Supplemental Material

This article contains the following supporting material:

• Supplemental Figure 1 - Vacuolar protein sorting mutants are defective for the sorting of Sna3-GFP. The collection of viable deletion mutants was transformed with Sna3-GFP and analyzed by fluorescence microscopy. Virtually all the known vps mutants were identified through this approach and are listed. While the total number of mutants identified through this procedure is less than identified through similar screens for mutants that missort the soluble vacuolar hydrolases CPY (Bonangelino et al., 2002), representatives identified in that effort were directly analyzed after they were not identified in our screen. These directed analyses failed to identify additional ORFs required for proper sorting of Sna3-GFP. This may have to do with the cargo we have utilized (Sna3-GFP) and the tremendously sensitive assays available for detecting the secretion of CPY. Phenotypes indicative of the defect observed in each class are presented for clarity. As can be seen in the case of vps classes C (vps11·) and B (vps39·), intact vacuoles are not obvious and for this reason these mutants are not thought to play a direct role in the function of the MVB pathway beyond fusion of intermediates. Class D mutants (vac1·) are defective for endosomal fusion of Golgi-derived transport and display a cytoplasmic haze as a result. Class A mutants (vps35·) have been shown to play a role in recycling from the endosome. Class E mutants (vps20·) display the characteristic of mislocalizing Sna3-GFP to the class E compartment. *While YKR035c was identified as having a phenotype consistent with designation as a class E vps mutant, this ORF is located on the opposite strand from the ORF encoding FTI1/DID2. For this reason, the most likely interpretation of the observed phenotype is that it is merely recapitulating deletion of FTI1/DID2 and was not further pursued. Scale bar is 5 microns.
• Supplemental Figure 2 - Kinetic analyses of MVB cargoes. A, B, C) Representative gels used for the quantitation provided in Figure 3. Wild type, mvb12· or vps23· cells were pulse-labeled with ³⁵S-Cys/Met and chased with excess Cys/Met for the indicated times, samples from the indicated time point were used to perform immunoprecipitations with the indicated antibody. Immunoprecipitated material was subjected to SDS-PAGE and phosphoimaging. D) anti-CPS imuunoprecipitaion reactions from 80 minute chase points from wild type, mvb12· and 3 repetitions of vps23· were run side by side to allow comparison of the migration of mature CPS.
• Supplemental Figure 3 - Analysis of Mvb12-GFP localization in a class E vps backgrounds. The indicated strains were transformed with plasmid-borne MVB12-GFP and visualized by fluorescence microscopy. A) Wild Type (SEY6210 + MVB12-GFP), hse1· (MBY44 + MVB12-GFP), fti1· (EEY17 + MVB12-GFP), bro1· (JPY9), and vps60· (JPY16) cells displayed indistinguishable Mvb12-GFP distribution, while vps37· cells displayed increased cytoplasmic distribution (DKY48 + MVB12-GFP). B) snf7· (EEY9 + MVB12-GFP), vps20· (EEY2-1 + MVB12-GFP), vps24· (BWY102 + MVB12-GFP), vps25· (JPY13), vps4· (MBY3 + MVB12-GFP) and vps28· (SRY28 + MVB12-GFP) cells displayed increased association with membrane compartments. This is in contrast to vps27·, vps23· (Figure 7) and vps37· cells (A), which display increased cytoplasmic distribution relative to Wild Type. C) GFP-Vps27 localization was analyzed in wild type, mvb12· and vps23· cells by fluorescence microscopy. Visualization of endosomes was facilitated by co-expression of DsRed-FYVE. Scale bar is 5 microns.
• Supplemental Figure 4 - Analysis of TAP-Mvb12 association with ECSRT-I. The indicated strains were transformed with pnTAP416 or pnTAP416-Mvb12, and were used to generate native lysates. IgG Sepharose was used to isolate TAP proteins, and isolated material was resolved by SDS-PAGE using 10% acrylamide for anti-Vps23 Western blotting or 15% acrylamide for anti-Vps28 Western blotting. The TAP tag contains Protein A, allowing detection by both anti-Vps23 and anti-Vps28 antisera. Positions of Vps23, Vps28 and TAP proteins are indicated, as is a degradation product of TAP itself (‡). Quantitation of Vps28 and Vps23 isolated from the vps23· and vps28· strains, lanes 9 and 10 respectively, revealed equivalent levels compared to the wild type strain (lane 8) when normalized to TAP-Mvb12 levels.