A noncanonical GTPase signaling mechanism controls exit from mitosis in budding yeast

Abstract

In the budding yeast Saccharomyces cerevisiae, exit from mitosis is coupled to spindle position to ensure successful genome partitioning between mother and daughter cell. This coupling occurs through a GTPase signaling cascade known as the mitotic exit network (MEN). The MEN senses spindle position via a Ras-like GTPase Tem1 which localizes to the spindle pole bodies (SPBs, yeast equivalent of centrosomes) during anaphase and signals to its effector protein kinase Cdc15. How Tem1 couples the status of spindle position to MEN activation is not fully understood. Here, we show that Cdc15 has a relatively weak preference for $\text{Tem}{1}^{\text{GTP}}$ and Tem1's nucleotide state does not change upon MEN activation. Instead, we find that Tem1's nucleotide cycle establishes a localization-based concentration difference in the cell where only $\text{Tem}{1}^{\text{GTP}}$ is recruited to the SPB, and spindle position regulates the MEN by controlling Tem1 localization. SPB localization of Tem1 primarily functions to promote Tem1-Cdc15 interaction for MEN activation by increasing the effective concentration of Tem1. Consistent with this model, we demonstrate that artificially tethering Tem1 to the SPB or concentrating Tem1 in the cytoplasm with genetically encoded multimeric nanoparticles could bypass the requirement of $\text{Tem}{1}^{\text{GTP}}$ and correct spindle position for MEN activation. This localization/concentration-based GTPase signaling mechanism for Tem1 differs from the canonical Ras-like GTPase signaling paradigm and is likely relevant to other localization-based signaling scenarios.

Figure 1:. Tem1 is a noncanonical GTPase with relatively low affinity to nucleotides.

(A) Illustration of the spindle position checkpoint in budding yeast and localization of the MEN GTPase Tem1 and its GAP. (B) Major components of the MEN and its inputs and outputs. (C) Molecular switch model for Tem1. (D) GTPase activity of Tem1. Recombinantly purified Tem1 was mixed with GTP and [$\text{α}$-³²P]GTP on ice and the reaction was initiated by shifting to 30 °C. Samples were taken at indicated time points and analyzed by TLC (Fig. S1B-C). Shown are the average of 3 technical replicates with a linear fit. Error bars represent ± standard deviation (SD). (E) The rate of nucleotide exchange (MANT-GDP to GTP) for Tem1 and N-Ras. Recombinantly purified Tem1 (5 μM) and N-Ras (5 μM) were first incubated with MANT-GDP (200 nM) and the nucleotide exchange was initiated by the addition of excess GTP (1 mM). Shown are the average of 3 technical replicates with an exponential fit for Tem1 and a linear fit for N-Ras. Error bars represent ± SD. (F) The binding affinity of Tem1 for MANT-GDP. 200 nM MANT-GDP was titrated with increasing amount of purified apo-Tem1. Shown are the average from 3 technical replicates with a quadratic fit for the dissociation constant ($K_D$). Error bars represent ± SD.

Figure 2:. Tem1’s effector Cdc15 has a weak preference for Tem1GTP.

