Skip to main content
NCBI home page
Elsevier Sponsored Documents logoLink to Elsevier Sponsored Documents
. 2020 Jun 16;31(11):107757. doi: 10.1016/j.celrep.2020.107757

Proline-Rich Motifs Control G2-CDK Target Phosphorylation and Priming an Anchoring Protein for Polo Kinase Localization

Mihkel Örd 1, Kait Kaarel Puss 1, Rait Kivi 1, Kaidi Möll 1, Tuuliki Ojala 1, Irina Borovko 1, Ilona Faustova 1, Rainis Venta 1, Ervin Valk 1, Mardo Kõivomägi 2, Mart Loog 1,3,∗
PMCID: PMC7301157  PMID: 32553169

Summary

The hydrophobic patch (hp), a docking pocket on cyclins of CDKs (cyclin-dependent kinases), has been thought to accommodate a single short linear motif (SLiM), the “RxL or Cy” docking motif. Here we show that hp can bind different motifs with high specificity. We identify a PxxPxF motif that is necessary for G2-cyclin Clb3 function in S. cerevisiae, and that mediates Clb3-Cdk1 phosphorylation of Ypr174c (proposed name: Cdc5 SPB anchor-Csa1) to regulate the localization of Polo kinase Cdc5. Similar motifs exist in other Clb3-Cdk1 targets. Our work completes the set of docking specificities for the four major cyclins: LP, RxL, PxxPxF, and LxF motifs for G1-, S-, G2-, and M-phase cyclins, respectively. Further, we show that variations in motifs can change their specificity for human cyclins. This diversity could provide complexity for the encoding of CDK thresholds to achieve ordered cell-cycle phosphorylation.

Keywords: cell cycle, cyclin-dependent kinase, Polo kinase, kinase specificity, short linear motif, SLiM, intrinsically disordered regions

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Budding yeast G2-cyclin Clb3 uses a PxxPxF motif for specific substrate targeting

  • •

    PxxPxF promotes Ypr174c phosphorylation to recruit Polo kinase Cdc5 to centrosomes

  • •

    Hydrophobic patch of cyclins can bind both exclusive and universal docking motifs

  • •

    Elements around the docking motifs tune their specificity for yeast and human CDKs

Örd et al. find that a specific cyclin docking motif mediates G2-CDK target phosphorylation in S. cerevisiae, including priming a docking site for Polo kinase Cdc5 centrosome recruitment. The study also shows how variations within cyclin motifs (RxL, PxxPxF, and ExLxF) result in different docking specificities for yeast and mammalian CDKs.

Introduction

Intrinsically disordered protein segments often contain sequence motifs less than 10 amino acids long that bind specific binding pockets in globular protein domains (Davey et al., 2012). Initially, when crystallography was the primary method for solving the structures of folded protein domains, the disordered regions were largely dismissed as less important elements. During the last decade, however, these sequence elements, also called short linear motifs (SLiMs), have been found in a wide variety of eukaryotic proteins, and the majority of protein signaling is now thought to be mediated by SLiMs (Davey et al., 2012, Iakoucheva et al., 2002). However, only a fraction of SLiMs has been mapped so far (Tompa et al., 2014).

A number of SLiMs mediate specific signaling interactions for key enzymes of the cell-cycle control system: protein kinases (Miller and Turk, 2018), phosphatases (Brautigan and Shenolikar, 2018, Hertz et al., 2016, Kataria et al., 2018), and ubiquitin ligases (Ravid and Hochstrasser, 2008). Among these are cyclin-dependent kinases (CDKs), the enzymes that are activated by waves of different cyclins, and function as master regulators of the cell cycle (Morgan, 2007). While cyclins were initially considered as merely the activating subunits of CDK, it is now clear that they also provide binding pockets for SLiMs in target proteins to facilitate cell-cycle-stage-specific phosphorylation (Bhaduri and Pryciak, 2011, Kõivomägi et al., 2011, Loog and Morgan, 2005, Örd et al., 2019a, Örd et al., 2019b, Swaffer et al., 2016).

Early studies revealed that an RxL motif (defined as R/K-x-L-Φ or R/K-x-L-x-Φ, where Φ stands for a large hydrophobic amino acid) facilitates specific phosphorylation by S-phase cyclin-CDK in mammalian and yeast cells (Figure 1A; Chen et al., 1996, Loog and Morgan, 2005, Lowe et al., 2002, Russo et al., 1996, Schulman et al., 1998, Takeda et al., 2001, Wilmes et al., 2004, Wohlschlegel et al., 2001). The binding pocket on cyclins, the hydrophobic patch (hp), was identified by mutational and crystallographic studies (Brown et al., 1999, Schulman et al., 1998), and a mutation in this patch (denoted as hpm) reduced specificity to the same extent as a mutation in the RxL motif in substrates. Recent studies in S. cerevisiae have broadened the range of SLiMs used by CDK, as substrate targeting by G1-, S-, and M cyclins were found to be facilitated by three different motifs. The G1 cyclins (Cln1 and Cln2) bind LP motifs (Bhaduri and Pryciak, 2011, Kõivomägi et al., 2011), the S cyclin (Clb5) and the G2 cyclin (Clb3) use RxL motifs (Kõivomägi et al., 2011), and the M-cyclin Clb2 uses an LxF motif for both substrate and inhibitor targeting (Örd et al., 2019a). Similar differential use of docking motifs can be predicted for mammalian CDKs. For example, a global analysis of the interactors of human cyclins E, A, and B revealed that the cyclin-CDK substrate interactions vary according to the cyclin and cell-cycle stage (Pagliuca et al., 2011).

