Spatiotemporal regulation of organelle transport by spindle position checkpoint kinase Kin4

In the budding yeast Saccharomyces cerevisiae (S. cerevisiae), organelle segregation during cell division is ensured by class V myosin motor-based transport to the emerging bud. The majority of organelles, including vacuoles (the yeast equivalent of mammalian lysosomes) and peroxisomes, are transported by Myo2 (Hill et al., 1996; Hoepfner et al., 2001). Organelle-specific receptors recruit Myo2 to initiate transport along the polarised actin cytoskeleton (Knoblach and Rachubinski, 2015). This directs the delivery of Myo2 cargoes to zones of cell growth, i.e. the growing bud or late in the cell cycle to the bud neck which is the site of cytokinesis (Bi and Park, 2012; Pruyne and Bretscher, 2000a,b; Wong and Weisman, 2021). However, vacuoles and peroxisomes are positioned away from the bud neck area as their transport is terminated prior to cytokinesis through detachment of Myo2 from the organelles (Yau et al., 2014). Organelle transport needs to be coordinated with the cell cycle; hence, transport is under both temporal and spatial control. The mechanisms controlling the transport of vacuoles have been studied intensively and many factors controlling the initiation and termination of vacuole transport have been identified (Fig. 1A) (Wong and Weisman, 2021).

Vac17 is the vacuole-specific Myo2 receptor and is essential for the transport of vacuoles to the emerging bud (Ishikawa et al., 2003). Cdk1-dependent phosphorylation of Myo2 and Vac17 is crucial for formation of the Myo2−Vac17 transport complex early in the cell cycle (Legesse-Miller et al., 2006; Peng and Weisman, 2008) (Fig. 1A). Vacuoles form segregation structures which are pulled from the tip by Myo2 motors into the growing bud whereas the rest of the vacuole is retained in the mother cell (Eves et al., 2012; Li et al., 2021). Vacuoles detach from Myo2 and are positioned near the centre of medium- to large-sized buds (Tang et al., 2003). Two parallel converging pathways control the detachment of vacuoles from Myo2. One pathway extracts Myo2 from vacuoles, and is dependent upon yeast casein kinase 3 (Yck3) and the vacuolar membrane protein Vps41 via an unknown mechanism (Wong et al., 2020). The second pathway is more characterised, and is regulated by an amino acid (aa) sequence rich in proline (P), glutamate (E), serine (S) and threonine (T), a so-called PEST motif, within Vac17 (residues 204−250 aa) (Tang et al., 2003). PEST motifs target proteins for rapid degradation (Rechsteiner and Rogers, 1996). The detachment process is initiated by phosphorylation of the Vac17 PEST motif at T240 in the mother cell by an unidentified kinase. In the bud neck, the E3-ubiquitin ligase Dma1 binds directly to Vac17 phosphorylated at T240 (Yau et al., 2014) (Fig. 1A). Subsequently, the p21-activated kinase (PAK) Cla4 phosphorylates Vac17 at S222 in the PEST motif, which activates Dma1 and results in ubiquitylation of Vac17 (Yau et al., 2017). This later phosphorylation event is spatially controlled as Cla4 localisation is restricted to the cortex of buds (Holly and Blumer, 1999; Peter et al., 1996). Mutation of S222 or T240 into alanine or deletion of the PEST motif (PESTΔ) interferes with spatially regulated Vac17 breakdown. In cells deficient in detachment of Myo2 from vacuoles, the vacuole remains attached to Myo2 till late in the cell cycle and follows Myo2 to the bud neck (Yau et al., 2014, 2017). Overexpression of Cla4 causes excessive degradation of Vac17 and a defect in vacuole transport to the bud. As this transport defect can be rescued by co-expression of the vac17-PESTΔ mutant (Bartholomew and Hardy, 2009), ectopic Cla4 activity may lead to aberrant Vac17 degradation, too, early in the cell cycle or at the wrong place, e.g. in the mother cell, thereby interfering with spatiotemporal control of Vac17 detachment and degradation. Dma2 and the PAK Ste20 provide minor contributions to vacuole-transport termination, and are partially redundant with Dma1 and Cla4, respectively (Yau et al., 2014, 2017). The current model of spatial and temporal regulation of vacuole transport is depicted in Fig. 1A. However, this model does not explain how degradation of Vac17 is prevented in small-budded cells.

Cla4 also plays important roles during the final stages of nuclear segregation and mitotic exit (Caydasi et al., 2017). Mitotic exit involves a cascade of events, also called the mitotic exit network (MEN) (Weiss, 2012). GTPase Tem1 is the master regulator of MEN and triggers mitotic exit only from the spindle pole body (SPB) on the elongated nucleus (spindle) that has entered the bud (Weiss, 2012) (Fig. S1A,B). Proper spindle alignment during anaphase is a prerequisite for successful SPB entry and subsequent Tem1 activation in the bud. If the spindle is misaligned during anaphase, Tem1 is kept inactive to prevent premature mitotic exit through the action of the spindle position checkpoint (SPOC). The SPOC kinase Kin4 activates the bipartite GTPase-activating protein (GAP) complex Bfa1−Bub2 that inhibits Tem1 GTPase activity in the mother cell (Caydasi and Pereira, 2012) (Fig. S1A,B). Regulation of MEN is explained by the zone model (Caydasi and Pereira, 2012; Falk et al., 2016) (Fig. S1C). This model proposes that, late in the cell cycle, large-budded cells are divided into two zones, i.e. a MEN inhibitory zone in the mother cell and a MEN-activating zone in the daughter cell. The inhibitory zone is controlled by activated Kin4 and mainly localised to the mother cell cortex at this phase of the cell cycle, whereas the activating zone is controlled by the Kin4 inhibitor Lte1, with the latter concentrated in the daughter cell. Lte1 also directly activates the MEN, and Cla4 is required for Lte1 activity in the daughter (Bertazzi et al., 2011; Caydasi et al., 2017) (Fig. S1B,C). In contrast to activation of MEN, Cla4-dependent breakdown of Vac17 in large buds occurs independently of Lte1 (Yau et al., 2017).

Although Kin4 is mainly localised to the mother cell cortex in large-budded cells, its localisation to other sites within the cell, including the cortex of small buds during S phase and early in G2 phase, suggest additional functions for Kin4 (Chan and Amon, 2009; Pereira and Schiebel, 2005). This is further supported by the observation that Elm1 kinase activates Kin4 by phosphorylation of Thr209 throughout the cell cycle and not just during anaphase when SPOC is active (Caydasi et al., 2010). Recently, we have shown that Kin4 and, to a lesser extent, its functional paralog Frk1 are required for peroxisome transport to the bud and, thus, for peroxisome inheritance (Ekal et al., 2023b). Kin4 and Frk1 maintain protein levels of the peroxisome-specific Myo2 receptor Inp2, but no mechanistic details have been reported to explain their contribution to peroxisome transport (Ekal et al., 2023b). In addition to peroxisome inheritance, vacuole inheritance is also affected in mutant cells lacking Kin4 and Frk1 (hereafter referred to as kin4Δfrk1Δ cells) (Ekal et al., 2023b).

