GPI anchor lipids of gup1Δ cells are less hydrophobic than the ones of wt. (A) BY4742 wt and gup1Δ cells were labeled with [³H]inositol, and GPI proteins were purified, digested with pronase, and purified by chromatography on octyl-Sepharose. The octyl-Sepharose column was washed with 5%, eluted with 25% and then 50% propanol in water. Radioactivity in eluted fractions was measured by scintillation counting. (B) The fractions 1–9 of the gup1Δ derived peptides of panel A were subjected to nitrous acid treatment and the thus liberated lipid moieties from indicated fractions were analyzed by TLC in solvent 1 (lanes 9–13). In parallel, the lipid extract of [³H]inositol–labeled wt cells was treated with PLA₂ and loaded onto the same octyl-Sepharose column, and eluted fractions were analyzed by TLC as above (lanes 2–7). Untreated lipid extract was run in lanes 1 and 8. Note that anchor peptides eluting with 25% propanol (lanes 10 and 11) are contaminated by hydrophobic peptides, which slightly distort the migration of lipids during TLC.

The GPI remodeling activity of Gup1p requires Histidine 447. (A) The indicated strains were labeled with 40 μCi of [³H]inositol, and GPI anchor peptides were purified using octyl-Sepharose column chromatography as in Figure 2A. The columns were eluted with 25% (fractions 2–7) and 50% propanol (fractions 8–12). Ten percent of each fraction was used to determine the radioactivity. (B) Aliquots of anchor peptides (100,000 cpm/strain) from 50% propanol eluates (fractions 8–10, panel A) were treated with HNO₂ (lanes 2–6) or control incubated (lane 7), and liberated lipids were run on TLC using solvent 2; the same amount of radioactivity was also taken from 25% propanol eluates (fractions 3–5 in panel A) and similarly analyzed (lanes 4′–6′). Lane 1 contains the free lipids of BY4742. The total amount of radiolabeled anchor peptides obtained in corresponding elution fractions for each strain (cpm ×10⁻⁵) is given below each lane. (C) Western blotting with anti-HA antibody of protein extracts from gup1Δ cells harboring the indicated plasmids.

The transport of Gas1p from the ER to the Golgi is delayed in gup1Δ cells. (A) Cells were pulse-labeled with [³⁵S]Cys/Met at 30°C and chased for the indicated times. Gas1p was immunoprecipitated and analyzed by SDS-PAGE/fluorography. (B) The fraction of mature 125-kDa Gas1p was obtained from the phosphorimager analysis of panel A, as well as a second, independent experiment, and was expressed as % of the total Gas1p (mature and immature). (C) CPY was similarly immunoprecipitated in lysates of panel A and was quantitated as Gas1p.

GUP1 homologues of T. cruzi and A. fumigatus partially complement gup1Δ cells. (A) The amino acid sequence of S. cerevisiae Gup1p is shown with its putative transmembrane domains (black lines, obtained with the TMHMM Server, v. 2.0 at http://www.cbs.dtu.dk/services/TMHMM/), the MBOAT motif (gray line, aa 271-510), and the predicted Histidine 447 active site (*). The closest homologues, namely 17 fungal and 13 protozoan sequences were aligned, each phylogentic group separately, with scGup1p using Clustalv W, and identities and similarities within each alignment were indicated with stars and points below the Gup1p sequence. (B) Calcofluor white resistance of BY4742 wt and gup1Δ cells was assayed as in Figure 5B. (C) Proteins of wt or gup1Δ mutants harboring different plasmids were extracted and analyzed in Western blots for the presence of Gas1p and Wbp1p. (D) Quantification of immature Gas1p (as a percentage of total Gas1p) in four independent experiments, one being the one shown in C. The Student's t test was utilized to compare the values obtained for gup1Δ cells harboring the empty vector with the ones obtained for gup1Δ cells harboring other plasmids: ∗p = 0.40; ∗∗p = 0.0034; ∗∗∗p = 0.188.