Figure 1.

Figure 1

Open in a new tab

Docking Specificity Is Essential for Clb3 Function in Spindle Formation and Mitotic Transcription

(A) Structural model of the cyclin-CDK complex with substrate peptides bound to the CDK active site (SP) and the cyclin hp (RxLxF), based on cyclin A-Cdk2 structure 2CCI (Cheng et al., 2006).

(B) Time-lapse microscopy images showing a cell cycle of a cell expressing Spc42-GFP and NLS-NES-mCherry Cdk1 activity sensor. Between 3 min and 6 min, NLS-NES-mCherry is exported from the nucleus due to a rise in CDK activity at G1/S. At 51 min, the SPBs are separated and at 69 min, the spindle elongates.

(C) The time from G1/S to spindle formation was measured in single cells in a microscopy experiment described in (B). The lines show median time with 95% confidence interval (CI). ∗∗∗∗p < 0.0001 and not significant (ns) > 0.05 by Mann-Whitney test for pairwise comparisons with wild type.

(D) The time from spindle formation to elongation as measured in single cells. The plot shows single cell values with median and its 95% CI. ∗∗∗∗p < 0.0001, ∗∗p < 0.01, and ns > 0.05 by Mann-Whitney test for comparisons with wild type.

(E) Fluorescence levels of Clb2-Citrine after Start defined by nuclear export of 50% of Whi5-mCherry were measured in single cells of wild-type, clb3Δ, and clb3(hpm) strains. The plot shows mean ± SEM of Clb2-Citrine fluorescence intensities.

(F) The fluorescence intensities of Clb3-Citrine and Clb3(hpm)-Citrine were measured using live cell microscopy. Plot shows the mean ± SEM of cells synchronized by the nuclear export of 50% of Whi5-mCherry.

See also Figure S1.

In contrast to mammalian cells, where S- and M phases are controlled by A- and B-type cyclins, respectively, in budding yeast, events from S phase to the end of mitosis are coordinated by a series of closely related B-type cyclins (Bloom and Cross, 2007). While the yeast non-B-type G1 cyclins Cln1 and Cln2 contain a unique binding site for LP motifs (Bhaduri et al., 2015), the B-type cyclins (Clb5 and Clb2) bind their respective SLiMs, RxL and LxF, via the conserved hp. Furthermore, Clb3, although having a hp-dependent specificity increase of about 3- to 5-fold for the RxL motif, has no known specific SLiM of its own. However, our previous studies revealed that there are several Cdk1 targets (Ypr174c, Tos4, and Ash1) that are specifically phosphorylated by Clb3-Cdk1 in a hp-dependent manner (Kõivomägi et al., 2011). This suggests that the hp, despite its conserved nature, can accommodate various RxL-like, and non-RxL-like SLiMs, to exclusively target a specific B-type cyclin.

Here, we addressed this hypothesis of overlapping and specific targeting of the major B-type cyclins in budding yeast. Our study adds the final missing element to the set of cyclin specificities by showing that the G2-cyclin Clb3 hp binds a specific “non-RxL” SLiM. We mapped the Clb3 specificity determinants of an uncharacterized protein Ypr174c and show how phosphorylation of Ypr174c facilitates pre-anaphase localization of Polo kinase Cdc5 at spindle pole bodies (SPBs). Using a set of synthetic substrates, we show that the SLiMs used for docking by yeast and mammalian cyclins, despite sharing some features, can utilize much wider sequence diversity than appreciated so far. We show that the flanking sequences around the motifs can be either more specific for S- or M cyclins, suggesting that temporal coordination of “early” and “late” phosphorylation by CDK can be tuned by sequence variation around and within the SLiMs. Furthermore, we were able to demonstrate fully orthogonal signaling specificity among closely related Clb cyclins.

Results

Hp Docking of G2 Cyclin Clb3 Mediates Spindle Formation and Activates Expression of Late Cyclins

To study the contribution of Clb3 and its hp to mitotic processes, we first addressed its effect on spindle dynamics. Previous work suggested that Clb3 and Clb4 promote spindle assembly (Fitch et al., 1992, Richardson et al., 1992, Schwob and Nasmyth, 1993, Segal et al., 2000). To study this further, we used time-lapse microscopy to analyze the timing of spindle assembly and elongation. As a marker for G1/S we used a CDK activity sensor based on nuclear-cytoplasmic shuttling of mCherry fused to a phospho-regulatable nuclear localization module (Liku et al., 2005). Spindle dynamics were detected using Spc42-GFP-tagged SPBs (Figure 1B). The median time from G1/S to spindle formation, defined by separation of SPBs, and from spindle formation to elongation, was 30 min in both cases (Figures 1C and 1D). CLB3 deletion led to a 12-min delay in spindle formation, while deletion of CLB4 had a minor effect (Figure 1C). Deletion of CLB4 in clb3 deletion background further delayed spindle formation, showing that both Clb3 and Clb4 contribute to SPB separation, as suggested earlier (Schwob and Nasmyth, 1993, Segal et al., 2000). Interestingly, mutation of Clb3 hp caused a similar delay in spindle formation as CLB3 deletion (Figure 1C). Therefore, Clb3 with a functional hp is necessary for timely spindle assembly. Importantly, several SPB proteins, such as the γ-tubulin ring complex (Spc97-Spc98-Tub4), are phosphorylated preferentially by Clb3-Cdk1 (Ear et al., 2013). In clb3 or clb4 mutant strains, the time from spindle formation to elongation is decreased compared to wild type (Figure 1D). Thus, the G2-CDK specificity is important for proper timing of these events, as in the absence of Clb3/4, mitotic cyclin Clb2 can promote both spindle assembly and elongation, but with altered timing.