In this current study, we demonstrate that the vacuole inheritance defect in kin4Δfrk1Δ cells is mainly due to increased Cla4-/Dma1-dependent Vac17 turnover. Further experiments investigating the interplay between Kin4/ Frk1 and Cla4/Dma1 support the idea that the zone model proposed for the regulation of mitotic exit could be extended to other spatiotemporal events, including polarised vacuole and peroxisome transport.

S. cerevisiae Kin4 and its paralogue Frk1 are redundant in peroxisome transport to the bud (Ekal et al., 2023b). As kin4Δfrk1Δ cells are defective in vacuole inheritance (Ekal et al., 2023b), we tested whether Kin4 and Frk1 are also redundant in vacuole inheritance. Cells were pulse-chase labelled with the fluorescent lipophilic dye FM4-64 that accumulates in the vacuolar membrane (Vida and Emr, 1995) (for details, see Materials and Methods ‘Vacuolar staining with FM4-64’). This pulse-chase experiment allowed the tracking of pre-existing vacuolar membranes over time and, therefore, their inheritance. In virtually all wild-type (WT) and frk1Δ cells, including buds, vacuoles were stained with FM4-64, indicating efficient transfer of vacuoles from mother to daughter cells (Fig. 1B,C). In vac17Δ cells, transport of vacuoles to the bud is blocked. Consequently, ∼20% of mother cells contain brightly stained vacuoles, but their buds are empty and ∼80% of the cells lack bright FM4-64 staining altogether. In kin4Δ cells, a significant number of buds (11.7% ±1.4%) showed strongly decreased or undetectable levels of FM4-64 fluorescence. This was mainly observed in small-budded cells, implying a partial defect in vacuole transport. However, cells lacking FM4-64 staining were observed rarely (0.7%), with a frequency not significantly different from that of WT cells (Fig. 1B,C). This shows that kin4Δ cells still inherit vacuoles but that this process may be delayed. Upon additional deletion of Frk1 (kin4Δfrk1Δ) >40% of the cell population either failed to transport vacuoles or only transported a strongly reduced amount of FM4-64 labelled vacuoles from mother cell to bud. Many of the kin4Δfrk1Δ cells lacked brightly stained vacuoles all together (42.7%) (Fig. 1B,C). The synergistic effect of KIN4 and FRK1 gene deletion indicates that Kin4 function in vacuole transport is partially backed up by that of Frk1.

Vacuoles are essential for cell growth and progression through the cell cycle (Weisman et al., 1990). In cells that fail to transport vacuoles to the bud, such as in vac17Δ cells, new vacuoles are formed in the bud (Jin and Weisman, 2015). Since vacuole inheritance is affected in kin4Δfrk1Δ cells, de novo formation of vacuoles was analysed in kin4Δfrk1Δ and compared to WT and vac17Δ cells. Cells constitutively expressing the GFP-tagged version of the vacuole membrane marker protein Vph1 (Vph1-GFP) were pulse-chase labelled with FM4-64 before being imaged. Vacuoles in the mother cell and bud of WT cells were labelled with both FM4-64 and Vph1-GFP (Fig. 1D) implying proper vacuole segregation during cell growth and division in line with previous observations (Jin and Weisman, 2015). In contrast, in some kin4Δfrk1Δ cells, vacuoles only showed Vph1-GFP but not FM4-64 fluorescence. Moreover, in many kin4Δfrk1Δ budding cells, buds contained vacuoles labelled with Vph1-GFP only (Fig. 1D). Similar observations were made in vac17Δ cells that were used as a control for de novo vacuole formation. We, therefore, conclude that many of the kin4Δfrk1Δ mother cells fail to pass on their vacuoles to their daughters and that, consequently, vacuoles were formed de novo in their buds (Fig. 1C,D).

The inheritance defect for vacuoles in kin4Δfrk1Δ cells resembles the defect observed in mutants that lack Vac17. Western blot analysis revealed that protein levels of Vac17 tagged with the Staphylococcus aureus protein A (ProtA) were significantly reduced in kin4Δfrk1Δ cells compared to in WT cells (Fig. 2A,B; Fig. S8). Next, we tested whether Kin4 and Frk1 overexpression would lead to increased levels of Vac17. As KIN4 and FRK1 overexpression is lethal for WT cells by blocking mitotic exit via Bfa1 (Ekal et al., 2023b; Maekawa et al., 2007), BFA1-knockout (bfa1Δ) cells expressing Vac17-ProtA were transformed with plasmids containing either KIN4 or FRK1 under control of a strong galactose-inducible promoter (pGAL-KIN4, pGAL-FRK1). Protein extracts from cells grown on galactose were analysed by western blotting. Vac17-ProtA levels were elevated in cells overexpressing either KIN4 or FRK1 compared to control cells (Fig. 2C). We conclude that Kin4 and Frk1 affect Vac17 steady-state levels.

To study the effect of KIN4 or FRK1 overexpression on the positioning of vacuoles and Vac17-GFP, bfa1Δ cells were transformed with a Vac17-GFP expression plasmid under the control of its endogenous promoter and vacuoles were visualised by FM4-64 staining. Overexpression of either KIN4 or FRK1 increased the Vac17-GFP signal (Fig. 2D,E). Furthermore, Vac17-GFP was not visible in the majority (92.8%) of bfa1Δ cells transformed with a control plasmid (bfa1Δ+CP cells) but was visible in those overexpressing KIN4 or FRK1, where it was located either at the bud tip in small-budded cells or at the bud neck in large-budded cells (Fig. 2D,E). In large-budded control cells, vacuoles were positioned away from the bud neck, whereas in cells overexpressing KIN4 or FRK1, vacuoles frequently localised to the bud neck where they colocalise with Vac17-GFP (Fig. 2D,E). This phenotype is reminiscent of mutants in which vacuoles – owing to the continued association of vacuoles with Myo2 − are not released in the bud but are transported back to the bud neck, (Wong et al., 2020; Yau et al., 2014, 2017) (Fig. S2).

