|
Gene name - jelly belly
Synonyms - Cytological map position - 48E9 Function - ligand |
Symbol - jeb
FlyBase ID: FBgn0086677 Genetic map position - 2- Classification - secretory signal sequence and LDL receptor repeat motif Cellular location - secreted |
Inductive interactions subdivide the Drosophila mesoderm into visceral, somatic, and heart muscle precursors. The muscle precursors form organs by executing tissue-specific migrations and cell fusions. jelly belly (jeb) is required for visceral mesoderm development. jeb encodes a secreted protein that contains an LDL receptor repeat. In jeb mutants, visceral mesoderm precursors form, but they fail to migrate or differentiate normally; no visceral muscles develop. Jeb protein is produced in somatic muscle precursors and taken up by visceral muscle precursors. jeb reveals a signaling process in which somatic muscle precursors support the proper migration and differentiation of visceral muscle cells. Later in embryogenesis, jeb is transcribed in neurons and Jeb protein is found in axons (Weiss, 2001).
A screen was performed to identify genes that are transcriptionally regulated by the homeodomain protein Tinman (Tin). Tin, a member of the NK family of homeodomain proteins, is required for organogenesis of the embryonic heart and visceral mesoderm. It is one of a number of transcription factors whose functions in mesoderm development are conserved from insects to mammals. The screening method relies on genetic selection in yeast for a protein-DNA interaction. A library was screened that represents 15% of the Drosophila genomic DNA and six DNA fragments were obtained that satisfied genetic criteria in yeast for Tin binding sites. Most of the genomic DNA fragments were isolated multiple times. Sequence analysis has confirmed the presence of core recognition sites for NK class homeodomains in all of the fragments. To show that these fragments function as Tin-responsive enhancers in vivo, it was asked if they could drive expression of a reporter gene in patterns consistent with Tin regulation (Weiss, 2001).
The screen is surprisingly specific for genes regulated by Tinman (or closely related genes), as demonstrated both by the reporter-construct results and the genes that are located adjacent to the Tinman binding sites. Four fragments identified in the screen were inserted upstream of a lacZ reporter. Three of the four reporter constructs, tested as transgenes, are active in patterns consistent with Tin regulation. One fragment lies adjacent to jelly belly (jeb), a gene expressed in ventral, early mesoderm. The Tin binding site that led to the identification of jeb contains two Tin/NK2 class homeodomain recognition sites oriented as an imperfect inverted repeat. This genomic fragment was mapped to interval 48E9 of polytene chromosome 2R by in situ hybridization and based on the Drosophila genome sequence. The Tin binding sites lie adjacent to a P element insertion within a large intron of the jeb gene (Weiss, 2001).
Analysis of the jeb mutant phenotype reveals that jeb is required for visceral mesoderm development, but not for somatic muscle, fat body, or hemocyte development. To understand how Jeb might function biochemically, it was determined where, within the mesoderm, jeb is expressed in relation to early visceral mesoderm. jeb is clearly expressed in ventral and medial mesoderm immediately adjacent to the visceral mesoderm cells that depend on Jeb function. The cells that express jeb are somatic muscle precursors. jeb mRNA is initially produced in clusters of cells ventral to clusters of bagpipe-expressing cells. At stage 10, jeb-expressing cells surround the visceral mesoderm and fill in the gaps between the clusters of bap expression. By mid stage 11, jeb- and bap-expressing cells lie in juxtaposed layers (Weiss, 2001).
The signal sequence and LDL receptor repeat predicted in Jeb protein imply that Jeb is secreted from somatic mesoderm precursor cells and acts in the extracellular compartment. Specific Jeb antisera were used to monitor a possible Jeb signal from somatic to visceral mesoderm precursors. The antisera do not stain embryos homozygous for the P element excision allele. bap-expressing visceral mesoderm precursor cells that are dependent on jeb function, but do not transcribe jeb, clearly contain Jeb protein (Weiss, 2001).
Jeb protein is secreted from tissue culture cells. Extracts of Drosophila tissue culture cells producing Jeb were compared to protein found in their medium. The bulk of the Jeb protein was found outside the cells. The secreted protein migrates as a broad band. Thus, Jeb protein is clearly detectable in the culture medium, evidently in a posttranslationally modified form (Weiss, 2001).
The P element that is integrated into the jeb locus interrupts the transcription unit in a large intron. Transcription of jeb upstream of the integration site should produce a protein of about 50 kDa. In mutant embryos, affinity-purified sera detect a truncated Jeb protein with an apparent molecular weight of 45 kDa. The predicted mutant protein would contain the secretory signal sequence but not the type A LDL receptor repeat. Antibody stains of jeb mutant embryos reveal two notable differences with respect to wild-type protein distribution: (1) the truncated, mutant protein accumulates to lower levels than wild-type protein; (2) visceral mesoderm precursors do not take up the truncated protein. The only detectable protein in mutant embryos is in or adjacent to the cells that make it. The type A LDL receptor repeat, missing from the mutant protein, thus appears to be necessary for Jeb function (Weiss, 2001).
The pattern of Jeb protein staining in the visceral mesoderm is qualitatively different from the staining observed in the Jeb-producing cells. It is exclusively punctate, in contrast to the diffuse staining observed in Jeb secreting cells. The punctate staining pattern suggests receptor-mediated endocytosis as a mechanism for Jeb accumulation in visceral mesoderm cells. To test this hypothesis, a temperature-sensitive allele of the gene shibire was used. shibire encodes a dynamin-related GTPase that is required for microtubule-mediated endocytosis. In shibire temperature-sensitive mutant embryos, raised at the nonpermissive temperature during the period of Jeb secretion and uptake, reduced or absent association of Jeb with the visceral mesoderm is found. This demonstrates that Shibire-mediated endocytosis is required for Jeb to accumulate in visceral mesoderm. It also suggests that a specific Jeb receptor may be required for uptake by the visceral mesoderm (Weiss, 2001).