We hypothesized that differential specificity of the consecutive cyclins Clb5, Clb3, and Clb2 could be important in the temporal control of cell-cycle transcriptional programs, as suggested previously (Linke et al., 2017). Cdk1 activates CLB2 transcription via phosphorylation of Fkh2 and Ndd1 (Linke et al., 2017, Reynolds et al., 2003). To study the role of Clb3 specificity in the transcriptional networks, we measured the protein levels of Clb3-Citrine and Clb2-Citrine in wild-type or clb3(hpm) strains (Figures 1E, 1F, and S1A). We found that the expression of both Clb3 and Clb2 is delayed by about 10 min in clb3(hpm) strains (Figures 1E, 1F, and S1B–S1E). Interestingly, the accumulation of Clb2 is delayed to the same extent in both clb3Δ and clb3(hpm) strains (Figures 1F and S1B–S1E), indicating that the docking specificity of Clb3 is critical. These data predict that Clb3 has multiple functions that are mediated by specific interactions of the hp. Also, we have previously found that the nuclear levels of Clb5 and Clb2 are about 3-fold higher than the level of Clb3 (Örd et al., 2019b), which suggests that instead of providing a major contribution to the bulk of Cdk1 activity for mitotic thresholds, relatively low levels of Clb3 act via specific targeting of a few key substrates.

A PxxPxF Motif Enables Clb3-Specific Phosphorylation

To reveal the mechanism of Clb3 specificity, we chose an uncharacterized protein, Ypr174c, which has exclusive specificity for Clb3-Cdk1 (Kõivomägi et al., 2011). Ypr174c is predicted to have a disordered C terminus with four CDK phosphorylation sites (Figure 2A). We made C-terminal deletions of Ypr174c (Figure 2A), and analyzed phosphorylation of these proteins by Clb3- and Clb2-Cdk1. Interestingly, while deletion of positions 198–221 did not affect phosphorylation of Ypr174c by Clb3-Cdk1, deletion of 192–221 reduced the phosphorylation rate significantly (Figures 2B, S2A, and S2B). The deletions had little effect on Clb2-mediated phosphorylation (Figure 2B), suggesting that Ypr174c positions 192–198 contain a docking motif for Clb3.

Figure 2.

Figure 2

Open in a new tab

Mapping of Clb3 Docking Motif in Ypr174c

(A) Scheme showing the disordered C terminus of Ypr174c, CDK phosphorylation sites, and the C-terminal truncations used in (B).

(B) Phosphorylation of Ypr174c truncations by Clb3- and Clb2-Cdk1 was analyzed in a kinase assay, an autoradiograph of the reactions is shown.

(C) Autoradiograph showing phosphorylation of Ypr174c mutants by Clb3-Cdk1. The construct labeled as pxf contains mutations P190A P193A F195A. WT/mut shows the phosphorylation rate of the substrate compared to wild-type Ypr174c.

(D) 6–10 amino acid segments containing the PxF motif from Ypr174c were introduced to a Sic1-based substrate containing one minimal consensus phosphorylation site (SPxA).

(E) Autoradiographs showing phosphorylation of the substrate constructs described in (D) by Clb3-Cdk1. Relative specificity shows the phosphorylation rate of the indicated substrate relative to phosphorylation of histone H1 by Clb3-Cdk1.

See also Figure S2.

As this region does not contain any previously described cyclin-docking motifs, we mapped the effects of single amino acid substitutions within the region of 189–198 of Ypr174c on Clb3-Cdk1-mediated phosphorylation. Mutations of P190, P193, and F195 markedly decreased the phosphorylation rate, while mutations in other positions had little effect (Figures 2C and S2C).

This suggested that PxxPxF could be a SLiM for Clb3-specific docking (later denoted as PxF motif). To test whether the sequence in Ypr174c is a modular linear docking motif, similar to RxL and LxF motifs (Örd et al., 2019a, Schulman et al., 1998), we introduced fragments from Ypr174c containing the PxF with different lengths of flanking sequences into a minimal CDK model substrate Sic1(1–33) containing one minimal consensus phosphorylation site (Figure 2D). We found that addition of a 6 amino acid motif—PKGPNF—only mildly enhanced phosphorylation by Clb3-Cdk1 (Figures 2E and S2D). This suggests that the flanking residues around PKGPNF in Ypr174c could also be critical for specificity. By attaching longer motifs to Sic1(1–33), we found two minimal segments PPKGPNF and PPKGPNFYAK that greatly increased the phosphorylation rate, by 65- and 160-fold, respectively (Figure 2E). These results lead to a possibility that a proline at position −1 from the PxxPxF and lysine at +3 contribute significantly to the docking. Importantly, the motifs are specific to Clb3-Cdk1, as no increase in phosphorylation rate was seen with Cln2-, Clb5-, and Clb2-Cdk1 (Figure S2E). This demonstrates that the closely related B-type cyclins in S. cerevisiae have evolved different SLiMs for specific substrate docking, while having some level of common specificity for the RxL motif (Kõivomägi et al., 2011).

