Fig. 1. The Myo5p tail interacts with actin. (A) Cytosolic fractions were prepared from strains bearing integrated versions of wt or mutant myo5 (SCMIG277, SCMIG278, SCMIG279, SCMIG280), or a strain expressing the TH2 and SH3 domains fused to LexA [EGY48 pEG202(TH2, SH3)] in a buffer containing 1 mM ATP. Purified actin (+ actin) or buffer (– actin) were added to the cell extracts. After allowing actin polymerization, one sample containing exogenously added actin was incubated with apyrase to hydrolyze ATP (+ apyrase). F-actin was recovered by centrifugation and the pellet analyzed by immunoblotting. 9E10 was used for detection of Myo5p, and EW/IK for detection of the LexA fusion protein. One quarter of the protein extract utilized per incubation was loaded as total. (B) Recombinant GST fusion proteins [GST–(TH2, SH3) or GST–(SH3, TH2c)] were incubated with cytosolic fractions of a wt strain (RH3975) and increasing concentrations of G-actin. After allowing actin polymerization, F-actin was recovered by centrifugation and the pellets were analyzed for the presence of the recombinant proteins by immunoblotting using the EW/IK antibody. The increasing amounts of actin in the pellet were monitored by Ponceau Red staining. The same amount of GST fusion proteins used per assay were loaded as total (T). (C) Different MYO5 tail fragments [pEG202MYO5, pEG202 myo5-(TH2, SH3)Δ, pEG202 myo5-(TH2n)Δ, pEG202 myo5-(SH3, TH2c)Δ, pEG202(TH2, SH3), pEG202(SH3, TH2c), pEG202(TH2n), pEG202(SH3), pEG202(TH2c)] were tested versus ACT1 (pJG4-5ACT1), or versus the B42 transcriptional activator (pJG4-5) as a control, in the two-hybrid assay on X-Gal-containing plates.

Fig. 2. The Myo5p tail–actin interaction is functionally relevant. (A) The indicated MYO5 tail fragments or END3 (as control) [pEG202MYO5, pEG202(TH2, SH3), pEG202 (SH3, TH2c), pEG202END3] were tested versus the indicated ACT1 alleles (pJG4-5act1-x) in the two-hybrid X-Gal-containing plate assay. The same actin mutants were tested for their ability to sustain endocytosis in vivo. The percentage of cell-bound α-factor internalized/min is indicated. (B) myo3Δ strains bearing integrated myc-tagged wt or mutant MYO5 (SCMIG277, SCMIG278, SCMIG279, SCMIG280) were tested for α-factor internalization. The percentage of cell-bound α-factor internalized/min is indicated. The expression level of the integrated proteins (lower panel) was monitored by immunoblotting. 9E10 was used for detection of wt and mutant Myo5p (lanes 1–4). C4 was used for detection of actin as internal reference.