(A) Illustration of the hybrid AlphaLISA assay used in this study. Recombinant biotinylated Tem1 were incubated together with cell lysate with eGFP tagged binding partners in the presence of $\text{GTPγS}$ or GDP. The interaction was monitored with streptavidin coated Alpha donor beads and anti-GFP conjugated accepter beads. (B-C) Interaction between recombinant biotinylated Tem1 and eGFP tagged Cdc15 (B) or Bub2 (C) in lysate monitored by AlphaLISA. Diluted lysate of y1275 (Cdc15-eGFP) or y1378 (Bub2-eGFP) were incubated with different concentrations of recombinant biotinylated Tem1 in the presence of $\text{GTPγS}$ or GDP. The Alpha signals were normalized to the maximum signal of each sample in the $\text{GTPγS}$ condition. Black lines represent the average of three biological replicates and error bars denote standard deviation (SD). The Hook point is when the added biotin-Tem1 exceeds the binding capacity of the Alpha Donor beads which results in a competition between biotin-Tem1 in solution and on beads for binding. Red line indicates 10% of the available eGFP tagged protein (maximum binding capacity defined by the $\text{GTPγS}$ condition) were bound by biotin-Tem1 on beads. (D) Dissociation constant of Tem1 binding to Cdc15 and the GAP estimated by a competition binding assay. Diluted lysate of y1275 (Cdc15-eGFP) or y1378 (Bub2-eGFP) were incubated with a fixed amount of biotinylated Tem1 and different concentrations of unlabeled Tem1 in the presence of $\text{GTPγ<S}$ or GDP. Signal for Bub2-eGFP with GDP was too low to be quantified reliably. The Alpha signals were double normalized to the signal without competing Tem1 (normalized signal = 1) and without biotin-Tem1 (normalized signal = 0). Lines represent the average of three biological replicates and error bars denote SD. (E-F) Interaction between biotinylated Tem1 and eGFP tagged Bfa1 (E) or Bub2 (F). Diluted lysate of y1374 (Bfa1-eGFP), y1378 (Bub2-eGFP), y3099 (Bfa1-eGFP in $\mathit{\text{bub2Δ}}$), or y3632 (Bub2-eGFP in $\mathit{\text{bfa1Δ}}$) were incubated with different concentrations of biotinylated Tem1 in the presence of $\text{GTPγS}$ or GDP, and the interaction was monitored with AlphaLISA. Lines represent the average of three biological replicates and error bars denote SD. (G) Point mutations generated in this study to modulate Tem1’s nucleotide state. (H-I) Interaction between different Tem1 mutants and the effector Cdc15 (H) or GAP complex (I) via co-immunoprecipitation. yEGFP-Tem1 was immunoprecipitated from exponentially grown cells of the indicated genotypes (y3461/y3452/y3472/y3457/y3456 or y2761/y2885/y2886/y3209/y3208) and the presence of Bfa1–3V5 or Cdc15–3HA were analyzed by Western blot analysis.

Figure 3:. Tem1’s Nucleotide state regulates its localization.

(A-C) Representative images and quantification of SPB localization in cells expressing yEGFP tagged wild-type Tem1 (y1748, n = 73 cells), GTP-locked hydrolysis mutant Q79L (y1824, n = 63 cells), GDP-locked/apo nucleotide binding mutant T34A/N (y3167/y3171, n = 29/21 cells), or effector binding mutant T52A (y3147, n = 32 cells) together with Spc42-mScarlet-I (SPB marker). Cells expressing tem1T34A/N and tem1T52A (marked with *) were kept alive with hyperactive DBF2-HyA that can rescue $\mathit{\text{tem1Δ}}$. Cells were grown at 25°C in SC medium + 2% glucose and imaged every 3 minutes for 4 hours. Single cell traces were aligned based on anaphase onset, as defined as spindle length > 3 μm (measured based on SPB marker Spc42-mScarlet-I), and averaged. Solid lines represent the mean, shaded areas represent 95% confidence intervals (B). For maximum enrichment at SPB (C), solid lines represent the median. ****p < 0.0001 by two-sided Wilcoxon rank sum test for comparing with WT.

Figure 4:. Localization of Tem1 to the SPB depends on its GAP and effector.

