical power of the significance tests, revealing patterns of TFBSs that otherwise could have been neglected (Busser, 2012).

Of note, when 5 of these orthologous sequences were tested in transgenic reporter assays in D. melanogaster, the overall expression pattern generated was similar to the D. melanogaster counterpart despite extensive evolutionary shuffling of known TFBSs. Similar binding site reorganization has been documented for the enhancers that regulate both the segmentation and mesodermal patterns of eve expression. Numerous other studies have shown that the order and spacing of TFBSs is critical for enhancer function. These results suggest that regulatory elements can direct similar expression patterns provided that the overall composition and order of collaborating TFs is maintained. The finding that enhancer function is preserved in the orthologous sequences examined in this study establishes the validity of the sequence conservation thresholds chosen for the present studies, and suggests that the incorporation of orthologous sequences to increase a training set without over-fitting the data will be a generally applicable approach (Busser, 2012).

To assess the accuracy of this method, 12 predicted FC enhancers were selected and their in vivo functions were tested. Seventy-five percent of the putative enhancers were experimentally validated as having transcriptional activity, demonstrating the effectiveness of this approach to identify regulatory sequences. However, of the sequences showing regulatory functions, only 4 of 9 were active in the mesoderm—including 2 in FCs—and 3 of 9 had nervous system activity. These data suggest that the model has been able to reliably recognize general properties of tissue-specific enhancers without specifically distinguishing an overall muscle FC code, even though numerous individual FC-specific motifs were identified. The former finding is similar to the results of Sinha and colleagues (Kantorovitz, 2009) who found that the majority of their classifier predictions were active enhancers, but only a minority were expressed in the predicted pattern. A number of confounding factors can explain this outcome (Busser, 2012).

First, most members of the enhancer training set are active in both FCs and other cell types, including additional mesodermal cells such as the cardiac and visceral mesoderm, as well as some cells of the nervous system. For example, the enhancer responsible for the FC activity of the hunchback gene is also active in the longitudinal visceral mesoderm, and enhancers directing the FC expression of the vestigial, big brain and king-tubby genes are also active in the peripheral nervous system. These results suggest that the regulatory networks specifying the somatic and visceral mesoderm share common features, which is consistent with both the available genetic and genomic evidence for the diverse developmental functions of key mesodermal transcription factors. Second, different members of a given TF family bind to similar motifs but have distinct tissue-specific expression patterns and developmental activities. Thus, combinations of motifs involved in the specification of muscle FCs and the nervous system may overlap. For example, this situation occurs with E-box and NK-homeodomain motifs. Third, some TFs are expressed and functional in the derivatives of more than one germ layer. Fourth, the sequence features characteristic of cell type-specific enhancers, such as those active in muscle FCs, are expected to be under-represented in available training sets owing to the diversity of combinatorial TF models required to specify such a heterogeneous cell type. Identification of many examples of a particular cell-specific signature is a major challenge since each of the approximately 30 FCs in each Drosophila hemisegement expresses a unique combination of cell-specific muscle identity TFs and downstream target genes. Thus, 30 distinct cell states exist, each governed by a different but partially overlapping set of regulatory TFs. In contrast to the difficulties involved in dissecting regulatory codes at single cell resolution, shared features that direct activity to the general level of tissues and organs have been more readily identified using a machine learning approach, as was found in this study for enhancers having mesodermal, although not necessarily FC, activity. This likely reflects the dominant role that some TFs play in the regulatory network specifying the identities of numerous tissues. Fifth, since there appears to be a regulatory signature for enhancers, it is likely that these aspects of enhancer structure will be more significantly over-represented than those features that specify individual FC activity patterns. Sixth, the use of phylogenetic profiling might have expanded the biological function of the training dataset by introducing additional enhancer functions acquired by the orthologs of the original D. melanogaster sequences during their evolution. While this study has been able to show that the phylogenetic profiling approach improves the accuracy of the classifier, one drawback of its use might be that the final classifier recognizes a broader biological domain than the function of the original training set of sequences derived from the reference species. Finally, classifier predictions may represent cis-regulatory elements other than enhancers, for example, silencers and insulators, which would not be detected by the transgenic reporter assays (Busser, 2012).

In summary, a number of confounding factors influenced the ability to identify an enhancer signature that is specific for individual muscle FCs. However, despite these challenges, the successful identification of novel TF binding motifs responsible for the cell type-specific activity of FC enhancers is encouraging the idea that this is a tractable problem that can be solved by an iterative approach to the computational analysis of this and other complex developmental systems. Thus, future studies must focus on obtaining a larger training set of sequences in which enhancers are categorized based on their activities at single cell resolution, combined with the appropriate weighting of newly validated motifs that contribute to the expression pattern of interest. In this manner, each experimental round would improve the accuracy of the classifier (Busser, 2012).

The motifs ranked by the classifier used in this study as having the highest discriminatory power are part of a large regulatory network that is known to be critical for mesoderm specification and myogenesis. These motifs include binding sites for JAK/STAT, Ets, bHLH, Wingless/Tcf, homeodomain and forkhead proteins. Furthermore, it has previously been suggested that Ets is part of a transcriptional code regulating the C1 subset of FC genes, which was validated in this study using site-directed mutational analysis of the Ndg enhancer, a previously characterized regulatory element associated with a C1 FC gene (Busser, 2012).

To extend the components of the myogenic regulatory network beyond these known TFs and motifs, the function was examined of the classifier-defined sequence motifs recognized by POU homeodomain and Myb proteins, transcription factors having no previously known role in Drosophila myogenesis. Mutagenesis of POUHD motifs attenuated the activity of the Ndg enhancer in many mesodermal cells. However, a zygotic loss-of-function mutation in acj6, the only POUHD that was found to be expressed in the mesoderm, had no effect on Ndg gene expression . Given the strong maternal contribution to this gene, RNAi was used to knock down both maternal and zygotic acj6 transcripts, but this manipulation had no effect on Ndg-GFP reporter activity. These findings leave unresolved the identity of the TF that binds to the motif in question. The future characterization of this TF, including exploring the possibility that it is not a POUHD protein, will require searching functional motifs against larger TF databases, combined with analysis of the embryonic expression and function of any new candidates that emerge (Busser, 2012).

Inactivating mutations of the Myb binding sites in the Ndg enhancer led to extensive de-repression of the reporter in other mesodermal cells. Myb is a ubiquitously-expressed DNA binding protein which plays a critical role in controlling regulatory decisions during proliferation and differentiation of progenitor cells. Identifying a putative role for Myb in myogenesis documents the power of this approach, since functional studies tend to focus on genes with restricted expression patterns. However, a definitive assessment requires examining the effect of loss-of-function mutations in Myb. In any event, as myogenesis in Drosophila occurs through a series of asymmetric and symmetric cell divisions, a role for Myb in regulating FC gene expression is entirely consistent with a transcriptional regulator acting at the interface between replication and transcription. Alternatively, Myb may cooperate with other TFs to activate cell or tissue-specific gene expression (Busser, 2012).

Interestingly, T-box motifs scored well in the classification, yet no role for T-box TFs has previously been described in Drosophila somatic muscle development, despite widespread functions of this TF class in mesoderm specification and myogenesis in vertebrates, as well as cardiogenesis in Drosophila and vertebrates. This study shows using both cis and trans tests of TF function, along with gene co-expression, that Optomotor-blind-related-gene-1 (Org-1) is a muscle identity TF. In particular, the cis effects of Org-1 were documented in the FC enhancers associated with two known muscle identity TFs, Slou and Lbl, and org-1 expression localizes to the SBM and VT1, muscles in which the lb genes and slou, respectively, are the only previously described determinants of muscle identity. Slou function is critical for the proper development of muscles LO1 and VT1 and is further required to repress the lb genes in these cells, suggesting a co-regulatory relationship between slou and lb. It is likely that org-1 acts upstream of slou and lb in this regulatory hierarchy since org-1 expression precedes slou and lb, and the ectopic expression of org-1 causes increased expression of slou and lb. In addition, the essential role of org-1 in this regulatory network is revealed by the effects of org-1 overexpression and RNAi knockdown on development of lb- and slou-expressing muscles. Interestingly, the mouse orthologs of org-1 and lb genes, Tbx1 and Lbx1, respectively, have been suggested to regulate myogenic differentiation in the limb. Given the high degree of sequence similarity, and the close correspondence of expression patterns and functions in Drosophila and mouse, the collaborative roles of these two TFs in myogenesis appear to have been conserved through evolution (Busser, 2012).

Computational prediction of regulatory elements requires a thorough understanding of the TFs and motifs that orchestrate gene co-expression patterns. In prior studies, it was established that 5-way and 3-way 'AND' combinations of 3 signal-activated (Tcf, Mad and Pnt) plus 2 tissue-restricted (Twi and Tin) TFs constitute distinct regulatory models for different FC enhancers. The present study significantly extend these prior combinatorial codes for FC gene regulation by identifying four additional classes of TFBSs that are critical for accurate FC enhancer activity, namely POUHD, Myb, Fkh and T-box motifs. Moreover, these findings provided an opportunity to examine the complete spectrum of regulatory motif usage across a collection of regulatory elements that are active in different muscle FCs, which led to the identification of 18 unique combinations of 11 TFBSs for the entire set of 18 known FC enhancers. Thus, unlike other cases that have been studied, a single enhancer archetype does not appear to exist for this subpopulation of myoblasts. This finding likely reflects the fact that although these elements all display FC activity, with some overlap at the level of individual cells, no two FC gene expression patterns directed by this enhancer set are identical (Busser, 2012).

The marked heterogeneity of FC enhancer architecture uncovered in this study reflects not only distinct combinations of various TF classes (including signal-activated, ubiquitous and both tissue- and cell type-specific TFs), but also diversity at other biological levels, including the unique identities of the thirty muscle FCs and their differentiated derivatives in each abdominal hemisegment, and the different gene expression patterns exhibited by those particular cells. Thus, TFBS combinatorics provide a plausible molecular explanation for the functional complexity of enhancers having related but non-identical activites at the resolution of individual cells in the context of the developing embryo (Busser, 2012).

This study has investigated the transcriptional regulatory network specifying individual muscle FCs using an integrated genomics approach that includes identification of orthologous enhancers, de novo motif discovery, classification of enhancer sequence features, empirical testing of candidate enhancers, and cis-trans tests of target gene regulation. It was also established that a small set of training sequences can be expanded with orthologous sequences. Moreover, motifs learned by the classifier were empirically found to be critical for the appropriate spatio-temporal activities of FC enhancers, and suggested new candidate TFs in the myogenic regulatory network. Using this approach, one such candidate TF, Org-1, was identified as a novel muscle identity TF, and further found that no two enhancers with related activities contain the same combination of TFBSs. The tools and strategy used in this study can be readily applied to other cell types to identify the motifs and trans-acting factors regulating a set of co-expressed genes. Finally, it is anticipated that an iterative application of this approach, which could include training on datasets of different epigenetic marks associated with active enhancers or previous ChIP studies of known mesodermally-relevant TFs , will lead to further refinements in the determination of cell type-specific transcriptional codes (Busser, 2012).

Notch and Ras signaling pathway effector genes expressed in fusion competent and founder cells during Drosophila myogenesis

Drosophila muscles originate from the fusion of two types of myoblasts -- founder cells (FCs) and fusion-competent myoblasts (FCMs). To better understand muscle diversity and morphogenesis, a large-scale gene expression analysis was performed to identify genes differentially expressed in FCs and FCMs. Embryos derived from Toll10b mutants were employed to obtain primarily muscle-forming mesoderm, and activated forms of Ras or Notch were expressed to induce FC or FCM fate, respectively. The transcripts present in embryos of each genotype were compared by hybridization to cDNA microarrays. Among the 83 genes differentially expressed, genes known to be enriched in FCs or FCMs, such as heartless or hibris, previously characterized genes with unknown roles in muscle development, and predicted genes of unknown function, were found. These studies of newly identified genes revealed new patterns of gene expression restricted to one of the two types of myoblasts, and also striking muscle phenotypes. Whereas genes such as phyllopod play a crucial role during specification of particular muscles, others such as tartan are necessary for normal muscle morphogenesis (Artero, 2003).

The Toll10b mutation gives rise to embryos composed primarily of somatic mesoderm. In these embryos FCs and FCMs are readily detected, and they respond to the Ras and Notch signaling pathways in the same way as their wild-type counterparts. Advantage was taken of this fact to enrich Toll10b mutant embryos for FCs or FCMs, which allowed a concentration on the transcription in these two specific cell types within the context of the entire embryo. Genes known to be expressed and regulated in FCs or FCMs emerged from the screen in the proper categories. Not all known FC/FCM genes were detected in the screen for several reasons: the high stringency set for interpretation of the array data; the presence of only about one-third of the genome on the arrays; the loss of Dpp in the Toll10b background, and the specific window of myogenesis (5- to 9-hours) that was the focus of this investigation. However, a plethora of potential new muscle regulators were uncovered, including known genes with no previously recognized function in the mesoderm (such as phyl and asteroid), and genes predicted from the Drosophila genome sequence but not previously analyzed (Artero, 2003).

Various tests were applied to ascertain the validity of the results. Available databases were analyzed to find evidence that the known and predicted genes were expressed at the correct time and place. In addition, Northern analysis with eleven genes tested the reliability of the microarray detection and selection criteria; the results from all genes tested agreed with the array data (Artero, 2003).

A Toll10b sample on the Northern blots allowed ascertainment of why a gene is enriched in a particular condition. For example, in the case of FC enriched genes, the signal in the Ras and Notch lanes can be compared with Toll10b alone to determine whether the Ras/Notch ratio for a gene is due to activation by Ras or repression by Notch. Those genes that are 'enriched under Notch conditions', for example, could reflect a variety of transcription mechanisms that would result in a ratio of less than 0.6. By Northern analysis, many of the 'Notch-regulated' genes, and hence the predicted FCM genes, were found to be repressed by Ras signaling and slightly activated by Notch. As a case in point, hibris is induced by Notch (2-fold) and repressed by Ras (10-fold), both by Northern analysis and by in situ hybridization in embryos (Artero, 2003).

A combination of in situ hybridization, immunostaining and confocal microscopy was used to verify that the differential expression changes that were observed in these overexpression embryos reflected true differential expression in the wild-type situation. The expression of nine genes from different functional categories was analyzed. For seven of these, expression was detected in the predicted type of myoblast. For two, asteroid (ast) and cadmus, no specific staining in embryos was detected by in situ hybridization. For those genes that fell into the category of 'specific role in muscle development uncertain', in situ hybridization of several (28%) showed expression in tissues other than somatic mesoderm that are present in the Toll10b background. These genes changed their expression levels in response to Ras or Notch, and may be Ras and Notch targets in non-mesodermal tissues (Artero, 2003).

The most stringent test, mutational analysis, was applied to a set of genes for which mutations are available. Preliminary analyses of another four FCM-enriched genes was carried out: Elongation factor Tu mitochondrial (EfTuM), Glutamine synthetase 1, cadmus and parcas. All four mutants have muscle defects, including muscle losses and aberrant muscle morphologies. Thus all the genes tested show some muscle defect, supporting the usefulness of the genetic and genomic approach (Artero, 2003).

Taken together, these data suggest that the majority of genes detected play important roles in FCs or FCMs during muscle development. Some of these genes might not have been found in traditional forward genetic screens because of partial or complete genetic redundancy. The data complement traditional forward genetic approaches for finding genes crucial for muscle morphogenesis (Artero, 2003).

Each of the thirty FCs per abdominal hemisegment is hypothesized to produce its own unique combination of transcriptional regulators, though the evidence for this is limited. In turn the combination of regulators would control the morphology of the final muscle. Although several transcriptional regulators have been linked to FC identity, the molecular description is far from complete. This screen contributed two more FC-specific genes. Previously known markers, such as slouch or eve, once induced in the muscle FC, are maintained throughout the remainder of development. Ubx, which emerged from this screen, is a similarly simple case, as its transcripts are steadily present in most FCs. By contrast, more complex patterns of gene expression have been identified in FCs, such as the transient transcription of asense in a subset of FCs. The subsequent transcriptional inactivation of asense may underlie temporal changes in cell properties (Artero, 2003).

Even less is known about transcriptional regulators controlling FCM differentiation. Only one gene, lame duck, has been shown to have a role in FCMs. This screen has uncovered three more potential players: delilah, E(spl)mß and CG4136, confirming that FCMs follow their own, distinct, myogenic program. Discovering what aspects of FCM biology are controlled by these transcriptional regulators awaits analysis of the loss-of-function phenotypes (Artero, 2003).

Notch and Ras signaling pathways interact during muscle progenitor segregation. The results suggest that phyl and polychaetoid (pyd) may be additional links between the two signaling pathways in FCs. phyl and pyd both interact genetically with Notch and Delta. The transcription of phyl, which promotes neural differentiation, is negatively regulated by Notch signaling during specification of SOPs and their progeny. This study shows a similar regulation in muscle cells, where Notch signaling represses phyl expression and Ras signaling increases phyl expression. Likewise, in the nervous system, the segregation of SOPs requires pyd, a Ras target gene, to negatively regulate ac-sc complex expression. Similarly, Pyd may restrict the muscle progenitor fate to a single cell, perhaps by regulating lethal of scute transcription. Thus, Pyd would collaborate with Notch signaling to restrict muscle progenitor fate to one cell (Artero, 2003).

FCMs appear to integrate Ras and Notch signaling differently. Two genes whose transcripts were enriched under activated Notch conditions, parcas and asteroid (ast), have been implicated in Ras signaling in other tissues, directly (ast) or indirectly (parcas). These data are suggestive of a role for Ras signaling in the FCMs, in addition to its role in FC specification. In addition, Notch signaling to FCMs may prime cells for subsequent Ras signaling during muscle morphogenesis, much as occurs in FCs where Ras signaling primes the cell for subsequent Notch signaling during asymmetric division of the muscle progenitor (Artero, 2003).

