Synthetic genetic analyses identified RSR1 as a CDC42-interacting gene. (A) Genetic interaction network representing synthetic lethal/sick interactions of cdc42-118 determined by SGA analysis. Genes are colored according to their YPD-defined cellular function. (B) Equivalent dilutions of rsr1Δ, cdc42, and rsr1Δ cdc42 mutants grown at 25, 30, and 34°C for 2d on SC-URA. The strains shown (left to right) are KKY 342, 343, 344, 345, 346, 297, and 298.

A multicopy RSR1 plasmid, but not plasmids carrying other rsr1 alleles, suppressed the growth defect of cdc42-118. (A) Dilution series compared growth at 25 and 36°C (SC-URA, 4d) of wild-type and mutant yeast transformed with a high-copy (YEplac195) 2-μm plasmid containing RSR1 or no insert. Each spot of cells represented a 10-fold serial dilution from left to right. From top to bottom, RSR1, 2 μm-plasmid containing strains are KKY 66, 68, 287, 289, 291, and 190; strains containing only 2 μm-plasmid are KKY 239, 240, 286, 288, 290, and 238. (B) Wild-type and cdc42-118 strains were transformed with 2-μm plasmids (shown in brackets) containing no insert, a genomic fragment containing RSR1, alleles of RSR1, or CDC24 and grown at 25 (top panel) or 37°C (bottom panel) for 3 d on SC-URA. The strains shown (clockwise from 1 o'clock) are KKY 238, 240, 1012, 68, 60, 201, 200, and 69. Note: RSR1 suppressed cdc24-118 better than CDC24. (C) The cdc42-118 and cdc24-4 mutants, transformed with 2-μm plasmids containing RSR1, rsr1-7, or no insert, were grown at 25°C (for 3 d) or 37°C (for 4 d) on SC-URA (for cdc42-118) or SC-URA + 1 M sorbitol (for cdc24-4). Tenfold serial dilutions of each cdc42-118 transformant and 2.5-fold serial dilutions of each cdc24-4 transformant are shown.

Rsr1p interacts with Cdc42p. (A) Cdc42p coimmunoprecipitated with HARsr1p from yeast extracts. CDC42, HA-RSR1, and HA-rsr1^T35A were expressed from multicopy plasmids in yeast. Association of Cdc42p and HA-Rsr1p was not detectable when HA-Rsr1p and Cdc42p were expressed at the endogenous level. After immunoprecipitation of HA-Rsr1p (lane 1) or HARsr1^T35Ap (lane 3) with an antibody against the HA epitope, association of Cdc42p was determined by immunoblotting with polyclonal antibodies against Cdc42p (top panel). Approximately equal amounts of HA-Rsr1p and HA-Rsr1^T35Ap were recovered for each reaction as judged by immunoblotting with polyclonal antibodies against Rsr1p (bottom panel). Control reactions without HA antibody were shown in lanes 2 and 4. (B) Association of GST-Rsr1p and His₆-Cdc42p in vitro. GST-Rsr1p (∼400 nM) preloaded with GTPγS (T) or GDP (D) was incubated with His₆-tagged Cdc42p (∼400 nM) preloaded with GTPγS or GDP. Rsr1p-GTPγS + Cdc42p-GTPγS (lane 1); Rsr1p-GTPγS + Cdc42p-GDP (lane 2); Rsr1p-GDP + Cdc42p-GTPγS (lane 3); Rsr1p-GDP + Cdc42p-GDP (lane 4). GST that had been preincubated with GTPγS (lanes 5 and 6) or GDP (lanes 7 and 8) was used as a control. Association of Cdc42p with Rsr1p was determined by immunoblotting with antibodies against Cdc42p (top panel). Approximately equal amounts of GSTRsr1 and GST proteins were recovered for each reaction as judged by immunoblotting with antibodies against GST (bottom panel). (C) Association of Rsr1p and GST-Cdc42p in vitro. Rsr1p (∼400 nM) purified after removal of GST was preloaded with GTPγS (T) or GDP (D) and incubated with GST-Cdc42p (∼400 nM) preloaded with GTPγS or GDP. Rsr1p-GTPγS + Cdc42p-GTPγS (lane 1); Rsr1p-GTPγS + Cdc42p-GDP (lane 2); Rsr1p-GDP + Cdc42p-GTPγS (lane 3); Rsr1p-GDP + Cdc42p-GDP (lane 4). Association of Rsr1p with GST-Cdc42p was determined by immunoblotting with antibodies against Rsr1p (top panel). Approximately equal amounts of GST-Cdc42p were recovered for each reaction as judged by immunoblotting with antibodies against Cdc42p (lower panel). (D) Quantification of the amounts of Rsr1p associated with GST-Cdc42p. Rsr1p preloaded with [³H]GTPγS (T) or [³H]GDP (D) was incubated with GST-Cdc42p preloaded with GTPγS or GDP. The amount of [³H]GTPγS- or [³H]GDP-bound Rsr1p that associated with GST-Cdc42p was determined by a filter binding assay. An average of five independent assays is shown as percentage of counts (cpm) divided by the initial input to each reaction.