To test whether the increase in the level of Vac17 observed upon overexpression of KIN4 is a consequence of an increase in Vac17 stability, we performed a cycloheximide (CHX) chase assay. CHX blocks protein synthesis and, thereby, permits analysis of the degradation kinetics of steady-state protein levels. A control experiment in WT cells showed a clear and strong reduction of Vac17-ProtA levels upon treatment with CHX; however, no reduction in Vac17-ProtA levels was observed in cells not treated with CHX (Fig. S3). To study Vac17 turnover upon KIN4 overexpression, bfa1Δ cells transformed with the plasmid pGAL-KIN4 or a negative control plasmid were grown in medium containing galactose as a carbon source (YM+Gal) to induce GAL-KIN4 expression for 6 h and then treated with CHX. Vac17-ProtA turnover was significantly reduced in cells that overexpressed KIN4 compared to those that did not (Fig. 2F,G). Next, we used a well-established in vivo ubiquitylation assay for Vac17 (Yau et al., 2014, 2017). Briefly, in cells expressing Vac17-GFP, expression of Myc epitope-tagged ubiquitin (Myc-Ub) was induced by addition of CuCl₂. Subsequently, Vac17-GFP was precipitated by using GFP-TRAP and precipitates were analysed by immunoblotting for the presence of Myc-Ub to reveal relative levels of Vac17 ubiquitylation in vivo. We observed a clear ubiquitylation pattern in precipitates from cells expressing Vac17-GFP, which was absent in cells without Vac17-GFP expression (Fig. 2H). Upon KIN4 overexpression there was a significant decrease in Myc-ubiquitylated Vac17-GFP compared to those cells not overexpressing KIN4 (Fig. 2H,I). The above results show that changes in Kin4 protein levels affect Vac17 ubiquitylation and, thus, its turnover.

Phosphorylation of aa residue T209 within the Kin4 kinase activation loop by Elm1 is crucial for Kin4 function in SPOC as well as for peroxisome transport (Caydasi et al., 2010; Ekal et al., 2023b; Moore et al., 2010). To test whether Kin4 kinase activity is also required for vacuole inheritance, kin4Δfrk1Δ cells expressing either wild-type KIN4 or kin4-T209A were pulse-chased with FM4-64 and imaged. Vacuole inheritance was restored by wild-type Kin4 but not by kin4-T209A. The activation loop of Kin4 and Frk1 are identical in amino acid sequence including the T209 residue (Ekal et al., 2023b). Moreover, as shown previously, toxicity caused by FRK1 overexpression can be rescued by the deletion of the ELM1 gene (Ekal et al., 2023b). This tempted us to postulate that Frk1 is also a potential substrate for Elm1. Therefore, Frk1 was also tested in this assay. Indeed, expression of wild-type Frk1 restored vacuole inheritance whereas frk1-T209A only partially restored inheritance, illustrating the importance of T209 for Frk1 function (Fig. 3A,B). Restoration of inheritance is accompanied by an increase of Vac17-ProtA levels to almost WT levels (Fig. 3C). Moreover, Vac17-ProtA is phosphorylated on many sites (Peng and Weisman, 2008; Yau et al., 2014; Zhou et al., 2021). During SDS-PAGE, Vac17-ProtA from WT samples migrates differently compared to that derived from kin4Δfrk1Δ cells. Reintroduction of KIN4 or FRK1 in kin4Δfrk1Δ cells, changes the Vac17-ProtA migration pattern whereas the introduction of kin4-T209A resembled that of kin4Δfrk1Δ cells. The introduction of frk1-T209A resulted in an intermediate migration pattern (Fig. 3C). These observations suggest that Vac17-ProtA is differentially modified depending on Kin4 and Frk1 activity. However, whether the change in migration pattern is a result of direct phosphorylation by Kin4 or Frk1 is unknown. So far, experiments addressing this, have failed to identify specific Kin4 phosphorylation sites in Vac17. Furthermore, many buds of elm1Δ cells lacked FM4-64 pulse-labelled vacuoles, implying a vacuole inheritance defect. Not many budding mother cells were found without any vacuoles (category III), but this could be a consequence of the slow progression of the cell cycle through the G2/M phase in elm1Δ cells (Moriya and Isono, 1999) (Fig. 3D; Fig. S4A,B). Moreover, western blot analysis showed that Vac17-ProtA levels were strongly reduced in elm1Δ cells and were comparable to those in kin4Δfrk1Δ cells (Fig. 3E). Combined, all these observations strongly suggest that Elm1 activates Kin4 as well as, albeit to a lesser extent, Frk1 by phosphorylation of T209 in its activation loops and that this is required for vacuole inheritance.

During SPOC, the bipartite GTPase complex Bfa1−Bub2 is activated by phosphorylation of the Bfa1 subunit through Kin4 (Maekawa et al., 2007). In addition, the PP2A phosphatase subunit Rts1 regulates Kin4 localisation to spindle pole bodies; this is required for Kin4 function at the SPOC (Chan and Amon, 2009). We analysed vacuole inheritance in cells lacking BFA1, BUB2 or RTS1 (bfa1Δ, bub2Δ or rts1Δ cells, respectively). None of the mutants showed a strong defect in vacuole inheritance (Fig. S4A,B). Kar9 is a Myo2-receptor for astral microtubule transport and plays an important role in maintaining the spindle alignment along the cell polarity axis (Beach et al., 2000; Miller and Rose, 1998). Thus deletion of KAR9 (kar9Δ) leads to spindle misalignment in many cells, which results in activation of SPOC (Pereira et al., 2000). WT, kar9Δ and kin4Δfrk1Δ cultures that express GFP-tagged tubulin1 (GFP-Tub1), a marker for the mitotic spindle, were grown to exponential growth phase and stained with FM4-64, before being analysed by epifluorescence microscopy. The large-budded kin4Δfrk1Δ cells defective in vacuole transport do not show defects in spindle alignment as those observed in kar9Δ cells (Fig. S4C). Furthermore, in contrast to Vac17 and Inp2, there is no reduction in Kar9 protein levels in kin4Δfrk1Δ cells compared to in those in WT cells (Fig. 2A,B; Fig. S4D) (Ekal et al., 2023b). Taken together, we conclude that SPOC is not required for vacuole inheritance and, therefore, the role of Kin4 in vacuole transport is independent of its role in SPOC. These results are in line with previous observations, where the function of Kin4 in peroxisome transport has been shown to be independent of its function in SPOC (Ekal et al., 2023b).

