Unformatted text preview:

Vol 441 8 June 2006 doi 10 1038 nature04829 LETTERS Social interactions among epithelial cells during tracheal branching morphogenesis Amin S Ghabrial1 Mark A Krasnow1 Many organs are composed of tubular networks that arise by branching morphogenesis in which cells bud from an epithelium and organize into a tube1 3 Fibroblast growth factors FGFs and other signalling molecules have been shown to guide branch budding and outgrowth4 7 but it is not known how epithelial cells coordinate their movements and morphogenesis Here we use genetic mosaic analysis in Drosophila melanogaster to show that there are two functionally distinct classes of cells in budding tracheal branches cells at the tip that respond directly to Branchless FGF and lead branch outgrowth and trailing cells that receive a secondary signal to follow the lead cells and form a tube These roles are not pre specified rather there is competition between cells such that those with the highest FGF receptor activity take the lead positions whereas those with less FGF receptor activity assume subsidiary positions and form the branch stalk Competition appears to involve Notch mediated lateral inhibition that prevents extra cells from assuming the lead There may also be cooperation between budding cells because in a mosaic epithelium cells that cannot respond to the chemoattractant or respond only poorly allow other cells in the epithelium to move ahead of them The Drosophila tracheal system develops from epithelial sacs of about 80 cells from which primary secondary and terminal branches sprout without cell division or cell death8 9 Primary branch sprouting is induced by Branchless Bnl FGF a chemoattractant secreted by clusters of cells surrounding each sac4 Fig 1a which activates Breathless Btl FGF receptor FGFR a receptor tyrosine kinase expressed on tracheal cells10 Primary branches contain 3 20 cells that organize into a tube as they migrate out from the sac Fig 1b Bnl also induces the expression of secondary branching genes such as the transcription factor pointed pnt 11 and specifies terminal cells at the ends of outgrowing branches4 8 Terminal cells ramify in the larva in response to Bnl to form fine terminal branches12 Other cells at the ends of primary branches become fusion cells that connect with neighbouring branches to form a continuous tracheal network Terminal and fusion cell fate decisions are also influenced by the Notch Dpp and Wingless signalling pathways13 17 Dorsal branches the primary branches that we focus on here typically consist of five or six cells two cells near the branch tip one DB1 that becomes a terminal cell and another DB2 that becomes a fusion cell and three or four cells DB3 DB6 that form the branch stalk Fig 1a b In a genetic mosaic screen A S G B P Levi and M A K unpublished observations six mutants 724 788 1118 1187 1476 and 1684 were identified with a subtle phenotype mosaic branches 2 2 2 cells were grossly normal yet homozygous mutant clones 2 2 cells rarely if ever included terminal cells Fig 1c e and Supplementary Tables S1 S4 data not shown for other primary branches These were neither general nor terminalcell specific lethal mutations because homozygous mutant cells were readily recovered in all other tracheal positions and there was no decrease in the overall number of cells in mosaic dorsal branches 1 5 3 1 1 mean s d versus 4 9 1 1 in contralateral control branches n 22 pairs of branches or the number of terminal cells 98 of both wild type n 127 and 724 or 788 mosaic dorsal branches n 291 had a terminal cell It was difficult to imagine how mutations could block clone generation in specific cells It seemed more likely that the mutations caused cells otherwise destined to become terminal cells to switch fates with other tracheal cells The six mutations compose a single lethal complementation group that mapped to the left arm of chromosome 3 and failed to complement breathless LG18 DNA sequencing identified a single nucleotide change in each mutant resulting in a nonsense or missense mutation in btl Fig 2a Five mutations 724 788 1118 1476 and 1684 appear to be null btl mutations whereas the sixth mutation 1187 causes partial loss of function Fig 2b and data not shown Thus the no mutant terminal cells gene is btl We quantified the distribution of cells homozygous for btl null mutations 724 and 788 or homozygous for a wild type btl allele as a control in mosaic dorsal branches Supplementary Tables S1 S4 Control clones were evenly distributed throughout the branch at the expected frequencies for example the ratio of stalk cell to terminalcell clones was about 3 1 By contrast btl 2 2 cells showed a nearly complete bias against the DB1 position the ratio of stalk cell to terminal cell clones was 51 1 The three exceptional mutant terminal cells may be cases in which the clone was induced after btl began to be expressed allowing wild type btl gene products to perdure in mutant cells We recovered hundreds of mosaic branches with one or more btl 2 2 cells present in positions DB2 DB6 without affecting cell or branch morphology Indeed branches composed largely or exclusively of btl 2 2 cells except for a wild type terminal cell were morphologically indistinguishable from wild type branches Fig 2c and Supplementary Tables S3 and S4 Thus although all tracheal cells normally express btl and the receptor is activated by Bnl in most or all cells of budding branches8 18 the receptor appears to be required in just a single leading cell DB1 All other cells can migrate normally and form tubes in the absence of btl We conclude that there are two functionally distinct classes of cells in budding primary branches lead cells which require Btl FGFR and directly respond to Bnl FGF and trailing cells which do not require Btl but follow the lead cell and form the stalk What does it take to become the leader The lead cell DB1 is specified to become a terminal cell by Bnl Btl signalling4 8 If terminal cell specification is required then null mutations in the downstream gene pnt which abolish this function8 should have the same effect as btl mutations Cell clones homozygous for pnt D88 or two new pnt alleles isolated in our screen 198 and 1318 failed to develop as terminal cells as expected However unlike btl mosaic branches pnt mosaic branches often lacked a terminal cell 26 of mosaic branches n 97 When a terminal cell was missing there was usually about 80 of the time a pnt 2 2 cell in the stalk position


View Full Document

UNC-Chapel Hill GNET 624 - Study Guide

Download Study Guide
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Study Guide and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Study Guide 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?