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

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 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 Laporte, J. Articles by Mandel, J.-L. Search for Related Content PubMed PubMed Citation Articles by Laporte, J. Articles by Mandel, J.-L.

The PtdIns3P phosphatase myotubularin is a cytoplasmic protein that also localizes to Rac1-inducible plasma membrane ruffles

Jocelyn Laporte, Francois Blondeau^*, Anne Gansmuller, Yves Lutz, Jean-Luc Vonesch and Jean-Louis Mandel

Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, 1 rue Laurent Fries, BP 10142, 67404 Illkirch Cedex, CU de Strasbourg, France
^* Present address: McGill Cancer Centre, McGill University, 3655 Promenade Sir William Osler, McIntyre Medical Sciences Building, Room 702, Montreal, Québec, Canada H3G 1Y6
Present address: INSERM U 338, Centre de Neurochimie, 5 rue Blaise Pascal, 67084 Strasbourg, France

View larger version (41K): [in a new window]   Fig. 1. Characterization of myotubularin antibodies. (A) `Structural epitope' means that the antibodies recognized only the full-length myotubularin but not overlapping fragments. Western blotting and immunocytochemistry results are from transfected cells as detection of endogenous myotubularin was unsuccessful with these two methods. Immunoprecipitation data were obtained for the endogenous myotubularin from muscle cells and are described elsewhere (Laporte et al., 2001b). +, the antibody is working with the corresponding technique; -, no signal has been detected. At least 2D2, 1G1 and R1208 crossreact with mouse myotubularin, while 1G6, 1D10, R929 and R1141 do not. 1C7 does not immunoprecipitate the mouse myotubularin. R1208 crossreacts with MTMR1 while none of these antibodies crossreact with hMTMR2 and hMTMR3 proteins. (B) Example of immunoprecipitant antibodies crossreacting with the endogenous mouse mMTM1 myotubularin. Mouse C2C12 myotube protein extract or buffer (/) were immunoprecipitated (IP) with the listed antibodies and the purified mouse myotubularin was detected by western blot with the 2D2 antibody (1/2000) followed by an anti-Kappa light chain (1/2500). Myotubularin has an estimated molecular weight of 70 kDa compared with size markers. Transfected COS cells with human myotubularin serve as a size control on the left. R1203 is a serum from a different rabbit immunized as for R1208. The dog myotubularin was also immunoprecipitated and detected with the 1G1 and 2D2 monoclonal antibodies respectively (not shown).

View larger version (50K): [in a new window]   Fig. 2. Tissue expression and isoforms of myotubularin. (A) Three mg of total protein extracts from different tissues of a 14-week-old mouse were immunoprecipitated with 1G1 antibody and loaded on a 8% SDS-PAGE gel. Myotubularin was detected after western blotting with the 2D2 antibody (1/2000). Below, a longer migration reveals the presence of a doublet specific to muscle (and heart, not shown). Different percentages of acrylamide and gradients tested did not resolve the two bands any better. The arrow indicates the migration of the 603 amino acid myotubularin construct transfected into cells, representing the common isoform. (B) The muscle-specific isoform appears after myotube formation. C2C12 mouse myoblasts were differentiated at confluence in DMEM+5% FCS into myotubes, and myotubes were stimulated with insulin like growth factor (IGF1 from Calbiochem at 25 ng/ml) as indicated. Protein extracts were prepared at different time points during differentiation and compared with adult muscle and liver tissues.

View larger version (46K): [in a new window]   Fig. 3. Myotubularin is cytoplasmic and associated with plasma membrane. (A) Wild-type untagged myotubularin overexpressed in COS cells was detected with the specific 1G6 monoclonal antibody (1/1000) followed by Cy3-conjugated goat anti-mouse antibody (1/300). Confocal microscopy analysis does not show any nuclear signal. The inactive C375S mutant was also localized as a dense cytoplasmic network in different cell types (COS, HeLa, 3T3 fibroblasts, myoblasts and myotubes). Also note the labeling of plasma membrane. Preimmune serum or immunizing peptide competition (100 µg, 30 minutes, room temperature) abolished the signal. (B,C) Localization in transfected C2C12 mouse myoblasts and myotubes respectively. (D) Transfected COS cell showing altered cell shape and presence of myotubularin in extended filopodia. This pattern was also observed with inactive myotubularin mutants.

View larger version (104K): [in a new window]   Fig. 7. Effect of myotubularin on endocytic markers and PtdIns3P distribution. (A) COS cells transfected with wild-type myotubularin do not show changes in endocytic markers such as dynamin (clathrin-coated vesicles), caveolin 1 (uncoated vesicles), and Rab5 (endosomes). Note the strong labeling of a membrane ruffle with anti-myotubularin antibody in the caveolin 1 co-staining. (B) COS cells co-transfected with wild-type or C-ter delSID myotubularin constructs and a myc-tagged 2XFYVE expression construct show no co-localization and no effect of myotubularin with the endosomal staining by 2XFYVE.