The defect in vacuole inheritance observed in kin4Δfrk1Δ cells is caused by a decrease in organelle transport. Decreased transport could be caused by failure to assemble transport complexes to initiate transport or to maintain transport. Previous studies have shown that, in cells with defective recruitment of Myo2 by Vac17, Vac17 levels increase as vacuoles and Vac17 associated with them remain in the mother cell, and are kept away from the bud-restricted Vac17 degradation mechanism (Eves et al., 2012; Tang et al., 2003). In kin4Δfrk1Δ cells, Vac17 levels are reduced even though there is an inheritance defect, and most vacuoles remain in the mother cell. This suggests that it is not the assembly of the Myo2−Vac17 complex that is affected. Indeed, co-immunoprecipitation experiments revealed that Vac17-ProtA binds Myo2 in kin4Δfrk1Δ cells (Fig. 4A). However, the complex is present at lower levels and this may lead to inefficient transport to the bud. Typically, vacuoles form a segregation structure during the early stages of transport and the structures were observed in 29.3% (64 out of 218) of cells with small- to medium-sized buds (Fig. 4B). In kin4Δfrk1Δ cells, segregation structures were only observed in 5.3% (10 out of 187) of the cells (Fig. 4B). Time-lapse imaging of FM4-64 pulse chased cells confirmed that segregation structures are formed in kin4Δfrk1Δ cells but that, in contrast to in WT cells, these structures failed to be maintained over a long period of time (Fig. 4C). Instead, as shown in the example of the budding kin4Δfrk1Δ cell, the segregation structure observed at t=0 and t=10 disappeared and vacuoles are being synthesised de novo in the bud (Vph1-GFP stained only) between 30 min and 80 min (Fig. 4C). Subsequently, some transport of pre-existing vacuolar membrane (stained with FM4-64) occurred between 80 and 90 min (Fig. 4C), although we did not capture the transient segregation structure. Thus, while in WT cells the segregation structure lasted >1 h, the structure was short-lived in kin4Δfrk1Δ cells. We conclude that kin4Δfrk1Δ cells can assemble Myo2−Vac17 transport complexes but fail to maintain the continued vacuolar segregation.

Since Kin4 activity is restricted to the mother cell during SPOC in medium- to large-budded cells (Caydasi and Pereira, 2012) (Fig. S1B,C), we tested whether stabilisation of Vac17 by Kin4 also occurs in the mother cell. We made use of the previous observation that − in mutants affecting the Vac17-Myo2 interaction, including myo2-D1297N − Vac17 remains in the mother cell at elevated levels (Eves et al., 2012; Tang et al., 2003). In line with previous observations, elevated levels of Vac17-ProtA were detected in myo2-D1297N cells compared with those in cells expressing wild-type MYO2 (Fig. S5A). Interestingly, myo2-D1297N cells lacking KIN4 and FRK1 failed to maintain these elevated levels of Vac17 (Fig. S5B). A CHX chase assay revealed an increase in Vac17 turnover in myo2-D1297N cells lacking KIN4 and FRK1, compared to in myo2-D1297N cells with KIN4 and FRK1 present (Fig. S5C,D).

Next, we sought to test whether the above observations are also reproducible with Vac17 mutants that fail to interact with Myo2. The structure of Vac17 (112−157 aa) in complex with the Myo2-cargo-binding domain (CBD) had been predicted and the Myo2 interaction site (MIS) in Vac17 had been narrowed down to aa residues 131−145 (Liu et al., 2022). Moreover, point mutations within three Vac17-MIS residues (i.e. R135, K138 and R142) reduce interaction with Myo2-CBD in vitro (Liu et al., 2022) (Fig. S6A). To further corroborate these results, we performed yeast two hybrid (Y2H) assays to study the Vac17 and Myo2-CBD interaction in vivo, as described previously by Eves et al. (2012). Here, we generated single- and double-point mutations in the above three Vac17 residues, and tested them for interaction with Myo2-CBD using growth in the absence of adenine and histidine as readout (Fig. S6B). A control plasmid without Vac17 (BD) did not show any growth; however, expression of wild-type Vac17 showed proper growth, suggesting strong and specific interaction with Myo2-CBD (MYO2). Interestingly, vac17-R135E showed interaction with Myo2-CBD similar to wild-type Vac17, whereas vac17-K138D and R142E showed reduced interaction (Fig. S6B). However, the expression of three double mutants vac17-R135E,K138D, vac17-R135E,R142E or vac17-K138D,R142E did not restore cell growth at all and hinted towards a strongly reduced interaction with Myo2 (Fig. S6B). Subsequently, Myo2-CBD mutants (i.e. myo2-E1293 K, myo2-D1297N, myo2-N1304D) known to interfere with binding of Vac17 were tested in a two-hybrid assay. Here, myo2-D1296N was used as a control that still interacts with Vac17 (Fig. S6B). Interestingly, vac17-R142E and vac17-R135E,R142E restored interaction with myo2-E1293 K but not myo2-D1297N or myo2-N1304D. The restoration of an interaction between two charge reversal mutants that, on their own, are each affected in their binding to WT partners is strong evidence that these residues interact in vivo and support the structural model in Fig. S6A). vac17Δ cells expressing vac17-R135E mutant restored vacuole transport comparable to that of wild-type, whereas expression of vac17-K138D and vac17-R142E mutants only partially restored vacuole inheritance (Fig. S6C,D). Expression of vac17-R135E,K138D, vac17-R135E,R142E or vac17-K138D,R142E double mutants showed a strong defect in vacuole transport (Fig. S6C,D). Western blot analysis revealed that protein levels of all single and double mutants were increased compared to WT Vac17 levels − except for the vac17-R135E mutant, which did not affect vacuole inheritance (Fig. S6E). These results corroborated the Y2H analysis and validated the function of the previously identified Vac17-MIS in vivo. We selected vac17-R135E, K138D MIS mutant for further studies.

To study Vac17 localisation, wild-type Vac17-GFP or vac17-R135E,K138D-GFP were expressed and visualised in vac17Δ cells. Expression of vac17-R135E,K138D-GFP failed to restore vacuole inheritance in vac17Δ cells and vacuoles in mother cells were decorated by GFP-signals. In contrast, Vac17-GFP restored inheritance of vacuoles and was hardly detectable by epifluorescence microscopy, although weak Vac17-GFP puncta were present in some small buds (Fig. 5A,B), in agreement with previous reports (Yau et al., 2014, 2017). Western blot analysis revealed elevated levels of vac17-R135E,K138D-ProtA compared to Vac17-ProtA (Fig. 5C,D; Fig. S6E). This is in line with earlier observations in cells expressing myo2-D1297N (Yau et al., 2014) (Fig. S5A). Next, we analysed whether the elevated protein levels of vac17-R135E,K138D are affected by deletion of KIN4 and FRK1. Indeed, vac17-R135E,K138D protein levels were strongly reduced in the absence of both Kin4 and Frk1 (Fig. 5E). Therefore, as vac17-R135E,K138D is associated with the vacuole in the mother cell (Fig. 5A), we concluded that Kin4 and Frk1 protect Vac17 from degradation in mother cells.