Embryos that lack or ectopically express phyl have morphological defects in specific muscles, for example, in LL1 and DO4 in response to diminished phyl function, and in DT1 and LT4 in response to increased phyl function. The morphological defects in the loss-of-function embryos appear to be due to a failure to specify particular FCs, a conclusion that is based upon missing or abnormal production of the FC marker Kr. In eye development and SOP specification, Phyl directs degradation of the transcriptional repressor Tramtrack. In a subset of the primordial muscle cells, Phyl may work similarly, targeting Tramtrack for degradation. The presence of Tramtrack would contribute to the specific identity program of the muscle. Since Tramtrack is expressed in the mesoderm, this possibility is likely. Alternatively, Phyl may be required for targeted degradation of some other protein in a subset of FCs. The molecular partner for Phyl during muscle differentiation is unknown, although preliminary data suggest that sina is also expressed in somatic mesoderm and thus may be its partner. These studies have identified a new role for Phyl in muscle progenitor specification and suggest the importance of targeted ubiquitination for proper muscle patterning (Artero, 2003).

A role for ubiquitination in muscle differentiation is further reinforced by the identification of the RING finger-containing protein Goliath (Gol), induced by activated Notch conditions, and CG17492, induced by activated Ras conditions. Several RING-containing proteins function as E3 ubiquitin ligases, with the ligase activity mapping to the RING motif itself. Ligase function has been experimentally confirmed for the Gol ortholog GREUL1 in Xenopus. Thus, targeted protein degradation during muscle morphogenesis could serve a host of crucial functions, such as protein turnover, vesicle sorting, transcription factor activation and signal degradation (Artero, 2003).

The simplest view of the 'founder cell' hypothesis is that each FC contains all the information for the development of a particular muscle. By contrast, FCMs have been seen as a naïve group of myoblasts, entrained to a particular muscle program upon fusion to the FC. This work indicates that these two groups of myoblasts have distinct transcriptional profiles. These data raise the possibility of a greater role for FCMs in determining the final morphology of the muscle and emphasize a need to characterize fully those FCM genes. For example, this screen identified a protein kinase of the SR splice site selector factors (SRPK) whose transcripts are enriched in FCMs, suggesting that regulation of the splicing machinery is important for muscle morphogenesis. The Mhc gene undergoes spatially and temporally regulated alternative splicing in body wall muscles conferring different physiological properties on these muscles. This FCM-specific expression of SRPK may indicate that the production of a particular Mhc isoform is regulated by the FCMs that contribute to that muscle, rather than by the particular FC that seeds the muscle. In addition, a number of observations suggest that FCMs may be a diverse population of myoblasts, with different subsets having different potential to contribute to the final muscle pattern. For example, hbs expression suggests that only a subset of FCMs express the gene, and twist expression in lame duck mutant embryos persists in a subset of FCMs. This study provides additional genes for exploring whether FCMs are a heterogeneous population of myoblasts as well as determining the nature of FCM contribution to the final muscle (Artero, 2003).

The molecular events underlying complex morphological changes, such as migration, cell fusion, cell shape changes or changes in the physiology of a cell, require a rich and dynamic program of transcription changes. This study has described approximately one-third of this transcriptional profile. The FC- or FCM-specific transcription of seven genes, and the mutant phenotype of four selected genes, allowed the definition of new muscle mutations that specifically affect the morphological traits of a subset of muscles (Artero, 2003).

Motoneurons regulate myoblast proliferation and patterning in Drosophila

Motoneurons directly influence the differentiation of muscle fibers, regulating features such as muscle fiber type and receptor development. Less well understood is whether motoneurons direct earlier events, such as the patterning of the musculature. In Drosophila, the denervation of indirect flight muscles results in a diminished myoblast population and smaller or missing muscle fibers. Whether the neuron-dependent control of myoblast number is due to regulation of cell division, motoneuron-dependent apoptosis, or nerve-dependent localization and migration of myoblasts, was examined. Denervation results in a reduced rate of cell division, as revealed by BrDU incorporation. There is no change in the frequency of apoptotic myoblasts following denervation. Using time lapse imaging of GFP-expressing myoblasts in vivo in pupae, it was observed that despite denervation, the migration and localization of myoblasts remains unchanged. In addition to reducing myoblast proliferation, denervation also alters the segregation of myoblasts into the de novo arising dorso-ventral muscles (DVMs). To address this effect on muscle patterning, the expression of the founder-cell marker Dumbfounded/Kirre (Duf) in imaginal pioneer cells was examined. There is a strong correspondence between cells that express Dumbfounded/Kirre and the number of DVM fibers, consistent with a role for these cells in establishing adult muscles. In the absence of innervation the Duf-positive cells are no longer detected, and muscle patterning is severely disrupted. These results support a model where specialized founder cells prefigure the adult muscle fibers under the control of the nervous system (Fernandes, 2005).

The motoneuron exerts a mitogenic influence on IFM myoblasts. Following unilateral denervation, the BrDU birthdating experiments revealed a significant decline in the rate of proliferation. This decline is likely sufficient to account for the reduced myoblast population observed in denervated hemisegments (previously quantified by morphometry; Fernandes, 1998). The smaller muscles that result are thus likely due to the smaller number of available myoblasts, and possibly to the absence of neuromuscular excitation following denervation (Fernandes, 2005).

Two alternate mechanisms were ruled out: there was no evidence for a change in the rate of myoblast cell death following denervation, indicating that the motoneuron does not provide an essential survival factor for the cells. Also, no significant change was observed in the migratory behavior of myoblasts following unilateral denervation, when examined at either the single cell or population level. This indicates that the reduced population size was not the result of myoblast emigration from denervated sites (Fernandes, 2005).

The second major effect of denervation was the gross disruption of normal muscle formation in the de novo arising DVMs. Normally, myoblasts coalesced into discrete primordia that prefigure the three sets of DVM muscle fibers. When denervated, the DVM myoblasts remain unpatterned, and the muscles fail to form. This may be due to a direct effect of the motoneuron on the myoblasts. However, denervation also disrupts the behavior of a potential intermediary player, the imaginal pioneer cells, that are thought to prefigure the DVM fibers as myoblast fusion targets (Fernandes, 2005).

BrDU birthdating experiments reveal a significant rise in the rate of DLM myoblast proliferation that normally occurs at 18-24 APF. In denervated regions, myoblast proliferation remains unchanged, holding steady at the earlier, basal level seen prior to the onset of myoblast fusion (at 12-16 h APF). Thus, DLM myoblast proliferation involves two components: a basal and nerve-independent phase (at 12-16 h APF) and a later incremental nerve-dependent phase (at 18-24 h APF). It is proposed that the nerve-dependent increase in myoblast proliferation regulates the number of myoblasts available for fusion, and thus is a way for motoneurons to control muscle size (Fernandes, 2005).

The nerve-dependent rise in DLM myoblast proliferation correlates with the expansion of motoneuronal terminal arbors on the muscle fiber surface. While it is possible that the expansion of motoneuron terminals and the change in myoblast proliferation are independent responses to exogenous hormone signals, the data argue that motoneurons strongly influence myoblast cell division, since proliferation is reduced following denervation. It is proposed that the growing motoneuron terminal either releases a factor that influences myoblast cell division, or alternatively potentiates myoblast responsiveness to available growth factors and mitogens. In either case, the nerve-dependent control of myoblast proliferation would in turn influence the growth of the muscle fiber (Fernandes, 2005).

The rise in myoblast proliferation observed during the nerve-dependent phase of DLM development resembles a feature seen during photoreceptor development in the Drosophila eye. During differentiation of the neuroepithelium, there is a rise in cell proliferation, referred to as the second mitotic wave (SMW), which produces additional precursors that are recruited to eventually form a complete ommatidium. It is likely that the rise in DLM myoblast proliferation similarly serves to maintain the size of the myoblast pool, so that cells can be continuously drawn from the pool until the desired muscle size is achieved (Fernandes, 2005).

That the two phases of DLM myogenesis differ in their nerve-dependence resembles events associated with vertebrate myogenesis. Mammalian skeletal muscles form in two waves: primary myotubes form first, and serve as scaffolds for secondary myotube formation. Primary myogenesis is independent of innervation, while secondary myogenesis is nerve dependent. This is due to the presence of a nerve-dependent population of myoblasts essential for secondary myotube formation (Fernandes, 2005).

The DVMs develop from the de novo fusion of myoblasts, and critically depend on the motoneuron for muscle fiber formation (Fernandes, 1998 and Fernandes, 1999). Denervation also results in a failure of myoblasts to segregate into distinct DVM primordia. Myoblast patterning and fiber development for the DVMs have been proposed to depend on specialized imaginal pioneer cells. The 'imaginal pioneer' (IP) cells lie in close association with motoneuron arbors, as demonstrated by EM analysis, and are thus potentially dependent on neurons for their normal function or survival. The IP cells are thought to serve as myoblast fusion targets, and thus to prefigure the mature muscle fibers. There have been, however, no reported molecular markers for these cells (Fernandes, 2005).

In the Drosophila embryo the mesodermal cells that generate the somatic muscles are critically dependent on specialized founder cells. Each embryonic founder cell is the precursor of a specific muscle fiber, and is the target of fusion-competent myoblasts. The founder cells each express Dumbfounded/Kirre, a key component of the cell fusion machinery. In loss of function duf mutations myoblast fusion is disrupted (Fernandes, 2005).

Intriguingly, it was found that there are Duf-positive cells within the DVM myoblast pool. Their location, number, and size indicate that they are likely to be the IP cells previously described. The Duf-positive cells are present in the DVM I and II primordia in direct correspondence to the final number of DVM fibers, as is the case for the IP cells. Duf-positive cells have also been reported for other pupal muscles, and a correlation exists between Duf-positive cells and the numbers of both IFM and abdominal muscle fibers. Significantly, it was found that denervation affects the Duf cells of the DVM primordia. Although Duf-positive cells are initially present in the denervated hemisegments in the normal pattern and number (at 12 h APF), following denervation they are no longer reliably observed by 18-20 h APF. By 24 h APF, when control hemisegments possess well patterned Duf-positive DVMs, Duf-positive cells on the denervated side are rarely observed (Fernandes, 2005).

These observations support a model where Duf-positive IP cells in the pupa serve as fusion targets of myoblasts, as is the case for the Duf-positive founder cells in the embryo. Since the Duf molecule is an essential component of the cell fusion machinery, its disappearance following denervation suggests that fusion events are severely disrupted and may explain the associated muscle patterning defects. It cannot as yet be determine whether the loss of Duf labeling is due to a loss of Duf expression in the IP cells, or due to apoptosis. Distinguishing between these scenarios will require in situ time lapse imaging of vitally labeled IP cells in denervated hemisegments (Fernandes, 2005).

DLM fibers arise from larval muscles that persist into the pupal stage. Like all embryonically established somatic muscle fibers, the persistent larval fibers also do not depend on the motoneuron for their formation or maintenance. When denervated, the larval fibers persist and DLM fibers still form, albeit at a slower rate (Fernandes, 1998; Fernandes. 1999). Duf expression is also detected in the persistent larval fibers, consistent with the fact that they function as myoblast fusion targets. However, unlike the DVMs, denervation does not result in a loss of Duf expression in the developing DLMs. The reason for this independence remains uncharacterized, but likely reflects the distinct origin of these cells from larval precursor muscles (Fernandes, 2005).

A dependence on motoneurons for the regulation of muscle size and patterning has been observed for several insect systems. When abdominal Drosophila muscles are denervated, the adult fibers are significantly reduced in mass. The most prominent effect involves the male-specific muscle (MSM) of the fifth abdominal segment of the adult. This muscle is larger than other body wall muscle fibers, a difference attributed to the enhanced recruitment of myoblasts from a common myoblast pool. When the abdominal myoblast pool is reduced experimentally through hydroxyurea treatment, a smaller muscle is present at the MSM location in segment A5. A BrDU labeling analysis remains to be performed to confirm the role of myoblast proliferation on MSM development (Fernandes, 2005).

Denervation studies in Manduca have similarly shown that proliferation of myonuclei is reduced in leg, abdominal, and DLM muscles. At the onset of metamorphosis, muscle precursors appear in the region of the future adult muscles and become associated with tendons (leg muscles) or persistent larval muscles (DLM). This accumulation is then followed by the appearance of proliferating 'myonuclei' within the developing primordia. By contrast, in the case of Drosophila DLMs, BrDU incorporation is restricted to myoblasts present outside the primordia, and there is no evidence of nuclear division within the muscle fibers (Fernandes, 2005).

In conclusion, it is proposed that the motoneuron critically influences the size of the myoblast pool through a direct effect on myoblast cell division, and that this helps regulate the final size of adult muscle fibers. The motoneuron has a second role in regulating the development of de novo forming fibers, where it is essential for the partitioning of myoblasts into muscle primordia. Moreover, continued Duf labeling within the primordia depends on the motoneuron's presence. Thus, the motoneuron influences both the number of cells available for fusion, as well as potentially regulates the fusion events themselves. This is an elegant mechanism for controlling muscle fiber differentiation during myogenesis, and may have evolved as a way to ensure that muscle primordia develop into muscles that meet the diverse demands placed on them by the nervous system (Fernandes, 2005).

Genetic control of the distinction between fat body and gonadal mesoderm

The somatic muscles, the heart, the fat body, the somatic part of the gonad and most of the visceral muscles are derived from a series of segmentally repeated primordia in the Drosophila mesoderm. This work describes the early development of the fat body and its relationship to the gonadal mesoderm, as well as the genetic control of the development of these tissues. The first sign of fat body development is the expression of serpent in segmentally repeated clusters within the trunk mesoderm in parasegments 4-9. Segmentation and dorsoventral patterning genes define three regions in each parasegment in which fat body precursors can develop. The primary and secondary dorsolateral fat body primordia are formed ventral to the visceral muscle primoridium in each parasegment. The ventral secondary cluster forms more ventrally in the posterior portion of each parasegment. Fat body progenitors in these regions are specified by different genetic pathways. Two dorsolateral regions require engrailed and hedgehog (within the even-skipped domain) for their development while the ventral secondary cluster is controlled by wingless. Ubiquitous mesodermal en expression leads to an expansion of the primary clusters into the sloppy-paired domain, resulting in a continuous band of serpent-expressing cells in parasegments 4-9. The observed effect of en on fat body development is seen not only on mesodermal overexpression but also when en is overexpressed in the ectoderm. Loss of wingless leads to an expansion of the dorsolateral fat body primordium. decapentaplegic and one or more unknown genes determine the dorsoventral extent of these regions. High levels of Dpp repress serpent, resulting in the formation of visceral musculature, an alternative cell fate (Reichmann, 1998).

In each of parasegments 10-12 one of these primary dorsolateral regions generates somatic gonadal precursors instead of fat body. The balance between fat body and somatic gonadal fate in these serially homologous cell clusters is controlled by at least five genes. A model is suggested in which tinman, engrailed and wingless are necessary to permit somatic gonadal develoment, while serpent counteracts the effects of these genes and promotes fat body development. In wg mutant embryos, all dorsolateral mesodermal cells, including those in parasegments 10-12, acquire fat body fate. This phenotype can be interpreted as the combined effects of two separate functions of wg: (1) wg is necessary to repress fat body development in the dorsolateral mesoderm underlying the wg domain in all parasegments; (2) wg is required in the primary cluster to permit somatic gonadal precursor instead of fat body development in parasegments 10-12. Loss of engrailed results in the absence of demonstrable somatic gonadal precursors, similar to the situation in tinman mutants. Ubiquitous mesodermal en expression leads to the formation of additional somatic gonadal precursor cells in parasegments 10-12. The homeotic gene abdominalA limits the region of serpent activity by interfering in a mutually repressive feed back loop between gonadal and fat body development. It is unlikely that abdA represses srp directly, since srp can be expressed in cells in which abdA is active. abdA might prevent srp from inhibition of a somatic gonadal precursor competence factor (Riechmann, 1998).

A nutrient sensor mechanism controls Drosophila growth

Organisms modulate their growth according to nutrient availability. Although individual cells in a multicellular animal may respond directly to nutrient levels, growth of the entire organism needs to be coordinated. This study provides evidence that in Drosophila, coordination of organismal growth originates from the fat body, an insect organ that retains endocrine and storage functions of the vertebrate liver. A genetic screen for growth modifiers discovered slimfast, a gene that encodes an amino acid transporter. Remarkably, downregulation of slimfast specifically within the fat body causes a global growth defect similar to that seen in Drosophila raised under poor nutritional conditions. This involves TSC/TOR signaling in the fat body, and a remote inhibition of organismal growth via local repression of PI3-kinase signaling in peripheral tissues. These results demonstrate that the fat body functions as a nutrient sensor that restricts global growth through a humoral mechanism (Colombani, 2003).

In multicellular organisms, the control of growth depends on the integration of various genetic and environmental cues. Nutrient availability is one of the major environmental signals influencing growth and, as such, has dictated adaptative responses during evolution toward multicellularity. In particular, complex humoral responses ensure that growth and development are properly coordinated with nutritional conditions (Colombani, 2003).

In isolated cells, amino acid withdrawal leads to an immediate suppression of protein synthesis, suggesting that cells are protected by active sensing mechanims that block translation prior to depletion of internal amino acid stores. In many mammalian cell types, changes in amino acid diet affect the binding of the translation repressor 4EBP1 to initiation factor eIF4E and the activity of ribosomal protein S6 kinase (S6K). These two signaling events require the activity of TOR (target of rapamycin), a conserved kinase recently shown to participate in a nutrient-sensitive complex both in mammalian cells and in yeast. Mutations in the Drosophila TOR homolog (dTOR) results in cellular and physiological responses characteristic of amino acid deprivation and establish that TOR is cell autonomously required for growth in a multicellular organism. Furthermore, the TSC (tuberous sclerosis complex) tumor suppressor, consisting of a TSC1 and TSC2 heterodimer (TSC1/2), as well as the small GTPase Rheb participate to the regulation of TOR function. Overall, these data suggest that TSC, Rheb, TOR, and S6K participate in a conserved pathway that coordinates growth with nutrition in a cell-intrinsic manner (Colombani, 2003).

