Video fluorescence microscopy studies of phospholipid vesicle fusion with a planar phospholipid membrane. Nature of membrane-membrane interactions and detection of release of contents.

Overview of Method

To study binding of vesicles and release of their contents, we loaded vesicles with the fluorescent dye calcein (4',5'-bis[N,N-bis(carboxymethyl)aminomethyl]fluorescein) at high concentrations and used a planar bilayer setup built to accommodate a fluorescence microscope with video equipment. The vesicles were sufficiently large (2-10 t~m)and the video camera was sufficiently sensitive for the fluorescent vesicles to be observed. This enabled us to view the binding of vesicles to the planar membrane directly. When fluorescent molecules were released, they diffused and became diluted, and the region occupied by dye increased. Release of dye resulting from a vesicle rupture could thus be NILESAND COHEN Vesicle-PlanarMembrane Interactions and Fusion 705 observed as a flash of light. The side of the planar membrane to which dye was released as determined by allowing cobalt, added to one side or the other, to quench the caicein fluorescence. Release of dye into cobalt-containing solution resulted in light flashes of short duration, whereas release into cobalt-free solution led to flashes of long duration. In this study, fusion is operationally defined as the transfer of dye to the trans side of the planar membrane.

Preparation of Vesicles

Large unilamellar vesicles filled with calcein were prepared by a modification of the double-emulsion method of Kim and Martin (1981). An outline of the procedure is shown in Fig. 1. In this method, the aqueous material to be entrapped is added to lipid dissolved in chloroform. Vortexing results in an emulsion of aqueous spherules suspended in chloroform with a monolayer of lipid at each water-chloroform interface (Fig. 1, step a). These spherules will become the internal contents of the vesicles. Lipid dissolved in diethyl ether is emulsified in the buffer that is to become the external medium of the vesicles; this emulsion is termed the "mother liquor" (step b). The spherules of entrapped material in chloroform are then injected into the mother liquor. This results in each spherule being surrounded by a shell of chloroform and ether solvent, which in turn is suspended in the outside buffer (step c). A lipid monolayer is present at each solvent-water interface. When the solvent is evaporated, the monolayers appose to form a lipid bilayer and vesicles form (step d). We used 10 mg of lipid mixture, consisting of 8 mg crude asolectin (type 2S, Sigma Chemical Co., St. Louis, MO) and 2 mg ergosterol recrystallized once from ethanol, dissolved in chloroform. This mixture was split equally into two l-dram vials. To one vial, marked "inside," enough chloroform was added to bring the volume to 0.5 ml. A water- in-chloroform emulsion was made in this vial by carefully adding, dropwise, 0.5 ml of dye solution, consisting of 200 mM caicein, 10 mM MES, 5 mM n-propy[ gallate, and 50 ~g] ml nystatin, pH 6.5 (666 mosmol). Calcein (Hach Chemicals, Ames, IA) was purified by the method of Ralston et al. (1981). The n-propyl gallate served as an antioxidant to retard fading of the dye under illumination (Giloh and Sedat, 1982). This vial was vortexed at a moderate rate for 40 s to produce spherules of dye. The final diameters of the vesicles are determined by this step: the longer the vortexing, the smaller the vesicles. The lipid mixture in the other vial, marked "outside," was dried under a stream of nitrogen gas and resuspended in 0.5 ml of diethyl ether by vortexing. 