Baumgart et al. 10.1073/pnas.0611357104. |
Fig. 4. Fluid phase segregation in GPMVs is temperature-dependent. Confocal fluorescence micrographs of cell-detached GPMVs prepared by formaldehyde/DTT treatment of RBL cells, labeled with R-DOPE and imaged at three different temperatures. (A) Representative large field image of GPMVs at 5°C shows that a large percentage of vesicles exhibit fluid phase coexistence at this temperature. (B) Representative large field image of GPMVs at 23°C. A significantly smaller percentage of vesicles are phase-separated at this temperature, typically ≈10-25% of all GPMVs. (C) Selected phase-separated GPMV imaged at 37°C. Optically resolvable fluid phase coexistence at this temperature is uncommon: <1% of GPMVs show detectable fluid phase coexistence. (Scale bar, 5 mm.)
Fig. 5. GPMV membrane domain geometries reveal mechanical aspects of fluid phase coexistence analogous to model membranes. (A) GPMVs from RBL cells labeled with R-DOPE and confocal imaged near top surface indicate lateral ordering of domains in the form of a hexagonal pattern. The image was obtained at 15°C. (B) GPMV labeled with CTB A488 bound to GM1 and confocal imaged in an equatorial plane. CTB-enriched membrane phases show invaginated geometries, indicating significant influence of cholera toxin binding on local membrane curvature. The image was obtained at 23°. (C) Time series of a phase-separated GPMV labeled with R-DOPE and imaged near the top surface at sequential 3-s intervals. Large-scale two-dimensional shape undulations are indicated at the arrow, indicating small line tension at this domain boundary. The image was obtained at 23°. (Scale bar, 5 mm.)
SI Materials and Methods
Lipid Probes.
R-DOPE was obtained from Avanti Polar Lipids, 1,1'-dihexadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI C16:0) was from Molecular Probes, and naphtho[2,3-a]pyrene (naphthopyrene) was obtained from Sigma--Aldrich. Membrane staining by lipid fluorophores added to the aqueous cell/GPMV dispersion allows only limited control of the amount of dye added to cellular membranes, a factor that could potentially influence membrane phase behavior.Antibodies and Protein Ligands.
Mouse monoclonal anti-2,4-dinitrophenyl (DNP) IgE was purified and labeled with Alexa Fluor 488 (Molecular Probes/Invitrogen) as previously described (1). Ox7 mAb, specific for rat glycosylphosphatidylinositol-linked Thy-1 (BD PharMingen), was labeled with Cy3 (Amersham Pharmacia Biotech) according to manufacturer's instructions. Alexa Fluor 488-CTB (A488-CTB), A546-CTB, A546-annexin V and A488-Concanavalin A (A488-Con A) were obtained from Molecular Probes. These fluorescent derivatives were added at concentrations of 1-2 mg/ml to GPMV or cell suspensions, and samples were incubated for 15 min at room temperature or on ice before imaging.GFP Constructs.
Vectors encoding the expression of plasma membrane-targeted EGFP (PM-GFP) (2), geranyl-geranyl EGFP (GG-GFP) (2), and Lyn-EGFP (Lyn-GFP) (3) were constructed as previously described.Cell Culture and Transfection.
RBL-2H3 cells were maintained in monolayer cultures for 4 - 6 days after passage as described (4), and either harvested with trypsin-EDTA (Life Technologies, Rockville, MD) and washed twice in PBS, or caused to form cell-free GPMVs as described below. NIH 3T3 cells were grown to confluency in DMEM with Penn/Strep and FBS, then passaged and plated at a density of ≈5 ´ 104 cells per squared centimeter in a four-well chamber (Nalge-NUNC) with 2 cm2 of area per well and a #1.5 German borosilicate glass coverslip bottom. The cells were then incubated at 37°C in phenol red-free media for 2.5 h and subsequently rinsed in PBS plus 10 mM glucose (PBS/G) before addition of solvent/dye solution to induce formation of cell-attached GPMVs as described below.RBL cells were transiently transfected with EGFP constructs using GenePORTER (Gene Therapy Systems, San Diego) as previously described (1). PM-GFP was stably expressed as reported in ref. 2. In some cases, A488-IgE was added to RBL cell monolayer cultures at 0.5 mg/ml and 37°C, 8 h before vesicles were induced.
Imaging.
For room temperature fluorescence microscopy imaging, 5 ml of cell/GPMV suspension was deposited onto microscopy slides and enclosed in a chamber consisting of an additional coverslip and a rectangular spacer of vacuum grease enclosing an area of ≈0.2 cm2. Temperature-controlled imaging was performed after sealing 5 ml of cells or GPMVs with nail polish ("clear coat"; Sally Hansen, Uniondale, NY). This sample chamber was immersed into a small water bath on top of a water immersion objective (´60, 1.2 N.A.; Olympus). Both objective and water bath were temperature controlled by means of a circulating water bath. The sample temperature was measured near the sample chamber by using a thermal couple (Bat-10, Physitemp) and was stable within ±0.1°C. In case of RBL cells, GPMVs were imaged with either a Leica TCS SP2 spectral confocal system (Leica Microsystems, Bannockburn, IL) (objective ´63, 1.2 N.A., water immersion) or by confocal and two-photon excitation microscopy (for naphthopyrene imaging) and using a homebuilt multiphoton laser scanning microscope based on a Bio-Rad 600 scanhead (Bio-Rad, Richmond, CA) (objective ´60, 1.2 N.A., water immersion; Olympus), using an excitation wavelength of l = 780 nm for two-photon excitation. Fibroblasts were imaged with a Zeiss 510 LSM/META confocal microscope (Zeiss, Thornwood, NY), with a 40´ PlanApo 1.2 N.A. water immersion coverslip-corrected objective under 543-nm laser excitation. In all cases, laser polarization was minimized to ensure homogenous fluorescence detection from fluorophores with orientational preference.1. Gosse JA, Wagenknecht-Wiesner A, Holowka D, Baird B (2005) J Immunol 175:2123-2131.
2. Pyenta PS, Holowka D, Baird B (2001) Biophys J 80:2120-2132.
3. Hess ST, Sheets ED, Wagenknecht-Wiesner A, Heikal AA (2003) Biophys J 85:2566-2580.
4. Pierini L, Holowka D, Baird B (1996) J Cell Biol 134:1427-1439.