In multicellular organisms, humoral controls are believed to buffer variations in nutrient levels. However, little is known about how growth of individual cells is coordinated. In vertebrates, growth-promoting action of the growth hormone (GH) is mostly relayed to peripheral tissues through the production of IGF-I. Binding of IGF-I to its cognate receptor tyrosine kinase (IGF-IR) induces phosphorylation of insulin receptor substrates (IRS), which in turn activate a cascade of downstream effectors. These include phospho-inositide 3-kinase (PI3K), which generates the second messenger phosphatidylinositol-3,4,5-P3 (PIP3), and thereby activates the AKT/PKB kinase. Genetic manipulation of IGF-I, IGF-IR, PI3K, and AKT in mice modulates tissue growth in vivo thus demonstrating a requirement of the IGF pathway for growth. In Drosophila, both loss- and gain-of function studies have also exemplified the role of a conserved insulin/IGF signaling pathway in the control of growth. Ligands for the unique insulin receptor (Inr) constitute a family of seven peptides related to insulin, the Drosophila insulin-like peptides (Dilps). Remarkably, three dilp genes (dilp2, dilp3, and dilp5) are expressed in a cluster of seven median neurosecretory cells (m-NSCs) in the larval brain, suggesting that they have an endocrine function. Indeed, ablation of the seven dilp-expressing mNSCs in larvae induces a systemic growth defect (Colombani, 2003).

Both in flies and mice, mutations in IRS provoke growth retardation as well as female sterility similar to what is observed in starved animals. Moreover, PI3K activity in Drosophila larvae depends on the availability of proteins in the food. Overall, this supports the notion that the insulin/IGF pathway might coordinate tissue growth with nutritional conditions. However, upon amino acid withdrawal, neither PI3K nor AKT/PKB activities are downregulated in mammalian or insect cells in culture, suggesting that this pathway does not directly respond to nutrient shortage. Hence, an intermediate sensor mechanism must link nutrient availability to insulin/IGF signaling (Colombani, 2003).

An intriguing possibility is that specific organs could function as nutrient sensors and induce a nonautonomous modulation of insulin/IGF growth signaling in response to changes in nutrient levels. This study used a genetic approach in Drosophila to assess both the cellular and humoral responses to amino acid deprivation in the context of a developing organism. The insect fat body (FB) has important storage and humoral functions associated with nutrition, comparable to vertebrate liver and adipose tissue. During larval stages, the FB accumulates large stores of proteins, lipids, and carbohydrates, which are normally degraded by autophagy during metamorphosis in order to supply the developing tissues but can also be remobilized during larval life to compensate transitory nutrient shortage. In addition to its storage function, the FB also has endocrine activity and supports growth of imaginal disc explants and DNA replication of larval brains in coculture. This study demonstrates that the FB operates as a sensor for variations in nutrient levels and coordinates growth of peripheral tissues accordingly via a humoral mechanism (Colombani, 2003).

In the course of a P[UAS]-based overexpression screen for growth modifiers, a P[UAS]-insertion line (UY681) was found to cause growth retardation upon ectopic activation. Sequence analysis revealed that P(UY)681 is inserted in a predicted gene (CG11128) that encodes a putative protein showing strong homology with amino acid permeases of the cationic amino acid transporter (CAT) family. The P[UAS] element is inserted in the first intron of the CG11128 gene, potentially driving transcription of an antisense RNA in a GAL4-dependent manner. To assess the function of this transporter, 3H-arginine uptake was measured in S2 cells. Results indicate that amino acid uptake is either enhanced by transfection of a CG11128 cDNA or suppressed by RNAi, indicating that the encoded protein presents CAT activity. In situ hybridization revealed basal levels of CG11128 expression in most larval tissues but much higher levels in the FB and the gut, two tissues involved in amino acid processing (Colombani, 2003).

By P element remobilization, an imprecise excision was obtained that deletes the sequences encoding the N-terminal half of the protein. 87% of homozygous mutant animals die during larval stages. The few viable adults emerged after a 2 day delay and were smaller and markedly slimmer than control animals. The associated gene was named slimfast (slif) and the excision allele slif1. Weight measurement indicated that homozygous slif1 adult males displayed a 16% mass reduction compared to control. Accordingly, adult wing size was reduced by 8% due to a reduction of both cell size and cell number. When the slif1 allele was in trans to Df(3L)Δ1AK, a deficiency covering the locus, larval lethality was slightly enhanced, suggesting that slif1 corresponds to a strong hypomorphic allele. The amino acid transporter function of slif, as well as the phenotypes observed upon reduction of slif function suggest that slif mutant animals might suffer amino acid deprivation. A major consequence of amino acid deprivation in larvae is the remobilization of nutrient stores in the FB, which typically results in aggregation of storage vesicles. Consistently, fusion of storage vesicles was observed in the FB of slif1 larvae and was indistinguishable from that observed in animals fed on protein-free media (Colombani, 2003).

GAL4 induction of P(UY)681 resulted in a growth-deficient phenotype similar to that of slif1 loss of function. The antisense orientation of P(UY)681 suggested that the growth defect following GAL4 induction was due to an RNAi effect. Indeed, Northern blot analysis revealed that ubiquitous GAL4-dependent activation of P(UY)681 using the daughterless-GAL4 (da-GAL4) driver strongly reduced slif mRNA levels. Only two of the three alternative first exons are potentially affected by the antisense RNA, possibly explaining the residual accumulation of slif mRNAs in da-GAL4; P(UY)681 animals. Most of these animals died at larval stage, similar to what was observed for slif1 mutants. Specific induction of P(UY)681in the wing disc using the MS1096-GAL4 driver provoked a reduction of the adult wing size, which could be either rescued by coactivation of a UAS-slif transgene or enhanced by reducing slif gene dosage with the heterozygous Df(3L)Δ1AK deficiency. Thus, GAL4-dependent activation of P(UY)681 reduces slif function and defines a conditional loss-of-function allele hereafter termed slifAnti (Colombani, 2003).

As expected, loss of slif function using the slifAnti allele also mimicked amino acid deprivation. Accordingly, ubiquitous slifAnti induction in growing larvae resulted in storage vesicle aggregation and strong reduction of global S6 kinase activity, similar to what was reported in animals raised on protein-free diet. Additionally, an increase in PEPCK1 gene transcription was observed, similar to the effect of amino acid withdrawal. In summary, this study has identified two loss-of-function alleles of the slif gene whose defects mimic physiological aspects of amino acid deprivation. Importantly, the conditional slifAnti allele provides a unique tool to mimic an amino acid deprivation in a tissue-specific manner (Colombani, 2003).

This study established that the FB is a sensor tissue for amino acid levels, as downregulation of the Slif amino acid transporter within the FB is sufficient to induce a general reduction in the rate of larval growth. In contrast, specific disruption of slif in imaginal discs, larval gut, or salivary glands did not induce a nonautonomous growth response, suggesting that these tissues do not participate in the systemic control of growth. The dilp-expressing median neurosecretory cells (m-NSCs) also affect growth control, since selective ablation of these cells in the larval brain induces an overall reduction of animal size. In response to complete sugar and protein starvation, the m-NSCs stop expressing dilp3 and dilp5 genes, suggesting that these neurons also sense nutrient levels. This study shows that the selective reduction of slif function in these cells has no obvious effect on tissue growth and animal development. This indicates that the seven dilp-expressing m-NSCs do not constitute a general amino acid sensor. In contrast, the role of m-NSCs in carbohydrate homeostasis and the observation that they stop expressing certain dilp genes when larvae are deprived of sugar rather suggests that these cells have a role in sensing carbohydrate levels (Colombani, 2003 and references therein).

This analysis also provides a framework in which to understand the phenotype of minidisc, a mutation in an amino acid transporter gene that exhibits nonautonomous growth defects in imaginal discs (Colombani, 2003).

In a number of model systems, both PI3K and TOR have been implicated in linking growth to nutritional status and, until recently, were considered as intermediates of a common regulatory pathway. In yeast, the TOR kinase is part of a cell-autonomous nutrient sensor, which controls protein synthesis, ribosome biogenesis, nutrient import, and autophagy. Genetic analysis in Drosophila indicates that dTOR is required for cell-intrinsic growth control. The results obtained using the slifAnti allele in the wing disc indicate that individual tissues have indeed the potential to respond to amino acid deprivation in a cell-autonomous manner. Nonetheless, this study also demonstrates that the TOR nutritional checkpoint participates in a systemic control of larval growth emanating from the FB. Within a developing organism, each cell may integrate these two distinct inputs regarding nutritional status, one originating from a systemically-acting FB sensor, and the other from TOR-dependent signaling in individual cells. One can further speculate that depending on the strength and duration of starvation, different in vivo nutritional checkpoints will be hierarchically recruited to protect the animal and that the systemic control might, in most physiological situations, override the cell-autonomous control. Indeed, as the data demonstrate, the FB sensor is sufficient to induce a general and coordinated response to starvation without calling individual cell-autonomous mechanisms into play (Colombani, 2003).

Several lines of evidence indicate that the PI3K pathway is not part of the sensor mechanism in FB cells. First, a sensor for PI3K activity in the FB is only marginally affected by amino acid deprivation in that tissue, indicating that the cell-autonomous response to amino acid starvation does not directly influence PI3K signaling. This is reminiscent of previous observations in mammalian cultured cells, showing that PI3K activity does not respond to variations in amino acid levels. Moreover, inhibition of PI3K signaling by dPTEN expression in the FB is not sufficient to trigger the sensing mechanism. Although, dPTEN overexpression causes a complete disappearance of the PI3K sensor accompanied by growth suppression of FB cells, the FB maintains a critical mass that allows for normal larval growth. In contrast, the regulatory subunit p60 whose overexpression potently inhibits PI3-kinase in flies has been shown to induce a systemic effect on larval growth when overexpressed in the FB using an Adh-Gal4 driver. This study found that a pumpless ppl-GAL4-directed expression of p60 also provokes a strong suppression of larval growth and a dramatic inhibition of FB development in young larvae. Thus, the systemic effect on growth observed upon p60 overexpression most likely results from a drastic reduction of FB mass, which then fails to support normal larval growth (Colombani, 2003).

These results further indicate that PI3K signaling is a remote target of the humoral message that originates from the FB in response to amino acid deprivation. This is in agreement with previous data showing that PI3K activity is downregulated by dietary amino acid deprivation and explains why global PI3-kinase inhibition mimics cellular and organismal effects of starvation. The existence of a humoral relay reconciles these in vivo studies with the absence of direct PI3K responsiveness to amino acid levels (Colombani, 2003).

The relative resistance of imaginal disc growth to the systemic control exerted by the FB correlates with maintenance of PI3K activity in these tissues. This is in agreement with previous observations that cells in the larval brain and in imaginal discs maintain a slow rate of proliferation under protein starvation, while larval endoreduplicating tissues (ERTs) arrest. This difference might be attributed to the basal levels of dilp2 expression observed in imaginal discs, allowing a moderate growth rate of these tissues through an autocrine/paracrine mechanism. It was recently shown that clonal induction of PI3K potently induces cell-autonomous growth response even in fasting larvae, indicating that some nutrients are still accessible to support cell growth within a fasted larva. The main function of a general sensor could be to preserve these limited nutrients for use by high priority tissues. In this context, local PI3K activation through an autocrine loop in imaginal tissues could favor the growth of prospective adult structures in adverse food conditions. Thus, the FB would have an active role in controlling the allocation of resources depending on nutritional status. In this respect, it is noteworthy that FB cells are relatively resistant to the FB-derived humoral signal, since the PI3K sensor is not drastically affected in the FB of ppl>slifAnti animals. Thereby, essential regulatory functions of the FB could be preserved even in severely restricted nutritional conditions (Colombani, 2003).

How does the FB signal to other tissues? This study suggests that a humoral signal relays information from the FB amino acid sensor and systemically inhibits PI3K signaling. In addition, this downregulation is not due to a direct inhibition of dilp expression by neurosecretory cells in the brain. Nevertheless, it cannot be ruled out that the secretion of these molecules is subjected to regulation in the mNSCs. Both in vivo and in insect cell culture, several imaginal discs growth factors (IDGF) secreted by the FB have been proposed to function synergistically with Dilp signaling to promote growth. However, this study did not find any modification of IDGF expression in the FB of larvae raised on water- or sugar-only diet, or upon FB induction of slifanti. In vertebrates, the different functions of the circulating IGF-I are modulated through its association with IGF-BPs and acid labile subunit (ALS). In particular, the formation of a ternary complex with ALS leads to a considerable extension of IGF-I half-life. The finding that a Drosophila ALS ortholog is expressed within the FB in an amino acid-dependent manner provides a new avenue to study the molecular mechanisms of nonautonomous growth control mediated by the FB (Colombani, 2003).

This study highlights the contribution that genetics can provide to unravel the mechanisms of physiological control. Using a genetic tool to mimic amino acid deprivation, it was demonstrated that nutrition systemically controls body size through an amino acid sensor operating in the FB. It is proposed that (1) in metazoans, a systemic nutritional sensor modulates the conserved TOR-signaling pathway, and (2) the response to sensor activation is relayed by a hormonal mechanism, which triggers an Inr/PI3K-dependent response in peripheral tissues (Colombani, 2003).

Role and regulation of starvation-induced autophagy in the Drosophila fat body

In response to starvation, eukaryotic cells recover nutrients through autophagy, a lysosomal-mediated process of cytoplasmic degradation. Autophagy is known to be inhibited by TOR signaling, but the mechanisms of autophagy regulation and its role in TOR-mediated cell growth are unclear. Signaling through TOR and its upstream regulators PI3K and Rheb is necessary and sufficient to suppress starvation-induced autophagy in the Drosophila fat body. In contrast, TOR's downstream effector S6K promotes rather than suppresses autophagy, suggesting S6K downregulation may limit autophagy during extended starvation. Despite the catabolic potential of autophagy, disruption of conserved components of the autophagic machinery, including ATG1 and ATG5, does not restore growth to TOR mutant cells. Instead, inhibition of autophagy enhances TOR mutant phenotypes, including reduced cell size, growth rate, and survival. Thus, in cells lacking TOR, autophagy plays a protective role that is dominant over its potential role as a growth suppressor (Scott, 2004).

Autophagy likely evolved in single-cell eukaryotes to provide an energy and nutrient source allowing temporary survival of starvation. In yeast, Tor1 and Tor2 act as direct links between nutrient conditions and cell metabolism. These proteins sense nutritional status by an unknown mechanism, and effect a variety of starvation responses including changes in transcriptional and translational programs, nutrient import, protein and mRNA stability, cell cycle arrest, and induction of autophagy. Autophagy thus occurs in the context of a comprehensive reorganization of cellular activities aimed at surviving low nutrient levels (Scott, 2004).

In multicellular organisms, TOR is thought to have retained its role as a nutrient sensor but has also adopted new functions in regulating and responding to growth factor signaling pathways and developmental programs. Thus in a variety of signaling, developmental, and disease contexts, TOR activity can be regulated independently of nutritional conditions. In these cases, autophagy may be induced in response to downregulation of TOR despite the presence of abundant nutrients and may potentially play an important role in suppressing cell growth rather than promoting survival. Identification of the tumor suppressors PTEN, and TSC1 and TSC2 as positive regulators of autophagy provides correlative evidence supporting such a role for autophagy in growth control. Alternatively, since TOR activity is required for proper expression and localization of a number of nutrient transporters, inactivation of TOR may lead to reduced intracellular nutrient levels, and autophagy may therefore be required under these conditions to provide the nutrients and energy necessary for normal cell metabolism and survival (Scott, 2004).

The results presented here provide genetic evidence that under conditions of low TOR signaling, autophagy functions primarily to promote normal cell function and survival, rather than to suppress cell growth. This conclusion is based on the finding that genetic disruption of autophagy does not restore growth to cells lacking TOR, but instead exacerbates multiple TOR mutant phenotypes. It is important to note that mutations in TOR do not disrupt larval feeding, and thus disruption of autophagy is detrimental in TOR mutants despite the presence of ample extracellular nutrients. The finding that autophagy is critical in cells lacking TOR further supports earlier studies suggesting that inactivation of TOR causes defects in nutrient import, resulting in an intracellular state of pseudo-starvation (Scott, 2004).

Can the further reduction in growth of TOR mutant cells upon disruption of autophagy be reconciled with the potential catabolic effects of autophagy? TOR regulates the bidirectional flow of nutrients between protein synthesis and degradation through effects on nutrient import, autophagy, and ribosome biogenesis. When TOR is inactivated, rates of nutrient import and protein synthesis decrease, resulting in a commensurate reduction in mass accumulation and cell growth. In addition, autophagy is induced to maintain intracellular nutrient and energy levels sufficient for normal cell metabolism. When autophagy is experimentally inhibited in cells lacking TOR, this reserve source of nutrients is blocked, leading to a further decrease in energy levels, protein synthesis, and growth. It is noted that autophagy may have additional functions in cells with depressed TOR signaling, including recycling of organelles damaged by the absence of TOR activity, or selective degradation of cell growth regulators, analogous to the regulatory roles of ubiquitin-mediated degradation (Scott, 2004).

Autophagy is required for normal developmental responses to inactivation of insulin/PI3K signaling in the nematode C. elegans. In response to starvation or disruption of insulin/PI3K signaling, C. elegans larvae enter a dormant state called the dauer. Autophagy has been observed in C. elegans larvae undergoing dauer formation: disruption of a number of ATG homologs interfers with normal dauer morphogenesis. Importantly, simultaneous disruption of insulin/PI3K signaling and autophagy genes results in lethality, similar to the results presented in this study. Thus despite significant differences in developmental strategies for surviving nutrient deprivation, autophagy plays an essential role in the starvation responses of yeast, flies, and worms (Scott, 2004).

