Article Figures & Data
Figures
- Figure 1. Interaction partners of Bem1.
(A) Yeast cells carrying Bem1CRU or either of its two mutants Bem1WK- and Bem1PB1∆CRU were independently mated four times with Nub fusion expressing strains. Interaction is indicated by growth of the four matings on SD medium containing 5-fluoro-orotic acid. Shown are the cut outs of the quadruplets expressing the Nub fusion of the interacting protein on the left, next to a fusion that does not interact. (B) The domains of Bem1 and the positions of the residue exchanges of the bem1-alleles used in this work. The domain-specific interaction partners of (A) are listed below the respective domains.
- Figure 2. Characterization of the Bem1-Bud6 interaction state.
(A) Yeast cells carrying the indicated mutations were co-expressing CRU fusions to Bem1 together with the indicated Nub fusions. Cells were grown to an OD600 of 1 and spotted in 10-fold serial dilutions onto medium containing 5-fluoro-orotic acid. Interactions are indicated by the growth of the yeast cells. (B) As in (A) but with yeast cells co-expressing CRU fusions to Bud61-364 or Bud6360-788, together with Nub fusions to Boi1 and Boi2. (C) Extracts of yeast cells expressing either Boi1-GFP (lanes 1, 3, 5, and 7) or Boi2-GFP (lanes 2, 4, 6, and 8) were incubated with GST- (lanes 3, 4, 7, and 8) or GST-Bud61-364–immobilized (lanes 1, 2, 5, and 6) sepharose beads. Bound (lanes 1–4) and unbound (lanes 5–8) fractions were analyzed by anti-GFP antibodies after SDS–PAGE and transfer onto nitrocellulose. (D) Model of a potential regulator of actin nucleation. Bud6 is known to homodimerize, whereas Boi1 and Boi2 either homo- or heterodimerize. Bud6 binds and stimulates the yeast formin Bni1 (not shown).
- Figure 3. Characterization of the Bem1-Nba1 interaction state.
(A) Split-Ub assay as in Fig 2A, but with cells co-expressing CRU fusions to Nba1 together with the indicated Nub fusions. Nub-Guk1: negative control. (B) a-yeast cells expressing the indicated CRU fusions were mated with α-yeast cells expressing the indicated Nub fusions and spotted on 5-fluoro-orotic acid medium as in Fig 1A. (C) As in (A) but with yeast cells co-expressing CRU fusions to Bem1, Boi1, or Boi2 together with Nub fusions to Nba1 or its mutants. (D) Surface plasmon resonance analysis of the interaction between 6xHIS-Nba1202-289-SNAP and the chip-coated SH3 domains of Boi1 or Boi2. Shown are representative plots of the surface plasmon resonance signals as response units against the concentrations of 6xHIS-Nba1202-289-SNAP. Corresponding sensograms are shown as insets. (E) As in (A) but with yeast cells containing or lacking NBA1 and co-expressing Gps1536-758CRU with the indicated Nub fusions. (F) Cartoon of the Nba1-Bem1 interaction state. The postulated indirect interaction between Bem1 and Gps1 was not experimentally observed but inferred from the Nba1-dependent interaction between Boi1/2 and Gps1 (see also Fig 7).
- Figure 4. Bem1 contains two functionally independent regions.
(A) bem1∆ cells expressing a vector-encoded BEM1 and carrying an additional gene deletion as well as an empty vector, or a vector expressing Gic2PBD, were incubated on media selecting against the presence of the plasmid-encoded BEM1. (B) bem1∆ cells carrying a vector-encoded BEM1 and a vector-expressing Bem1 or the indicated fragments of BEM1 (left panel) were incubated on medium selecting against the vector-encoded BEM1 (right panel). (C) bem1∆ cells expressing BEM1 or the indicated fragments of BEM1 were spotted in 10-fold serial dilutions onto the SD medium selecting for the presence of the plasmids and incubated at 37°C. (D) Haploid cells (upper panel), or diploid cells (lower panel) carrying the indicated alleles of BEM1 were incubated in 10-fold serial dilutions on media inducing the expression of Gic2PBD to moderate (left panels) or high (right panels) levels. (E) bem1∆ cells carrying a vector-encoded BEM1 and additionally expressing the full-length BEM1 with the indicated residue exchanges (upper panel), or fragments of BEM1 with the indicated residue exchanges (lower panel), were incubated on medium selecting against the vector-encoded BEM1.