The PxF Motif Is Present in Several Clb3-Cdk1 Targets

Next, we aimed to determine if the PxF motif also directs Clb3-specific phosphorylation of other targets. We searched for potential PxF motifs in the disordered regions of S. cerevisiae proteins. Based on the mapping in Figure 2, we searched for the motif [PILVM]PxxPxFxx[KR], with the key residues fixed in the core motif PxxPxF, but allowing a selection of hydrophobic residues in the N-terminal position −1, and basic residues in C-terminal position +3. We found the motif in four proteins: Ypr174c, helicase Sen1, lipase Tgl5, and polar growth protein Boi1 (Figure 3A). We purified protein fragments containing the disordered regions with the predicted motifs and CDK phosphorylation sites and analyzed the cyclin specificity of their phosphorylation (Figures 3B and S3A). A docking-independent substrate histone H1 was used as a control (Kõivomägi et al., 2011). Tgl5 and Boi1 were phosphorylated most efficiently by Clb3-Cdk1, whereas Sen1 had high specificity with both Cln2- and Clb3-Cdk1 (Figure 3B). As with Ypr174c, mutation of Clb3 hp decreased the rate of Tgl5, Sen1, and Boi1 phosphorylation markedly (Figures 3C and S3B). Importantly, mutation of the predicted PxF motif led to a similar decrease in the phosphorylation rate, further suggesting that the motif functions as a specific SLiM for the Clb3 hp (Figures 3D and S3C).

Figure 3.

Figure 3

Open in a new tab

PxF Motif Mediates Clb3-Specific Phosphorylation

(A) Multiple sequence alignment of predicted PxF motifs in Ypr174c, Tgl5, Boi1, and Sen1.

(B) Cyclin specificity in phosphorylation of Cdk1 substrates was tested in a kinase assay with Cln2-, Clb5-, Clb3-, and Clb2-Cdk1. Autoradiographs of the reactions are shown.

(C) Clb3-Cdk1 targets were phosphorylated with either wild-type or hydrophobic patch mutant (hpm) Clb3-Cdk1.

(D) The importance of PxF motif in Clb3-specific phosphorylation was analyzed in a kinase assay with Clb3-Cdk1 and either wild-type Ypr174c, Tgl5, Boi1, Sen1, or the pxf mutants, where the predicted PxF shown in (A) was mutated.

See also Figure S3.

We have identified more hp-dependent Clb3-specific Cdk1 targets, such as Ndd1, Plm2, Spc29, Tos4, and Ash1 (Figure 3B; Kõivomägi et al., 2011). However, they do not contain an exact match to the PxF motif, suggesting that there could be additional Clb3 docking motifs, or variations in the PxF motif.

Ypr174c Is Phosphorylated by Clb3-Cdk1 In Vivo

Next, we set out to analyze the function of Ypr174c phosphorylation. Both mutation of Clb3 hp and deletion of CLB3 partially reduced the phosphorylation shifts of Ypr174c-6HA in asynchronous cultures, while CLB4 had no effect (Figure 4A), indicating that Ypr174c could be phosphorylated specifically by Clb3-Cdk1 in vivo. Further, mutation of the Ypr174c PxF motif led to a similar effect on Ypr174c phosphorylation as seen in clb3 deletion and clb3(hpm) strains (Figures 4A and 4B). Mutation of phosphorylation sites located N-terminally from the PxF, S158, S170, and S175, all reduced the phosphorylation shifts, indicating that all three sites are phosphorylated, as also suggested by the in vitro assays with Clb3-Cdk1 (Figures 4B and S4A). However, since all mobility shifts did not disappear in strains with clb3 deletion or pxf mutation in Ypr174c, other proline-directed kinases or other CDK complexes apparently also contribute. To study the Clb3 dependency of different sites in vivo, we analyzed the effect of clb3 deletion in single-phosphorylation-site mutants (Figure 4C). While deletion of CLB3 decreased the multisite phosphorylation of Ypr174c(S158A), the phosphorylation pattern of S170A and S175A mutants was less affected, indicating that the optimal site S158 is efficiently phosphorylated also in the absence of Clb3, whereas the suboptimal sites S170 and S175 are more dependent on Clb3 (Figure 4C).

Figure 4.

Figure 4

Open in a new tab

Clb3-PxF Docking Mediates Phosphorylation of Ypr174c and Localization of GFP-NES-Clb3

(A–C) Western blot images showing phosphorylation of Ypr174c-6HA in indicated strains in asynchronous cultures. The proteins were separated using either Phos-tag SDS-PAGE or conventional SDS-PAGE.

(A) Analysis of Ypr174c-6HA phosphorylation in wild-type, clb3Δ, clb4Δ or clb3(hpm) strains.

(B) Phosphorylation analysis of different Ypr174c mutants. Ypr174c(AP) carries mutations S69A S158A S170A S175A S211A, Ypr174c(pxf) is P190A P193A F195A mutant.

(C) The effect of clb3 deletion on the phosphorylation pattern of Ypr174c-6HA mutants, where the indicated phosphorylation site has been mutated.

(D) Microscopy images showing localization of Ypr174c-GFP different Cdk1 phosphorylation site mutants.

(E) Images showing localization of GAL1-expressed GFP-NES-Clb3 and Ypr174c-mCherry.

