Supplementary Online Information GTP-dependent twisting of dynamin implicates constriction and tension in membrane fission

Transcription

1 1 Materials & Methods Materials All lipids were purchased from Avanti Polar Lipids Inc., USA ( Two lipid preparations were used. Brain Polar Lipids (BPL)(Avanti product # ), and a Synthetic Lipid Mixture (SLM) composed of 85 % (m/m) phospholipids [30% brain phosphatidylethanolamine (PE), 5% liver phosphoinositides (PI), 30% palmitoyl-oleyl phosphatidylserine (POPS) and 35% palmitoyl-oleyl phosphatidylcholine (POPC)], and 15% (m/m) cholesterol. In some experiments (as noted in figure legends), POPS was 50% and POPC 15%. Each preparation was supplemented with 5% (m/m) final phosphatidylinositol 4,5 bisphosphate [PtdIns(4,5)P 2 ]). Nucleotides were obtained from Boehringer-Mannheim or Sigma (GDP). Dynamin purification and labelling Dynamin was purified from 6 rat brains using the GST-tagged SH3 domain of rat amphiphysin 2 as an affinity ligand 1,2. After elution with low ph and salt, the two fractions most enriched in dynamin were pooled (2 ml total), dialysed against storage buffer (20 mm HEPES ph 7.4, 100 mm NaCl, 50% v/v glycerol)(final volume about 0.5 ml, typical concentration approximately 2 mg/ml), flash-frozen in liquid nitrogen and stored at 80 C. This material contained small amount of synaptojanin 3, but results obtained with this material were confirmed by selected experiments using purified baculovirus expressed human dynamin 1. Before any labelling with either Alexa 488 or biotin, dynamin was dialysed against PBS in 50% glycerol. Conjugation of dynamin to Alexa 488 was carried out with the protein labeling kit from Molecular Probes Inc., USA (product # A10235). For conjugation to biotin, 1.2 mg of Sulfo-NHS-LC-Biotin (product # 21335, from Pierce Inc.) was dissolved into ~0.5 ml of dynamin containing solution. After one hour under agitation at room temperature, the reaction was stopped by adding 0.5 ml of 100 mm glycine in PBS. Labeled dynamin was dialysed against storage buffer, aliquoted, flash-frozen and stored at 80 C. Conjugation to Alexa 488 did not appear to affect the property of dynamin in our in vitro assays. Biotinylation, however, partially impaired tubulation. Hence biotinylated dynamin was always diluted into unlabeled dynamin at a 1:5 ratio or more. Formation of membrane sheets 22x40 mm glass coverslips were cleaned by sonicating them (5 min) in 1% 7X (MP biomed., Germany) in distilled water followed by thorough washing and sonication (5 min) in distilled water to remove any trace of detergent and a final wash with 100% ethanol prior to storage in ethanol. Coverslip were dried under a N 2 flux, and 1 µl droplets of lipid solution (10 mg/ml in pure chloroform) were deposited and allowed to dry on the coverslip. Typically, two drops were deposited at different sites on a same coverslips (see Fig. 2 in 4 ). The use of pure chloroform was essential to allow lipid droplet drying in a way that was optimal for the subsequent formation of membrane sheets upon hydration. Coverslips were then dried again under vacuum (0.2 milli-torr) for at least one hour, and kept up to several days under vacuum. Before use, coverslips were placed for min in a cell incubator (37 C, 10% CO 2, 100% humidity) to allow partial hydration of the lipids. Next, a small chamber (approximately 15 µl volume) was built by placing the coverslip onto a glass slide, with the lipids facing the glass

