QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 4 May 2005
doi: 10.1242/dev.01838

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Serpe, M. Articles by O'Connor, M. B. Search for Related Content PubMed PubMed Citation Articles by Serpe, M. Articles by O'Connor, M. B.

Matching catalytic activity to developmental function: Tolloid-related processes Sog in order to help specify the posterior crossvein in the Drosophila wing

¹ Department of Genetics Cell Biology and Development, and the Developmental Biology Center, University of Minnesota and the Howard Hughes Medical Institute, Minneapolis, MN 55455, USA
² Department of Zoology, University of Wisconsin, 250 North Mills Street, Madison, WI 53706, USA

View larger version (20K): [in a new window]   Fig. 1. Biochemical characterization of the Tlr protein. (A) Western blotting analysis of the Tld-HA, Tlr-HA proteins and various modified forms. Shown are supernatants (s) and soluble cell fraction (c) from transfected S2 cells probed with anti-HA 12CA5 antibody. The bracket on the left indicates position of unprocessed and partially processed forms. The amount of secreted Tld is always low and the band is often smeared due to glycosylation (Marqués et al., 1997). (B) Comparison of Sog processing by Tld and Tlr, in the absence or presence of Dpp and Tsg. The indicated combinations of proteins were incubated for 16 hours at 25°C and Sog fragments were visualized with anti-Myc A14 antibody. Under these conditions, processing is incomplete, enabling the different fragments to be seen. The molecular weight of full-length Sog is 120 kDa and the C-terminally Myc tagged products shown are of 110 kDa, 50 kDa and 25 kDa. The 25 kDa band in the Tld lane is not visible with this level of exposure. (C) Schematic representation of the Tld and Tlr cleavage sites. The position of cleavage site II is indicated but this site is very weak for Tld (Shimmi et al., 2003) and undetectable for Tlr. Sog processing by Tlr requires an active, enzymatically intact astacin domain. (D) Equivalent amounts of wild-type and modified Tlr proteins were tested for their Sog processing activity by incubation for 40 hours at 25°C in the combinations indicated. Processing at site I only (full-length Sog versus the 110 kDa fragment) is shown. pmTlr, processed mutant Tlr; cdTlr, catalytically dead Tlr.

View larger version (40K): [in a new window]   Fig. 2. Tlr exhibits slow Sog processing kinetics compared with Tld. (A) Time course of Sog processing by Tlr and Tld. The amount of Tlr in the reactions was fivefold more than the amount of Tld, as estimated by quantitative fluorescence measurements of the anti-HA immunoreactivity of the equivalently C-tagged Tld-HA and Tlr-HA contained in the bracketed area. The Dpp (R&D Systems) concentration was 3 x10–10 M. The processing reactions were incubated at 25°C. Aliquots were removed at the indicated times, and were analyzed by immunoblotting with anti-Myc A14 antibody. The processing products were quantified using Odyssey Infrared Imaging System. The graphs below indicate the quantitation of the indicated bands at each time point. (B) Time course processing of Sog with increasing amounts of Tlr protein as indicated (10- and 25-fold excess when compared with Tld). (C) Time course of Sog processing in the presence of 1 x10–9 M Dpp and variable amounts of Tlr protein as indicated. (D) Time course assays in which the Tld and Tlr protease domains have been swapped. In TldastTlr the Tld astacin domain has been replaced with the Tlr astacin domain, while TlrastTld is the converse construct. (E) Sog co-immunoprecipitates with Tlr and catalytically inactive Tld variants but not with active Tld. Equivalent amounts of wild-type and modified Tld-HA and Tlr-HA proteins were mixed with Sog-Myc for 3 hours at room temperature followed by immunoprecipitation with the anti-HA 12CA5 antibody. Sog was detected with anti-Myc A14 antibody.