(F) Co-localization of GFP-NES-Clb3 and Ypr174c-mCherry was quantified using Pearson’s correlation coefficient (Dunn et al., 2011, Pearson, 1896). The plot shows mean ± 95% CI correlation coefficients from eight, six, and six images with 556, 428, and 909 cells for wild type, ypr174c(pxf) and clb3(hpm), respectively. ∗p < 0.05 and ∗∗∗∗p < 0.0001, with two-tailed t test for comparisons with wild type.

(G) Images showing localization of GFP-NES-Clb3 and mCherry-Tgl5 expressed from GAL1 promoter.

(H) Plot showing mean ± 95% CI of Pearson’s correlation coefficient of GFP-NES-Clb3 and mCherry-Tgl5 localization. Data are from seven, four, and four images containing 143, 485, and 164 cells for wild-type, tgl5(pxf), and clb3(hpm), respectively. ∗∗∗p < 0.001 with two-tailed t test for comparisons with wild type.

See also Figure S4.

Ypr174c is a protein of unknown function that localizes to the SPBs and nuclear envelope (Sundin et al., 2004). To explore the function of Ypr174c phosphorylation, we first tested how mutations in the CDK phosphorylation sites affect its localization. Mutation of all five CDK consensus sites (Ypr174c(AP)) abolished both SPB and nuclear envelope localization of Ypr174c (Figure 4D). Single-site mutations, however, showed similar localization as the wild-type protein, suggesting that different combinations of phosphorylation sites are sufficient for localization. Although dependent on phosphorylation sites, Ypr174c-GFP still localized to the nuclear envelope in G1 cells, prior to CDK activation (data not shown); this suggests that other kinases can contribute to its phosphorylation, consistent with the presence of residual gel shifts after deleting CLB3 or inhibiting Cdk1 (Figures 4A, 4C, and S4B).

Further, we observed that overexpression of GFP-tagged Clb3 with an added nuclear export signal (NES, to better visualize nuclear envelope signal relative to net nuclear signal) leads to accumulation of Clb3 in the nuclear envelope and SPBs, similar to Ypr174c (Figure 4E). To analyze if the co-localization of Clb3 and Ypr174c was mediated by the PxF motif, we mutated the docking motif in Ypr174c or the hp in Clb3. Both mutations caused a loss of GFP-Clb3 accumulation in the nuclear envelope (Figures 4E, 4F, and S4C). Cytoplasmic GFP-Clb3 also localizes to discrete points in the cytoplasm (Figures 4E and 4G). As the Clb3-specific target lipase Tgl5 has been shown to accumulate to lipid particles (Athenstaedt and Daum, 2005), we tested if the GFP-Clb3 was co-localized with Tgl5. We found a high overlap between GFP-Clb3 and mCherry-Tgl5 signals (Figures 4G and 4H). Importantly, when the PxF motif in Tgl5 was mutated, the strong co-localization of Clb3 and Tgl5 was lost (Figures 4G and 4H). We also detected discrete GFP-Clb3(hpm) signals, but these were not co-localized with Tgl5 or Ypr174c (Figure S4D). Therefore, the PxF motif can lead to increased localization of Clb3 to the nuclear envelope, SPBs, and lipid particles.

Ypr174c Recruits Cdc5 to SPBs in Metaphase

To further uncover the function of Ypr174c phosphorylation, we looked for additional SLiMs in the C terminus of Ypr174c (Figure 5A). Interestingly, the sequence 165PLVTS170SP matches the consensus of the phospho-dependent Polo-box binding motif (Elia et al., 2003). This motif contains the Clb3-Cdk1 phosphorylation site S170 (Figure 4C), and an interaction between Ypr174c and Polo kinase Cdc5 has been detected in a two-hybrid screen (Yu et al., 2008). Cdc5 localizes to the nuclear side of SPBs from S phase to early anaphase, when it relocates to the cytoplasmic side of SPBs in a Bfa1- and Cdc14-dependent manner (Botchkarev et al., 2014, Botchkarev et al., 2017, Song et al., 2000). The nuclear SPB-binding partner of Cdc5, however, is unknown. Therefore, we asked whether Ypr174c affects the localization of Cdc5. Consistent with previous reports, we found that in metaphase, Cdc5-Venus accumulated to the SPBs and nucleus (Figure 5B). In the ypr174c(AP) strain, however, Cdc5-Venus was detectable only as a nucleoplasmic signal prior to anaphase, indicating that the metaphase SPB localization of Cdc5 is dependent on Ypr174c phosphorylation (Figure 5B). When the CDK site in the Polo-box binding motif was mutated (Ypr174c(S170A)), only a faint Cdc5-Venus signal was detectable at SPB in metaphase, whereas in anaphase the SPB localization was similar to the wild type (Figure 5B). We used Spc42-mCherry-tagged SPBs to quantify the localization of Cdc5-Venus to the SPBs during the cell cycle. This revealed a strong decrease in Cdc5 signal at the SPBs in ypr174c, ypr174c(AP), and ypr174c(S170A) strains, whereas the nuclear Cdc5 signal remained similar (Figures 5C, 5D, and S5A–S5C). Mutation of Ypr174c PxF motif resulted in a slight decrease of Cdc5 SPB accumulation (Figures 5C and S5A–S5C). The presence of S170 in a background where all other CDK sites are mutated (Ypr174c(AP S170)) was sufficient to restore Cdc5 SPB localization (Figure 5E). Interestingly, mutation of PxF motif or deletion of CLB3 in this background caused a drop in Cdc5 SPB levels compared to the nuclear levels, suggesting that the PxF motif promotes Clb3-dependent phosphorylation of S170 (Figures 5E and S5D). However, clb3 deletion in wild-type background led to a minor increase in both nuclear and SPB-localized Cdc5, and did not affect the ratio of nuclear to SPB-localized Cdc5 (Figures S5E and S5F), suggesting that phosphorylation by bulk Cdk1 activity may offer a backup when Clb3 expression is less pronounced.

