Fus2p-GFP is functional and localizes to the shmoo tip. (A) FUS2 encodes a 79-kD protein with a Dbl Rho-GEF domain (DBH). A putative NLS is present at amino acid residues 313–319. A putative coiled-coil domain is present near the C terminus. (B) Plate mating assays show complementation of the fus2Δ by FLAG and GFP-tagged FUS2. Patches of MY9182, MY9183, MY9184, JM269, and JM362 were mated to JY429 and replica printed to selective media (SC-leu-trp) to select for diploids. (C) Quantitative mating assays. Ratios of diploids formed/total cells (±SEM) are reported for matings of MY9182, MY9183, MY9184, and JM269 to JY429. Experiments were preformed in triplicate. (D) Fus2p-GFP transport in live cells. MY9184 was treated with α-factor for 1.5 h. Images were acquired every 1.5 s. (E) Fus2p-GFP localization during shmoo formation. MY9184 was treated with α-factor at 0 min. Images were acquired every 10 min. The localization to the shmoo tip (arrowhead), rapid movement through the cell (filled arrow), faint nuclear localization (open arrow), and unstable cortical localization (asterisk) are indicated in D and E. Bars, 1 μm.

Fus2p localization is regulated by the mating response. (A) Mitotically expressed Fus2p-GFP localizes to the nucleus at all stages of mitotic growth. MY9201 containing P_GAL-FUS2-GFP was grown in media containing galactose (top, DIC; bottom, GFP). (B) α-factor treatment causes Fus2p-GFP to exit the nucleus. MY9201 was grown in galactose-containing media and induced with α-factor. As the cell completed mitosis, nuclear Fus2p exited the nucleus and localized to the shmoo tip in pheromone-responding cells (left, DIC at 0 min; remaining panels, GFP). (C) Preexisting nuclear Fus2p-GFP exits the nucleus. MY9201 grown in galactose was transferred to media containing dextrose for the indicated times (10, 20, or 30 min) to shut off expression before treatment with pheromone for 1.5 h. Fus2p-GFP was nuclear in mitotic cells but localized to the shmoo tip in shmoos (top, DIC; bottom, GFP). (D) Fus2p-GFP nuclear exit is dependent on high levels of pheromone. MY9184 was treated with 6 μM α-factor (top) or 0.6 μM α-factor (bottom) and imaged at various times (left, transmitted or DIC at 0 min; right, GFP). (E) Nuclear exit requires the Fus3p MAP-kinase. MY9211 was treated with α-factor for 1.5 h. Transmitted light (first panel), DAPI of the nucleus (second panel), and GFP (remaining panels) are shown. Bars, 1 μm.

Rvs161p is required for Fus2p movement and localization. (A) Fus2p-GFP is not transported or localized in an rvs161Δ mutant. MY9214 was induced with pheromone for 1.5 h. (B) Fus2p-GFP localization is dependent on Rvs161p's cell fusion function. MY9231 (rvs161Δ), MY9233 (cell fusion defective allele rvs161-P203Q), and MY9239 (endocytosis-defective allele rvs161-R113K) were treated with α-factor for 1.5 h. Bars, 1 μm. (C) Fus2p interacts with Rvs161p but not Rvs167p. Wild-type cells expressing Rvs167p-HA, Fus2p-FLAG, or both epitope-tagged proteins were treated with α-factor, and proteins were immunoprecipitated with anti-HA or anti-FLAG antibodies. Fus2p-FLAG and Rvs167p-HA could immunoprecipitate Rvs161p but the two tagged proteins did not coimmunoprecipitate with each other. (D) Rvs161p association with Fus2p is dependent on pheromone induction. Fus2p-GFP was expressed from the P_GAL1 promoter (MY9201) by induction with galactose, mock treated, or treated with α-factor for 1.5 h. Protein extracts were prepared and Fus2p-GFP was immunoprecipitated.