View larger version (29K): [in a new window]   Fig. 3. Sog processing is enhanced by a Dpp/Gbb heterodimer. (A) Western blot showing a representative sample of ligand produced by cells expressing either Dpp–HA alone (lane 1), Gbb (lane 3) or co-transfected Dpp-HA and Gbb (lane 2). The immunoblots were simultaneously probed for Dpp-HA with anti-HA 12CA5 antibody and Gbb with rat anti-Gbb C terminus antibody. The blots were scanned with Odyssey Infrared Imaging System. (B) Co-expression of Dpp-HA and Gbb-Flag leads to heterodimer formation. The top panel is probed for the presence of Dpp-HA, while the bottom panel is probed for Gbb-Flag using the anti Flag M2 antibody. Lanes 1-4 show the input levels of ligands present in the starting samples. Lanes 5-8 are immunoprecipitations with the anti-HA antibody. Lanes 9-12 show immunoprecipitated material using the anti-Flag antibody. The red arrows indicate co-immunoprecipitation of Dpp and Gbb which is only seen in the samples (lanes 7 and 11) where the two ligands where co-transfected into the same S2 cells and not in samples where homodimers where mixed together (lanes 8 and 12). The 25 kDa band (black arrowhead) present in all anti-Flag M2 immunoreactions is due to the mouse IgG light chain detached from the beads under our experimental conditions. (C) Sog processing by Tld and Tlr (16 hours at 25°C) in the presence of Dpp or Gbb homodimers was compared with the processing in the presence of Dpp and Gbb heterodimers (Dpp/Gbb). Equivalent amounts of Bmp ligands estimated by their immunoreactivity (A) were included in the processing reactions as indicated. (D) Percentage of remaining full-length Sog relative to the no-enzyme containing processing reaction (lane 1) was quantified using the Odyssey Infrared Imaging System (see Materials and methods).

View larger version (81K): [in a new window]   Fig. 4. Tlr suppresses Sog gain-of-function activities in vivo. (A) Injections of sog mRNA in one of the ventral four-cell stage blastomeres of Xenopus embryos induces secondary axis formation (red arrowheads) in tadpoles shown at development stages 19 and 28. Co-injection of tld and tlr inhibit Sog-mediated secondary axis formation. To ensure equimolecular amounts of mRNA, we used 2 ng capped mRNA for sog and tld and 4 ng for tlr, which had a transcript approximately twice as long as tld because of the longer pro-peptide and UTR sequences. (B) Wing phenotypes produced by ectopic expression of Sog and Tsg are suppressed by addition of activated Tld (atld) (UAS-tld27) or wild-type Tlr (tlr) (UAS-tlr3).

View larger version (64K): [in a new window]   Fig. 5. tlr is expressed in the pupal wings and is required for the Bmp signaling in the posterior crossvein. (A) Expression of tld and tlr in 30 hours AP pupal wings is compared by in situ hybridization with antisense tld and tlr DIG-labeled probes for the same length of time. Inserts show control hybridization of embryos at the early cellularization stages using the same probes. (B) Accumulation of pMad during the pupal wing development in wild-type and tlr mutant animals 19 hours after pupariation (AP), 25 hours AP and 30 hours AP. The bottom panel shows the corresponding adult wings, with the tlr mutant wing missing the posterior crossvein.

View larger version (64K): [in a new window]   Fig. 6. The loss of PCV in tlr mutants is rescued by reducing endogenous Sog levels or by overexpressing Tlr, but not by overexpressing Tld. (A) sog heteroallelic combinations of the indicated genotype restores the PCV structure in tlr mutant animals. (B) Overexpression of UAS-tlr(2) with da>Gal4 driver rescues Bmp signaling in the PCV of developing tlr mutant pupal wings, as well as the PCV in the adult wings. (C) Overexpression of UAS-tld(16) with da>Gal4 driver does not rescue the PCV in the tlr mutant escapers.

View larger version (75K): [in a new window]   Fig. 7. Reduction of Sog levels perturbs the PCV formation. (A) Overexpression of UAS-tld(16) with da>Gal4 driver causes expanded pMad in the PCV region and expanded PCV material and detachment of PCV from L5. (B) Overexpression of UAS-tld(16) with en>Gal4 in a wild-type background causes widening or absence of PCV. (C) sogYL26/sogP129D results in detached PCV. (D,D') In the stronger allelic combination sogP1/sogP129D, partial to more complete loss of PCV structures is observed in a larger percentage of animals (see text for numbers). (E) Overexpression of two copies of UAS-tlr (X) with A9>Gal4 does not affect PCV formation. (F) Overexpression of one copy of UAS-tld(16) with A9>Gal4 driver produces detached PCV in 16% of the animals (n=102). (G) Overexpression of UAS-tld(16) and UAS-tlr(X) with A9>Gal4 driver produces a range of crossvein defects, from detached PCV shown to almost complete loss of PCV structures in 82% (n=192) of the adults.