- Figure S1. Influence of Cdc42 levels on growth of bem1-, cla4-, or boi1-allele-expresssing cells.
(A) bem1∆bem3∆ cells harboring BEM1 with its native promotor on a centromeric, URA3-expressing plasmid and containing the indicated GFP fusion alleles of BEM3 on an additional plasmid were streaked on medium containing 5-fluoro-orotic acid and selecting for the presence of the bem3 containing plasmids. Growth of the cells indicates alleles that do not complement the function of BEM3. bem3PHmut harbors the residue exchanges R646S, R645S, and K647D that are known to impair the binding of the PH domain to phospholipids (Mukherjee et al, 2013). bem3R950G contains an exchange in the catalytic site of the GAP domain that eliminates the activity of Bem3. (B) Cells containing the indicated bem1 alleles and an empty vector or overexpressing Bem3-GFP or Bem3R950G-GFP from a centromeric plasmid under the control of the PMET17 promoter were spotted in 10-fold serial dilutions on media without methionine to induce the expression of the bem3 alleles. Cells were incubated for 2 d at 37°C. (C) bem1∆ cells expressing BEM1 or bem1KA from a centromeric plasmid under control of the PMET17 promoter and containing an empty vector (upper lanes), or Gic2PBD expressed from the PMET17-promoter (lower lanes), were spotted in 10-fold serial dilutions on medium-lacking methionine and incubated for 2 d at 30°C. bem1KA harbors the residue exchanges K482A that is known to impair the binding of Cdc24. (D) Left panel: cla4∆ ste20∆ cells expressing CLA4 or cla4AAF451L and either the PH domain of Boi1 (Boi1745-980) or an empty vector were spotted on plates containing 70 μM methionine or no methionine to fully express Boi1745-980. Right panel: boi1∆ boi2∆ cells expressing Boi1 or boi1∆PxxP and either the PH domain of Boi1 (Boi1745-980) or an empty vector were spotted on plates containing 70 μM methionine or no methionine to fully express Boi1745-980. Cells were incubated for 2 d at 35°C. (D, E) Cells as in left panel of (D) but additionally expressing Cdc42 on a centromeric vector under the control of the PGAL1 promoter (left panel) or containing an empty vector (right panel). Cells were spotted on media containing glucose (Glc) and lacking methionine to overexpress Boi1745-980 and repress Cdc42 or on media lacking methionine and containing galactose (Gal) to simultaneously overexpress Boi1745-980 and Cdc42. Note the partial rescue of cla4AAF451L cells overexpressing Cdc42.
- Figure S2. Expression levels of different BEM1 alleles.
N-terminal GFP-tagged fragments of Bem1 and mutants thereof were expressed under control of the PMET17 promoter from a centromeric plasmid. Yeast cells were grown in media containing 70 μM methionine to an OD600 of 1.5–2. Extracts containing equal amounts of proteins were separated by SDS–PAGE and stained with anti-GFP antibody after transfer onto nitrocellulose. Equal loading was controlled by staining the blot with anti-alkaline phosphatase antibody (anti-ALP). Lanes showing fragments that are not measurably expressed were framed by a red box.
- Figure S3. Characterization of mutations in Cla4 and Ste20 that impair the interactions with Bem1 but not with Nbp2.
