A clustering-outgrowth model for filopodia formation, based on study using Xenopus egg extracts and liposome

Science 10 September 2010:
Vol. 329. no. 5997, pp. 1341 - 1345
DOI: 10.1126/science.1191710

Self-Assembly of Filopodia-Like Structures on Supported Lipid Bilayers

Kwonmoo Lee,1,2,* Jennifer L. Gallop,1,* Komal Rambani,1 Marc W. Kirschner1,

Filopodia are finger-like protrusive structures, containing actin bundles. By incubating frog egg extracts with supported lipid bilayers containing phosphatidylinositol 4,5 bisphosphate, we have reconstituted the assembly of filopodia-like structures (FLSs). The actin assembles into parallel bundles, and known filopodial components localize to the tip and shaft. The filopodia tip complexes self-organize—they are not templated by preexisting membrane microdomains. The F-BAR domain protein toca-1 recruits N-WASP, followed by the Arp2/3 complex and actin. Elongation proteins, Diaphanous-related formin, VASP, and fascin are recruited subsequently. Although the Arp2/3 complex is required for FLS initiation, it is not essential for elongation, which involves formins. We propose that filopodia form via clustering of Arp2/3 complex activators, self-assembly of filopodial tip complexes on the membrane, and outgrowth of actin bundles.

1 Department of Systems Biology, Harvard Medical School, Boston, MA 02115, USA.
2 Department of Physics, Massachusetts Institute of Technology (MIT), Cambridge, MA 02139, USA.

* These authors contributed equally to this work.
To whom correspondence should be addressed. E-mail: marc@hms.harvard.edu

PMID: 20829485

The paper is quite interesting - I believe it is the first time to make filopodia like structures in vitro using Xenopus egg extracts. Really hope they did it with just purified proteins! I don't understand why they started with toca-1 in the mixture. The F-Bar induced membrane curvature may be the reason why the drawing in their model is like hair follicle. And, it is recruited first, even before actin! Their immunodepletion of toca-1 had minimal effect, and their explanation is that there are a lot of F-Bar proteins in the extract could substitute. The questions are:

• Is it possible that the F-Bar protein making those follicle like base first, which then recruits N-WASP? The recruitment of N-WASP might very well be the consequence of local lipid composition.
• Does the follicle like membrane structure restrict the branching direction and force the merging of the actin branches?
• Unlike the bead assay, minimal branched actin structure is observed with this FLS assay. But, the authors have not shown an EM image of these FLS bases (follicle region), where I believe the branches are...

FLS assays

To make the supported bilayers, No. 1.5 glass coverslips were incubated with freshly prepared liposomes containing 45 % PC, 45 % PI, and 10 % PI(4,5)P2 in XB buffer for 20 min at room temperature, followed by extensive washing with XB buffer. Membrane phase separation was variable and was largely influenced by the particular batch of glass. Rigorous washing of the coverslips with hot detergent also promoted the liquid disordered phase. All assays were carried out at room temperature (~22 °C).

For the purified system experiments, prenylated Cdc42.GTPrS was supplied to the lipid bilayer from 100 nM Cdc42-RhoGDI in solution using the EDTA exchange reaction (S9). The reaction mixture including N-WASP-WIP, toca-1, Arp2/3 complex, and actin as previously described (S2) was added after Cdc42 loading.

Typical FLS reactions (50 μl volume) contained a 2-fold dilution of Xenopus egg extract, 4 μM Alexa 647 actin (10% labeling efficiency, rabbit skeleton muscle actin), 0.35 M sucrose, 1 mM ATP, 1 mM MgCl2, 7.5 mM phosphocreatine in XB buffer. The reaction mixtures were added on top of the freshly prepared supported bilayer and monitored with a spinning disk confocal microscope.

For the pulse chase experiments, the second reaction (5 μl volume) Xenopus egg extract, 12 uM Alexa-488 actin (5% labeling efficiency, rabbit skeletal muscle actin), 1 mM ATP, 7.5 mM phosphocreatine in XB and 5 μl was added gently on top of the first reactions.

For dose response of FLS initial elongation, the reaction mixture was supplement with different dose of GST-CA and images were taken after 7 min.

For Arp2/3 complex independent elongation experiments, the first reaction was supplemented with 50 nM Alexa568-Arp2/3 complex. The second reaction is assembled similarly to the pulse-chase experiments with different doses of GST-CA. For the GFP-PH domain experiments, 50-300 nM was used.