Though Jeb protein is secreted from somatic muscle precursors and taken up by visceral muscle precursors, Jeb might act in somatic muscle precursors to produce a signal that is not Jeb. This possibility was ruled out by expressing Jeb in visceral muscle precursors in a jeb mutant background. Production of Jeb in the visceral mesoderm of mutants rescues early visceral mesoderm development. Robust Fas3 staining is restored in the visceral mesoderm of these rescued, mutant embryos. Despite the restoration of Fas3 production, subsequent visceral mesoderm migration is frequently abnormal. Longitudinal migration to form continuous bands is incomplete, resulting in gaps in the pattern of Fas3. Expression of Jeb in the visceral mesoderm is sufficient to rescue the differentiation and, to a lesser extent, migration, of visceral mesoderm precursors (Weiss, 2001).
The migration defect observed in the rescue experiment could mean that the normal location of the Jeb source conveys positional information to visceral muscle precursors. Consistent with this hypothesis, misexpression of jeb in the visceral mesoderm of jeb heterozygotes produces visceral mesoderm defects. Fas3 expression in these embryos is frequently discontinuous, in contrast to the linear expression in jeb heterozygotes in the absence of Jeb misexpression. These results show that jeb misexpression is sufficient to perturb the migration of visceral muscle precursors and support the model of Jeb functioning as a signal (Weiss, 2001).
The data are consistent with Jeb functioning primarily in visceral mesoderm migration, but it may also be required for visceral mesoderm differentiation. When jeb mutants were rescued by producing Jeb in discrete clusters of visceral mesoderm cells, local rescue of differentiation and subsequent gaps in the normally continuous, longitudinal bands of Fas3 expression were observed; this is presumably a defect in migration. This is consistent with ectopic jeb in discrete clusters of visceral mesoderm cells in a nonmutant embryo causing longitudinal gaps in the visceral mesoderm. The result is most readily explained if Jeb acts as a positive, positional cue for visceral mesoderm migration. An alternative would be that Jeb provides a permissive differentiation function necessary for migration (Weiss, 2001).
Jeb has a single LDL receptor repeat. LDL receptor repeats are found in several functional classes of proteins. One large class consists of a group of receptors and coreceptors (reviewed in Cooper, 1999; Tamai, 2000; Wehrli, 2000). All these proteins, many of which function cell autonomously in signaling systems, have transmembrane and intracellular domains. The absence of a transmembrane domain from Jeb, its non-cell-autonomous phenotype, and its translocation from synthesizing to responding cells argue against a similar receptor function for Jeb (Weiss, 2001).
Some secreted proteases and protease inhibitors contain LDL receptor repeats. The Drosophila protein Nudel, a secreted protease that carries out one step of a localized, signaling, protease cascade, contains an LDL receptor repeat that is highly related to the one in Jeb. Though Jeb has no apparent similarity to known proteases or protease inhibitors other than the type A LDL receptor repeat, it is possible that Jeb acts through a second, unknown signaling protein or protease (Weiss, 2001).
A mammalian protein, the 8D6 antigen, structurally resembles Jeb in that it is secreted and contains two LDL receptor repeats. 8D6 is synthesized in follicular dendritic cells of the immune system and stimulates germinal center B cell proliferation. 8D6 may function as a signal from follicular dendritic cells to B cells in immune responses (Li, 2000; Weiss, 2001).
One other well-characterized LDL receptor repeat-containing protein may be functionally related to Jeb -- the product of the C. elegans gene Mig-13, which, like Jeb, contains a single LDL receptor repeat (Sym, 1999). Structurally, Mig-13 differs from Jeb in that it contains both a CUB and a transmembrane domain not found in Jeb. Mig-13 function, however, resembles Jeb in two notable ways: (1) Mig-13 is required non-cell-autonomously, like Jeb; (2) Mig-13 is a positive migratory factor necessary for anterior migration of developing neurons in C. elegans, a function similar to Jeb's. Mig-13 is produced locally along the anterior-posterior body axis under the control of specific Hox genes, and appears to guide migrations in a concentration-dependent manner (Weiss, 2001).
Whether Jeb signaling is conserved in evolution is not simple to determine. Outside the LDL receptor repeat, a motif shared by a number of extracellular proteins, no unambiguous vertebrate Jeb homologs have been identified in the public databases. Either the LDL receptor repeat is the essential functional, and therefore conserved, portion of Jeb, or Jeb signaling is not widespread in the animal kingdom. The former hypothesis if favored because every known signaling system in Drosophila has also been found in vertebrates. The sequence of Sco-Spondin, a secreted protein, is significantly similar to Jeb (Gobron, 1996). Jeb signaling may therefore be an evolutionarily conserved process, a possibility that is now being investigated using the mouse Sco-spondin gene (Weiss, 2001).
The cDNA sequences and developmental RNA blots of jeb demonstrate two size classes of transcripts derived from the jeb locus during early to mid embryogenesis. Later in embryogenesis, a third, larger, transcript is detected. The two early embryonic transcripts contain the same open reading frame. They differ only in 5' and 3' untranslated regions. The predicted protein product of the jeb locus contains a secretory signal sequence and a single LDL receptor repeat motif. In the region of the LDL receptor repeat, Jeb is most similar to two bovine proteins, Sco-spondin and enterokinase (Weiss, 2001).
Currently jeb corresponds to two predicted genes of the Berkeley Drosophila genome project, CG13180 and CG13182. The LDL receptor repeat is found in CG13183.
date revised: 15 December 2001
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.