Fus2p movement and localization are independent of microtubules. MY9184 expressing Fus2-GFP expressed from its own promoter was treated with α-factor for 1.5 h and either mock treated (A and C) or treated with nocodazole (B and D) for 10 min. Live cell image stacks were acquired (A and B). Arrows indicate directed movement of Fus2p-GFP puncta. Alternatively, cells were fixed and stained with anti-tubulin antibody (C and D) to demonstrate microtubule depolymerization. Bars, 1 μm.

Fus2p localization and transport is dependent on actin organization. (A) Fus2p-GFP localization is disrupted in cell polarity mutants. Strains (MY9184 MY9208, MY9213, and MY9216) were treated with α-factor for 1.5 h (top, DIC or transmitted; bottom, GFP). (B) Quantitation of Fus2p-GFP localization in polarity mutants. See text for details. Numbers indicate the percentage of cells with indicated patterns of localization (50 < n < 100). (C–E) Actin depolymerization disrupts Fus2p-GFP movement but not maintenance. MY9184 was treated with α-factor for 1.5 h and either mock-treated (C) or treated with lat-A (D) for 10 min. Fus2p-GFP moved rapidly in mock-treated cells (arrows) and was stably localized to the shmoo tip. Lat-A–treated cells showed no movement of Fus2p-GFP dots, but Fus2p-GFP stayed localized at the shmoo tip. (E) Cells in C and D were fixed was stained with Texas red–phalloidin to demonstrate actin depolymerization. Bars, 1 μm.

Cortical localization is dependent on both actin and Fus1p. (A and B) Fus2p-GFP localization is disrupted by lat-A in a fus1 mutant. MY9217 was treated with α-factor for 1.5 h and either mock treated or treated with lat-A as in Fig. 5. (A) The fus1 mutation had no effect on the transport or localization of Fus2p-GFP. Arrows indicate rapid movement of Fus2p-GFP dots. (B) Treatment with lat-A (10 min) leads to a rapid loss of Fus2p-GFP localization at the shmoo tip. (C) Texas red–phalloidin staining of fixed cells after lat-A treatment demonstrating actin depolymerization. Bars, 1 μm. (D) Distribution of GFP fluorescence in wild-type and fus1 mutant shmoos. The fraction of total fluorescence was measured for every 1/10 of the cell, starting at the shmoo tip. 50 cells were measured for each strain and condition. Bars represent the mean values; error bars indicate 95% confidence intervals.

Fus2p-GFP localizes of to an expanding ring at the ZCF. (A) Mating of wild type (MY9184) expressing Fus2p-GFP to wild type (MY8093) for 1 h. Flattened projections of deconvolved images are shown (top). To reveal the ring, deconvolved images were rotated ∼150° around the x axis. Fus2p-GFP localizes first to a restricted patch at the ZCF (left zygote, 0 min), dilates into a ring (10–40 min), and then accumulates in the nucleus (50 min). (B) Mating of wild type (MY9184) expressing Fus2p-GFP to a fus1 fus2 mutant (JY429). Deconvolved images were rotated slightly around the y axis to show the ring. (C) Indicated strains expressing Fus2p-GFP (MY9184, MY9213, MY9216, MY9208, MY9217, and 9211) were mated to a fus1 fus2 mutant (JY429). Bars, 1 μm.

A model for the role of Fus2p in cell fusion. Cytoplasmic Fus2p associates with Rvs161p and attaches to vesicles (ves) as it is transported to the shmoo tip along actin cables by a myosin (myo). Fus2p/Rvs161p would be tethered to the cortex by Fus1p. Fus2p then acts as a Rho-GEF, activating a putative Rho-GTPase to trigger vesicle fusion with the plasma membrane. Cell wall (cw) hydrolytic enzymes, carried by the vesicles, would be released into the extracellular space to degrade the intervening cell wall, allowing plasma membranes (pm) to fuse in subsequent steps.