For timelapse movies of FLS growth, an oxygen scavenger mix was added which contained: 4.5 mg/ml glucose, 0.5 % 2-mercaptoethanol, 0.2 mg/ml glucose oxidase (Sigma-Aldrich), 35 ug/ml catalase (Sigma-Aldrich).

Light microscopy

Microscopy for figures 1, 2 and 5 was performed using an inverted Nikon TE2000U microscope with a 100x, 1.4 NA Plan Apochromat objective lens and motorized stage and focus motor from Prior. Confocal images were obtained using a Yokogawa CSU-10 spinning disk confocal head with Prairie laser launch with a 2.5 W water-cooled Coherent Argon-Krypton laser. Excitation and emission wavelengths were selected and attenuated with an AOTF and a triple 488/568/647 dichroic mirror from Chroma. GFP and Alexa-488 were visualized using the 488 laser line and 525/50 emission filter; Alexa-568 was visualized using the 568 laser line and 600/45 emission filter; Alexa-647 was visualized by the 647 laser line and 700/75 emission filter (Chroma). Images were collected with a ORCA-AG cooled CCD camera from Hamamatsu and Metamorph software v7.6 (Molecular Devices). Exposure times were typically 100~400 ms using 25~50% laser power and a bin of 2x2. Z-stacks were collected with a step size of 0.5 μm.

Light microscopy for figure 3 and supplementary figures 2, 3 and 4 was performed using an inverted Nikon Ti-E microscope with a 100x, 1.4 NA Plan Apochromat objective lens and motorized stage from Prior. Confocal images were obtained using a Yokogawa CSU-10 spinning disk confocal head with 100 mW Argon-Krypton laser from Melles Griot. Excitation and emission wavelengths were selected using Sutter filter wheels and a triple 488/568/647 dichroic mirror from Chroma. Images were collected with an ORCA-ER cooled CCD camera from Hamamatsu and Metamorph software v7.6 (Molecular Devices). GFP was visualized using the 488 laser line selected with a 488/10 excitation filter and 525/50 emission filter; rhodamine and Alexa-568 were visualized using the 568 laser line selected with a 568/10 excitation filter and 620/60 emission filter; Alexa-647 was visualized by the 647 laser line selected with a 647/10 filter, and 647/10 emission filter (Chroma). Exposure times were typically 200 ms using a bin of 2x2.

For time-lapse experiments of FLS initiation, the Perfect Focus System (Nikon) was used to maintain focus, and images were acquired every 10 s for 10 minutes. Z-stacks were acquired with a step size of 1 μm. For the fluorescence recovery after photobleaching experiments of the supported bilayer, wide-field epifluorescence illumination was used (with a Hamamatsu ORCA-R2 cooled CCD camera and an X-Cite series 120 light source) and rhodamine-PE was photobleached to 80-90% of initial intensity with 515 nm light from a nitrogen pulse laser (Photonic Instruments Micropoint system) focused to a spot less than 1 micron in diameter. The filter was Y-2E/C (excitation: 560/40 dichroic: 595 emission; 630/60) from Nikon. The exposure time was 25 ms, and images were typically acquired every 1 s for 1-20 min.

For the multispectral total internal reflection fluorescence microscopy in figure 4, we used a Nikon Ti-E inverted motorized microscope with integrated Perfect Focus System, Nikon 100x 1.49 NA TIRF DIC objective lens, Nikon halogen trans illuminator with 0.52 NA LWD and 0.85 NA Dry condenser, Nikon dual-port TIRF/Epi illuminator with motorized laser incident angle adjustment and motorized switching between TIRF and epi-illumination. For lasers, a Solamere laser launch was used with 100mW 491nm, 75mW 561nm and 30mW 640nm solid state lasers with a fiber-optic delivery system and 4-channel AOTF. A Prior Proscan II controller was used for fast excitation and emission filter wheels, fast transmitted and epi-fluorescence light path shutters, and a linear-encoded motorized stage. A Chroma zet405/491/561/638 dichroic mirror was used with a 491 nm laser line and a 525/50 emission filter for GFP; a 561 laser line and 600/50 emission filter for Alexa568; and a 640 laser line and a 700/75 emission filter for Alexa647. In addition to emission filters, a custom Chroma laser notch filter was used in the emission path to further block the illumination light from reaching the camera and to minimize interference patterns. Images were collected with a Hamamatsu ImagEM 512x512 back-thinned electron multiplying cooled CCD camera and MetaMorph v7.7 (Molecular Devices). Exposure times were typically ~100 ms using 25~50% laser power.

Previous: IRES and co-localization analysis reveal the mechanism behind SNARE mediated membrane fusion
Next: Weekly Tweets for 2010-09-20