gup1Δmutant is deficient in GPI anchor remodeling. (A–C) Cells of indicated genotype were labeled at 30°C with [³H]inositol, and lipids were extracted and analyzed by TLC/fluorography using solvent 2, either directly (B) or after deacylation with NaOH (C). GPI anchors were isolated from delipidated proteins, and their lipid moiety was released with nitrous acid and analyzed by TLC/fluorography (A) along with the free lipids of wt (SCY325) cells (wt, FL) using solvent 1. 2Δ are = are1Δ are2Δ; 2Δ gup = gup1Δ gup2Δ; 4Δ = are1Δ are2Δ gup1Δ gup2Δ. pG1 is a remodeled form of PI, containing C26:0 in sn2. IPC/B contains PHS-C26:0; IPC/C contains PHS-C26:0-OH. (D) wt or gup1Δ cells were labeled at 24 or 37°C with [³H]inositol for 2 h. The delipidated labeled proteins were analyzed by SDS-PAGE/fluorography.

gup1Δcells contain a novel base-resistant GPI lipid. The [³H]inositol–labeled GPI anchor lipids were isolated from indicated strains as in Figure 2, but omitting the purification over octyl-Sepharose except in lanes 5 and 6, which contain 50% propanol eluates. Samples were treated or not with HNO₂ and/or NaOH, as indicated. The novel base resistant anchor lipid of gup1Δ is indicated by an asterisk (∗, lanes 1 and 2).

The gup1Δ cells are hypersensitive to calcofluor white. (A and B) Tenfold dilutions of the indicated strains harboring the indicated plasmids were spotted on plates containing 0 or 50 μg/ml calcofluor white (CFW). The plates were incubated at 30°C for 3 d. 2Δ gup is gup1Δ gup2Δ and 4Δ signifies the are1Δ are2Δ gup1Δ gup2Δ strain, and wt is BY4742. (C) BY4742 wt cells harboring the indicated plasmids were plated either on glucose (left) or galactose in the presence of CFW (right) and incubated at 30°C for 3 d.

Intracellular transport and sorting of GPI proteins in gup1Δ cells. (A) Cells were grown in YPD, harvested, and lysed using glass beads. The lysate was cleared of unbroken cells by centrifugation and extracted with 1% Triton X-100 for 30 min on ice. The samples were then centrifuged for 40 min at 15,000 × g to yield a detergent-resistant pellet (P) and a soluble (S) fraction. Proteins were then precipitated with TCA and analyzed by Western blotting using antibodies against the indicated proteins. (B) Cells were grown at 30°C to exponential phase, except for gup1Δ pre1-1 pre2-2 and gup1Δ cim5-1 thermosensitive strains that were grown at 24°C and incubated 2 h at 37°C. Proteins were extracted (Kushnirov, 2000) and analyzed by Western blotting using anti-Gas1p antibody. (C) Cells were grown at 30°C to exponential phase and harvested at a density (A₆₀₀) of 1.6 and 1.98 for gup1Δ and BY4742, respectively. Proteins were extracted from the cells (Kushnirov, 2000), and secreted proteins were precipitated from the medium using TCA. Proteins were analyzed by Western blotting using anti-Gas1p antibody. (D) Cells harboring pBC322 (obtained from Brendan Cormack), a plasmid containing HA-tagged EPA1, were grown to exponential phase. Cells were broken with glass beads, the lysate was spun at 15,000 × g, and the pellet was first extracted by boiling in SDS (Frieman and Cormack, 2003). After extensive washing of SDS-insoluble cell walls, covalently attached cell wall proteins were liberated by β1,3-glucanase treatment as described (Frieman and Cormack, 2003). Cellular proteins extracted by boiling in SDS or cell wall proteins extracted with glucanase (GLU) were analyzed by Western blotting using Gas1p (top panel) or anti-HA antibody (bottom panel). Material from equivalent numbers of cells was loaded in lanes 1–4, whereas lanes 5 and 6 contained three times the amount of the material loaded in lanes 2 and 4, respectively (GLU 3×).

GUP1 homologues of T. cruzi and A. fumigatus partially complement the bud site selection defect of gup1Δ/gup1Δ cells. The indicated diploid strains were grown exponentially for 24 h, stained with CFW, and viewed under the microscope. Two hundred cells present on randomly taken pictures were analyzed for each strain. The percentage of cells showing the normal bipolar budding pattern is indicated on the right side.