Figure 5.

Figure 5

Open in a new tab

Ypr174c Recruits Cdc5 to SPBs

(A) Scheme showing the predicted SLiMs in Ypr174c.

(B) Microscopy images showing localization of Cdc5-Venus in metaphase and anaphase cells of wild-type, ypr174c(AP), and ypr174c(S170A) strains. Ypr174c(AP) carries mutations S69A S158A S170A S175A S211A.

(C and D) Quantified nuclear and SPB-localized Cdc5-Venus fluorescence signals. Spc42-mCherry was used to detect spindle elongation.

(C) Mean ± SEM fluorescence intensities of Cdc5-Venus co-localized with Spc42-mCherry.

(D) Mean ± SEM nuclear Cdc5-Venus fluorescence levels from the anaphase onset.

(E) Plots showing the peak nuclear or SPB-localized Cdc5-Venus levels in single cells. The error bars show 95% CI of the median. ∗∗∗∗p < 0.0001, ∗∗∗p < 0.001, and ns > 0.05 by Mann-Whitney test for pairwise comparisons of either nuclear or SPB-localized Cdc5 signal intensities with ypr174c(AP S170).

(F) Co-immunoprecipitation of Cdc5-6HA from cell extracts using GST-tagged Ypr174c as bait. GST-Ypr174c was phosphorylated with Clb3-Cdk1 in the indicated lanes prior to addition of yeast lysate. Ponceau S staining shows the amount and phosphorylation of GST-Ypr174c. The co-immunoprecipitation was performed in two individual replicate experiments, a representative example is shown.

(G) Competitive Ypr174c phosphorylation and dephosphorylation assay with Clb3-Cdk1 and Cdc14. Ypr174c was phosphorylated with Clb3-Cdk1, then different concentrations of Cdc14 were added and the phosphorylation of Ypr174c was followed by autoradiography.

(H) Western blots showing multisite phosphorylation of Ypr174c-6HA and Ypr174c(pxl)-6HA following the release from metaphase arrest induced by Cdc20 depletion.

See also Figure S5.

To further investigate the Ypr174c-Cdc5 interaction, we performed pull-down experiments from Cdc5-6HA-expressing strains using immobilized GST-Ypr174c. We found that Ypr174c directly binds Cdc5 via phosphorylated S170, as the interaction was enhanced by prior phosphorylation of Ypr174c by Clb3-Cdk1 and the presence of site S170 (Figure 5F). These data suggest that Ypr174c mediates localization of Cdc5 to SPBs in metaphase.

We found an additional feature of SLiM-based regulation of Ypr174c by the discovery of the PxL motif, recognized by phosphatase Cdc14, which counteracts Cdk1 in anaphase (Kataria et al., 2018). Mutation of this motif in Ypr174c, 200PKL, greatly decreased the dephosphorylation rate by Cdc14 in vitro (Figure 5G). Next, we analyzed Ypr174c phosphorylation shifts after release from metaphase arrest induced by Cdc20 depletion. The PxL motif was essential for Ypr174c dephosphorylation in early anaphase, as the wild-type Ypr174c showed a rapid decrease in phosphorylation shift upon anaphase entry, but the phosphorylation of Ypr174c(pxl) remained high until Clb2 was degraded (Figure 5H). This is reminiscent of a previous finding (Hirschi et al., 2010) that the phosphorylation turnover can be accelerated if the substrate contains docking motifs for both the kinase and the phosphatase.

Specificity of Cyclin hp-Docking Motifs

There is now evidence of specific SLiMs for substrate docking by G1-, S-, G2-, and M cyclins (Bhaduri and Pryciak, 2011, Loog and Morgan, 2005, Örd et al., 2019a, Wilmes et al., 2004). Strikingly, the three B-type cyclins driving the cell cycle from S phase to the exit from mitosis (Clb5, Clb3, and Clb2) all use the conserved hp to accommodate different SLiMs. While some motifs (PxF, LxF) are unique for two paralogous cyclins (Örd et al., 2019a), the RxL motif is shared by S- and G2 cyclins, although showing lower specificity for Clb3 compared to Clb5. Motifs are partially overlapping; for example, some RxL and LxF motifs differ only in one position (Figure 6A).

Figure 6.

Figure 6

Open in a new tab

Hp Specificity of B-type Cyclins

(A) Sequence alignment of Orc6 RxL, Cdc6 LxF, and Ypr174c PxF motifs. The numbers above the alignment show the position relative to the L in RxL.

(B–G) In vitro phosphorylation experiments of Sic1(1–33 SPxA) combined with different docking motifs by Clb5-, Clb3-, and Clb2-Cdk1. The reactions were performed at initial velocity conditions. The relative specificity below each lane shows the phosphorylation rate of the substrate relative to phosphorylation of histone H1 by the respective kinase. SDS-PAGE 32P autoradiographs are shown.

(B) The importance of the residue in position −2 from the L in RxL or LxF motifs.

(C) Contribution of the N-terminal flanking residues from RILFP motif on phosphorylation rate.

