Vesicle formation depends on cytosol concentration, time, and hydrolyzable ATP and GTP. (A) The assay for the formation of endosomal vesicles was performed using increasing concentrations of cytosol at 37°C in the presence of ATP, GTP, and an ATP-regenerating system. Recovery of biotinylated H1 in the supernatant as determined by immunoblot analysis was quantified and plotted relative to the maximal value. The data represent the average and SD of three independent determinations. (B) Biotinylated broken cells were incubated in the presence of ATP, GTP, and an ATP-regenerating system at 37°C with or without cytosol (12.5 mg/ml) for increasing times. Recovery of biotinylated H1 in the supernatant was plotted relative to the maximal value (•, with cytosol; ×, without cytosol). (C) Vesicle formation was assayed in the presence of cytosol (12.5 mg/ml) with GTP or GMP-PNP (GN), and ATP or AMP-PNP (AN) or without nucleotides. Lanes 1–7 and 8–12 are from separate experiments.

Depletion of AP-2 or AP-3 adaptors does not affect in vitro vesicle formation from labeled endosomes. Bovine brain cytosol was immunodepleted for AP-2 (A) or AP-3 adaptors (C) by using anti-α- and anti-δ-adaptin antibodies, respectively, as in Figure 3A for AP-1. Depletion efficiency was >90% in both cases. Biotinylated permeabilized cells were incubated in the presence of ATP, GTP, and ATP-regenerating system without cytosol (no cyt), with untreated cytosol (cyt; 1.2 mg/ml), with adaptor-depleted cytosol (cyt–AP-2 or -3), or with mock-depleted cytosol (B for AP-2 and D for AP-3). Immunoblot analysis of biotinylated H1 in the supernatant after cell pelleting is shown for a representative experiment. Quantitation of three independent experiments (average with SD) is presented below.

Electron microscopic visualization of H1-containing endosomal vesicles. (A) H1 at the cell surface was decorated with nanogold-coupled anti-H1 F(ab′) fragments and allowed to internalize for 10 min. Cells were stripped of surface antibody; permeabilized; incubated with cytosol, nucleotides, and ATP-regenerating system; and sedimented. Membranes in the supernatant were collected by high-speed centrifugation and subjected to cryoelectron microscopy and silver enhancement of nanogold. (B and C) As in A, but in addition decorated with anti-γ-adaptin and a secondary anti-mouse IgG antibody coupled to 10-nm colloidal gold. Arrowheads indicate colloidal gold particles, which have a more dense appearance than silver-enhanced nanogold particles. Bar, 200 nm.

Rabaptin-5/rabex-5 inhibits in vitro formation of recycling vesicles. Bovine brain cytosol was immunodepleted for rabaptin-5/rabex-5 (Rbpt5) by using an antibody directed against rabaptin-5 as in Figure 3A for AP-1 (A). Biotinylated permeabilized cells were incubated in the presence of ATP, GTP, and ATP-regenerating system without cytosol, with untreated cytosol (cyt; 1.2 mg/ml), and with modified cytosol. In B, the effect of the rabaptin-5/rabex-5-depleted cytosol was tested, in C the effect of cytosol supplemented with rabaptin-5/rabex-5 partially purified from calf brain and immunodepleted for AP-1 (approximately doubling the amount of rabaptin-5 already in the cytosol), and in D the effect of adding 12 or 24 μg/ml purified rabaptin-5/rabex-5 produced by the baculovirus system. As controls, cytosol containing the buffer of the corresponding rabaptin-5/rabex-5 preparation was analyzed in parallel. Immunoblot analysis of biotinylated H1 in the supernatant after cell pelleting is shown for a representative experiment, as well as the quantitation of three independent experiments (average with SD) in B and D.