(A) Comparison of SPB localization for Bfa1, Tem1 and Cdc15 with or without BUB2. Cells with the indicated BUB2 alleles (columns) and yEGFP-labeled proteins (rows) were grown at 25 °C in SC medium + 2% glucose and imaged every 3 minutes for 4 hours. Strains imaged were y1374 (Bfa1-yEGFP, n = 104 cells), y3099 (Bfa1-yEGFP in $\mathit{\text{bub2Δ}}$, n = 17 cells), y1748 (yEGFP-Tem1, n = 65 cells), y1929 (yEGFP-Tem1 in $\mathit{\text{bub2Δ}}$, n = 72 cells), y1275 (Cdc15-yEGFP, n = 79 cells), and y1693 (Cdc15-yEGFP in $\mathit{\text{bub2Δ}}$, n = 76 cells). (B) Dependency of SPB localization for two different modes of localization patterns observed. (C) Cdc15 was efficiently depleted with the auxin-inducible degron (CDC15-AID). Cells with the indicated genotypes (y3229 (osTIR-/BFA1), y3224 (osTIR+/BFA1), y3228 (osTIR-/$\mathit{\text{bfa1Δ}}$), y3223 (osTIR+/ $\mathit{\text{bfa1Δ}}$)) were grown at 25 °C in SC medium + 2% glucose + 200 μM IAA for 2 hours before harvesting for Western blot analysis (left). Anaphase progression (right) was monitored in these cells by measuring spindle length (distance between two SPBs). Cells with Cdc15 depleted arrested in anaphase with elongated spindles whereas non-depleted cells disassembled their spindle (as indicated by the shortening of the distance between SPBs in late anaphase) to exit from mitosis. (D) Representative images (left) and quantification (right) of Tem1 localization at the SPB with or without depleting Cdc15 (CDC15-AID +/− osTIR) in the presence or absence of BFA1. Cells of y3229 (Tem1 control, n = 16 cells), y3224 (Tem1 with Cdc15 depletion, n = 16 cells), y3228 (Tem1 in $\mathit{\text{bfa1Δ}}$, n = 12 cells), y3223 (Tem1 in $\mathit{\text{bfa1Δ}}$ with Cdc15 depletion, n = 10 cells) were grown at 25 °C in SC medium + 2% glucose + 200 μM auxin IAA and imaged every 3 minutes for 4 hours. For all graphs, single cell traces were aligned based on anaphase onset, as defined in Fig. 3, and averaged. Solid lines represent the average. Shaded areas represent 95% confidence intervals.

Figure 5:. Tem1’s nucleotide state does not change upon MEN activation.

(A) Illustration of the different routes for Tem1 and the GTP-locked Tem1Q79L at dSPB. Once $\text{Tem}{1}^{\text{GTP}}$ is recruited to the dSPB by the GAP, its departure can occur either via $\text{Tem}{1}^{\text{GTP}}$ dissociation (${k_{off}}^T$) or GAP-stimulated GTP hydrolysis ($k_{hyd}$) followed by dissociation (${k_{\mathit{\text{off}}}}^D$). Assuming ${k_{\mathit{\text{off}}}}^D>>k_{\mathit{\text{hyd}}}$, the residence time of Tem1 at the dSPB (${t_{\mathrm{1/2}}}^{\text{Tem}1}$) is set by the sum of two rates: ${t_{\mathrm{1/2}}}^{\text{Tem}1}=\ln 2/\left({k_{off}}^T+k_{hyd}\right)$. For Tem1Q79L, its residence time (${t_{\text{1/2}}}^{\text{Tem1Q79L}}$) is simply ${t_{1/2}}^{\text{Tem1Q79L}}=\text{ln}2/{k_{off}}^T$ assuming that $\text{Tem}{1}^{\text{GTP}}$ and Tem1Q79L dissociate at similar rates. (B) FRAP analysis of Tem1 and TemQ79L at dSPB in anaphase. Cells of y3900 (TEM1, n = 9 cells) and y3903 (TEM1Q79L, n = 15 cells) were grown and imaged at room temperature in SC medium + 2% glucose + 100 μM IAA. Circles represent the average normalized fluorescence intensities after correcting for photo-bleaching during acquisition. Solid lines are the average fit and shaded areas represent SD. Half recovery time t_1/2 ± SD are indicated. (C) FRAP analysis of Tem1 at dSPB in metaphase and anaphase. Cells of y3900 were grown and imaged at room temperature in SC medium + 2% glucose + 100 μM IAA. (D) Proposed model for GAP and Tem1 localization at the SPB. Under wildtype condition (low [$\text{Tem}{1}^{\text{GTP}}$] in the cell), only a fraction of the GAP complex at the dSPB is bound to $\text{Tem}{1}^{\text{GTP}}$ in both metaphase and anaphase. While more GAP complex is recruited to dSPB in anaphase, the fraction of GAP bound to $\text{Tem}{1}^{\text{GTP}}$ does not change between metaphase and anaphase. In cells with high [$\text{Tem}{1}^{\text{GTP}}$] such as with TEM1Q79L, most of the GAP complex are bound to [$\text{Tem}{1}^{\text{GTP}}$] because Tem1 is more abundant than the GAP. (E) dSPB localization of Tem1 (WT or Q79L) and Bfa1 in cells with Cdc15 depleted. Cells of y3900 (Tem1, n = 16 cells), y3903 (Tem1Q79L, n = 35 cells), and y4123 (Bfa1, n = 24 cells) were grown at 25 °C in SC medium + 2% glucose + 100 μM IAA and imaged every 3 minutes for 4 hours. Lines represent the average. Shaded areas represent 95% confidence intervals. (F) Ratio of dSPB intensities of Tem1 to Tem1Q79L from (E) was plotted to estimate the cellular level of $\text{Tem}{1}^{\text{GTP}}$.