2 2 slide, using a double-sided Scotch (3M) tape as a spacer (see Fig. 2 in 4 ). The lipids were fully rehydrated by applying to the side of the chamber µl of GTPase buffer (20 mm HEPES ph 7.4, 100 mm NaCl, 1 mm MgCl 2 ) containing 0.1 mg/ml casein (C7078, Sigma, USA)(casein buffer) which were taken up into the chamber by capillarity. Rehydration generated membrane sheets, as previously described (see also 4-6 ). The glass slide was placed on the stage of an Axiovert 200 ZEISS (Germany) microscope for observation with a JAI Pulnix (USA) TM1400CL camera and DVR software (Advanced Digital Vision Inc. USA). 5 µl of a dynamin containing solution were applied to one side of the chamber and the deformation of membrane sheets produced by its diffusion into the chamber was recorded at normal video rate (30 fps) with high resolution (1300x1024) imaging under Differential Interference Contrast (DIC) settings. 5µl of nucleotides containing casein buffer were added after formation of the tubes. In experiments involving streptavidin-coated polystyrene beads (260 nm diameter, Bangs Labs, USA), biotinylated dynamin was used and the dynamin solution also contained beads at an approximately fold dilution relative to the commercial stock solution. For the experiments, only tubules not adherent to the glass surface throughout their length were selected for observation. For experiment involving COOH-coated beads (negatively charged at neutral ph, 320 nm diameter, Bangs Labs, USA), beads were diluted fold in the dynamin solution prior to injection. Kinesin-induced lipid tubulation Microtubule-coated chambers between glass-slides were prepared as described 7, but using double sided tape as the spacer and washed with kinesin motility buffer (see below). Giant liposomes were generated using modifications of the protocol from 8. One mg of BPL supplemented with 5% PtdIns(4,5)P 2 and 5% biotin-lc-dope lipid (Avanti Polar Lipids, USA) was dried in a glass vial under vacuum for at least half an hour. Lipids were re-suspended by vortexing for 5 min in 1 ml GTPase buffer containing 1% v/v glycerol. Small unilamellar vesicles were obtained by sonicating the solution on ice with a tip sonicator for 5 min (0.3 Hz cycles) and stored at 20 C. For the experiments, 5 µl of solution were spotted and dried on a glass slide in the vacuum oven for half an hour. Giant liposomes were obtained by re-hydrating the lipid spot for min with 10 µl 220 mm sucrose. These liposomes were then carefully aspirated from the glass slide surface with a micropipette, transferred to an Eppendorf tube, clarified by mixing with GTPase buffer (100 µl final volume) and spun at 800 rpm for 2 min. The top 50 µl of the solution were replaced by fresh GTPase buffer and, after resuspension, this material was further spun at 800 rpm for 2 min. The top 75 µl were discarded and the remaining material (giant liposomes) was used as follows as starting material for the generation of kinesin-pulled tubules (see also 9,10 ). 3 µl of giant liposomes were mixed with 2 µl of 0.1 mg/ml streptavidin in GTPase buffer and incubated for 3 min. Two µl of biotinylated recombinant drosophila kinesin (typically from a 1 µm stock, cdna for was kind gift of Patricia Bassereau, Institut Curie, Paris) was then added to this mixture, incubated for 5 min and added to 7 µl of motility buffer (2 mm ATP, 0.4 mg/ml glucose oxidase, 0.2 mg/ml catalase, 240 mm glucose, 5 mm DTT, 10 µm taxol, in GTPase Buffer). The resulting material was added to the microtubule-coated microchambers after their prewashing with 20 µl of motility buffer. After a min incubation to allow tubules to grow, 5 µl of Alexa 488-dynamin in GTPase buffer were applied to the chamber. Following an additional incubation, 5 µl 1mM GTP in GTPase buffer were applied.

3 3 Movie processing and compression Uncompressed DIC movies (AVI files) were resized, contrasted and accelerated using the virtualdub freeware ( Raw movies were compressed using the DivX codec ( to ensure good quality compression for data storage. For the analysis of beads movement, movies were contrasted using virtualdub, transformed to 8 bit grayscale stack (.stk) files using the ImageJ freeware, and the spinning beads were tracked using the Tracking Function of the Metamorph software (Molecular Devices Corp., USA). For each concentration of GTP, the average maximal angular speed was calculated from the maximal speeds (which in turn is an average of at least three turns) of 15 to 30 beads. For the 40 µm GTP concentration, where a high proportion of beads did not move, a fraction of the maximum speeds used for the statistics were actually zero. Electron microscopy Negative staining electron microscopy was performed according to 11,12. Bibliography 1. Stowell, M. H., Marks, B., Wigge, P. & McMahon, H. T. Nucleotide-dependent conformational changes in dynamin: evidence for a mechanochemical molecular spring. Nat Cell Biol 1, (1999). 2. David, C., McPherson, P. S., Mundigl, O. & de Camilli, P. A role of amphiphysin in synaptic vesicle endocytosis suggested by its binding to dynamin in nerve terminals. Proc Natl Acad Sci U S A 93, (1996). 3. Farsad, K. et al. Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J Cell Biol 155, (2001). 4. Itoh, T. et al. Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. Dev Cell, in press (2005). 5. Suzuki, K. & Masuhara, H. Growth of giant membrane lobes mechanically driven by wetting fronts of phospholipid membranes at water-solid interfaces. Langmuir 21, (2005). 6. Tsafrir, I. et al. Pearling Instabilities of Membrane Tubes with Anchored Polymers. Phys Rev Lett 86, (2001). 7. Roux, A. et al. A minimal system allowing tubulation with molecular motors pulling on giant liposomes. Proc Natl Acad Sci U S A 99, (2002). 8. Moscho, A., Orwar, O., Chiu, D. T., Modi, B. P. & Zare, R. N. Rapid preparation of giant unilamellar vesicles. Proc Natl Acad Sci U S A 93, (1996). 9. Leduc, C. et al. Cooperative extraction of membrane nanotubes by molecular motors. Proc Natl Acad Sci U S A 101, (2004). 10. Koster, G., VanDuijn, M., Hofs, B. & Dogterom, M. Membrane tube formation from giant vesicles by dynamic association of motor proteins. Proc Natl Acad Sci U S A 100, (2003). 11. Takei, K., Slepnev, V. I., Haucke, V. & De Camilli, P. Functional partnership between amphiphysin and dynamin in clathrin- mediated endocytosis. Nat Cell Biol 1, 33-9 (1999). 12. Danino, D., Moon, K. H. & Hinshaw, J. E. Rapid constriction of lipid bilayers by the mechanochemical enzyme dynamin. J Struct Biol 147, (2004).