Cla4/Dma1 and Yck3/ Vps41 have each been implicated in the spatially controlled breakdown of Vac17 (Wong et al., 2020; Yau et al., 2017). We, therefore, hypothesised that the defect in kin4Δfrk1Δ cells is a consequence of misregulation of one of these processes. To test this, kin4Δfrk1Δ cells lacking either CLA4 or STE20 were generated and vacuole inheritance was analysed in these strains. As the S. cerevisiae genome encodes the Cla4 paralog Skm1, we also included the SKM1 gene in our analysis. Deletion of CLA4 significantly restored transport of vacuoles to the bud in kin4Δfrk1Δ cells, whereas deletion of either STE20 or SKM1 failed to do so (Fig. 6A,B). Deletion of DMA1 in kin4Δfrk1Δ cells rescued the inheritance defect to almost WT level (Fig. 6C,D) but deletion of either YCK3 or VPS41 in kin4Δfrk1Δ cells did not. In contrast, kin4Δfrk1Δyck3Δ and kin4Δfrk1Δvps41Δ cells showed increased (additive) defects in vacuole inheritance compared to that in kin4Δfrk1Δ cells (Fig. S7A,B). Moreover, in contrast to those in yck3Δ and vps41Δ cells, observed Vac17-GFP levels were less at the bud tip/bud neck in kin4Δfrk1Δyck3Δ and kin4Δfrk1Δvps41Δ cells (Fig. S7C). Notice that, in line with earlier reports, yck3Δ and vps41Δ cells show a weak inheritance defect within some strain backgrounds, in addition to increased levels of Vac17 and a transport termination defect (Fig. S7B,D) (LaGrassa and Ungermann, 2005; Wong et al., 2020). These results indicate that Cla4 and Dma1 but not Yck3 and Vps41 function antagonistically to Kin4 and Frk1.

Cla4 and Dma1 act in multiple cellular pathways, including septin dynamics, which are crucial for bud morphogenesis and recruitment of Elm1 at the bud neck during SPOC (Merlini et al., 2012; Versele and Thorner, 2004). Hence, to rule out the possibility of a pleiotropic effect caused by deletion of CLA4 and DMA1, we analysed the Vac17 point mutants S222A and T240A, which are blocked in Cla4- and Dma1-dependent ubiquitylation, respectively (Yau et al., 2014, 2017). GFP-tagged VAC17, vac17-S222A and vac17-T240A were expressed in kin4Δfrk1Δvac17Δ cells, and vacuole inheritance was analysed using FM4-64 pulse-chase labelling. As expected, kin4Δfrk1Δvac17Δ cells expressing Vac17-GFP showed a severe defect in vacuole inheritance but expression of either vac17-S222A-GFP or vac17-T240A-GFP restored vacuole inheritance to WT levels (Fig. 6E,F). In addition, vac17-T240A-GFP and vac17-S222A-GFP signals were strongly increased and, in many cells, vacuoles were found to position inappropriately at the bud neck of large-budded cells, indicating a failure of timely vac17-T240A-GFP and vac17-S222A-GFP breakdown, respectively, as well as vacuole detachment from Myo2 (Fig. 6E). Moreover, Vac17-ProtA levels in S222A and T240A mutant cells were not affected by the absence of FRK1 and KIN4 (Fig. 6G). These results support a model, in which the decreased levels of Vac17 observed in kin4Δfrk1Δ cells are the result of Cla4-/Dma1-dependent Vac17 breakdown.

Since vacuole inheritance and Vac17 protein levels in kin4Δfrk1Δ cells are restored upon inhibition of Cla4-/Dma1-dependent turnover of Vac17, we hypothesised that Cla4 and Dma1 can act in the mother cell to stimulate Vac17 breakdown, but that this is normally counteracted by Kin4 and Frk1. To test this directly, we expressed the Vac17 MIS mutant vac17-R135E,K138D-ProtA in vac17Δ or kin4Δfrk1Δvac17Δ cells. This Vac17 mutant accumulated in mother cells at an increased level dependent upon the presence of KIN4 and FRK1 (Fig. 5A,E). Subsequently, we introduced point mutation S222A or T240A into vac17-R135E,K138D and expressed each triple mutant in kin4Δfrk1Δvac17Δ cells. Importantly, western blot analysis revealed that inhibition of Cla4-/Dma1-dependent degradation increased the levels of Vac17 (Fig. 6H). Similarly, as shown above, if Vac17 accumulated in mother cells because of a Myo2 mutation (myo2-D1297N) that inhibits binding to Vac17 and, consequently, blocks vacuole inheritance, Vac17 levels increase (Fig. S5A) (Yau et al., 2014). In myo2-D1297N cells lacking KIN4 and FRK1 (myo2-D1297N kin4Δfrk1Δ), this increase was no longer observed as a consequence of increased turnover (Fig. S5B,C,D). However, in contrast to Vac17, vac17-S222A levels were increased in myo2-D1297N kin4Δfrk1Δ cells (Fig. S5E). We conclude that both Cla4 and Dma1 stimulated the degradation of Vac17 in the mother cell but that Kin4 and Frk1 antagonise this activity, thereby protecting Vac17 from breakdown.

A defect in the timely degradation of Vac17 resulted in vacuoles mispositioning at the bud neck late during the cell cycle. Interestingly, in dma1Δdma2Δ cells, mispositioning of peroxisomes has also been reported (Yau et al., 2014). However, whether this is a result of increased Inp2 levels has not been reported. To study this, we analysed Inp2-ProtA protein levels in WT and dma1Δdma2Δ cells in western blots. Inp2-ProtA levels were clearly elevated in dma1Δdma2Δ cells compared to those in WT cells (Fig. 7A). Moreover, analysis of dma1Δdma2Δ cells expressing Inp2-GFP and the red peroxisomal marker mKate2-PTS1 showed an altered distribution of Inp2-GFP and peroxisomes. In dma1Δdma2Δ cells, peroxisomes were frequently lacking in mother cells and, when they were present, localised close to the bud neck. While Inp2-GFP in WT cells was mainly localised to peroxisomes within the bud, in dma1Δdma2Δ cells, Inp2-GFP was observed in both mother and daughter cells as long as the mother cells still contained peroxisomes (Fig. 7B,C). These results suggest that Inp2 levels are controlled by Dma1/2, and that this is important for proper peroxisome partitioning and positioning. Next, we analysed the effect of CLA4 overexpression on peroxisome segregation. Interestingly, high expression of CLA4 led to a clear defect in peroxisome transport to the bud (Fig. 7D,E) and CLA4 overexpression led to reduced Inp2 levels compared to those in control (Fig. 7F). Previously, we have reported that kin4Δfrk1Δ cells fail to maintain steady-state protein levels of Inp2 (Ekal et al., 2023b), similar to the effect on Vac17 (Fig. 2A,B). Interestingly, Inp2 levels were restored in kin4Δfrk1Δ cells upon additional deletion of DMA1 (Fig. 7G). These results demonstrated that transport of vacuoles and peroxisomes not only shares the requirement of Kin4 and Frk1 as regulators, but also Cla4 and Dma1/2.