2.5 ml of outside buffer, consisting of 700 mM sucrose (856 mosmol), was added to this vial, and the emulsion of mother liquor was made by vortexing for 30 s. The emulsion of the inside vial was pulled up into a Pasteur pipette with a tip drawn to a narrow diameter (515 nm. The emitted light was magnified by a 15x eyepiece placed in tandem with a 0.3x projection lens, and focused onto the photocathode of the SIT camera. The binocular eyepiece assembly contained partially silvered prisms that allowed emitted light to be directed either to the eyepieces or to the video camera. The video camera was operated at settings that enabled detection of 200 mM calcein in the vesicles. The gain was manually set at 75% of maximum, thereby disabling a reflex circuit that automatically scaled the gain. Contrast was fixed at maximum. When required, the voltage-clamped membrane current was digitized, and a digital display generated by a modified video date-time generator (Video Timer VTG-33, For-A Corp. Ltd., West Newton, MA) was superimposed on the video image. The composite video signal was displayed on a video monitor (V20, Electrohome Ltd., Kitchener, Ontario, Canada). The total magnification of the optical and video system was 1,000. The video signal was recorded on video tape (3/4-in. U-marie) by a video tape recorder (VO-5800H, Sony Corp. of America, Park Ridge, NJ). All data analysis was performed from the video tape. Delivery of Vesicles The pipette used to deliver vesicles to the planar membrane was fashioned after those used to patch-clamp black lipid membranes (Andersen, 1983). A 6-in.-Iong shank of glass capillary tubing, 0.8 mm o.d. and 0.5 mm i.d. (7740, Coming Glass Works, Coming, NY), was pulled over a gas flame to yield a long, tapered tip. The pipette was cut to yield a right circular break with an outer diameter of 0.2 mm and then forged to make a right- angle bend 5 mm back from the tip. The tip was fire-polished until its inner diameter was 50 ~m. Vesicles in sucrose buffer were introduced into the pipette by back-filling from a 3-in.-iong 30-gauge needle. The pipette was placed in an electrode holder (EH-2R, E. W. NILESANDCOHEN Vesicle-PlanarMembraneInteractionsand Fusion 709 Wright, Guilford, CT)joined by a Lucite block to a hydraulic micromanipulator (MO- 103, Narishige Scientific Instruments). The back end of the pipette, which protruded through the top of the holder, was connected by polyethylene tubing to an electric valve (Picospritzer, General Valve Corp., E. Hanover, NJ), which was connected to a supply of pressurized nitrogen. The pipette was lowered into the rear compartment of the chamber and the tip was brought close to the membrane (5-10 #m) by means of the microma- nipulator. Vesicles were pressure-ejected toward the membrane by opening the valve for 3-4 ms. Chamber Buffers The following buffers were used in the chamber: KCi buffer: 400 mM KCI, 10 mM HEPES, 3 mM MgCI~, 1 mM EDTA, pH 7.5,760 mosmol; KCI plus cobalt citrate buffer: 350 mM KCi, 40 mM cobalt citrate, 10 mM HEPES, 3 mM MgC12, 1 mM EDTA, pH 7.5, 770 mosmol; sucrose buffer: 650 mM sucrose, 10 mM HEPES, 3 mM MgCi2, 1 mM EDTA, pH 7.5, 830 mosmoi; sucrose plus cobalt citrate buffer: 600 mM sucrose, 40 mM cobalt citrate, 10 mM HEPES, 3 mM MgCl~, 1 mM EDTA, pH 7.5, 820 mosmol. Cobalt citrate was made by dissolving citric acid in water and then slowly adding cobalt hydroxide powder (Aifa Products, Danvers, MA) in very small increments with vigorous stirring to prevent precipitation. Vesicle swelling was induced by the addition of aliquots from a stock of either 50 mg/ ml nystatin (Mycostatin, Sigma Chemical Co., St. Louis, MO) or 6 M urea to the rear compartment as described in the Results. Planar Membrane Formation A planar membrane was formed in the orifice of the black Teflon septum by the union of two phospholipid monolayers (Montal and Mueller, 1972). The lipid used in most experiments was asolectin (type 4S, Sigma Chemical Co.) washed in acetone and ether (Kagawa and Racker, 1971). The composition of this lipid mixture has been determined by Miller and Racker (1976). In experiments in which nystatin channels were incorporated into the planar membrane, the lipid mixture consisted of 4:1 wt/wt asolectin/ergosterol. In all experiments, the hole was precoated with 2% squalene in pentane and the lipid monolayers spread from a 1% solution in hexane on the surfaces of buffer in the two compartments. Membrane thinning was detected electricallyby the growth of capacitance, and optically, with bright-field illumination, by the development of a well-defined torus between the edge of the orifice and the planar bilayer. When the monolayers had been raised over the hole, the front compartment contained 0.7 ml of buffer and the rear compartment volume was 1.5 ml.