Biotinylated, internalized ASGP receptor H1 is released from broken cells in a temperature-, energy-, and cytosol-dependent manner in sealed membranes. (A) Cells expressing H1 were surface labeled with sulfo-NHS-SS-biotin on ice, washed, and then incubated with reduced glutathione (+GSH) to release the coupled biotin (lanes 1 and 2) or without glutathione (–GSH; lanes 3 and 4). On cell lysis, biotinylated proteins were isolated with avidin-agarose. Twenty and 80% of each sample were separated by SDS-gel electrophoresis and analyzed by immunoblotting by using anti-myc antibody. Stripping of surface-bound biotin by glutathione is essentially complete. (B) Surface-biotinylated cells were incubated at 37°C for 10 min to allow internalization, chilled in ice, and surface biotin was stripped with reduced glutathione. The cells were then broken and incubated for 30 min at 4 or 37°C, with or without cytosol (1.2 mg/ml) or nucleotides and energy-regenerating system as indicated. Cells were pelleted and the supernatant was lysed and analyzed by avidin precipitation of biotinylated proteins, SDS-gel electrophoresis, and immunoblotting by using anti-myc antibody. Values were normalized to the amount of total labeled H1 by analyzing 10% of a sample before centrifugation as a standard (10% std; lane 5). Average and SD of at least four independent determinations are shown. (C) The supernatant of cells treated as in B, lane 4, was incubated at 4°C for 45 min with or without trypsin in the presence or absence of detergent (NP-40). Trypsin was stopped with soybean trypsin inhibitor. Biotinylated proteins were then isolated, separated by SDS-gel electrophoresis, and analyzed by immunoblotting for myc-tagged H1. The positions of molecular weight markers are indicated (in kilodaltons).

AP-1 adaptors are necessary for endosomal vesicle formation in vitro. (A) Bovine brain cytosol was immunodepleted of AP-1 by using anti-γ-adaptin antibody and protein A/G-Sepharose beads. Total cytosol (T) and corresponding aliquots of the depleted cytosol (unbound fraction, U) and the bound material (B) were analyzed by immunoblotting with anti-γ-adaptin antibody. Depletion efficiency was >80%. The positions of molecular weight markers are indicated (in kilodaltons). (B) Biotinylated permeabilized cells were incubated in the presence of ATP, GTP, and ATP-regenerating system without cytosol (no cyt), with untreated cytosol (cyt; 1.2 mg/ml), with AP-1– depleted cytosol (cyt–AP-1), with AP-1– depleted cytosol that had been supplemented with 12 μg/ml purified AP-1 (cyt–AP-1+AP-1), or with mock-depleted cytosol. Immunoblot analysis of biotinylated H1 in the supernatant after cell pelleting is shown for a representative experiment. Quantitation of three independent experiments (average with SD) is presented below.

Clathrin stimulates and BFA inhibits in vitro formation of recycling vesicles. (A) Biotinylated permeabilized cells were incubated in the presence of ATP, GTP, and ATP-regenerating system without cytosol, with cytosol (1.2 mg/ml), or with cytosol supplemented with up to 120 μg/ml purified clathrin (0 indicates the addition of solvent only). In a further reaction, cytosol preincubated for 15 min at 4°C with 40 μg/ml anti-clathrin antibody X22 was used. Immunoblot analysis and quantitation of biotinylated H1 in the supernatant after cell pelleting is shown for a representative experiment. (B) Biotinylated and permeabilized cells were incubated with nucleotides, ATP-regenerating system with and without cytosol, and the indicated amount of BFA first at 4°C for 15 min and then for 10 min at 37°C. Immunoblot analysis of biotinylated H1 in the supernatant after cell pelleting is shown together with the quantitation of three independent experiments (average with SD). (C) Biotinylated and permeabilized cells were incubated with nucleotides, ATP-regenerating system with and without cytosol, and up to 100 μM phosphatidyl inositol 3-kinase inhibitor LY294002 (LY; 0 indicates the addition of solvent only). Immunoblot analysis of biotinylated H1 in the supernatant after cell pelleting is shown together with the quantitation of two independent experiments (average with range).

Rab4, but not Rab5, is required for in vitro formation of recycling vesicles. Bovine brain cytosol was immunodepleted for rab4 (A) or rab5 (B) by using specific antibodies as in Figure 3A for AP-1. Depletion efficiency was >90% in both cases. The positions of molecular weight markers are indicated (in kilodaltons). Biotinylated permeabilized cells were incubated in the presence of ATP, GTP, and ATP-regenerating system without cytosol (no cyt), with untreated cytosol (cyt; 1.2 mg/ml), with rab-depleted cytosol (cyt–rab4; cyt–rab5), or with mock-depleted cytosol (C). Immunoblot analysis of biotinylated H1 in the supernatant after cell pelleting is shown for a representative experiment. Quantitation of three independent experiments (average with SD) is presented below.