Previously, we have found that inheritance of vacuoles is strongly affected in kin4Δfrk1Δ cells (Ekal et al., 2023b). In this current study, we presented evidence supporting a model in which vacuole transport is controlled through the antagonistic relationship between Kin4/Frk1 and Cla4/Dma1. We showed that Kin4 and Frk1 are redundant in vacuole inheritance, and are required to maintain Vac17 steady-state levels. Our experiments showed that levels of the Vac17−Myo2 complex are, indeed, decreased in kin4Δfrk1Δ cells. However, they do not reveal whether this effect was solely due to a reduction in Vac17 levels or to possibly reduced affinity between Myo2 and Vac17. In line with this, in kin4Δfrk1Δ cells, vacuole segregation structures were formed but not − as observed in WT cells − maintained over long periods. Thus, a vacuole inheritance defect was observed in small- and large-budded kin4Δfrk1Δ cells. Vacuoles are essential for cell-cycle progression from G1 to S phase (Jin and Weisman, 2015), hence, in many buds of kin4Δfrk1Δ cells that fail to inherit vacuoles, they are formed de novo. This illustrates that Kin4 in SPOC is not required for de novo vacuole formation and maturation. Furthermore, we showed that SPOC is not required for vacuole inheritance and, therefore, that the role of Kin4 in this process is independent of its role in SPOC-dependent regulation of MEN (Fig. S4). Further analysis revealed that Elm1-dependent activation of Kin4 kinase activity is required for vacuole inheritance and to maintain Vac17 steady-state levels. Interestingly, similar to Kin4-T209A, Frk1-T209A failed to restore Vac17 protein levels and, thus, vacuole inheritance in kin4Δfrk1Δ cells. In addition, previously we have reported that cell toxicity caused by Frk1 overexpression can be alleviated by additional deletion of either ELM1 or BFA1 (Ekal et al., 2023b). These observations strongly hint towards Frk1 being a direct substrate for Elm1. Next, we performed Y2H and FM4-64 pulse-chase assays to validate previously characterised residues in the Vac17-MIS motif, which are crucial for interaction with Myo2 in vivo. Expression of vac17-R135E,K138D (MIS mutant) in vac17Δ cells not only failed to transport vacuoles to the growing bud but also resulted in elevated Vac17 protein levels compared to Vac17 WT levels and those accumulated on the vacuole within the mother cell. We found that kin4Δfrk1Δ cells failed to maintain these elevated MIS mutant protein levels in mother cells.

Next, we hypothesised that Cla4/Dma1 causes premature Vac17 breakdown in kin4Δfrk1Δ mother cells. In agreement with this, we showed that blockage of Cla4-/Dma1-dependent Vac17 breakdown − by deletion of either CLA4 or DMA1 − rescued the inheritance defect in kin4Δfrk1Δ cells. Moreover, expression of the S222A or T240A Vac17 mutants that are resistant to Cla4-/Dma1-dependent breakdown restored both vacuole inheritance and Vac17 protein levels in kin4Δfrk1Δ cells. Furthermore, by using the Vac17-MIS mutant, we showed that Vac17 is broken down in mother cells in a Cla4-/Dma1-dependent manner but that this is prevented by the presence of Frk1 and Kin4. In addition, high levels of Kin4 and Frk1 not only lead to decrease in Vac17 ubiquitylation and increased levels of Vac17 but also cause mispositioning of the vacuole in the bud, as reported previously for cells defective in timely Cla4-/Dma1-dependent Vac17 breakdown in the bud (Yau et al., 2014, 2017). We, therefore, conclude that Kin4 and Frk1 negatively regulate Cla4-/Dma1-dependent Vac17 degradation and transport termination. This conclusion was further supported by genetic data showing that disruption of Cla4-/Dma1-dependent Vac17 breakdown restored vacuole inheritance in kin4Δfrk1Δ cells. Thus, kin4Δfrk1Δ cells may initiate but fail to maintain vacuole transport, as Kin4 is not inhibiting Cla4-/Dma1-dependent Vac17 breakdown in mother cells and, consequently, Myo2 is prematurely released from vacuoles before they reach the bud.

Previously we have found that Kin4 and Frk1 are required to maintain steady-state levels of Inp2, as observed for Vac17 (Ekal et al., 2023b). In this current study, we found that Inp2 is stabilised in dma1Δdma2Δ cells. Interestingly, CLA4 overexpression led to failure in peroxisome transport to the growing bud. Moreover, steady-state protein levels of Inp2 are reduced upon CLA4 overexpression. The above results support a model in which the machinery that regulates vacuole transport also – and in a very similar manner − regulates peroxisome transport, although further detailed studies are required to corroborate this hypothesis. Furthermore, Cla4 and Dma1/2 also regulate turnover of the mitochondrial Myo2 receptor Mmr1 and, thus, maintain mitochondrial homeostasis (Nayef et al., 2024; Obara et al., 2022), which suggests a generic mode of action for Cla4/Dma1/2 in the regulation of multiple organelles within the cell.

Although Cla4 activity has been shown to trigger vacuole transport termination (VTT) in large buds through stimulating ubiquitylation of Vac17, our observations show that Cla4-/Dma1-dependent Vac17 breakdown is not limited to large buds. It can also occur in mother cells but this is counteracted in WT cells by Kin4. Similar to the zone model explaining regulation of MEN (Fig. S1B,C), a model can be proposed to explain VTT in large-budded cells (Fig. 8A,B). In this model Kin4 prevents Vac17 degradation and premature VTT in the mother cell, whereas Cla4 stimulates VTT through Dma1-dependent Vac17 degradation in the bud. Kin4 is concentrated at the cortex of large-budded mother cells and, therefore, unable to inhibit Cla4-/Dma1-dependent degradation in large buds resulting in VTT (Fig. 8A,B).