4 4 Legends for movies Movie S1 Growth of dynamin coated tubules on sheets of lipid membranes observed by DIC microscopy. Same field as for Fig. 1a. Speed, 30X. Movie S2 Effect of 1 mm GTP on a network of dynamin-coated lipid tubules. Same field as for Fig. 2a. Real-time (15 fps) DIC microscopy. Movie S3 Effect of 1 mm GTP on a dynamin-coated lipid tubule anchored at static point at both ends. Same field as for Fig. 2b. Real-time (15 fps) DIC microscopy. Movie S4 Effect of 1 mm GTP on a dynamin-coated lipid tubule free to retract. Same field as for Fig. 2c. Real-time (15 fps) DIC microscopy. Movie S5 Effect of 1 mm GTP, in the absence of ATP, on membrane tubules generated from giant liposomes by kinesin on microtubules and subsequently coated with dynamin. The fluorescence of Rhod-PE is shown. Note tubule fragmentation and small fragments of lipids floating away from the tubules. The field shown includes the field of Fig. 3b. Speed, 60X Movie S6 Effect of 1mg/ml dynamin co-added with 0.5 mm GTP in the presence of ATP on membrane tubules generated by kinesin on microtubules. The fluorescence of Rhod-PE is shown. Note tubule fragmentation and small fragments of lipids floating away from the tubules. The field shown includes the field of Fig. 3c. Speed, 15X Movie S7 1 mm GTP induces supercoiling of loops of dynamin-coated tubules. Same field as Fig. 4a. Realtime (15 fps) DIC microscopy. Movie S8 Rotation of a streptavidin bead (diameter 260 nm) attached to a biotin-dynamin coated tubule after addition of 200 µm GTP. Same field as for Fig. 4b. Real time (30fps) DIC microscopy.

5 5 Figure S1: Average number (5 experiments per condition) of tubes before and approximately 10 min. after addition of 0.4 mg/ml dynamin on SLM membrane sheets supplemented or not with PtdIns(4,5)P 2. Error bars are s.d.

6 6 Figure S2: a, No fragmentation is observed after addition of 1 mm GDP or GTPγS on dynamin coated tubule network grown on membrane sheets. b, Fluorescence time lapse series of dynamin (A-488 Dyn) coated tubules on Rhod-PE labeled membrane sheets after 1 mm GTP injection. c and d, Statistics of breaks for anchored and non anchored tubules, respectively.

7 7 Figure S3: a, Rhod-PE fluorescence images and plots of fluorescence intensity along a kinesinpulled tubule precoated with dynamin, before and after addition of GTP. The occurrence of fragmentation is demonstrated by the drop of fluorescence intensity to background level in the space among fragments. Pink arrows point to the tip of the tubules. b, Break densities (break/micron of tube) after GTP addition to tubules generated by dynamin versus tubules generated by kinesin and subsequently coated by dynamin.

8 8 Figure S4: a, Instantaneous angular speed (rotations per second) of one of the red trace tracked in c (1 mm experiment) versus the number (N) of rotations. b, the great majority of plectonemes have a plus configuration.

9 9 Figure S5: a, DIC sequence of a negatively charged bead attached to a dynamin tubule showing its rotation after addition of 1 mm GTP. Bar, 1 micron. b, Computational tracking of the bead.