Figure 6:. SPB localization modulates effective concentration of Tem1 for MEN activation.

(A)Complementation analysis of Tem1 mutants with or without tethering to the SPB via CNM67-GBP. 5-fold serial dilutions of strains y2891/y3202/y3203/y3204/y3214 and y3303/y3304/y3305/y3306/y3310, with the indicated TEM1 alleles, were spotted onto plates with or without 5’-fluoroorotic acid (5-FOA) and incubated at 25°C for 2 to 3 days. The presence of 5-FOA selects cells that are viable after losing the TEM1 (URA3/CEN) plasmid. GBP = GFP binding protein. (B) Distribution of anaphase duration for different yEGFP-TEM1 alleles in the presence of CNM67-GBP (y3289, y3287, and y3291; n =46, 43, and 47 cells respectively). Cells were grown at 25°C in SC medium + 2% glucose and imaged every 3 minutes for 4 hours. Solid lines represent the median. (C) Complementation analysis of tem1T34A with and without tethering to different GBP fusion proteins. 5-fold serial dilutions of strains y3203/y3305/y3500/y3499/y3958/y3956/y3502 were spotted onto plates with or without 5-FOA and incubated at 25°C for 2 to 3 days. (D) Representative localization of GEM-tethered Tem1T34A in the presence or absence of BFA1. Cells of y3630 and y3557 were grown at 25 °C in SC medium + 2% glucose and imaged every 3 minutes for 4 hours. (E) Complementation analysis of different Tem1 mutants tethered to GEMs in the presence or absence of BFA1. 5-fold serial dilutions of strains y2891/y3202/y3203/y3204/y3214, y3970/y3973/y3502/y3976/y3985, y3971/y3974/y3948/y3977/y3986, and y3972/y3975/y3945/y3978/y3987 were spotted onto plates with or without 5-FOA and incubated at 25°C for 2 to 3 days.

Figure 7:. SPoC is sensitive to Tem1 concentration.

(A-B) Evaluation of SPoC integrity in cells with different amount of Tem1. Cells of y3733 (1xTEM1, n = 102 cells), y3729 (2xTEM1, n = 72 cells), y3816 (3xTEM1, n = 64 cells), y3730 (4xTEM1, n = 102 cells), y3732 (9xTEM1, n = 101 cells), and y3815 ($\mathit{\text{bfa1Δ}}$, n = 65 cells) were grown at 25 °C in SC medium + 2% glucose + 100 μM IAA to induce spindle mispositioning by depleting Dyn1 and Kar9 and imaged every 5 minutes for 5 hours. Status of the spindle and cell cycle stages were monitored with GFP-Tub1. For time to exit in mother (mispositioned spindle disassembled in the mother cell) in D, solid lines represent the median, red dashed line indicate the median time to exit in bud (normal anaphase progression with correctly positioned spindle). (C) Evaluation of SPoC integrity in cells expressing Tem1Q79L. Cells of y3817 (yEGFP-TEM1, n = 55 cells), y3818 (yEGFP-TEM1Q79L, n = 47 cells), and y4311 (yEGFP-TEM1Q79L-AID*, n = 108 cells) were grown and monitored as described in (A). (D) Proposed model for Tem1 regulation. (E) Comparison between the canonical molecular switch model for Ras and the concentration model for Tem1.