Our results also provide an explanation for the conundrum as to how the transport of vacuoles is maintained in cells with small buds over extended periods of time (Fig. 8C). During S phase, in small-budded cells, localisation of both Cla4 and Kin4 is highly polarised to the growing bud cortex (Bartholomew and Hardy, 2009; Falk et al., 2011; Holly and Blumer, 1999; Pereira and Schiebel, 2005), and we propose that Kin4 prevents premature Cla4-/Dma1-dependent Vac17 breakdown in the emerging bud and in the mother cell during this period of the cell cycle (Fig. 8C). In support of this model, overexpression experiments show that Kin4 can act in the bud to protect Vac17 from Cla4-dependent degradation (Fig. 8D). However, as the cell cycle progresses, Kin4 becomes more confined to the cortex of the mother cell, and Cla4 − located in the growing bud cortex − will be less inhibited by Kin4, allowing phosphorylation and degradation of Vac17 in medium- to large-sized buds (Fig. 8C). In this model, the two opposing kinases act as a sort of intracellular morphogen, directing VTT. Changes in the distribution of these opposing kinases, either during progression through the cell cycle or as a result of genetic manipulation, would either allow or prevent Cla4-/Dma1-dependent degradation of Vac17 (Fig. 8D). However, our model may be too simple, and additional regulatory factors might transfer spatial information and integrate it with temporal information. Many questions remain. For instance: What is the target for Kin4? Although we have been unable to identify Kin4-specific phosphorylation sites on Vac17, it remains possible that Kin4 phosphorylates Vac17 directly. Alternatively, Kin4 could inhibit Cla4 activity or Dma1 activity on Vac17 and Inp2 by regulating another factor.

The mechanisms unravelled by studying organelle dynamics in yeasts are of general relevance. For instance, in humans, PAKs have multiple important functions that are crucial for smooth mitotic progression. In some cancers, PAKs are hyperactivated, and this causes defects in chromosome segregation leading to multipolar spindle formation (Kumar et al., 2017). Melanosomes are organelles that synthesise and store melanin pigment. The dynamics of melanosomes between melanocytes and keratinocytes are crucial for hair and skin colour. The class V myosin Myo5a and its receptor Slac2 (melanophilin) play an important role in melanosome transport in actin-rich dendrites of melanocytes (Marks and Seabra, 2001). Slac2 harbours a PEST motif and the slac2-PESTΔ mutant is defective in degradation and leads to perinuclear aggregation of melanosomes (Fukuda and Itoh, 2004). Thus, control of melanosome transport and positioning resembles that of yeast vacuoles. Moreover, melanosome biogenesis is similar to that of lysosomes (Marks and Seabra, 2001). In conclusion, the regulatory principles of organelle dynamics seem to be conserved from the yeast to higher eukaryotes. Therefore, the study of organelle maintenance in S. cerevisiae is likely to provide molecular insights that could be extrapolated to higher eukaryotes including humans.

Yeast strains used in this study are derivatives of S. cerevisiae strains, BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) and BY4742 (MATα his3Δ1 leu2Δ0 lys2Δ0 ura3Δ0) and they are listed in Table S1. BY4741 and BY4742 were referred to as wild-type (WT) cells throughout the paper. Single to multiple gene deletions were generated by replacing the entire coding sequence of the gene of interest as described in (Ekal and Hettema, 2023). The pFA6a-GFP(S65T)-spHIS5 plasmid was used as a template for PCR to introduce the GFP-tag at the C-terminal of the MYO2 open reading frame (ORF) in the genome.

Yeast expression plasmids used in this study were generated as described previously (Ekal et al., 2023a) and are listed in Table S2. Expression of Vph1-GFP was carried out under endogenous VPH1 promoter. Constitutive expression of Kin4 (and its Kin4-T209A mutant) or Frk1 (and its Frk1-T209A mutant) was carried out under endogenous KIN4 or FRK1 promoters, respectively, and conditional expression of Kin4 and Frk1 was under the GAL1/10 promoter. Expression of Vac17-ProtA/GFP or Inp2-ProtA/GFP was achieved using the native promoters of Vac17 or Inp2, respectively. C-terminal tagging of Vac17 and Inp2 does not affect their role in vacuole and peroxisome transport (Ekal et al., 2023b; Peng and Weisman, 2008; Tang et al., 2006). Vac17 point mutants were generated by sit-directed mutagenesis using PCRs and confirmed by sequencing analysis. Expression of the peroxisomal markers mKate2-PTS1 and mNG-PTS1 (Ekal et al., 2023a) was under the HIS3 promoter, and expression of Inp2-ProtA and Inp2-GFP was performed as described previously (Ekal et al., 2023b).

Yeast cells were grown at 30°C in either rich yeast peptone (YP) medium (carbon source, 1% yeast extract, 2% peptone), yeast minimal (YM) medium 1 (carbon source, 0.17% yeast nitrogen base without amino acids and ammonium sulphate, 0.5% ammonium sulphate) for the selection of all prototrophic markers, or YM medium 2 (carbon source, 0.17% yeast nitrogen base without amino acids and ammonium sulphate, 0.5% ammonium sulphate, 1% casamino acids) for the selection of the uracil prototrophic marker. As carbon sources, 2% (w/v) of either glucose or raffinose or galactose were added. The amino acid and nucleic acid were prepared as 100× stocks and added to the minimal medium as required. For immunoblotting or fluorescence microscopy analysis, a preculture from overnight-grown yeast cells was diluted with the appropriate medium to OD₆₀₀=0.1, and further grown to log phase (OD₆₀₀=0.5−0.6). To induce galactose-based protein expression, precultured cells were diluted to a selective YM medium containing galactose at an OD₆₀₀=0.3 and grown for 6−8 h before harvesting. For cycloheximide (CHX) assays, CHX (C-6255, Sigma-Aldrich) was added to yeast cell cultures at a final concentration of 35 µg/ml. After addition of CHX, cells were harvested at indicated time points and used for subsequent analysis. For quantification analysis, budding cells were considered as single cells.

Cells were grown to log phase before analysis using epifluorescence microscopy. Image acquisition was performed as described previously (Ekal et al., 2023a). Images were processed further using either Fiji-ImageJ-windows 64-bit software (v1.54f) (Schindelin et al., 2012) or Adobe Photoshop (version 24.7.0). To highlight the cell circumference, brightfield images were collected in one plane and processed where necessary in blue channel using Adobe Photoshop. Representative images from imaging experiments are shown.

Logarithmically growing cells (1−2 ml) were centrifuged (8000 g, 2 min) and the cell pellet was resuspended in 200 μl yeast extract peptone dextrose (YPD) medium containing FM4-64 (Invitrogen, T3166, 1 ng/μl final concentration). The FM4-64 staining was performed at 30°C for 1 h, following which the cells were centrifuged (8000 g, 2 min) and the supernatant was removed. The remaining cell pellet was washed thrice in YM medium. Subsequently, the cells were resuspended in 3−4 ml of fresh yeast minimal medium and incubated at 30°C for 4−5 h before imaging with an epifluorescence microscope.