(A) Cartoons of Cla4 (left panel) and Ste20 (right panel) indicating the positions of the kinase domains and the motifs for binding Cdc42GTP (PBD) and Bem1 (PxxP). Below are indicated the studied alleles of the PAKs and their corresponding residue exchanges. (B) Extracts of E. coli cells expressing a 6His–SNAP fusion protein of the N-terminal PxxP-motif of Cla4 (6His-Cla41-33-SNAP) containing no mutation (lanes 1, 6), a phenylalanine to leucine exchange at position 15 (6His-Cla41-33F15A-SNAP; lanes 2, 7), an alanine to proline exchange at positions 22 and 25 (6His-Cla41-33AA-SNAP; lanes 3, 8), the triple mutations (6His-Cla41-33F15AAA-SNAP; lanes 4, 9), or a 6His-SNAP fusion (lanes 5, 10) were incubated with Glutathione-coated beads exposing GST-Bem1140-271 (lanes 1–5), or GST (lanes 6–19). Shown is the anti-His Western blot of the glutathione eluates of the beads after SDS–PAGE and transfer onto nitrocellulose. Lane 11 documents the input of the 6His–SNAP fusion. (C) Analysis as in (B) but with extracts of E. coli cells expressing a 6His–SNAP fusion protein of C-terminal PxxP-motif of Cla4 (6His-Cla4437-471-SNAP) containing no mutation (lanes 1–3) or a phenylalanine to leucine exchange at position 451 (6His-Cla4437-471F451L-SNAP; lanes 4–6). Extracts were incubated with beads exposing GST-Bem1140-271 (lanes 1, 4), GST-Nbp2 (lanes 2, 5), or GST (lanes 3, 6). The input is shown in lane 7. (D) A Myc-tagged fragment of Cla4 containing both binding sites and no mutations (Cla41-4719xMyc; lanes 1–3), the indicated mutations (Cla41-471F15A AA/AA; lanes 4–6), or Cla41-471F15A AA/PP F451L (lanes 7–9) were expressed in yeast from a centromeric vector under control of the PMET17 promoter. Extracts of these cells were incubated with bead-coupled GST-Bem1140-271 (lanes 1, 4, and 7), GST-Nbp2 (lanes 2, 5, and 8), or GST (lanes 3, 6, and 9). The inputs of the Myc-tagged fusions are shown in lanes 10–12. The eluted proteins were separated by SDS–PAGE and stained with anti-Myc antibody after transfer onto nitrocellulose. The protein doublet below 60 kD is occasionally observed in lanes displaying GST-Nbp2 and might arise by a cross-reactivity of the antibodies. (E) Split-Ub assay of yeast cells co-expressing Nub-Bem1 (upper panel) or Nub-Nbp2 (lower panel) together with CRU fusions to Cla4 or its indicated mutants, expressed from a centromeric vector under control of the PMET17 promoter. 4 μl of cells of OD600 of 1 was spotted in 10-fold serial dilutions on media lacking (left panels) or containing 5-fluoro-orotic acid. Growth on media containing 5-fluoro-orotic acid indicates interaction between the respective Nub and CRU fusion. (F) As in (D) but with cells co-expressing 9xMyc fusions to Ste20, to Ste20 containing alanine exchanges for proline at positions 477 and 489 (Ste20PP477,480AA) or to Ste20 containing an phenylalanine to leucine exchange at position 470 and a proline to threonine exchange at position 475 (Ste20F470L P475T). E, Glutathione eluate; S, Supernatant. (E, G) Split-Ub assay as in (E) but with cells co-expressing the indicated Nub fusions with CRU fusions to Ste20 or to the indicated mutants of Ste20. The CRU fusions were expressed from a centromeric vector under the control of the PMET17 promoter.
- Figure 5. The connection between Bem1 and the PAKs is essential upon overexpression of Gic2PBD.
(A) Yeast cells carrying the indicated alleles and either Gic2PBD or an empty vector were spotted in 10-fold serial dilutions on medium inducing high levels of Gic2PBD. (B) Cells expressing Shs1-GFP, the indicated alleles of BEM1, and Gic2PBD were incubated under conditions of low (70 μM Met, gray bars) or high expression levels (0 Met, blue bars) of Gic2PBD. Cells (500 < n < 600) were classified according to their native-like or abnormal distribution of the Shs1-GFP. (C) As in (B) but with cells (500 < n < 600) carrying the indicated alleles of STE20 or CLA4. (D) Microscopy of the cells of (C). Upper panel: DIC channel. Lower panel: GFP channel. Scale bar indicates 3 μM. (E) cyk3∆ hof1∆ cells expressing HOF1 from an extra-chromosomal vector and carrying the indicated alleles of STE20 or an empty vector were incubated on media selecting against the HOF1-containing vector.
- Figure S4. Phenotypes of boi1, or ste20-allele-expressing cells.