(D) Comparison of the effect of +2F and +2L in PKKLQF and PEKLQF motifs.

(E) Analysis of cyclin specificity of different RxL variants.

(F) Mapping of Clb5-Cdk1 specificity determinants in LKGKNLLVEL docking motif by single amino acid mutations.

(G) Cyclin specificity of various docking motifs.

See also Figure S6.

We made a computational 3D model of Clb3 hp-PxF (PPKGPNF) interaction based on the alignment in Figure 6A and a crystal structure of cyclin A in complex with RxL peptide (Cheng et al., 2006), using FlexPepDock for refinement (Raveh et al., 2010; Figure S6A). The model suggests that PxF docks to the hp similarly to the LxF part of RxLxF, but the docking peptide is pulled closer to the hp because the P side chain is shorter than that of L. This could lead to a clash of the side chain of the residue in −1 position from PxF with cyclin, suggesting there could also be selectivity in −1 position, as the residue in the Ypr174c peptide is G. Also, the salt bridge mediated by R of the RxLxF is lost in the Clb3-PxF computational model, which is consistent with experimental evidence showing no effect of alanine mutation of the −2K in PKGPNF (Figure 2C). The loss of this interaction could be partially compensated by the −3P, which, as the model suggests, could interact with Y270 and E273 of Clb3, and promote phosphorylation by Clb3-Cdk1 (Figure 2C).

To understand the key determinants of hp-docking specificity, we performed a systematic mutational analysis of the motifs using a Sic1-based model substrate containing one minimal consensus phosphorylation site (SPxA) and a linker region (Figure 2D). We first studied the importance of the residue in position −2 from the L in RxL or LxF motifs (Figure 6A). Inserting an LxF motif from Cdc6 (PEKLQF) into this substrate enhanced the phosphorylation rate by Clb2-Cdk1 but showed a minor or no effect with Clb3- or Clb5-Cdk1, respectively (Figure 6B). Substitution of the −2E in motif PEKLQF with either K or R creates a potential RxL motif (KKLQF or RKLQF), and indeed, these motifs strongly promoted phosphorylation by Clb5- and Clb3-Cdk1 (Figures 6B and S6B). This highlights the importance of the positively charged residue at the −2 position for Clb5 and Clb3. Although substitution of −2E with K or R decreased the phosphorylation rate by Clb2-Cdk1, these motifs still promoted phosphorylation compared to the construct with no docking motif. We also observed differences in specificity with respect to K or R, with a strong preference for R in the case of Clb5. In fact, even the few-fold difference achieved by K/R swapping could enable a temporal fine-tuning of early phosphorylation events. Furthermore, using an RxL motif from Sic1, VNRILFP, we found that negatively charged residues upstream of RxL may have a strong unfavorable effect on specificity with Clb5 and Clb3 (Figures 6C and S6C).

While the Clb2- and Clb3-specific motifs LxF and PxF contain F, different RxL motifs have been shown to have either F, L, M, or P in position +1 or +2 from the L (Lowe et al., 2002). To test the importance of F in +2, we replaced the +2F with L in motifs PKKLQF and PEKLQF. Interestingly, the motifs containing +2L (PKKLQL and PEKLQL) did not promote phosphorylation by Clb3- and Clb2-Cdk1, while PKKLQL slightly improved phosphorylation by Clb5-Cdk1 (Figures 6D and S6D). This indicates that F is optimal in position +2, and that functional motifs lacking the +2F might contain additional interactions that compensate for its absence. A Clb5-specific substrate Fin1 contains an RxL without the F (LKGKNLLV; Loog and Morgan, 2005). To test if the flanking residues contribute to the docking, we replaced the three N-terminal flanking residues (LKG) from Fin1 motif into PKKLQL. This led to even more increased specificity for Clb5- compared to Clb3- and Clb2-Cdk1 (Figure 6D).

Furthermore, moving F into position +1 instead of +2, did not change the specificity for Clb5-Cdk1, but reduced the docking potency for Clb3- and Clb2-Cdk1 (Figures 6E and S6E). We also analyzed RxL motifs from Clb5-specific targets Fin1 and Spc110 (Loog and Morgan, 2005). The motifs from Fin1 (LKGKNLLVEL) and Spc110 (SRKRNLIDDL) strongly promoted phosphorylation by Clb5-Cdk1, while having no effect for Clb3- and Clb2-Cdk1 (Figure 6E). This shows that while some RxL motifs are used by B-type cyclins in general (R/KxLxF), some can be used only by a specific cyclin. To understand the Clb5 specificity determinants, we analyzed the Fin1 motif by mutating the flanking residues. We found that −4K, −1N, and +4L all contributed to the docking (Figures 6F and S6F). This shows the complexity of RxL motifs and highlights the importance of the residues flanking the RxL core in determining specificity. Further, R/KxL sequences are abundant, as they can be found in disordered regions of over 65% of yeast proteins (3,969 of 6,049), whereas only 8% has been estimated to be Cdk1 substrates (Ubersax et al., 2003). This indicates that most R/KxL sequences are not cyclin docking sites and that specificity must arise from flanking residues. Therefore, S. cerevisiae B-type cyclins have evolved highly orthogonal substrate docking specificity (Clb5-specific RxL variants, Clb3 PxF and Clb2 LxF), and retained common specificity of some RxL motifs (R/KxLxF; Figure 6G).