The C-terminal half of Cdc24p enhances interaction between Rsr1p-GTP and Cdc42p-GDP. (A) GST-Rsr1p (∼400 nM) preloaded with GTPγS was incubated with His₆-tagged Cdc42p (∼400 nM) preloaded with GDP in the absence (lane 1) or in the presence of MBP-Cdc24^C (lane 2: 2 nM; lane 3: 10 nM; lane 4: 20 nM). GST-Rsr1p was collected with glutathione Sepharose. As a control, GST was incubated with MBP-Cdc24 ^C and Cdc42p-GDP, and MBPCdc24 ^C was brought down with amylose resin (lanes 5–7). Approximately equal amounts of GST-Rsr1p were recovered for each reaction as judged by immunoblotting with Rsr1p-specific antibodies (bottom panel: lanes 1–4). Association of Cdc42p and MBP-Cdc24 ^C with GST-Rsr1p or GST was determined by immunoblotting with Cdc42p-specific antibodies (top panel) and with polyclonal antibodies against Cdc24p (middle panel), respectively. (B) GST-Cdc42p (∼400 nM) preloaded with GDP was incubated with Rsr1p (after cleavage of GST; ∼400 nM) preloaded with GTPγS in the absence (lane 1) or in the presence of MBP-Cdc24^C (lane 2: 2 nM; lane 3: 10 nM; lane 4: 20 nM). As a control, GST-Cdc42p preloaded with GDP was incubated with MBP-Cdc24^C (2–20 nM) without Rsr1p (lanes 5–7). Approximately equal amounts of GST-Cdc42p were recovered for each reaction as judged by immunoblotting with GST-specific antibodies (bottom panel). Association of Rsr1p with Cdc42p was determined by immunoblotting with Rsr1p-specific antibodies (top panel). Presence of MBP-Cdc24^C in the eluents was determined by immunoblotting with Cdc24p-specific antibodies (middle panel). (C) His₆-Cdc42p preloaded with [³H]GTP (T) or [³H]GDP (D) was used in a filter binding assay with GST-Rsr1p preloaded with GTPγS (T) or GDP (D) in the presence of various amount of Cdc24^C (0–40 nM). The data are averages of three independent assays.

The effector domain of Rsr1p is not directly involved in interaction with Cdc42p. (A) The Rsr1p^T35A protein interacts with His₆-tagged Cdc42p similarly to the wild-type Rsr1p in vitro. Rsr1p-GTPγS + Cdc42p-GDP (lane 1); Rsr1p^T35A-GTPγS + Cdc42p-GTPγS (lane 2); Rsr1p^T35A-GTPγS + Cdc42p-GDP (lane 3); Rsr1p^T35A-GDP + Cdc42p-GTPγS (lane 4); Rsr1p^T35A-GDP + Cdc42p-GDP (lane 5). Proteins were analyzed as in Figure 3B. (B) Interaction between GST-Rsr1p^T35A-GTP and His₆-Cdc42p-GDP in the presence of C-terminal half of Cdc24p. GST-Rsr1p^T35Ap (∼400 nM) preloaded with GTPγS was incubated with His₆-Cdc42p (∼400 nM) preloaded with GDP in the absence (lane 1) or in the presence of MBP-Cdc24^C (lane 2: 2 nM; lane 3: 10 nM; lane 4: 20 nM). GST was used as a control instead of GST-Rsr1p in the presence of MBP-Cdc24 ^C and Cdc42p-GDP, and MBP-Cdc24 ^C was brought down with amylose resin (lanes 5–7). Proteins were analyzed as in Figure 4A.

A model of how bud site selection couples to polarity establishment in late G1. Step 1: Bud5p exchanges GDP for GTP on Rsr1p. Step 2: Cdc24p and Cdc42p associate with GTP-bound Rsr1p at the bud site. Step 3: Bud2p activates Rsr1p to hydrolyze GTP bound to Rsr1p. Step 4: Dissociation of GDP-bound Rsr1p from Cdc24p may activate Cdc24p. Activated Cdc24p exchanges GDP for GTP on Cdc42p. GTP-bound Cdc42 then triggers actin assembly and exocyst localization to establish an axis of polarity. Bud5p at the bud site would then return Rsr1p to a GTP bound state (dashed line), allowing for another cycle of signal transduction. See DISCUSSION for details of the model.