The prevailing view that S6K acts to suppress autophagy was founded on correlations between induction of autophagy and dephosphorylation of rpS6 in response to amino acid deprivation or rapamycin treatment. However, the genetic data presented in this study argue strongly against a role for S6K in suppressing autophagy: unlike other positive components of the TOR pathway, null mutations in S6K do not induce autophagy in fed animals. It is suggested that the observed correlation between S6K activity and suppression of autophagy is due to common but independent regulation of S6K and autophagy by TOR. Thus, autophagy suppression and S6K-dependent functions such as ribosome biogenesis represent distinct outputs of TOR signaling (Scott, 2004).

How might TOR signal to the autophagic machinery, if not through S6K? In yeast, this is accomplished in part through regulation of Atg1 kinase activity and ATG8 gene expression (Kamada, 2000 and Kirisako, 1999). The demonstration of a role for Drosophila ATG1 and ATG8 homologs [see TG8a (CG32672) and ATG8b (CG12334)] in starvation-induced autophagy, and the genetic interaction observed between ATG1 and TOR, are consistent with a related mode of regulation in higher eukaryotes. However, it is noted that other components of the yeast Atg1 complex such as Atg17 and Atg13, whose phosphorylation state is rapamycin sensitive, do not have clear homologs in metazoans, indicating that differences in regulation of autophagy by TOR are likely (Scott, 2004).

In addition to excluding a role for S6K in suppression of autophagy, these results reveal a positive role for S6K in induction of autophagy. S6K may promote autophagy directly, through activation of the autophagy machinery, or indirectly through its effects on protein synthesis. The latter possibility is consistent with previous reports that protein synthesis is required for expansion and maturation of autophagosomes. Interestingly, despite being required for autophagy, S6K is downregulated under conditions that induce it, including chronic starvation and TOR inactivation. Consistent with this, it was found that lysotracker staining is significantly weaker in chronically starved animals or in TOR mutants than in wild-type animals starved 3-4 hr. Furthermore, expression of constitutively activated S6K has no effect in wild-type, but restores lysotracker staining in TOR mutants to levels similar to those of acutely starved wild-type animals. It is suggested that downregulation of S6K may limit rates of autophagy under conditions of extended starvation or TOR inactivation and that this may protect cells from the potentially damaging effects of unrestrained autophagy (Scott, 2004).

Co-culture and conditioned media experiments have shown that the Drosophila fat body is a source of diffusible mitogens. The fat body has also been shown to act as a nutrient sensor through a TOR-dependent mechanism and to regulate organismal growth through effects on insulin/PI3K signaling. The results in this study extend these findings by showing that this endocrine response is accompanied by the regulated release of nutrients through autophagic degradation of fat body cytoplasm. Preventing this reallocation of resources, either through constitutive activation of PI3K or through inactivation of ATG genes, results in profound nutrient sensitivity. Thus, in response to nutrient limitation, the fat body is capable of simultaneously restricting growth of peripheral tissues through downregulation of insulin/PI3K signaling and providing these tissues with a buffering source of nutrients necessary for survival through autophagy (Scott, 2004).

The two origins of hemocytes in Drosophila

As in many other organisms, the blood of Drosophila consists of several types of hemocytes, which originate from the mesoderm. By lineage analyses of transplanted cells, two separate anlagen have been defined that give rise to different populations of hemocytes: embryonic hemocytes and lymph gland hemocytes. The anlage of the embryonic hemocytes is restricted to a region within the head mesoderm between 70% and 80% egg length. In contrast to all other mesodermal cells, the cells of this anlage are already determined as hemocytes at the blastoderm stage. Unexpectedly, these hemocytes do not degenerate during late larval stages, but have the capacity to persist through metamorphosis and are still detectable in the adult fly. A second anlage, which gives rise to additional hemocytes at the onset of metamorphosis, is located within the thoracic mesoderm at 50% to 53% egg length. After transplantation within this region, clones were detected in the larval lymph glands. Labeled hemocytes are released by the lymph glands not before the late third larval instar. The anlage of these lymph gland-derived hemocytes is not determined at the blastoderm stage, as indicated by the overlap of clones with other tissues. These analyses reveal that the hemocytes of pupae and adult flies consist of a mixture of embryonic hemocytes and lymph gland-derived hemocytes, originating from two distinct anlagen that are determined at different stages of development (Holz, 2003).

The origin of the embryonic hemocytes (EH) can be traced back to the head mesoderm of late stage 11 embryos by morphological criteria. Owing to the fact that srp is expressed in a narrow stripe within the cephalic mesoderm at the blastoderm stage and that a loss of srp function leads to a complete loss of embryonic hemocytes, this domain is considered to be the primordium of the EH. By homotopic single-cell transplantations it was possible to restrict the anlage to a sharply delimitated region located at 70% to 80% EL within the mesoderm, exactly corresponding to the cephalic expression domain of srp. The fact that none of the EH clones overlapped with other tissues indicates that the hemocytes are already determined at the blastoderm stage. This was confirmed by heterotopic transplantations from the EH anlage into the abdominal mesoderm; these transplanted cells give rise to hemocytes. Since mesodermal cells transplanted into the EH anlage do not transform into embryonic hemocytes, the determining factor is not able to induce a hemocyte fate within these cells and seems to function cell-autonomously. A good candidate for such a factor is Srp. However, since srp is also expressed in many other tissues that do not give rise to hemocytes, there must be additional genes that lead to a determination of the EH at the blastoderm stage. The early determination of the EH is quite unusual, since all other mesodermal tissues analyzed to date -- including the anlage of the lymph gland-derived hemocytes -- are not restricted to a tissue-specific fate prior to the second postblastodermal mitoses. This might be a developmental adaptation of the EH, which at stage 12 are already differentiated into functional macrophages and are responsible for the removal of apoptotic cells within developing tissues (Holz, 2003).

It is commonly believed that in Drosophila during larval development the EH population is entirely replaced by hemocytes that have been released by the larval lymph glands. However, it is possible to trace hemocytes originating from the head mesoderm through all stages of development until 14-day-old adult flies. The number of hemocytes progressively rises during larval life, from less than 200 to more than 5000 per individual. Cell lineage analyses unambiguously demonstrate that this increase is due to postembryonic proliferation of the EH. The contribution of the lymph glands to the hemocyte population was determined by means of cell lineage analyses. These studies reveal that the lymph glands do not release blood cells into the hemocoel during all larval stages but exclusively at the end of the third larval instar (Holz, 2003).

With the onset of metamorphosis, additional hemocytes are released from the lymph glands. Although the lymph glands do not persist through metamorphosis, the marked hemocytes released by the labeled lymph glands are still detectable in adult flies. Hence, all hemocytes found throughout larval life originate solely from the EH anlage, whereas the pupal and imaginal blood is made up of two different populations: EH and LGH (Holz, 2003).

The two populations of hemocytes share many functional, morphological and genetic similarities. In both cases, the determination of hemocytes depends on srp, while the specification towards the distinct blood cell types is induced by the expression of lozenge (lz) glia cells missing (gcm) and the gcm homolog gcm2. Both EH and LGH differentiate into podocytes, crystal cells and plasmatocytes. Hemocytes of both populations have the capability to adopt macrophage characteristics. However, despite all similarities, the history of the two populations is quite different, since they originate from two different mesodermal regions and are determined at different developmental stages. In view of the fact that the lymph glands do not release hemocytes before the onset of metamorphosis under nonimmune conditions, all hemocytes found in the larval hemocoel represent EH (Holz, 2003).

The many similarities between EG and LGH raise the question why there are two populations at all. A massive release of hemocytes by the lymph glands is seen just at the onset of pupation. The lymph glands additionally have the capacity to differentiate and release a special type of hemocytes, the lamellocytes, under immune conditions even before the onset of metamorphosis. Thus, because under nonimmune conditions the lymph glands do not release any cells before the onset of pupation, it might be their primary role to provide a reservoir of immune defensive hemocytes. The massive apoptosis and accumulation of cell debris might be a secondary trigger to stimulate proliferation and release of the lymph gland hemocytes (Holz, 2003).

The Drosophila lymph gland as a developmental model of hematopoiesis

Drosophila hematopoiesis occurs in a specialized organ called the lymph gland. In this systematic analysis of lymph gland structure and gene expression, the developmental steps in the maturation of blood cells (hemocytes) from their precursors are defined. In particular, distinct zones of hemocyte maturation, signaling and proliferation in the lymph gland during hematopoietic progression are described. Different stages of hemocyte development have been classified according to marker expression and placed within developmental niches: a medullary zone for quiescent prohemocytes, a cortical zone for maturing hemocytes and a zone called the posterior signaling center for specialized signaling hemocytes. This establishes a framework for the identification of Drosophila blood cells, at various stages of maturation, and provides a genetic basis for spatial and temporal events that govern hemocyte development. The cellular events identified in this analysis further establish Drosophila as a model system for hematopoiesis (Jung, 2005).

In the late embryo, the lymph gland consists of a single pair of lobes containing ~20 cells each. These express the transcription factors Srp and Odd skipped (Odd), and each cluster of hemocyte precursors is followed by a string of Odd-expressing pericardial cells that are proposed to have nephrocyte function. These lymph gland lobes are arranged bilaterally such that they flank the dorsal vessel, the simple aorta/heart tube of the open circulatory system, at the midline. By the second larval instar, lymph gland morphology is distinctly different in that two or three new pairs of posterior lobes have formed and the primary lobes have increased in size approximately tenfold (to ~200 cells. By the late third instar, the lymph gland has grown significantly in size (approximately another tenfold) but the arrangement of the lobes and pericardial cells has remained the same. The cells of the third instar lymph gland continue to express Srp (Jung, 2005).

The third instar lymph gland also exhibits a strong, branching network of extracellular matrix (ECM) throughout the primary lobe. This network was visualized using several GFP-trap lines in which GFP is fused to endogenous proteins. For example, line G454 represents an insertion into the viking locus, which encodes a Collagen IV component of the extracellular matrix. The hemocytes in the primary lobes of G454 (expressing Viking-GFP) appear to be clustered into small populations within pockets or chambers bounded by GFP-labeled branches of various sizes. Other lines, such as the uncharacterized GFP-trap line ZCL2867, also highlight this branching pattern. What role this intricate ECM network plays in hematopoiesis, as well as why multiple cells cluster within these ECM chambers, remains to be determined (Jung, 2005).

Careful examination of dissected, late third-instar lymph glands by differential interference contrast (DIC) microscopy revealed the presence of two structurally distinct regions within the primary lymph gland lobes that have not been previously described. The periphery of the primary lobe generally exhibits a granular appearance, whereas the medial region looks smooth and compact. These characteristics were examined further with confocal microscopy using a GFP-trap line G147, in which GFP is fused to a microtubule-associated protein. The G147 line is expressed throughout the lymph gland but, in contrast to nuclear markers such as Srp and Odd, distinguishes morphological differences among cells because the GFP-fusion protein is expressed in the cytoplasm in association with the microtubule network. Cells in the periphery of the lymph gland make relatively few cell-cell contacts, thereby giving rise to gaps and voids among the cells within this region. This cellular individualization is consistent with the granularity of the peripheral region observed by DIC microscopy. By contrast, cells in the medial region were relatively compact with minimal intercellular space, which is also consistent with the smoother appearance of this region by DIC microscopy. Thus, in the late third instar, the lymph gland primary lobes consist of two physically distinct regions: a medial region consisting of compactly arranged cells, which was termed the medullary zone; and a peripheral region of loosely arranged cells, termed the cortical zone (Jung, 2005).

Mature hemocytes have been shown to express several markers, including collagens, Hemolectin, Lozenge, Peroxidasin and P1 antigen. The expression of the reporter Collagen-gal4 (Cg-gal4), which is expressed by both plasmatocytes and crystal cells, is restricted to the periphery of the primary lymph gland lobe. Comparison of Cg-gal4 expression in G147 lymph glands, in which the medullary zone and cortical zone can be distinguished, reveals that maturing hemocytes are restricted to the cortical zone. In fact, the expression of each of the maturation markers mentioned above is found to be restricted to the cortical zone. The reporter hml-gal4 and Pxn, which are expressed by the plasmatocyte and crystal cell lineages, are extensively expressed in this region. Likewise, the expression of the crystal cell lineage marker Lozenge is restricted in this manner. The spatial restriction of maturing crystal cells to the cortical zone was verified by several means, including the distribution of melanized lymph gland crystal cells in the Black cells background and analysis of the terminal marker ProPOA1. The cortical zone is also the site of P1 antigen expression, a marker of the plasmatocyte lineage. The uncharacterized GFP fusion line ZCL2826 also exhibits preferential expression in the cortical zone. Last, it was found that the homeobox transcription factor Cut is preferentially expressed in the cortical zone of the primary lobe. Although the role of Cut in Drosophila hematopoiesis is currently unknown, homologs of Cut are known to be regulators of the myeloid hematopoietic lineage in both mice and humans. Cells of the rare third cell type, lamellocytes, are also restricted to the cortical zone, based upon cell morphology and the expression of a msn-lacZ reporter (msn06946). In summary, based on the expression patterns of several genetic markers that identify the three major blood cell lineages, it is proposed that the cortical zone is a specific site for hemocyte maturation (Jung, 2005).

The medullary zone was initially defined by structural characteristics and subsequently by the lack of expression of mature hemocyte markers. However, several markers have been identified that are exclusively expressed in the medullary zone at high levels but not the cortical zone. Consistent with the compact arrangement of cells in the medullary zone, it was found that Drosophila E-cadherin (DE-cadherin or Shotgun) is highly expressed in this region. No significant expression of DE-cadherin was observed among maturing cells in the cortical zone. E-cadherin, in both vertebrates and Drosophila, is a Ca2+-dependent, homotypic adhesion molecule often expressed by epithelial cells and is a crucial component of adherens junctions. Attempts to study DE-cadherin mutant clones in the medullary zone where the protein is expressed were unsuccessful since no clones were recoverable. The reporter lines domeless-gal4 and unpaired3-gal4 are preferentially expressed in the medullary zone. The gene domeless (dome) encodes a receptor molecule known to mediate the activation of the JAK/STAT pathway upon binding of the ligand Unpaired. The unpaired3 (upd3) gene encodes a protein with homology to Unpaired and has been associated with innate immune function. These gal4 lines are in this study only as markers that correlate with the medullary zone and, at the present time, there is no evidence that their associated proteins have a role in lymph gland hematopoiesis. Other markers of interest with preferential expression in the medullary zone include the molecularly uncharacterized GFP-trap line ZCL2897 and actin5C-GFP. Cells expressing hemocyte maturation markers are not seen in the medullary zone. It is therefore reasonable to propose that this zone is largely populated by prohemocytes that will later mature in the cortical zone. Prohemocytes are characterized by their lack of maturation markers, as well as their expression of several markers described as expressed in the medullary zone (Jung, 2005).

The posterior signaling center (PSC), a small cluster of cells at the posterior tip of each of the primary (anterior-most) lymph gland lobes, is defined by its expression of the Notch ligand Serrate and the transcription factor Collier. During this analysis, several additional markers were identified that exhibit specific or preferential expression in the PSC region. For example, it was found that the reporter Dorothy-gal4 is strongly expressed in this zone. The Dorothy gene encodes a UDP-glycosyltransferase, which belongs to a class of enzymes that function in the detoxification of metabolites. The upd3-gal4 reporter, which has preferential expression in the medullary zone, is also strongly expressed among cells of the PSC. Last, three uncharacterized GFP-gene trap lines, ZCL2375, ZCL2856 and ZCL0611 were found, that are preferentially expressed in the PSC. This analysis has made it clear that the PSC is a distinct zone of cells that can be defined by the expression of multiple gene products (Jung, 2005).

The PSC can be defined just as definitively by the characteristic absence of several markers. For example, the RTK receptor Pvr, which is expressed throughout the lymph gland, is notably absent from the PSC. Likewise, dome-gal4 is not expressed in the PSC, further suggesting that this population of cells is biased toward the production of ligands rather than receptor proteins. Maturation markers such as Cg-gal4, which are expressed throughout the cortical zone, are not expressed by PSC cells. Additionally, the expression levels of the hemocyte marker Hemese and the Friend-of-GATA protein U-shaped are dramatically reduced in the PSC when compared with other hemocytes of the lymph gland. Taken together, both the expression and lack of expression of a number of genetic markers defines the cells of the PSC as a unique hemocyte population (Jung, 2005).

In contrast to primary lobes of the third instar, maturing hemocytes are generally not seen in the secondary lobes. Correspondingly, secondary lobes often have a smooth and compact appearance, much like the medullary zone of the primary lobe. Consistent with this appearance, secondary lymph gland lobes also express high levels of DE-cadherin. The size of the secondary lobe, however, varies from animal to animal and this correlates with the presence or absence of maturation markers. Smaller secondary lobes contain a few or no cells expressing maturation markers, whereas larger secondary lobes usually exhibit groups of differentiating cells. Direct comparison of DE-cadherin expression in secondary lobes with that of Cg-gal4, hml-gal4 or Lz revealed that the expression of these maturation markers occurs only in areas in which DE-cadherin is downregulated. Therefore, although there is no apparent distinction between cortical and medullary zones in differentiating secondary lobes, there is a significant correlation between the expression of maturation markers and the downregulation of DE-cadherin, as is observed in primary lobes (Jung, 2005).