Fig. 3. Vrp1p is required for the Myo5p tail–actin interaction. (A) Upper left panel: the most C-terminal MYO5 fragment containing the SH3 domain [pEG202 (SH3, TH2c); wt] or the same fragment bearing a point mutation in the SH3 domain (W1223S) was tested versus ACT1 (pJG4-5ACT1) or VRP1 (pJG4-5VRP1) using the two-hybrid assay on X-Gal-containing plates. Lower left panel: the MYO5 tail (pEG202MYO5) was tested versus ACT1 (pJG4-5ACT1) or RVS167 (pJG4-5RVS167) in wt (RH2884) or isogenic end3Δ (SCMIG45) or vrp1Δ (SCMIG48) strains bearing the β-galactosidase reporter gene (pSH18–34). Enzyme activity was monitored on X-Gal-containing plates. Right panel: the MYO5 tail (pEG202MYO5) was tested versus ACT1 (pJG4-5ACT1) in a vrp1Δ (SCMIG48) strain bearing the β-galactosidase reporter gene (pSH18–34), and either an empty plasmid (Ycplac111) or the same plasmid bearing wt or the indicated mutant vrp1 genes. Enzyme activity was monitored on X-Gal-containing plates. (B) Upper panel: recombinant purified GST fusion proteins containing the wt TH2 and SH3 domains [GST–(TH2, SH3), wt] or the same construct bearing the W1223S mutation in the SH3 domain were incubated with cytosolic fractions from wt (RH3975) or vrp1Δ (SCMIG48) strains and increasing concentrations of G-actin. After allowing actin polymerization, F-actin was recovered by centrifugation and the pellets analyzed for the presence of the recombinant fusion proteins by immunoblotting using the EW/IK antibody. The increasing amounts of actin in the pellet were monitored by Ponceau Red staining. The same amounts of GST fusion proteins used per assay were loaded as total (T). Lower panel: a cytosolic fraction from a strain bearing a plasmid with an HA-tagged VRP1 (SCMIG277 p181VRP1HA) was prepared in a buffer containing 1 mM ATP. Either purified G-actin (+ actin) or G buffer (– actin) was added to the cell extracts. After allowing actin polymerization, one of the samples containing actin was incubated with apyrase to hydrolyze ATP (+ apyrase). F-actin was recovered by centrifugation and the pellet analyzed by immunoblotting using the anti-HA antibody. One-quarter of the protein extract utilized per incubation was loaded as total (T). (C) The MYO5 tail (pEG202MYO5) or VRP1 (pEG202VRP1) was tested versus the indicated ACT1 alleles (pJG4-5act1-1x) in the EGY48 strain using the two-hybrid assay on X-Gal-containing plates.

Fig. 4. Myo5p might regulate localized actin polymerization. (A) A myo5Δ strain (RH3976) bearing plasmids carrying ACT1 (p413ACT1) and VRP1 (p181VRP1) or the corresponding control plasmids (pRS413 and Ycplac181, respectively) was tested for internalization of α-factor. The percentage of cell-bound α-factor internalized/min is shown. A wt (RH3975) was assayed as control. (B) arp3-62 (SCMIG286) and vrp1Δ (RH2892) strains were crossed to myo5Δ strains (SCMIG276 and RH3976, respectively) bearing a centromeric plasmid with the MYO5 and URA3 genes (p33MYO5), to construct the corresponding single or double mutations carrying p33MYO5, which were streaked on 5′-FOA plates. A representative of each genotype is shown. Lack of synthetic lethality (SL) (see table, –) between myo3Δ or myo5Δ mutations and the indicated mutants (vrp1Δ, arp3-62, sac6Δ, rvs161Δ, rvs167Δ, end3Δ and sla2Δ) was assessed by standard tetrad analysis. *Even though the myo5Δ sla2Δ spores failed to germinate, the double mutant covered with the p33MYO5 plasmid was able to grow on 5′-FOA plates. (C) The C-terminal fragment of the Myo5p tail is able to induce actin polymerization in vitro. Glutathione–Sepharose beads coated with either GST or GST fused to the MYO5 TH2 and SH3 domains [GST–(TH2, SH3)], or the same fusion protein bearing the W1123S mutation, were incubated with extracts from either a wt (RH3975), vrp1Δ (SCMIG48) or arp2-2 (SCMIG4165) strain, or buffer (none), in the presence of rhodamine–actin. Samples were incubated at 30°C and the fluorescent signal visualized using a fluorescence microscope. Latrunculin A (LatA) or Dnase I was added to the (wt + LatA) or (wt + Dnase I) samples, respectively, during the incubation. For the (wt + Pha + LatA) sample, actin filaments polymerized in the absence of beads were stabilized with phalloidin (Pha) prior to the addition of GST–(TH2, SH3)-coated beads and latrunculin A. Representative (+) and (–) signals are recorded in the left panel [GST–(TH2,SH3) and GST, respectively]. The appearance of the beads in a given sample was homogeneous and the assays were repeated at least three times independently.