(A) Cells of the indicated genotypes and expressing Gic2PBD-9myc under control of the PMET17 promoter were grown to an OD600 of 1 in medium containing 70 μM Met. Cells were spun down and resuspended in fresh medium containing either no or 70 μM Met. After 4 h of growth, equal cell numbers were lysed by boiling in SDS sample buffer, and the extracts were separated by SDS–PAGE. Shown are the anti-Myc antibody staining (upper panel) and Ponceau staining (lower panel) of the gels after transfer onto nitrocellulose. (B) a- and α-cells containing the same alleles of STE20 and expressing Gic2PBD were incubated together in media inducing low (70 μM Met, blue bars) or high expression (0 Met, gray bars) of Gic2PBD and plated on media selecting for the presence of diploid cells. (C) The disruption of the Boi1/2-Bem1 connection does not induce defects in septin organization. Quantification of wild type and boi2∆ boi1PxxP∆ cells as in Fig 5C. 500–600 cells containing a GFP fusion to the septin SHS1 were inspected under the fluorescence microscope under conditions of low (70 μM methionine, blue bar) or high (no methionine, gray bar) expression of Gic2PBD. No increased septin mislocalization is detectable in boi2∆ boi1PxxP∆ cells. (D) The bud length of cells (n = 38; SEM) carrying the indicated alleles of BOI1 and co-expressing Bem1-GFP and Shs1-mCHERRY were measured every min starting with a bud length of 0.65 μm and ending at the time point of septin splitting (left panel). Red box indicates the time window with a significant difference between ∆boi2 BOI1- and ∆boi2 boi1PxxP∆ cells. Right panel shows the first derivatives of the growth curves of the left panel and compares the rates of bud extension in μm/min. (E) Time-lapse analysis of cells co-expressing Bem1-GFP and Myo1-mCherry. Shown are the quantifications of 36 cells. Plotted are the times of Bem1-GFP appearances in relation to the disappearance of Myo1-mCherry. Bem1 appears in average 0.4 min before Myo1 is removed from the site of cell separation. (F) Split-Ub analysis as in Fig 1 but of cells expressing Bem1CRU together with the indicated NUb fusions expressed from their native promoters. * indicates Nub fusions expressed from the PCUP1 promoter. Interaction was scored after 4 d of growth on media containing 5-fluoro-orotic acid. (G) SPLIFF analysis of a diploid cell formed by the fusion of a Bem1CCG-expressing a-cell and a Nub-Cla4–expressing α-cell. Shown are the stills from the different cell cycle phases of cell fusion, bud formation, bud growth, and cell separation (see cartoon at the top). Red arrows point to regions where the fluorescence intensities were quantified. Note that the α-cell is not visible before cell fusion has occurred (first frame, 0 min).
- Figure 6. Boi1 and Boi2 localize Bem1-Cdc24 at bud tip and neck.
(A) Wild-type cells expressing GFP-Bem1145-268 (left panel) or GFP-Bem1145-268 carrying the ND (middle panel) or WK (right panel) exchange were inspected by fluorescence microscopy. (B) Cells of the indicated genotypes expressing GFP-Bem1145-268 were inspected by fluorescence microscopy (left panel). Only boi2∆ boi1PxxP∆ cells show a clear misdistribution of GFP-Bem1145-268. (C) Left panel: bem1∆ cells expressing Bem1-GFP or Bem1WK-GFP were inspected by fluorescence microscopy. Right panel: Quantification of the intensity ratios of Bem1-GFP (n = 58), Bem1WK-GFP (n = 72), and Bem1ND-GFP (n = 57) in bud and mother cells. (D) Half-times of fluorescence recovery after photo-bleaching the bud of cells expressing Bem1-GFP or its mutants. Left panel: Bem1-GFP (n = 11), Bem1WK-GFP (n = 18), or Bem1ND-GFP (n = 24). Middle panel: Bem1-GFP in boi2∆ BOI1 cells (n = 20) or boi2∆ boi1∆PxxP cells (n = 16). Right panel: Bem1-GFP in ste20∆ CLA4 cells (n = 16) or ste20∆ cla4PPAAFL cells (n = 14). (E) As in (D) but with BOI1 BOI2 cells (n = 23), boi2∆ BOI1 cells (n = 24), or boi2∆ boi1PxxP∆ cells (n = 22) expressing GFP fusions to BOI1 or boi1PxxP∆. (F) Bem1 cells (left panel), or Bem1WK cells (right panel) expressing GFP fusions to Boi1 (upper row), Ste20 (middle row), or Cla4 (lower row), were inspected by fluorescence microscopy. (F, G) The ratios of the fluorescence intensities of bud and mother cells from (F) were quantified in BEM1- and bem1WK cells expressing Boi1-GFP (n = 40, 43), Ste20-GFP (n = 43, 40), or Cla4-GFP (n = 34, 34). ns, not significant. * = P < 0.05, ** = P < 0.01. Scale bars indicate 3 μM.
- Figure 7. Two receptor systems attach Bem1-Cdc24 to the bud neck.