The relatively late 'snapshot' of lymph gland development in the third larval instar establishes the existence of spatial zones within the lymph gland that are characterized by differences in structure as well as gene expression. In order to understand how these zones form over time, lymph glands of second instar larvae, the earliest time at which it was possible to dissect and stain, were examined for the expression of hematopoietic markers. As expected, Srp and Odd are expressed throughout the lymph gland during the second instar since they are in the late embryo and third instar lymph gland. Likewise, the hemocyte-specific marker Hemese is expressed throughout the lymph gland at this stage, although it is not present in the embryonic lymph gland (Jung, 2005).

To determine whether the cortical zone is already formed or forming in second instar lymph glands, the expression of various maturation markers were examined in a pair-wise manner to establish their temporal order. Of the markers examined, hml-gal4 and Pxn are the earliest to be expressed. The majority of maturing cells were found to be double-positive for hml-gal4 and Pxn expression, although a few cells were found to express either hml-gal4 or Pxn alone. This indicates that the expression of these markers is initiated at approximately the same time, although probably independently, during lymph gland development. The marker Cg-gal4 is next to be expressed since it was found among a subpopulation of Pxn-expressing cells. Finally, P1 antigen expression is initiated late, usually in the early third instar. Interestingly, the early expression of each of these maturation markers is restricted to the periphery of the primary lymph gland lobe, indicating that the cortical zone begins to form in this position in the second instar. Whenever possible, each genetic marker was directly compared with other pertinent markers in double-labeling experiments, except in cases such as the comparison of two different gal4 reporter lines or when available antibodies were generated in the same animal. In such cases, the relationship between the two markers, for example dome-gal4 and hml-gal4, was inferred from independent comparison with a third marker such as Pxn (Jung, 2005).

By studying the temporal sequence of expression of hemocyte-specific markers, one can describe stages in the maturation of a hemocyte. It should be noted, however, that not all hemocytes of a particular lineage are identical. For example, in the late third instar lymph gland, the large majority of mature plasmatocytes (~80%) expresses both Pxn and hml-gal4, but the remainder express only Pxn (~15%) or hml-gal4 (~5%) alone. Thus, while plasmatocytes as a group can be characterized by the expression of representative markers, populations expressing subsets of these markers indeed exist. It remains unclear at this time whether this heterogeneity in the hemocyte population is reflective of specific functional differences (Jung, 2005).

In the third instar, Pxn is a prototypical hemocyte maturation marker, while immature cells of the medullary zone express dome-gal4. Comparing the expression of these two markers in the second instar reveals an interesting developmental progression. A group of cells along the peripheral edge of these early lymph glands already express Pxn. These developing hemocytes downregulate the expression of dome-gal4, as they do in the third instar. Next to these developing hemocytes is a group of cells that expresses dome-gal4 but not Pxn; these cells are most similar to medullary zone cells of the third instar and are therefore prohemocytes. Interestingly, there also exists a group of cells in the second instar that expresses neither Pxn nor dome-gal4. This population is most easily seen in the medial parts of the gland, close to the centrally placed dorsal. These cells resemble earlier precursors in the embryo, except they express the marker Hemese. These cells are called pre-prohemocytes. Interpretation of the expression data is that pre-prohemocytes upregulate dome-gal4 to become prohemocytes. As prohemocytes begin to mature into hemocytes, dome-gal4 expression is downregulated, while the expression of maturation markers is initiated. The prohemocyte and hemocyte populations continue to be represented in the third instar as components of the medullary and cortical zones, respectively (Jung, 2005).

The cells of the PSC are already distinguishable in the late embryo by their expression of collier. It was found that the canonical PSC marker Ser-lacZ is not expressed in the embryonic lymph gland and is only expressed in a small number of cells in the second instar. This relatively late onset of expression is consistent with collier acting genetically upstream of Ser. Another finding was that the earliest expression of upd3-gal4 parallels the expression of Ser-lacZ and is restricted to the PSC region. Finally, Pvr and dome-gal4 are excluded from the PSC in the second instar, similar to what is seen in the third instar (Jung, 2005).

To determine whether maturing cortical zone cells are indeed derived from medullary zone prohemocytes, a lineage-tracing experiment was performed in which dome-gal4 was used to initiate the permanent marking of all daughter cell lineages. In this system, the dome-gal4 reporter expresses both UAS-GFP and UAS-FLP. The FLP recombinase excises an intervening FRT-flanked 'STOP cassette', allowing constitutive expression of lacZ under the control of the actin5C promoter. At any developmental time point, GFP is expressed in cells where dome-gal4 is active, while lacZ is expressed in all subsequent daughter cells regardless of whether they continue to express dome-gal4. In this experiment, cortical zone cells are permanently marked with ß-galactosidase despite not expressing dome-gal4 (as assessed by GFP), indicating that these cells are derived from a dome-gal4-positive precursor. This result is consistent with and further supports independent marker analysis that shows that dome-gal4-positive prohemocytes downregulate dome-gal4 expression as they initiate expression of maturation markers representative of cortical zone cells. As controls to the above experiment, the expression patterns of two other gal4 lines, twist-gal4 and Serrate-gal4 were determined. The reporter twist-gal4 is expressed throughout the embryonic mesoderm from which the lymph gland is derived. Accordingly, the entire lymph gland is permanently marked by ß-galactosidase despite a lack of twist-gal4 expression (GFP) in the third instar lymph gland. Analysis of Ser-gal4 reveals that PSC cells remain a distinct population of signaling cells that do not contribute to the cortical zone (Jung, 2005).

Genetic manipulation of Pvr function provides valuable insight into its involvement in the regulation of temporal events of lymph gland development. To analyze Pvr function, FLP/FRT-based Pvr-mutant clones were generated in the lymph gland early in the first instar and then examined during the third instar for the expression of maturation markers. It was found that loss of Pvr function abolishes P1 antigen and Pxn expression, but not Hemese expression. The crystal cell markers Lz and ProPOA1 are also expressed normally in Pvr-mutant clones, consistent with the observation that mature crystal cells lack or downregulate Pvr. The fact that Pvr-mutant cells express Hemese and can differentiate into crystal cells suggests that Pvr specifically controls plasmatocyte differentiation. Pvr-mutant cells do not become TUNEL positive but do express the hemocyte marker Hemese and can differentiate into crystal cells, all suggesting that the observed block in plasmatocyte differentiation within the mutant clone is not due to cell death. Additionally, Pvr-mutant clones were large and not significantly different in size from their wild-type twin spots. Thus, the primary role of Pvr is not in the control of cell proliferation. Targeting Pvr by RNA interference (RNAi) revealed the same phenotypic features, confirming that Pvr controls the transition of Hemese-positive cells to plasmatocyte fate (Jung, 2005).

Entry into S phase was monitored using BrdU incorporation and distinct proliferative phases were identified that occur during lymph gland hematopoiesis. In the second instar, proliferating cells are evenly distributed throughout the lymph gland. By the third instar, however, the distribution of proliferating cells is no longer uniform; S-phase cells are largely restricted to the cortical zone. This is particularly evident when BrdU-labeled lymph glands are co-stained with Pxn. Medullary zone cells, which can be identified by the expression of dome-gal4, rarely incorporate BrdU. Therefore, the rapidly cycling prohemocytes of the second instar lymph gland quiesce as they populate the medullary zone of the third instar. As prohemocytes transition into hemocyte fates in the cortical zone, they once again begin to expand in number. This is supported by the observation that the medullary zone in white pre-pupae does not appear diminished in size, suggesting that the primary mechanism for the expansion of the cortical zone prior to this stage is through cell division within the zone. Proliferating cells in the secondary lobes continue to be distributed uniformly in the third instar, suggesting that secondary-lobe prohemocytes do not reach a state of quiescence as do the cells of the medullary zone. These results indicate that cells of the lymph gland go through distinct proliferative phases as hematopoietic development proceeds (Jung, 2005).

This analysis of the lymph gland revealed three key features that arise during development. The first feature is the presence of three distinct zones in the primary lymph gland lobe of third instar larvae. Two of these zones, termed the cortical and medullary zones, exhibit structural characteristics that make them morphologically distinct. These zones, as well as the third zone, the PSC, are also distinguishable by the expression of specific markers. The second key feature is that cells expressing maturation markers such as Lz, ProPOA1, Pxn, hml-gal4 and Cg-gal4 are restricted to the cortical zone. The medullary zone is consistently devoid of maturation marker expression and is therefore defined as a region composed of immature hemocytes (prohemocytes). The finding of different developmental populations within the lymph gland (prohemoctyes and their derived hemocytes) is similar to the situation in vertebrates where it is known that hematopoietic stem cells and other blood precursors give rise to various mature cell types. Additionally, Drosophila hemocyte maturation is akin to the progressive maturation of myeloid and lymphoid lineages in vertebrate hematopoiesis. The third key feature of lymph gland hematopoiesis is the dynamic pattern of cellular proliferation observed in the third instar. At this stage, the vast majority of S-phase cells in the primary lobe are located in the cortical zone, suggesting a strong correlation between proliferation and hemocyte differentiation. Compared with earlier developmental stages, cell proliferation in the medullary zone actually decreases by the late third instar, suggesting that these cells have entered a quiescent state. Thus, proliferation in the lymph gland appears to be regulated such that growth, quiescence and expansion phases are evident throughout its development (Jung, 2005).

Drosophila blood cell precursors, prohemocytes and maturing hemocytes each exhibit extensive phases of proliferation. The competence of these cells to proliferate seems to be a distinct cellular characteristic that is superimposed upon the intrinsic maturation program. Based on the patterns of BrdU incorporation in developing primary and secondary lymph gland lobes, it is possible to envision at least two levels of proliferation control during hematopoiesis. It is proposed that the widespread cell proliferation observed in second instar lymph glands and in secondary lobes of third instar lymph glands occurs in response to a growth requirement that provides a sufficient number of prohemocytes for subsequent differentiation. The mechanisms promoting differentiation in the cortical zone also trigger cell proliferation, which accounts for the observed BrdU incorporation in this zone and serves to expand the effector hemocyte population. The quiescent cells of the medullary zone represent a pluripotent precursor population because they, similar to vertebrate hematopoietic precursors, rarely divide and give rise to multiple lineages and cell types (Jung, 2005).

Based on this analysis a model is proposed by which hemocytes mature in the lymph gland. Hematopoietic precursors that populate the early lymph gland are first distinguishable as Srp+, Odd+ (S+O+) cells. These will eventually give rise to a primary lymph gland lobe where the steps of hemocyte maturation are most apparent. During the first or early second instar, these S+O+ cells begin to express the hemocyte-specific marker Hemese (He) and the tyrosine kinase receptor Pvr. Such cells can be called pre-prohemocytes and, in the second instar, cells expressing only these markers occupy a narrow region near the dorsal vessel. Subsequently, a subset of these Srp+, Odd+, He+, Pvr+ (S+O+H+Pv+) pre-prohemocytes initiate the expression of dome-gal4 (dg4), thereby maturing into prohemocytes. The prohemocyte population (S+O+H+Pv+dg4+) can be subdivided into two developmental stages. Stage 1 prohemocytes, which are abundantly seen in the second instar, are proliferative, whereas stage 2 prohemocytes, exemplified by the cells of the medullary zone, are quiescent. As development continues, prohemocytes begin to downregulate dome-gal4 and express maturation markers (M; becoming S+O+H+Pv+dg4lowM+). Eventually, dome-gal4 expression is lost entirely in these cells (becoming S+O+H+Pv+dg4-M+), found generally in the cortical zone. Thus, the maturing hemocytes of the cortical zone are derived from prohemocytes previously belonging to the medullary zone. This is supported by lineage-tracing experiments that show cells expressing medullary zone markers can indeed give rise to cells of the cortical zone. In turn, the medullary zone is derived from the earlier, pre-prohemocytes. Early cortical zone cells continue to express successive maturation markers (M) as they proceed towards terminal differentiation. Depending on the hemocyte type, examples of expressed maturation markers are Pxn, P1, Lz, L1, msn-lacZ, etc. These studies have shown that differentiation of the plasmatocyte lineage requires Pvr, while previous work has shown that the Notch pathway is crucial for the crystal cell fate. Both the JAK/STAT and Notch pathways have been implicated in lamellocyte production (Jung, 2005).

Previous investigations have demonstrated that similar transcription factors and signal transduction pathways are used in the specification of blood lineages in both vertebrates and Drosophila. Given this relationship, Drosophila represents a powerful system for identifying genes crucial to the hematopoietic process that are conserved in the vertebrate system. The work presented here provides an analysis of hematopoietic development in the Drosophila lymph gland that not only identifies stage-specific markers, but also reveals developmental mechanisms underlying hemocyte specification and maturation. The prohemocyte population in Drosophila becomes mitotically quiescent, much as their multipotent precursor counterparts in mammalian systems. These conserved mechanisms further establish Drosophila as an excellent genetic model for the study of hematopoiesis (Jung, 2005).

Subdivision and developmental fate of the head mesoderm in Drosophila

This paper defines temporal and spatial subdivisions of the embryonic head mesoderm and describes the fate of the main lineages derived from this tissue. During gastrulation, only a fraction of the head mesoderm (primary head mesoderm; PHM) invaginates as the anterior part of the ventral furrow. The PHM can be subdivided into four linearly arranged domains, based on the expression of different combinations of genetic markers (tinman, heartless, snail, serpent, mef-2, zfh-1). The anterior domain (PHMA) produces a variety of cell types, among them the neuroendocrine gland (corpus cardiacum). PHMB, forming much of the'T-bar' of the ventral furrow, migrates anteriorly and dorsally and gives rise to the dorsal pharyngeal musculature. PHMC is located behind the T-bar and forms part of the anterior endoderm, besides contributing to hemocytes. The most posterior domain, PHMD, belongs to the anterior gnathal segments and gives rise to a few somatic muscles, but also to hemocytes. The procephalic region flanking the ventral furrow also contributes to head mesoderm (secondary head mesoderm, SHM) that segregates from the surface after the ventral furrow has invaginated, indicating that gastrulation in the procephalon is much more protracted than in the trunk. This study distinguishes between an early SHM (eSHM) that is located on either side of the anterior endoderm and is the major source of hemocytes, including crystal cells. The eSHM is followed by the late SHM (lSHM), which consists of an anterior and posterior component (lSHMa, lSHMp). The lSHMa, flanking the stomodeum anteriorly and laterally, produces the visceral musculature of the esophagus, as well as a population of tinman-positive cells that is interpreted as a rudimentary cephalic aorta ('cephalic vascular rudiment'). The lSHM contributes hemocytes, as well as the nephrocytes forming the subesophageal body, also called garland cells (de Velasco, 2005).

The mesoderm is a morphologically distinct cell layer that can be recognized in early embryos of most bilaterian phyla and that gives rise to tissues interposed between ectodermal and endodermal epithelia, including muscle, connective, blood, vascular, and excretory tissue. Besides the differentiative fate of tissues derived from it, the mesoderm shares several common properties in regard to its formation during gastrulation. The anlage of the mesoderm is sandwiched in between the anlage of the endoderm and the neurectoderm. This has been documented in most detail in anamniote vertebrates, where signals from the vegetal blastomeres (the anlage of the endoderm) act on the adjacent marginal zone of the future ectoderm to induce mesoderm. Although gastrulation proceeds quite differently in arthropods from the way it does in chordates, the proximity of the mesodermal anlage to future endoderm and neurectoderm is conserved, and numerous signaling pathways and transcriptional regulators that share similar function and expression patterns in arthropods and chordates have been identified (de Velasco, 2005 and references therein).

Following gastrulation, the mesoderm is subdivided along the dorso-ventral axis into several subdivisions laid out in a distinct dorso-ventral order. In vertebrates, cells located in the dorsal part of the mesoderm anlage give rise to notochord and somites, which in turn produce muscular, skeletal, and connective tissue. Next to the somitic mesoderm is the intermediate mesoderm that will form the excretory and reproductive system. The ventral mesoderm (lateral plate) gives rise to blood, vascular system, visceral musculature, and coelomic cavity. In arthropods, fundamentally similar mesodermal subdivisions can be recognized, and similarities extend to the relative positions these domains obtain relative to each other and relative to the adjacent neurectoderm. For example, precursors of visceral muscles, vascular system, and blood are at the edge of the mesoderm facing away from the neural primordium (ventral in vertebrates, dorsal in arthropods (de Velasco, 2005 and references therein).

The subdivision of the vertebrate mesoderm into distinct longitudinal tissue columns with different fates is seen throughout the trunk and head of the embryo. However, several significant differences between the head and the trunk are immediately apparent. For example, cells derived from the anterior neurectoderm form the neural crest that migrates laterally and gives rise to many of the tissues that are produced by mesoderm in the trunk. As a result, the fates taken over by the head mesoderm are more limited than those of the trunk mesoderm. In contrast, the head mesoderm produces several unique lineages, such as the heart (cardiac mesoderm) and a population of early differentiating macrophages. Moreover, some of the signaling pathways responsible for inducing different mesodermal fates in the trunk appear to operate in a different manner in the head. A recently described example is the Wnt signal that induces somatic musculature in the trunk, but inhibits the same fate in the head (de Velasco, 2005 and references therein).

The head mesoderm of arthropods, like that of vertebrates, also appears to deviate in many ways from the trunk mesoderm. For example, specialized lineages like embryonic blood cells and nephrocytes forming the subesophageal body (also called garland cells) arise exclusively in the head. That being said, very little is known about how the arthropod head mesoderm arises and what types of tissues derive from it. The existing literature mainly uses histology, which severely limits the possibilities of following different cell types forward or backward in time. In this paper, several molecular markers have been used to initiate more detailed studies of the head mesoderm in Drosophila. The goal was to establish temporal and spatial subdivisions of the head mesoderm and, using molecular markers expressed from early stages onward, to follow the fate of the lineages derived from this embryonic tissue. Besides hemocytes and pharyngeal muscles described earlier, the head mesoderm also gives rise to several other lineages, including visceral muscle, putative vascular cells, nephrocytes, and neuroendocrine cells. The development of the head mesoderm is discussed in comparison with the trunk mesoderm and in the broader context of insect embryology (de Velasco, 2005).