WT and kin4Δfrk1Δ cells expressing Vph1-GFP under an endogenous promoter were grown to log phase. Cells were harvested and stained with FM4-64 as described above. For time-lapse imaging, 20 μl cell suspension in YM medium was immobilised within a (2% w/v) agarose gel pad in 35 mm μ-dish (Ibidi). Cells were spread uniformly by gently pressing on top of the gel pad. The agarose gel pads were prepared as described by Ekal et al., (2023a,b). Fluorescence images were collected as 0.5 μm z-stacks for every 10 min time point. Images were processed using Fiji-ImageJ software and Adobe Photoshop.

To analyse steady-state protein levels, protein extracts from logarithmically growing cells were prepared as described by Ekal et al., (2023a). Briefly, cells were lysed in a buffer containing 0.2 M NaOH and 0.2% β-mercaptoethanol, followed by protein precipitation using 5% trichloroacetic acid. Precipitated protein pellets were obtained by centrifugation and the pellets were resuspended in 10 μl 1 M Tris-HCl pH 9.4 and 90 μl 1×SDS–PAGE loading buffer, and denatured by boiling. Protein samples (OD₆₀₀=0.25–1 equivalent) were used for SDS–PAGE and further analysed by immunoblotting as reported previously (Ekal et al., 2023a). Myo2-GFP detection was with a monoclonal anti-GFP antibody (mouse IgG monoclonal antibody clone 7.1 and 13.1; 1:3000; Roche, #11814460001). ProtA-tagged Vac17 and Inp2 were detected by the peroxidase-anti peroxidase (PAP) antibody (rabbit; 1:4000; Sigma-Aldrich, #P1291). Myc-tagged ubiquitin (Myc-Ub) was detected by anti-Myc antibody clone 9E10 (mouse; 1:5000; Sigma-Aldrich, #M4439). Pgk1 was used as loading control and detected by a monoclonal anti-Pgk1 antibody (anti-mouse; 1:7000; Invitrogen, #459250). The secondary antibody was an HRP-linked anti-mouse polyclonal (goat; 1:4000; Bio-Rad, 1706516). Blots were incubated using enhanced chemiluminescence reagents (ECL, GE Healthcare) and protein bands were visualised by chemiluminescence imaging. See Fig. S8 for uncropped blot images.

For immunoprecipitation experiments, we transformed Myo2–GFP-expressing cells with a centromeric plasmid containing Vac17-ProtA under VAC17 promoter or an empty plasmid (Ycplac33). Myo2-GFP was affinity purified using GFP Trap agarose resins (GFP-Trap Agarose, GTA, ChromoTek) as described earlier (Ekal et al., 2023a). Affinity-purified samples were denatured by boiling and analysed by western blotting. Myo2-GFP was detected using anti-GFP antibody and Vac17–ProtA was detected using PAP. For further information see also ‘Immunoblotting’ section.

For yeast-two hybrid analysis, MATa and MATα of the S. cerevisiae strain PJ69-4A were used (Eves et al., 2012). MATa cells were transformed with plasmid encoding an activation domain fused to either wild-type Myo2 or the Myo2 point mutants as well as the LEU2 gene for auxotrophic selection. MATα cells were transformed with plasmid encoding a binding domain fused to either wild-type Vac17 or the Vac17 point mutants as well as TRP1 gene for auxotrophic selection. Transformed MATa and MATα cells were mated on a plate containing YPD-rich medium for 1 day, shifted to YM+Glu medium lacking leucine and tryptophan, and grown for another 2 days to select diploids. Selected diploid cells were grown for 3−4 days on YM+Glu (Leu-Trp-Ade-His-) medium supplemented with variable concentrations of 3-aminotriazole (3AT) (3 mM, 6 mM and 10 mM) or not. Finally, plates were imaged using the same setting for all images.

To detect ubiquitylated Vac17, in vivo ubiquitylation assay was performed as described before (Yau et al., 2014). Briefly, bfa1Δ cells expressing Kin4 under the endogenous KIN4 promoter or the inducible GAL1/10 promoter were transformed with VAC17-GFP and Myc-Ub plasmids. Cells were grown overnight in an appropriate YM medium supplemented with 2% (w/v) raffinose as a carbon source. The next day, the cells were diluted into YP rich medium containing 2% (w/v) galactose to induce GAL-KIN4 expression. Cells were grown for 4 h during the day, Myc-Ub expression was induced by addition of CuCl₂ (100 µM final concentration) and cells grown for another 4 h. After induction, 100 OD₆₀₀-equivalent cells were harvested by centrifugation. Vac17-GFP was affinity purified from cell extracts using GFP Trap agarose resins, as described previously (Ekal et al., 2023a) with a minor modification in the buffer, i.e. containing 25 mM Tris-Cl pH 7.2, 150 mM NaCl, 100 mM β-glycerol phosphate, 25 mM NaF, 1 mM EGTA, 1 mM MgCl2, 0.15% Tween-20, protease inhibitor cocktail and 20 mM N-ethylmaleimide. The immunoprecipitated proteins were denatured by boiling, followed by western blotting. Vac17-GFP was detected with anti-GFP antibody, Myc-Ub was detected using anti-Myc antibody. For more details see the ‘Immunoblotting’ section above.

For epifluorescence imaging experiments, budding cells were considered to be individual cells. Images were inspected manually using the Volocity software and (version 7.0.0, Quorum technologies Inc). For quantification of protein levels, western blots showing unsaturated protein bands were analysed in Image Lab software (version 6.1, Bio-Rad). For plotting graphs and statistical analysis GraphPad Prism 10.0.0 (153) (accessed on 12 June 2023) software was used. Statistical analysis was performed using a paired t-test or a one-way ANOVA test or a two-way ANOVA test as indicated. ****P<0.0001; ***P<0.005; **P<0.01; *P<0.05; non-significant (ns), P>0.05.

We thank Lois Weisman (University of Michigan, Ann Arbor, MI, USA) for providing the plasmid constructs for Myo2 mutants, Vac17-GFP and Myc-Ubiquitin, and for valuable discussions. We would also like to thank Maya Schuldiner (Weizmann Institute of Science, Rehovot, Israel) for sharing TEF2-CLA4 overexpression strains.

The peer review history is available online at https://journals.biologists.com/jcs/lookup/doi/10.1242/jcs.261948.reviewer-comments.pdf