View larger version (45K): [in a new window]   Fig. 8. Myotubularin and plasma membrane remodeling. (A) HeLa cells transfected with either wild-type or D278A substrate-trap mutant were left untreated (-) or were treated with cytochalasin D at 400 µM for 2 hours (+). Co-localization with endogenous actin showed disruption of the actin filaments, while both myotubularin constructs still showed labeling of the plasma membrane. (B) D278A mutant (in green) transfected in HeLa cells showed no co-localization with focal adhesion labeled by an anti-vinculin antibody (in red). (C) Co-localization of different myotubularin constructs with Rac1-induced plasma membrane ruffles in COS cells co-transfected with constitutively activated Rac1 V12 (flag-tagged). Wild-type myotubularin and its N-terminal part localized to the ruffles while the del(233-237) construct did not.

View larger version (56K): [in a new window]   Fig. 4. Subcellular distribution of endogenous myotubularin in (A) lymphoblasts from normal (L1421) or an XLMTM patient deleted for exons 1-13 of the hMTM1 gene (89-441), (B) HeLa cells and (C) mouse myotubes. Subcellular fractions were prepared as described in Materials and Methods and enrichment was confirmed under the microscope (example below the western blot in A) and with protein markers (tubulin is present in the same fraction as myotubularin). Myotubularin was immunoprecipitated from the different fractions with 1G1 antibody and detected on a western blot by 1G6 (1/10,000) for the human cell lines and 2D2 (1/2000) for the mouse myotubes. P1, nucleus; P2, big organelles; S1, cytoplasm and all organelles; S2, cytoplasm and small organelles; T, total extract; TR, myotubularin overexpressed in COS cells. (D) Cytoskeleton fractionation of HeLa cells transfected with wild-type (WT) and substrate-trap mutant (D278A) myotubularin. Actin-based microfilaments and intermediate filaments containing vimentin were recovered in the cytoskeletal (P) fraction, whereas actin monomers and tubulin were recovered in the cytosolic (S) fraction. (E) The same cytoskeleton fractionation applied to mouse C2C12 myotubes. Myotubularin was immunoprecipitated and detected as above.

View larger version (30K): [in a new window]   Fig. 5. Effect of protein domains and XLMTM mutations on the subcellular localization of myotubularin. (A) Schematic representation of myotubularin showing protein domains. GRAM, glucosyltransferase, Rab-like GTPase activator and myotubularin common domain (Doerks et al., 2000) (aa 29-97); RID, Rac1-induced localization to membrane ruffles (around aa 233-237); PTP, tyrosine phosphatase-like signature implicated in the lipid phosphatase activity (aa 371-385) with catalytic residues D278, C375 and R381; SID, SET-interacting domain (aa 435-486); PEST, (aa 581-598) with a significant PESTfind score of +8.23; PDZ-BS, putative PDZ-binding site functional in the hMTMR1 homolog (Fabre et al., 2000) (aa 599-603). Highly conserved regions through evolution correspond to a high frequency of missense mutations in XLMTM patients and are also indicated (aa 170-330, 45% identity with the S. cerevisiae protein; and aa 370-490, 55% identity with the S. cerevisiae protein). Below are indicated the subcellular localization and the Rac1-induced localization to membrane ruffles for some constructs. The subcellular localization of the depicted constructs were obtained from immunofluorescence experiments with either the N-terminal 1G6 or the C-terminal 1D10 anti-myotubularin antibodies on transfected COS and HeLa cells. Other constructs produced unstable proteins (aggregates in the cytoplasm and near the nucleus probably in the Golgi and very low myotubularin levels on western blot): del(1-95), del(97-122), del(183-245), del(224-245), del(308-325), del(396-406), del(437-469), del(482-494) and amino acid changes G378R, D394A, G402A, E404K, E410A, D443A, C444Y and H469P. Mutation of the conserved aspartate at position 257 (D257A) did not affect the localization of myotubularin. Note that missense mutations affecting R241 (mild phenotype) impaired the in vitro enzymatic activity toward PtdIns3P (Taylor et al., 2000) and lead to a decrease in protein level in a patient cell line (Laporte et al., 2001b), while mutations G378R (severe phenotype) impaired the in vitro enzymatic activity and G402A (probably severe) leads to a decrease in protein level in a patient cell line. (B) Confocal microscopy analysis of a truncated myotubularin shows nuclear localization (C-ter construct). Deletion of the SET-interacting domain from this construct abolished the nuclear localization. (C) Co-localization of the C-ter construct with PML in more than 50% of co-transfected cells (confocal microscopy). In the same experiment, other co-transfected cells showed no obvious co-localization.

View larger version (43K): [in a new window]   Fig. 6. Turn-over of myotubularin. COS cells transfected with wild-type myotubularin were labeled with [35S]methionine/cysteine and cold chased for 0, 15 or 30 minutes and 1, 2, 3 or 5 hours. Myotubularin was immunoprecipitated with the 1G1 monoclonal antibody. The period of cold chase is indicated above the lanes. Myotubularin is indicated by an arrow. Above, an additional protein is trapped probably by the beads as it is also present in the first lane without immunoprecipitant antibody. The estimated half-life is about 4 hours.