The Drosophila head mesoderm, as traditionally defined, includes all mesoderm cells originating anterior to the cephalic furrow. The formation of the head mesoderm is complicated by the fact that (unlike the mesoderm of the trunk) only part of it invaginates with the ventral furrow; by far, the majority of head mesoderm cells, recognizable in a stage 10 or 11 embryo, segregate from the surface epithelium of the head after the ventral furrow has formed. Another complicating factor is that head mesoderm cells derived from different antero-posterior levels adopt very different fates, unlike the situation in the trunk where mesodermal fates within different segments along the AP axis are fairly homogenous, with obvious exceptions such as the gonadal mesoderm that is derived exclusively from a subset of abdominal segments. Using several different markers, this study has followed the origin, migration pathways, and later, fates of head mesoderm cells (de Velasco, 2005).

The anterior part of the ventral furrow, called primary head mesoderm (PHM) in the following, includes cells that will contribute to diverse tissues, including muscle, hemocytes, endoderm, and several ill-defined cell populations closely associated with the brain and neuroendocrine system. For clarification, the anterior ventral furrow will be divided into the following domains:

The anterior lip of the T-bar (PHMA) is the source of the corpus cardiacum, as well as other gt-positive cells that at least in part end up as nerve cells flanking the frontal connective and frontal ganglion. These cells continue the expression of giant throughout late embryonic development; they represent a hitherto unknown class of nonneuroblast-derived neurons (de Velasco, 2005).

The posterior lip of the T-bar (PHMB) can be followed towards later stages by its continued expression of htl. These cells, called the procephalic somatic mesoderm, form a bilateral cluster that moves dorso-anteriorly into the labrum and becomes the dorsal pharyngeal musculature. Htl expression almost disappears in these cells around late stage 11, but is reinitiated at stage 12 and stays strong until stage 14, when the dorsal pharyngeal muscles differentiate. Many of the genes expressed in the somatic musculature of the trunk and its precursors (Dmef2, beta-3-tubulin) are also expressed in the procephalic somatic mesoderm (de Velasco, 2005).

The part of the ventral furrow posteriorly adjacent to the T-bar (PHMC) expresses srp, forkhead (fkh), and other endoderm/hemocyte markers. After the ventral furrow closes in the ventral midline (stage 7/8), these cells form a compact median mass, most of which represents part of the anterior endoderm that gives rise to the midgut epithelium. Starting at around this stage, the lateral part of the hemocyte-forming 'secondary head mesoderm' ingresses in between the endoderm and the surface ectoderm. It is likely that some of the PHMC cells invaginating already with the ventral furrow, along with the cells that form the anterior endoderm, also give rise to hemocytes. Precursors of hemocytes and midgut are difficult to distinguish during and shortly after ventral furrow invagination since both express srp and other markers shared between hemocytes and midgut precursors. At around stage 9, the two populations of precursors disengage. The endoderm remains a compact mesenchyme attached to the invaginating stomodeum; hemocyte precursors move dorsally and take on the shape of expanding vertical plates interposed in between endoderm and ectoderm (de Velasco, 2005).

Domain PHMD, the short portion of the ventral furrow situated posterior to the endoderm, along with a considerable portion of the mesoderm behind the cephalic furrow, forms the mesoderm of the three gnathal segments (mandible, maxilla, labium). The gnathal mesoderm in many ways behaves like the mesoderm of thoracic and abdominal segments. It gives rise to somatic muscle (the lateral pharyngeal muscles), visceral muscle, and fat body. Unlike trunk mesoderm, gnathal mesoderm does not produce cardioblasts and pericardial cells. Instead, a large proportion of gnathal mesoderm cells, joining the anteriorly adjacent secondary procephalic mesoderm, adopt the fate of hemocytes (de Velasco, 2005).

Besides the ventral furrow, other parts of the ventral procephalon produce head mesoderm in a complex succession of delamination and ingression events. The head mesoderm that forms from outside the ventral furrow will be called 'secondary mesoderm' (SHM) in the following. Based on the time of formation and the position relative to the stomodeum, the following phases and domains of secondary head mesoderm development can be distinguished.

Following the obliteration of the ventral furrow at stage 8, the eSHM delaminates from the ventral surface 'meso-ectoderm' (considering that this epithelium still contains mesodermal progenitors!) flanking the endodermal mass. The eSHM forms two monolayered sheets that gradually move dorsally and posteriorly; by stage 9, the eSHM cells line the basal surface of the emerging head neuroblasts. An undefined number of primary head mesoderm cells derived from domain PHMC of the ventral furrow are mingled together with the eSHM cells. The ultimate fate of the eSHM is that of hemocytes: they express srp, followed slightly later by other blood cell markers (e.g., peroxidasin and asrij). A subset of hemocytes, called crystal cells, derive from precursors that form a morphologically conspicuous cluster at the dorsal edge of the eSHM, identifiable from early stage 10 onward by the expression of lz. The mechanism by which at least part of the eSHM delaminates is unique. Thus, it is formed by the vertically oriented division of the surface epithelium, whereby the inner daughters will become eSHMe and the outer ones ectoderm. The focus of vertical mitosis has named the procephalic domain in which it occurs 'mitotic domain #9' (de Velasco, 2005).

From late stage 9 onward, the early SHMs are followed inside the embryo by the closely adjacent posterior late SHMs. One cluster of posterior late secondary head mesoderm (lSHMp) cells delaminates from the surface epithelium flanking the posterior lip of the stomodeum; a second lSHMp cluster appears at the same stage at a slightly more posterior level. The first cluster seems to contribute to the hemocyte population; the posterior cluster gives rise to the nephrocytes forming the subesophageal body (also called garland cells; labeled by CG32094). Garland cell precursors are initially arranged as a paired cluster latero-ventrally of the esophagus primordium; subsequently, the clusters fuse in the midline and form a crescent underneath the esophagus. Garland cells are distinguished from crystal cells by their size, location, and arrangement: crystal cells are large, round cells grouped in an oblong cloud dorso-anterior to the proventriculus. Garland cells are smaller, closely attached to each other, and lie ventral of the esophagus (de Velasco, 2005).

During stages 10 and 11, cells delaminate beside and anterior to the stomodeum, originating from the anlage of the esophagus and the epipharynx (labrum). These cells, called anterior late secondary head mesoderm cells (lSHMa), can be followed by their expression of tin. Two groups can be distinguished. The tin-positive cells delaminating from the esophageal anlage (es) give rise to the visceral musculature (vm) surrounding the esophagus. These cells lose tin expression soon after their segregation, but can be recognized by other visceral mesoderm markers such as anti-Connectin. More dorsally, in the anlage of the clypeolabrum (cl) delaminate, the dorsal subpopulation of the lSHMas, which rapidly migrates posteriorly on either side and slightly dorsal of the esophagus, can be found. These cells retain expression of tin into the late embryo. They assemble into two longitudinal rows stretching alongside the roof of the esophagus primordium. During late embryogenesis, they move posteriorly along with the esophagus towards a position behind the brain commissure. Many of the tin-positive SHMs apparently undergo apoptosis: initially counting approximately 25 on either side, they decrease to 12-15 at stage 14 to finally form a single, irregular row of about 15 cells total in the late embryo. These cells come into contact with the anterior tip of the dorsal vessel. This formation of previously undescribed cells, for which the term 'procephalic vascular cells', is proposed, is interpreted as a rudiment of the head aorta, which forms a prominent part of the dorsal vessel in many insect groups (de Velasco, 2005).

On the basis of additional molecular markers, the tin-positive procephalic vascular cells are further subdivided into two populations. The first subpopulation expresses the muscle and cardioblast-specific marker Dmef2; the second type is Dmef2-negative. In the dorsal vessel of the trunk, tin-positive cells also fall into a Dmef2-positive and a Dmef2-negative population. Dmef2-positive cells of the trunk represent the cardioblasts, myoendothelial cells lining the lumen of the dorsal vessel. Dmef2-negative/tin-positive cells form a somewhat irregular double row of cells attached to the ventral wall of the dorsal vessel. The ultimate fate of these cells has not been explored yet. However, preliminary data suggest that they develop into a muscle band that runs alongside the larval dorsal vessel. This would correspond to the situation in other insects in which such a ventral cardiac muscle band has been described (de Velasco, 2005).

The role of tinman in the formation of the procephalic vascular rudiment was investigated by assaying tin-mutant embryos for the expression of Dmef2. Similar to the cardioblasts of the trunk, the Dmef2-positive cells of the procephalic vascular rudiment are absent in tin mutants. It is quite likely that the (Dmef2-negative) remainder of the procephalic vascular rudiment is affected as well by loss of tin, but in the absence of appropriate markers (besides tin itself, which is not expressed in the mutant), it was not possible to substantiate this proposal (de Velasco, 2005).

At the time of appearance of the ventral furrow, segmental markers such as hh do not allow the distinction between distinct 'preoral' segments. Thus, hh is expressed in a wide procephalic stripe in front of the regularly sized mandibular stripe. During stage 7, the procephalic hh stripe splits into an anterior, antennal stripe and a posterior, short, intercalary stripe. The anterior lip of the ventral furrow (domain PHMA) coincides with the anterior boundary of the antenno-intercalary stripe. Thus, the primary head mesoderm and endoderm originating from within the anterior ventral furrow can be considered a derivative of the antennal and intercalary segments. This interpretation is supported by the expression of the homeobox gene labial (lab) found in the intercalary segment. The labial domain covers much of the anterior ventral furrow, including domains PHMB-C (de Velasco, 2005).

Morphogenetic movements in the ventral head, associated with the closure of the ventral furrow, the formation of the stomodeal placode, and the subsequent invagination of the stomodeum result in a shift of head segmental boundaries. The antennal segment tilts backward, as can be seen from the orientation of the antennal hh stripe that from stage 8 onward forms an almost horizontal line, connecting the cephalic furrow with the sides of the stomodeal invagination (which falls within the ventral realm of the antennal segment, in Drosophila as well as other insects). Since the expression of hh, like that of engrailed (en), coincides with the posterior boundary of a segment, the territory located ventral to the antennal hh stripe falls within the intercalary segment. This implies that most, if not all, of the posterior late SHM, is intercalary in origin. It is further plausible to consider that the anterior lSHM belongs to the intercalary and antennal segment. The vascular cells of the head, a conspicuous derivative of the anterior lSHM in Drosophila, are derived from the antennal mesoderm in other insects. The labrum, with which much of the anterior lSHM is associated, represents a structure that has always been difficult to integrate in the segmental organization of the head. Most likely the labrum represents part of the intercalary segment; this would help explain some of the unusual characteristics of the head mesoderm (de Velasco, 2005).

In conclusion, several fundamental similarities are found between the mesoderm of the head and that of the trunk regarding the tissues they give rise to, and possibly the signaling pathways deciding over these fates. After an initial phase of structural and molecular homogeneity, the trunk mesoderm becomes subdivided into a dorsal and a ventral domain by a Dpp-signaling event that emanates from the dorsal ectoderm. The dorsal domain, characterized by the Dpp-dependent continued expression of tinman, becomes the source of visceral and cardiogenic mesoderm, among other cell types. A role of Dpp/BMP signaling in cardiogenesis seems to be conserved among insects and vertebrates. Subsequent signaling steps, involving both Wingless and Notch/Delta, separate between these two fates and further subdivide the cardiogenic mesoderm into several distinct lineages, such as cardioblast, pericardial cells, and secondary hemocyte precursors (lymph gland). As a result of these signaling events, Tinman and several other fate-determining transcription factors become restricted to their respective lineages: tin to the cardioblasts, odd to pericardial cells and hemocyte precursors, zfh1 and srp to hemocyte precursors and fat body. Dmef2 and several other transcription factors become restricted to various combinations of muscle types (somatic, visceral, cardiac) (de Velasco, 2005).

In the head mesoderm, the above genes are associated with similar fates. Tin and Dmef2 appear widely in the procephalic ventral furrow and the anterior lSHM before getting restricted to the procephalic vascular rudiment and/or the pharyngeal musculature, respectively. In contrast with the initially ubiquitous expression of Tin and Dmef2 in the trunk mesoderm, those parts of the head mesoderm giving rise to hemocytes (PHMC, posterior lSHM) never express these mesodermal genes. Previous work has shown that the head gap gene buttonhead (btd) is responsible for the early repression of tin in the above mentioned domains of the head mesoderm. The early absence of Tin and Dmef2 in the head mesodermal hemocyte precursors is paralleled by the presence of Srp and Zfh1 in these cells. Interestingly, Srp/Zfh-positive cells of the head produce only hemocytes and no fat body, suggesting that an as-yet-uncharacterized signaling step prevents the formation of fat body in the head. It is tempting to speculate that there exists within the mesoderm a 'blood/fat body equivalence group'. Blood cells and fat body share not only the expression of fate-determining genes such as srp and zfh1, but also, later, functional properties that have to do with immunity. In the trunk, the blood/fat body equivalence group gives rise mostly to fat body, producing only a limited number of hemocyte precursors in the dorsal mesoderm of the thoracic segments. In the head, on the other hand, all cells of the equivalence group become hemocytes (de Velasco, 2005).

Attention is drawn to another mesodermal lineage that produces related, yet not identical, cell types in the trunk and the head: the nephrocytes. Nephrocytes are defined by their characteristic ultrastructure (membrane invaginations sealed off by junctions) that attests to their excretory function. In the trunk, nephrocytes are represented by the pericardial cells that settle beside the cardioblasts; a newly discovered nephrocyte population ('star cells') invading the Malpighian tubules is derived from the mesoderm of the tail segments. In the head, nephrocytes aggregate near the junction between esophagus and proventriculus as the subesophageal body, also called garland cells. The fact that from the early stages of development onward different transcription factors are expressed in garland cells and pericardial cells suggests that these cells perform similar, yet not fully overlapping, functions (de Velasco, 2005).

Mononuclear muscle cells in Drosophila ovaries revealed by GFP protein traps

Genetic analysis of muscle specification, formation and function in model systems has provided valuable insight into human muscle physiology and disease. Studies in Drosophila have been particularly useful for discovering key genes involved in muscle specification, myoblast fusion, and sarcomere organization. The muscles of the Drosophila female reproductive system have received little attention despite extensive work on oogenesis. This study used newly available GFP protein trap lines to characterize of ovarian muscle morphology and sarcomere organization. The muscle cells surrounding the oviducts are multinuclear with highly organized sarcomeres typical of somatic muscles. In contrast, the two muscle layers of the ovary, which are derived from gonadal mesoderm, have a mesh-like morphology similar to gut visceral muscle. Protein traps in the Fasciclin 3 gene produced Fas3::GFP that localized in dots around the periphery of epithelial sheath cells, the muscle surrounding ovarioles. Surprisingly, the epithelial sheath cells each contain a single nucleus, indicating these cells do not undergo myoblast fusion during development. Consistent with this observation, the Flp/FRT system was used to efficiently generate genetic mosaics in the epithelial sheath, suggesting these cells provide a new opportunity for clonal analysis of adult striated muscle (Hudson, 2008).

Two protein trap lines showed a striped pattern in ovarian muscle. The GFP transposon in these lines was inserted into the genes for two uncharacterized proteins, CG30084 and CG6416, both of which contain a PDZ domain and a ZASP motif (ZM). CG30084 also contains four LIM domains in its C-terminus. Both CG30084 and CG6416 proteins show homology to the human Z-disc Alternatively Spliced Protein (ZASP) that localizes to the Z-disc in skeletal muscle. Like the ZASP gene, both CG30084 and CG6416 encode a number of alternative splice forms, almost all of which could be tagged based on the locations of the protein trap insertions. An additional isoform of CG30084 has been annotated that encodes only LIM domains and cannot be tagged. Due to the homology of these genes to ZASP, CG30084 was designated as Zasp52, and CG6416 as Zasp66 based on the cytological location of the genes (Hudson, 2008).

The availability of protein trap lines was especially useful for describing intercellular contacts (e.g., Fas3::GFP, Ilk::GFP inserted into Integrin linked kinase) and for showing the location of 'new' sarcomere components (e.g., Zasp52::GFP, Zasp66::GFP). The muscle cells of the oviducts, which are derived from genital disks, have highly organized sarcomeres and a myotube organization typical of somatic muscles in insects, and likely arise by myoblast fusion during development. In contrast, the visceral muscle layers surrounding ovarioles and ovaries have a mesh-like shape similar to the visceral muscles of the gut. Surprisingly, it was found that the epithelial sheath muscles contain one nucleus rather than two as in gut visceral muscles, making them the only known adult somatic or visceral muscle that does not arise by myoblast fusion. It remains to be determined whether the peritoneal sheath cells are mononuclear as well (Hudson, 2008).

The nature of the epithelial sheath muscle makes it ideally suited for analysis of muscle function. The morphology of the cells can be viewed in great detail with an array of available sarcomere and cellular markers. Importantly, the ability to produce cells with a defined genotype by mitotic clone induction provides a new opportunity for studying the phenotypes in adult muscles caused by homozygous mutations in muscle genes, even if the mutations are lethal during early development. Finally, it is straightforward to do short-term in vitro culturing of ovarioles with the epithelial sheath intact and contracting; thus, the behavior of cells bearing mutations can be studied in live tissue under controlled conditions (Hudson, 2008).