(A) Ratios of bud neck to cytosolic fluorescence intensities of Nba1-GFP in wild type and gps1∆ cells. (B) As in (A) but in cells of the indicated genotypes expressing Boi1-GFP, Boi2-GFP, SH3Boi1-GFP, and SH3Boi2-GFP. (C) As in (A) but in cells of the indicated genotypes expressing Bem1-GFP or Cdc24-GFP. (D) Fluorescence recovery after photo bleaching the bud neck of boi1WK boi2WK cells expressing SH3Boi1-GFP (n = 11), of wild type cells expressing Boi1-GFP (n = 15), of boi2∆ cells expressing Boi1-GFP (n = 17), of boi2∆ cells expressing Boi1PxxP∆-GFP (n = 27). (E) Split-Ub analysis as in Fig 2A but with cells expressing Fir1CRU or Fir1PxxP∆CRU together with indicated Nub fusions. Nub-Cbk1: positive control; Nub-Guk1: negative control. (F) Anchoring the Bem1–Cdc24 complex at the bud neck in wild type- (upper panel), nba1∆- (middle panel), and fir1∆ cells (lower panel). Nba1 (red dots) and Fir1 (black dots) recruit Boi1 (blue dots) (for simplicity only Boi1 is shown), which recruits the Bem1–Cdc24 complex (orange and green dots). To explain the impact of the mutations on the distributions of the different proteins, we assume that Nba1 outnumbers Fir1 at the neck, that Nba1 and Fir1 are saturated by Boi1/2, the number of Nba1 and Bem1 molecules are similar, and that Nba1 reduces the affinity between Cdc24 and Bem1 (Meitinger et al, 2014).
- Figure 8. SPLIFF analysis of yeast zygotes formed by the fusion of a-cells expressing Bem1CCG and α-cells expressing the indicated Nub-fusions.
(A) Plotted are the conversions of Bem1CCG to Bem1CC (%) over time. GFP- and mCherry fluorescence intensities were measured at sites of polarized Bem1 locations as indicated in red in the cartoons of the upper panel. Shown are fitted lines calculated from single-cell experiments. (B) Interaction profiles for Bem1. Blue boxes indicate a significant increase in conversion over the indicated time. For values below 70% of conversion, significant slopes have to be 1 (% conversion/min) or larger. For values above 70%, slopes have to be positive and significant. To be counted as interaction, the criteria must not be met by the negative control in the respective time window (*** = P-value < 0.001; ** = P-value < 0.01; * = P-value < 0.05).
- Figure S5. Calculated slopes of the conversions of Bem1-, Boi1-, and Epo1CCG during PCDI, PCDII, and PCDIII.
Data are derived from Table S1. Dotted blue lines represent the nonparametric local regression (loess) fitted lines, whereas the corresponding confidence intervals (95%) are shown in light blue. Generalized additive model fitted over two time window intervals are represented by the colored segments. Numbers of cells are indicated in parentheses.
- Figure 9. Epo1 interacts with Boi1 during budding.
(A) As in Fig 1A but with cells expressing Epo1CRU or Epo1654-661∆ (Epo1PxxP∆) CRU with the indicated Nub fusions. (B) Extracts containing a 6xHIS SNAP-tag fusion to Epo1640-670 were incubated with GST-Boi1-, GST-Boi2-, or GST-coupled beads. Glutathione-eluates were stained with Ponceau (lower panel), and anti-His antibodies (upper panel) after SDS–PAGE and transfer onto nitrocellulose. (C) SPLIFF analysis: a-cells expressing Epo1CCG or Boi1CCG were mated with α-cells expressing the indicated Nub fusions. Upper panel: The measured fluorescence intensities were processed and visualized as in Fig 8A. Significance of slopes of fitted lines across time intervals are shown in the lower panel (*** = P-value < 0.001; ** = P-value < 0.01; * = P-value < 0.05).
- Figure 10. Interaction networks of Bem1 during PCDI (10–20 min), early and late PCDII (50–55 min, 60–80 min), and PCDIII (104–110 min).
Green color indicates proteins that promote, orange color indicates proteins that reduce active Cdc42. Blue color indicates effectors of Cdc42 or proteins that bind to effectors (Bud6-Bni1). Epo1 binds to the same site of Boi1 as Nba1 or Fir1 but at a different cell cycle phase. *The time point of interaction between Bem1-Cdc24-Cdc11 was obtained from a previous study (Chollet et al, 2020). **The time point of Fir1- and Gps1 binding to Bem1 were indirectly derived through their effects on Bem1 localization (Fig 7).
Supplementary Materials
