In vitro budding of ER-derived vesicles from wild-type and emp24Δ membranes. Spheroplasts from wild-type (EMP24) and mutant (emp24Δ) cells were pulse-labeled (3 min), permeabilized, and incubated with an energy source with or without cytosol (2 h, 25°C). Vesicles were purified by differential centrifugation and flotation on Nycodenz^® gradients. Budding of indicated proteins was assayed by immunoprecipitation, SDS-PAGE, visualization, and quantification using a PhosphorImager. The budding efficiency for a given protein refers to the percentage of the total radiolabeled protein that was recovered in the purified vesicular fraction.

Cross-linking of Emp24p and Gas1p. (A) ER-derived vesicles generated in vitro from wild-type (EMP24) and emp24-E178A mutant (mE178A) membranes were incubated with DSP. The samples were denatured and immunoprecipitated with antibodies against the tail of Emp24p. The precipitates were denatured and reprecipitated with antibody against Gas1p or Gap1p (C). 10 or 20% of the vesicle preparation was immunoprecipitated directly with antibodies against Gas1p or Gap1p, without addition of DSP, to use as a standard (S) for recovery. In control experiments, cross-linking with DSP did not affect recovery of Gap1p by immunoprecipitation. Samples were incubated with 5% 2-mercaptoethanol, analyzed by SDS-PAGE, and visualized using a PhosphorImager. (B) Vesicles generated from wild-type membranes were incubated with or without DSP and processed as above.

Direct cross-linking of Emp24p and Erv25p with Gas1p. (A) Vesicles generated in vitro from wild-type (EMP24) and emp24-E178A mutant (mE178A) membranes were incubated with 0.25 mM of DSG and processed as in Fig. 3. (B) Vesicles generated from wild-type membranes were incubated with the indicated amounts of DSG, denatured, immunoprecipitated with antibody against Erv25p, then reprecipitated with antibody against Gas1p and processed as in Fig. 3.

Inhibition of Gas1p budding by anti-Emp24p tail antibodies. (A) Permeabilized spheroplasts were prepared from wild-type (EMP24) and emp24-E178A (mE178A) cells as in Fig. 1. They were preincubated in the presence or absence of antibodies against the cytosolic domain of Emp24p (1 h, 4°C). Budding reactions were performed and analyzed as in Fig. 1. (B) The rate of Gas1p maturation in wild-type and emp24 mutants was analyzed after a pulse-chase immunoprecipitation protocol. Emp24p was visualized by Western blotting using antibodies against the c-myc epitope or against the cytosolic tail of Emp24p. Mutant (emp24Δ) cells were transformed with empty vector (emp24Δ), the wild-type allele (EMP24), myc-tagged Emp24p (mEMP24), or the myc-tagged emp24-E178A (mE178A) allele. (C) After pulse-chase labeling of wild-type (EMP24) and mutant (emp24Δ) strains, cells were lysed and extracted with Triton X-114 followed by detergent phase separation. Aqueous (A) and detergent (D) phases were subjected to immunoprecipitation and SDS-PAGE. Gas1p (both forms combined) was quantified using a PhosphorImager. Immature Gas1p (105) was converted to the mature form (125) with delayed kinetics in the emp24Δ mutant, but was efficiently extracted into the detergent phase.

Immunoisolation of Emp24p containing vesicles. ER-derived vesicles generated in vitro from sec18 membranes and cytosol were immunoisolated with or without Emp24p anti-tail antibody. The supernatants (S) and pellets (P) were processed for immunoprecipitation.