Variation in mesoderm specification across drosophilids is compensated by different rates of myoblast fusion during body wall musculature development

It has been shown that species separated by relatively short evolutionary distances may have extreme variations in egg size and shape. Those variations are expected to modify the polarized morphogenetic gradients that pattern the dorso-ventral axis of embryos. Currently, little is known about the effects of scaling over the embryonic architecture of organisms. This problem was examined by asking if changes in embryo size in closely related species of Drosophila modify all three dorso-ventral germ layers or only particular layers, and whether or not tissue patterning would be affected at later stages. This paper reports that changes in scale affect predominantly the mesodermal layer at early stages, while the neuroectoderm remains constant across the species studied. Next, the fate of somatic myoblast precursor cells that derive from the mesoderm was examined to test whether the assembly of the larval body wall musculature would be affected by the variation in mesoderm specification. The results show that in all four species analyzed, the stereotyped organization of the body wall musculature is not disrupted and remains the same as in D. melanogaster. Instead, the excess or shortage of myoblast precursors is compensated by the formation of individual muscle fibers containing more or less fused myoblasts. These data suggest that changes in embryonic scaling often lead to expansions or retractions of the mesodermal domain across Drosophila species. At later stages, two compensatory cellular mechanisms assure the formation of a highly stereotyped larval somatic musculature: an invariable selection of 30 muscle founder cells per hemisegment, which seed the formation of a complete array of muscle fibers, and a variable rate in myoblast fusion that modifies the number of myoblasts that fuse to individual muscle fibers (Belu, 2011).

The data shows that within evolutionary distances as short as 5 mya of divergence time, there can be extreme variations in mesoderm specification. In most species analyzed, the width of mesodermal domain decreases or increases according to embryo size. However, this is not an absolute rule as D. simulans has a larger mesodermal domain than D. melanogaster, despite the fact that those two species vary less in their DV axis (Belu, 2011).

In contrast to the plasticity seen in the mesoderm specification, the width of the neuroectoderm remains constant across species. This latter result is in agreement with two important findings regarding the development of the ventral nerve cord. First, neuroblast maps are nearly identical in a broad range of insect species, sharing similarities even with the crustacean phylum. Second, experiments of genetic manipulation that altered the width of D/V expression domains within the neuroectoderm resulted in the duplication or elimination of neuroblasts of particular identity. Thus, the stable width in the neuroectoderm appears to be essential for the generation of correct neural lineages and axonal scaffolds within Drosophila species. However, the mechanisms that protect the neuroectoderm from scaling effects remain elusive (Belu, 2011).

Based on the stereotyped arrays of muscle fibers and innervation patterns observed in the different species, the data indicate that the mesodermal alterations can be compensated later in development. These corrections would involve an invariable selection of 30 FCs per hemisegment, and a variable rate of myoblast fusion that allows more cells to be incorporated to each muscle fiber. These two cellular mechanisms cooperate and prevent supernumerary or lack of muscle fibers (Belu, 2011).

What protects the development of the somatic body wall musculature from variations in the mesoderm size? This study highlights some key differences between the myogenesis and neurogenesis that may explain why the assembly of the somatic body wall has more alternate ways to cope with the early variations in mesodermal specification than does the ventral nerve cord (Belu, 2011).

The initial steps of both myogenesis and neurogenesis are similar and rely on the formation of groups of equivalent cells, the promuscular and proneural groups, from which a single progenitor cell is selected through lateral inhibition. In the case of the neural progenitor cell, or neuroblast, its identity is determined once it delaminates from the proneural group, when it initiates stereotyped divisions giving rise to a defined number and types of neurons/glial cells. In contrast, the progenitor of somatic muscles undergoes additional asymmetric cell divisions before it gives rise to FCs and adult muscle progenitor cells. Thus, modifications in the specification of muscle progenitor cells and/or their asymmetric cell divisions could generate an identical outcome of 30 embryonic FCs in the different Drosophila species (Belu, 2011).

Another difference between mesodermal and neural tissue specification is the fact that the entire neuroectodermal domain contributes to the formation of a stereotyped tissue, whereas the mesodermal domain is further subdivided and gives rise to non-stereotyped tissues as well, such as the fat body, hematopoietic system and visceral musculature. Therefore, species with reduced mesodermal domain might still be able to assemble the same numbers of promuscular groups at the expense of other mesodermal precursor cells that form non-stereotyped tissues (Belu, 2011).

Finally, the present study reveals that the myoblast fusion step, which is unique to myogenesis, is an important compensatory mechanism for the formation of the somatic body wall musculature (Belu, 2011).

The data shows that during myogenesis of D. busckii and D. pseudoobscura, fewer myoblasts are fused together to form slender muscle fibers in comparison to D. melanogaster. In contrast, more myoblasts fuse into single fibers in D. simulans and D. sechellia, resulting in fibers of increased size. The differential regulation of fusion events appears to be the only characteristic of FC identity that is unique to each species (Belu, 2011).

One of the main regulators of myoblast fusion is the adhesion molecule Kin of Irre/Dumbfounded (Kirre/Duf), which is expressed exclusively by FCs and functions as an attractant to FCMs. The expression of kirre/duf is down-regulated once the correct number of fused FCMs is achieved for a given muscle fiber. If this down-regulation of kirre/duf is modulated by the number of FCM that are aggregated, then there are two ways of increasing myoblast fusion. One would be if inhibitory signals released from fused FCMs are weaker in strength and the other would be if the sensitivity of kirre/duf to these signals is lower. In either case, more myoblasts would be added to the fiber. Recently, the cis-regulatory region of kirre/duf gene was identified in a group of Drosophila species, including D. pseudoobscura and D. simulans, and was found to have stretches of sequence divergence. These results support the view that modifications in the cis-regulatory sequence of kirre/duf could be responsible for different rates of myoblast fusion observed in these Drosophila species. However, further tests would be needed to determine whether constructs with kirre/duf from D. simulans and D. pseudoobscura inserted in D. melanogaster respond as expected by creating fibers with more or less myoblasts, respectively (Belu, 2011).

Variations in embryo size impose challenges to developing organisms, which must be overcome to ensure viability. In all species investigated in this study, some separated by several million years and others by only several thousand years, it was noted that alterations in mesodermal size were resolved by a common mechanism that increases or decreases the rate of myoblast fusion to generate the same stereotyped array of muscles. Since the variation in mesodermal domain and myoblast fusion rates occurred within very short evolutionary distances, these are fast evolving traits. Consistent with this view, there is evidence from the literature that genes belonging to the Toll and Dorsal/NFkappaB pathway, which participate in both immune response and D/V patterning, are fast evolving within twelve Drosophila species. This finding can be explained as adaptation to new pathogens found in the particular niches these species occupy. However, a recent comparison of the genomes of three melanogaster sister species identified components of the Dorsal/NFkappaB pathway that diverged the most in D. melanogaster, but the least in the pair D. simulans/D. sechellia, despite the fact that the latter two species do occupy completely different niches (i.e. one is cosmopolitan and the other is restricted to the plant Morinda, respectively). These data provide further evidence that D/V patterning itself, and not only immunity, evolves fast and point out to specific candidates in the Dorsal/NFkappaB pathway undergoing those changes (Belu, 2011).

Founder cells regulate fiber number but not fiber formation during adult myogenesis in Drosophila

During insect myogenesis, myoblasts are organized into a pre-pattern by specialized organizer cells. In the Drosophila embryo, these cells have been termed founder cells and play important roles in specifying muscle identity and in serving as targets for myoblast fusion. A group of adult muscles, the dorsal longitudinal (flight) muscles, DLMs, is patterned by persistent larval scaffolds; the second set, the dorso-ventral muscles, DVMs is patterned by mono-nucleate founder cells (FCs) that are much larger than the surrounding myoblasts. Both types of organizer cells express Dumbfounded, which is known to regulate fusion during embryonic myogenesis. The role of DVM founder cells as well as the DLM scaffolds was tested in genetic ablation studies using the UAS/Gal4 system of targeted transgene expression. In both cases, removal of organizer cells prior to fusion, causes formation of supernumerary fibers, suggesting that cells in the myoblast pool have the capacity to initiate fiber formation, which is normally inhibited by the organizers. In addition to the large DVM FCs, some (smaller) cells in the myoblast pool also express Dumbfounded. It is proposed that these cells are responsible for seeding supernumerary fibers, when DVM FCs are eliminated prior to fusion. When these cells are also eliminated, myogenesis fails to occur. In the second set of studies, targeted expression of constitutively active RasV12 also resulted in the appearance of supernumerary fibers. In this case, the original DVM FCs are present, suggesting alterations in cell fate. Taken together, these data suggest that DVM myoblasts are able to respond to cues other than the original founder cell, to initiate fusion and fiber formation. Thus, the role of the large DVM founder cells is to generate the correct number of fibers, but they are not required for fiber formation itself. Evidence is also presented that the DVM FCs may arise from the leg imaginal disc (Atreya, 2008).

Founder cells (FCs) play an important role during myogenesis as they usually represent a pre-pattern prior to the onset of myoblast fusion. In the Drosophila embryo, each of the 30 muscle fibers in a hemisegment is seeded by a single founder cell. A group of adult flight muscles in Drosophila, the dorso-ventral muscles (DVMs) is patterned by founder cells which unlike the embryo are much larger than surrounding myoblasts. These cells have been referred to as imaginal pioneers and express the embryonic founder cell marker Dumbfounded, as detected by reporter activity. Duf is known to serve as an attractant for myoblasts and to thereby promote fusion. This study examined a role for the DVM FCs in organizing fiber formation. First the FCs were eliminated prior to the onset of fusion through targeted expression of the cell death gene reaper. Removal of these cells does not abolish fiber formation, but rather, excessive numbers of thinner fibers are formed. This outcome resembles what is observed when larval scaffolds for a related group of muscles, the dorsal longitudinal muscles (DLMs), are ablated. These scaffolds also express Duf, and the outcomes of their genetic ablation served as a useful comparison of muscle patterning in a set of functionally related fibers. In examining DVM myogenesis closely, it was found that in addition to the large Duf-expressing FCs, some (smaller) cells in the myoblast pool also express Duf, and it is proposed that these are the alternate/replacement founder cells that can seed fiber formation in absence of the large FCs. Interestingly, supernumerary fibers do not develop when reaper expression is maintained through the period of myogenesis, indicating that Duf expression is necessary in the alternate/replacement founder cells for fiber formation. In a second set of experiments, the Ras-signaling pathway was constitutively activated in using the rP298-Gal4 driver, and this manipulation also resulted in an increase in fiber number. The increase was not due to proliferation of pre-existing founder cells, and the supernumerary fibers develop in presence of the existing FCs. These outcomes for ablations of DVM organizers suggest that (1) DVM fibers can arise in the absence of the large elongate FCs; (2) a subset of cells in the myoblast pool have the potential to initiate fiber formation; (3) signals from the FC normally suppress this capability, so that the correct pattern of muscle fibers can be generated; and (4) interfering with the signaling results in the formation of supernumerary fibers. Thus, the most critical role of the DVM FCs is to regulate fiber number (Atreya, 2008).

An important consequence of eliminating the DVM FCs is that multiple fibers are formed, suggesting that cells from the myoblast pool are recruited to initiate fusion. Formation of supernumerary fibers is also observed when larval scaffolds that pattern the functionally opposing muscle group, the DLMs are ablated, or eliminated using reaper. Thus, although the two muscle groups develop using different modes of development, the larval scaffolds and the single celled FCs serve a similar function-specification of the number of adult fibers. These organizers serve as a pre-pattern, partitioning myoblasts through attractive cues, and subsequently when fusion begins, the final muscle pattern begins to emerge. The size of the myoblast pool remains unaffected by the manipulation (targeted reaper expression) and this is borne out by the fact that muscle volume in the adult thorax is not altered. These outcomes are similar to what is observed after DLM scaffolds are genetically ablated or laser ablated (Fernandes and Keshishian, 1996).

If a role for the large DVM founder cells is to initiate fusion, how does it occur in their absence? Two possibilities are suggested: that myoblasts randomly fuse with each other, or that 'replacement' founder cells begin to seed fiber formation. If the first scenario were true, many more than the 6-7 supernumerary fibers would be expected. Also, it would be a departure from the well understood separation of 'founder' and 'fusion-competent' fates (reviewed in Taylor, 2003) and the normal tendency of myoblasts to only fuse with a founder/scaffold. The current results do suggest the recruitment of 'alternate/replacement' founder cells. The rP298 promoter is active in a small subset of cells in the myoblast pool as early as 6 h APF, as detected by using the rP298-Gal4/UAS mCD8GFP. These cells are the same size as other cells in the myoblast pool and thus much smaller than the DVM FCs. It is proposed that these smaller Duf-expressing cells are recruited to initiate fusion when FCs are eliminated. They are capable of sustaining fusion, as suggested by multiple EWG-positive nuclei within a supernumerary fiber. It was also observed that fiber formation is disrupted upon prolonged exposure to reaper, which is most likely due to elimination of the smaller Duf-expressing cells as well. Thus, it appears that Duf expression is necessary in these cells to generate supernumerary fibers (Atreya, 2008).

It is useful to compare another manipulation which also eliminates Duf-lacZ expressing FCs, but which has different outcomes with respect to fiber formation. When the mesothorax is denervated at the third larval instar, the expression of Dumbfounded in DVM FCs is gradually abolished during the period of fusion, 12-24 h APF. Although the size of the myoblast pool is unaffected by the denervation, and Duf-expressing FCs are present prior to fusion (0-12 h APF), DVM fibers fail to form. In light of the reaper studies, it is reasonable to propose that during the 0-12 h period, DVM FCs engage in a 'lateral inhibition' process that is sufficient to prevent cells of the myoblast pool from seeding fibers during the subsequent fusion phase. Moreover, just as innervation is needed to maintain Duf expression in the FCs during the fusion phase, it may also be necessary to maintain expression in the smaller 'replacement/alternate' founder cells (Atreya, 2008).

In considering a role for the smaller Duf-positive cells during normal development, it is proposed that a subset of cells in the myoblast pool have the capability to initiate fiber formation, and that interactions with the founder cell prior to fusion are responsible for active suppression of this 'organizer' competency. When the FCs are eliminated, it is conceivable that the suppression is incomplete and they can go on to seed fibers. Thus, the smaller Duf-expressing cells serve as a reserve pool of founder cells to initiate fusion when necessary. The suppression could involve signaling mechanisms such as those that have been described for the embryonic mesoderm, wherein an equivalence group is first defined, from which a single founder cell is then selected, and the potential of the remaining cells is inhibited (Atreya, 2008).

Supernumerary fibers can arise despite the presence of pre-existing rP298-lacZ positive FCs Overexpression of RasV12 with the rP298-Gal4 driver prior to the onset of fusion also results in the development of supernumerary fibers. Two additional phenotypes are seen compared to what is seen under conditions of reaper expression (1) the large rP298-lacZ expressing FCs are still present and (2) many more rP298-lacZ expressing nuclei are seen in the myoblast pool than in controls. The formation of additional rP298-lacZ expressing cells is considered first. The BrDU studies demonstrate that they do not arise by proliferation. Thus the cells are being generated from the myoblast pool due to a change in cell fate. Two possibilities are suggested: one that activation of the Ras/Map kinase pathway in FCs signals cells in the myoblast pool to turn on rP298, and the second and more intriguing possibility considers the small subset of cells that express the rP298-reporter at low levels. There may be additional cells that are below the level of detection with the lacZ reporter; if the Ras/Map kinase pathway is activated in several of these cells, it would similarly result in activation of the rP298 promoter, giving rise to the increased number of rP298-positive cells. The levels of RasV12 activated in these unfused myoblasts may be just enough to alter cell fate, conferring founder cell like properties. In the Drosophila embryo, when RasV12 is targeted to the embryonic mesoderm an overproduction of progenitor cells is seen, which is brought about as a result of a change in cell fate of mesodermal cells. Although RasV12 is also being targeted to the large DVM FC, it does not alter cell fate there since that cell is already expressing high levels of rP298 and it continues to seed fiber formation, as in controls (Atreya, 2008).

Ras regulates intracellular signaling through the ERK/MAP kinase pathway, and RasV12 is known to increase levels of activated MAPK within cells. A recent report on founder cell specification during adult abdominal myogenesis in Drosophila has shown that restricted activation of the Ras/Map kinase pathway by the FGF receptor Htl, in a few myoblasts reinforces founder cell properties in them, leading to the upregulation of rP298-lacZ levels, and the eventual loss of reporter expression in the other myoblasts. While expression of Htl has not been reported in IFM myoblasts, a similar mechanism that uses the ERK/MAPK pathway might be at play. It is also considered that overexpression of RasV12 causes death of the founder cells. But this is clearly not the case, since TUNEL labeling does not show any cell death, and the original founder cells are present and seed fiber formation (Atreya, 2008).

Expressing RasV12 in DLM scaffolds has two prominent outcomes. First, there is an increase in the number of rows of nuclei within the developing fibers. The increase in number of nuclei is suggestive of rapid fusion, and an outcome of enhanced Ras signaling. Another interesting aspect is in regard to muscle splitting. The six DLM fibers are generated as a collective outcome of the longitudinal splitting of the three larval scaffolds and fusion of myoblasts with these scaffolds. It is thought that as myoblasts fuse with the scaffolds there is an increase in surface area, which is manifested as splitting. Under conditions of RasV12 expression, there is an increase in the number of nuclei within each DLM fiber, indicative of rapid fusion. However, in 45% of animals, splitting does not proceed normally. It follows therefore, that fusion occurs too rapidly, and becomes uncoupled from a sustained incorporation of new membrane, which is then manifested as a lag in splitting. Under conditions of nerve ablation, a lag in muscle splitting is observed, but in that case, a reduced pool of myoblasts and slower fusions are responsible (Atreya, 2008).

Which of the known organizer cells do they resemble — grasshopper pioneer, larval scaffolds, or embryonic FCs? The cells that organize the DVM myoblasts share features that are common to embryonic founder cells and to grasshopper pioneers. Like the embryonic FCs and grasshopper pioneer cells, they serve as fusion targets and prefigure a muscle fiber. They bear a closer resemblance to grasshopper pioneers with respect to their large size including cellular expanse. However, they are unique with respect to the manner in which they interact with surrounding myoblasts, a feature that has been revealed as a result of manipulations carried out in the present study. They are not necessary for fiber formation, per se, as in their absence, a reserve pool of 'alternate/replacement' FCs is recruited. When these cells are removed as well, fiber formation fails and this scenario is the most similar to the outcome of targeting reaper to embryonic founder cells (Atreya, 2008).

A distinction between adult and embryonic myogenesis is exemplified by the manner in which founder cells are selected during abdominal myogenesis. rP298-lacZ expressing founder cells in the adult abdomen are specified by signaling through the FGF receptor, Heartless, which is under positive as well as negative regulation through Sprouty and Heartbroken. This is different from founder cell specification in the embryo which uses Notch-mediated lateral inhibition. This difference between embryonic and adult abdominal muscles is particularly striking since the two muscle groups are more similar to each other with respect to the occurrence of segmentally repeated patterns, a feature that is not present in the adult thorax. The manner in which founder cells of the adult thoracic muscles are specified is not known, but it does not include a Notch-mediated lateral inhibition (Atreya, 2008).

In considering how a myoblast pool is patterned, these studies thus far have shown that an important property of organizer cells that regulate DVM patterning, is the regulation of fiber number. This property is related to the expression of Duf, which is known to serve as an attractant for myoblasts. Thus, the pre-pattern of large, elongate rP298-lacZ positive cells enables myoblast segregation in a manner that the fidelity of fiber number is ensured. There is a scattering of smaller rP298-lacZ expressing adult myoblasts in the pool, and these have the potential to seed fiber formation. The manipulations have shown that the DVM FCs suppress these smaller myoblasts from seeding fibers, and that interfering with this capacity can result in the formation of supernumerary fibers being generated. The interference can take the form of eliminating FCs or by activating the Ras/Map kinase pathway. Interestingly, cues from the epidermis are able to guide the nascent supernumerary fibers to attach appropriately. The organizer cells are contacted by neurons prior to the onset of fusion and this is the basis for another property that is distinct from embryonic founder cells -- the nerve dependence of rP298-lacZ expression. The innervating neurons continue to exert an influence on myogenesis during the stages of fusion and proliferation as well (Atreya, 2008).

Future work will be aimed at examining the mechanisms by which the correct number of FCs is established, the signaling mechanisms underlying founder cell-myoblast interactions, and how supernumerary founder cells are normally prevented from arising in the myoblast pool (Atreya, 2008).

The bHLH transcription factor Hand is required for proper wing heart formation in Drosophila

The Hand basic helix-loop-helix transcription factors play an important role in the specification and patterning of various tissues in vertebrates and invertebrates. This study has investigated the function of Hand in the development of the Drosophila wing hearts which consist of somatic muscle cells as well as a mesodermally derived epithelium. Hand was found to be essential in both tissues for proper organ formation. Loss of Hand leads to a reduced number of cells in the mature organ and loss of wing heart functionality. In wing heart muscles Hand is required for the correct positioning of attachment sites, the parallel alignment of muscle cells, and the proper orientation of myofibrils. At the protein level, α-Spectrin and Dystroglycan are misdistributed suggesting a defect in the costameric network. Hand is also required for proper differentiation of the wing heart epithelium. Additionally, the handC-GFP reporter line is not active in the mutant suggesting an autoregulatory role of Hand in wing hearts. Finally, in a candidate-based RNAi mediated knock-down approach Daughterless and Nautilus were identified as potential dimerization partners of Hand in wing hearts (Togel, 2013).

In hand null mutants, wing hearts are formed but exhibit severe morphological defects resulting in loss of wing heart function. Consequently, almost all individuals display opaque wings and are unable to fly. Moreover, over time many of the mutant flies accumulate hemolymph in their wings. This long term effect occurs also very frequently in flies that completely lack wing hearts. During wing inflation, hemolymph is forced into the wings by elevated hemolymph pressure in the thorax which is effectuated by rhythmic contractions of the abdomen. However, this does not result in uncontrolled hemolymph accumulation as observed in animals lacking wing heart function since the epidermal cells of the wings still interconnect the opposing wing surfaces at this stage. Only after their delamination, more hemolymph may accumulate resulting in balloon-like wings which explains the long term character of this phenotype. However, since a rather large amount of hemolymph may accumulate in the wings some mechanism must exist that prevents backflow into the body cavity. In the tubular connection between wing and wing heart, a back-flow valve exists that prohibits hemolymph flow from the body cavity into the wing and thus is unsuitable to maintain a large amount of hemolymph inside the wing. In the region of the hinge, no valves are present and hemolymph may freely enter or leave the wings. It is therefore assumed that the apoptotic epidermal cells that remain in the wings due to loss of wing heart function form clots in the inflow and outflow tracts and thereby block hemolymph passage. Animals exhibiting these long term effects are probably affected in various ways. Most obviously, flies with filled wings have difficulties moving around and tend to fall during climbing. However, there are probably also physiological effects since the amount of hemolymph trapped in the wings must be lacking in the body cavity and should therefore affect internal hemolymph pressure as well as tissue homeostasis. Thus, it is proposed that the long term effects on wing morphology may contribute to the observed shortened life span of adult hand mutants (Togel, 2013).

Based on handC-GFP reporter activity, wing hearts express hand throughout their entire development and probably also during their mature state. However, the requirement for Hand seems only critical during early pupal stages at the time when the wing hearts are formed. Similarly, hand mutants display a phenotype in the adult only with regard to the heart and the midgut indicating that Hand is likewise required only during metamorphosis in these organs. In the adult heart, loss of hand leads to disorganized myofibrils, a phenotype that was also observed in mature wing heart muscles. Additionally, it was found that the attachment sites are less regular leading to a disruption of the dorso-ventral order of the muscle cells and loss of their parallel alignment. In many cases, muscle cells even form ectopic attachment sites in an area where they never occur in the wild-type. In an attempt to characterize the phenotype at the protein level, it was found that α-Spectrin and Dystroglycan are not properly distributed. Both proteins constitute components of the costameric network and are enriched at the membrane overlying Z-discs in the wild-type. In hand mutants, however, their pattern is altered to a more or less homogenous distribution at the membrane. In knock-downs of α- or β-Spectrin in the postsynaptic neuromuscular junction (NMJ), it was shown that spectrins are required for normal growth of NMJs and normal distribution of Dlg at the junctions. The enlarged NMJs visible in the Dlg staining, support the observation that α-Spectrin is misdistributed in hand mutants. The similar localization of α-Spectrin and Dystroglycan at the membrane raises the question whether their misdistribution in the mutant is somehow interconnected or independent of each other. Spectrins are organized in tetramers, consisting of two α/β-Spectrin heterodimers, which bind actin and are connected to the plasma membrane via Ankyrins. Ankyrins, in turn, have binding sites for Dystroglycan and E-Cadherin and together Ankyrin and the actin/spectrin network are thought to stabilize cell-cell and cell-matrix attachments. A hint that the misdistribution of α-Spectrin and Dystroglycan may be interconnected comes from observations in Dystrophin deficient mdx mice. There, Dystroglycan and β-Spectrin are both irregularly distributed but always co-localize. This let the authors conclude that their organization is coordinated. A possible explanation for the general loss of costamere organization may be that the costameric γ-Actin, although expressed normally, does not form a stable link between Z-discs and the membrane in mxd mice. However, loss of Dystrophin results mostly in the disruption of the linear arrangement of the proteins at the Z- and M-lines and does not lead to their homogenous distribution as observed in hand mutants. Nevertheless, the data obtained in this study and the phenotypic analysis of loss of function studies strongly suggest that the Spectrin and Dystroglycan phenotypes in hand mutants are interconnected caused by a, yet unknown, defect in the costameric network. Moreover, the misdistribution of these two proteins suggests that other proteins might be affected in a similar way including receptors required for directed outgrowth of muscle cells and proper targeting of tendon cells. This would explain why many muscle cells attach at improper positions or are misaligned. However, the cytoskeleton is not affected as a whole since the muscle cells still attain an elongated shape with attachment sites at their ends and a wild-typic βPS-Integrin pattern (Togel, 2013).

In the epithelium, loss of hand results in the failure of cells to integrate into the developing epithelium leading to gaps and the loss of cells. In mature organs, cells are predominantly missing in the area that dorsally extends the muscle cells suggesting that epithelial cells have greater difficulties attaching to their own type than to the muscle cells. On the protein level, it was found that Arm does not localize to the periphery of the cells except for small dot-like areas in the remaining filopodia-like cellular interconnections. Arm (β-Catenin) constitutes an intracellular adapter protein that links the transmembrane receptor E-Cadherin to actin filaments in adherens junctions. Adherens junctions are predominantly found between cells of the same type whereas Integrin based hemiadherens junctions connect to the ECM and additionally form specialized junctions between different cell types (e.g. myotendinous junctions). Based on the correct distribution of βPS-Integrin and the absence of Arm at the cell borders in hand mutants, it could be that the epithelial cells are able to form hemiadherens junctions towards the muscle cells but fail to establish a sufficient number of adherens junctions towards other epithelial cells. However, an alternative explanation would be that the formation of hemiadherens junctions is not affected and the remaining cells are simply too far apart to establish proper cell-cell contacts. Further experiments are needed to clarify this point (Togel, 2013).

It has been suggested that Hand proteins are involved in the inhibition of apoptosis based on the observation that loss of Hand function leads to hypoplasia and that block of apoptosis in the mutant background, at least partially, rescues the hand phenotype. This study observed a similar effect with respect to wing heart cell number. However, live cell imaging showed that also in the wild-type muscle cells are removed by apoptosis suggesting that this is a normal process during regulation of muscle cell number. Consequently, block of apoptosis in the controls led to an increase in muscle cell number. This suggests further that wing hearts in general have the potential to form more functional muscle cells. In hand mutants, the same removal of muscle cells occurs indicating that hand does not in general block apoptosis in wing hearts. Moreover, since the inhibition of apoptosis by P35 also affects the apoptosis involved in regulation of muscle cell number it cannot be excluded that the observed effect is actually induced hyperplasia in the mutant background mimicking a rescue instead of a real rescue of the hand phenotype. Additionally, live cell imaging showed that the cells of the wing heart epithelium forming the dorsal extension arise at their correct position in a sufficient number so that no gaps are visible. Only after they fail to establish proper cell–cell contacts they are removed from the wing hearts. It is therefore proposed that loss of cells by apoptosis in hand mutants is only a secondary effect caused by the inability of cells to integrate into the forming wing hearts (Togel, 2013).

It has been shown that Hand proteins can function as transcriptional activators in vertebrates and invertebrates. However, no direct targets have been identified in Drosophila so far. This study reports that the handC-GFP reporter line shows almost no activity in wing heart progenitors during postembryonic development suggesting that hand itself is a direct target of Hand. Remarkably, in some individuals a few nuclei of the wing hearts still show reporter activity indicating that hand is not the only transcription factor involved in postembryonic activation of the reporter. Moreover, the fact that the hand null mutant is not always a null with respect to reporter activation makes it a variable phenotype. Similarly, the severity of the phenotype observed in individual wing hearts (e.g. left and right side of the same animal) may differ considerably. So, how can a null mutation cause variable phenotypes? The answer may lie in the fact that all bHLH transcription factors need to form homo- or heterodimers for DNA binding. It was therefore proposed that the absence of a bHLH transcription factor not only affects its direct downstream targets but also the entire bHLH factor stoichiometry within the cell suggesting that the pool of bHLH dimers might be dynamically balanced. In the absence of Hand, new and presumably also artificial bHLH dimers are formed which consequently can cause a variety of delicate differentiation defects. The scenario is becoming even more complicated by the observation that the dimerization property of Hand is modulated by its phosphorylation state as well as by the finding that Hand can inhibit the dimerization of other transcription factors by blocking their protein interaction sites (Togel, 2013).

A crucial prerequisite for understanding the bHLH network in wing hearts is therefore the identification of dimerization partners. In a candidate based RNAi approach, two bHLH proteins, Da and Nau, were identified which evoke a phenotype very similar to the hand mutant. In order to verify the indication that these factors are interacting with Hand, Y2H analysis was applied and and an interaction between both Hand and Da as well as Hand and Nau was confirmed at the protein level. Furthermore, in vertebrates it was shown that these proteins are also able to form heterodimers with each other and that Hand is able to compete for heterodimer formation and DNA binding. Thus, based on Y2H interaction as well as phenotype similarity, the potential bHLH network in wing hearts likely includes Hand/Da and Hand/Nau heterodimers which activate different sets of downstream genes. In hand null mutants, the balance may be shifted to Da/Nau heterodimers or even Da/Da or Nau/Nau homodimers which may be able to activate some of Hand's target genes but with lower or higher efficiency. The competition of all these dimers with different transcriptional activation efficiency for the hand targets might explain the variations observed in the hand mutants (Togel, 2013).


Artero, R., et al. (2003). Notch and Ras signaling pathway effector genes expressed in fusion competent and founder cells during Drosophila myogenesis. Development 130: 6257-6272. PubMed ID: 14602676

Atreya, K. B. and Fernandes, J. J. (2008). Founder cells regulate fiber number but not fiber formation during adult myogenesis in Drosophila. Dev. Biol. 321(1): 123-40. PubMed ID: 18616937

Belu, M. and Mizutani, C. M. (2011). Variation in mesoderm specification across drosophilids is compensated by different rates of myoblast fusion during body wall musculature development. PLoS One 6(12): e28970. PubMed ID: 22194964

Busser, B. W., Taher, L., Kim, Y., Tansey, T., Bloom, M. J., Ovcharenko, I. and Michelson, A. M. (2012). A machine learning approach for identifying novel cell type-specific transcriptional regulators of myogenesis. PLoS Genet 8: e1002531. PubMed ID: 22412381

Colombani, J., et al. (2003). A nutrient sensor mechanism controls Drosophila growth. Cell 114: 739-749. PubMed ID: 14505573

de Velasco, B., Mandal, L., Mkrtchyan, M. and Hartenstein, V. (2006). Subdivision and developmental fate of the head mesoderm in Drosophila melanogaster. Dev. Genes Evol. 216(1): 39-51. PubMed ID: 16249873

Fernandes, J. J. and Keshishian, H. (1998). Nerve-muscle interactions during flight muscle development in Drosophila. Development 125: 1769-1779. PubMed ID: 9521914

Fernandes, J. J. and Keshishian, H. (1999). Development of the adult neuromuscular system, Int. Rev. Neurobiol. 43: 221-239. PubMed ID: 10218161

Fernandes, J. J. and Keshishian, H. (2005). Motoneurons regulate myoblast proliferation and patterning in Drosophila. Dev. Biol. 277(2): 493-505. PubMed ID: 15617689

Furlong, E. E. M., Andersen, E. C., Null, B., White, K. P. and Scott, M. P. (2001). Patterns of gene expression during Drosophila mesoderm development. Science 293: 1629-1633. PubMed ID: 11486054

Holz, A., et al. (2003). The two origins of hemocytes in Drosophila. Development 130: 4955-4962. PubMed ID: 12930778

Hudson, A. M., Petrella, L. N., Tanaka, A. J. and Cooley, L. (2008). Mononuclear muscle cells in Drosophila ovaries revealed by GFP protein traps. Dev. Biol. 314(2): 329-40. PubMed ID: 18199432

Jung, S. H., Evans, C. J., Uemura, C. and Banerjee, U. (2005). The Drosophila lymph gland as a developmental model of hematopoiesis. Development 132(11): 2521-33. PubMed ID: 15857916

Kamada, Y., Funakoshi, T., Shintani, T., Nagano, K., Ohsumi, M. and Ohsumi, Y. (2000). Tor-mediated induction of autophagy via an Apg1 protein kinase complex. J. Cell Biol. 150: 1507-1513. PubMed ID: 10995454

Kantorovitz, M. R., Kazemian, M., Kinston, S., Miranda-Saavedra, D., Zhu, Q., Robinson, G. E., Gottgens, B., Halfon, M. S. and Sinha, S. (2009). Motif-blind, genome-wide discovery of cis-regulatory modules in Drosophila and mouse. Dev Cell 17: 568-579. PubMed ID: 19853570

Kirisako, T., Baba, M., Ishihara, N., Miyazawa, K., Ohsumi, M., Yoshimori, T., Noda, T. and Ohsumi, Y. (1999). Formation process of autophagosome is traced with Apg8/Aut7p in yeast. J. Cell Biol. 147: 435-446. PubMed ID: 10525546

McMahon, A., Supatto, W., Fraser S. E. and Stathopoulos, A. (2008). Dynamic analyses of Drosophila gastrulation provide insights into collective cell migration. Science 322: 1546-1550. PubMed ID: 19056986

McMahon, A., Reeves, G. T., Supatto, W. and Stathopoulos, A. (2010). Mesoderm migration in Drosophila is a multi-step process requiring FGF signaling and integrin activity. Development 137(13): 2167-75. PubMed ID: 20530544

Riechmann, V., et al. (1998). The genetic control of the distinction between fat body and gonadal mesoderm in Drosophila. Development 125(4): 713-723. PubMed ID: 9435291

Rouault, H., Mazouni, K., Couturier, L., Hakim, V. and Schweisguth, F. (2010). Genome-wide identification of cis-regulatory motifs and modules underlying gene coregulation using statistics and phylogeny. Proc Natl Acad Sci U S A 107: 14615-14620. PubMed ID: 20671200

Scott, R. C., Schuldiner, O. and Neufeld, T. P. (2004). Role and regulation of starvation-induced autophagy in the Drosophila fat body. Dev Cell. 7: 167-178. PubMed ID: 15296714

Togel, M., Meyer, H., Lehmacher, C., Heinisch, J. J., Pass, G., Paululat, A. (2013). The bHLH transcription factor Hand is required for proper wing heart formation in Drosophila. Dev Biol 381: 446-459. PubMed ID: 23747982

Return: Genes expressed in mesoderm

Genes involved in tissue development

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

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