We performed a peptide competition assay to confirm the specificity of PxF docking. The PxF peptide (PPKGPNFYAK from Ypr174c) inhibited Clb3-Cdk1-driven phosphorylation of hp-dependent substrates by an order of magnitude more efficiently compared to the inhibition of Clb5- and Clb2-Cdk1 with RxL- and LxF-containing substrates, respectively (Figure S6G). The effect of the PxF peptide on phosphorylation of RxL- and LxF-dependent substrates by Clb5- and Clb2-Cdk1 was comparable to the effect on docking-independent target, histone H1, indicating that the observed mild effect was not related to hp docking (Figures S6G and S6H).

Human Cyclins E, A, and B Also Exhibit Diverse Substrate-Docking Specificity

As yeast cyclins contain different and extended SLiM specificity determinants (Figures 6B–6F), we asked if similar differential specificity could also be the case with mammalian CDKs. CDK complexes with cyclins E and A, and to a lesser extent cyclin B, all target substrates using RxL motif, whereas cyclin D uses an LxCxE motif and a helical motif in pRB (Brown et al., 2007, Chen et al., 1996, Dowdy et al., 1993, Schulman et al., 1998, Topacio et al., 2019, Wohlschlegel et al., 2001). Therefore, while cell cycle entry is controlled by cyclin D-Cdk4/6 with a unique docking interaction, most cell cycle events are governed by CDK complexes that use the RxL interaction. However, mutational analysis of RxL motif has revealed differences in specificity of RxL variants for cyclin E- and cyclin A-Cdk2 (Wohlschlegel et al., 2001). We performed in vitro phosphorylation experiments with human cyclin E-Cdk2, cyclin A-Cdk2, and cyclin B-Cdk1 using similar minimal model substrates as above (Figure 2D). First, to assess the importance of the positive charge at −2 position from L in RxLxF, we used a motif with −2E from Cdc6 (PEKLQF). This motif promoted phosphorylation only slightly by cyclin B-Cdk1 and not by cyclin E- or cyclin A-Cdk2 (Figures 7A and S7A). When the −2E is replaced with either K or R, resulting in an RxL motif, it leads to an increase in phosphorylation rate by all tested complexes (Figure 7A). Importantly, cyclins E and A showed considerable preference for −2R in these motifs.

Figure 7.

Figure 7

Open in a new tab

Hp-Docking Specificity of Human Cyclin-CDK Complexes

(A–C) Different docking motifs were analyzed in a kinase assay with cyclin E-Cdk2, cyclin A-Cdk2, or cyclin B-Cdk1 using Sic1(1–33)-based model substrate with a minimal consensus phosphorylation site combined with indicated docking motifs. The reactions were performed at initial velocity conditions. The 32P autoradiographs showing substrate phosphorylation are shown. Relative specificity shows the phosphorylation rate of the substrate relative to the rate of histone H1 phosphorylation by the respective CDK complex.

(A) The importance of positive charge in -2 position in RxLxF motif on docking specificity.

(B) The effect of N-terminal flanking residues from RILFP motif on phosphorylation potentiation.

(C) Cyclin specificity analysis of different cyclin docking motifs.

See also Figure S7.

Next, to analyze the flanking positions preceding the RxL, we used the motif from Sic1 (VNRILFP) as the starting point. Addition of the VNRILFP motif to the substrate strongly promoted phosphorylation by all tested complexes (Figures 7B and S7B). When the sequence SVN was replaced with KKK, it led to a decrease in phosphorylation by cyclin-E and cyclin-B complexes, but an increase by cyclin A-Cdk2 (Figure 7B). Substitution of these residues with EEE sequence abolished docking by cyclins E and A; however, cyclin B-Cdk1 did not have a strong preference for either KKK or EEE (Figure 7B). These results show that the flanking residues N-terminal of RxL can contain differential specificity determinants for human cyclins.

As with yeast cyclins, F was found to be optimal at the L+2 position for human cyclins, as PKKLQF showed stronger docking effect than PKKLQL (Figures 7C and S7C). Finally, we compared the effect of the F in either L+1 or L+2 position (Figure 7C). Interestingly, while both +1 and +2F motifs showed similar docking potency with cyclin A and B complexes, cyclin E-Cdk2 preferred +1F (Figure 7C). Therefore, while cyclins A and B dock efficiently to a wider variety of RxL motifs, cyclin E seems to target a more limited set of RxL motifs, with the preference for R over K in L-2 (Figures 7A and 7C), and F in L+1 positions (Figures 7A–7C). Negative charged N-terminal of RxL inactivate docking for cyclins A and E, but not cyclin B (Figure 7B), providing clues about differential docking specificity between cyclins A and B. The highly Clb5-specific RxLs from Fin1 (LKGKNLLVEL) and Spc110 (SRKRNLIDDL) do not show a consistent specificity pattern with human cyclins, as LKGKNLLVEL promotes cyclin B dependent phosphorylation and SRKRNLIDDL is more efficiently used by cyclin A-Cdk2 (Figure 7C). Finally, the Clb3-specific PxF motif (PPKGPNFYAK) did not show strong specificity for the mammalian complexes (Figure 7C), which further emphasizes the flexibility of the hp in achieving high relative specificity between different SLiMs. Taken together, the selected substrate constructs indicate that variations at several positions extend the core element of SLiMs and therefore can be used to assign cell-cycle timing of CDK phosphorylation switches using cyclin-specific hp docking.

Discussion

By discovering a specific SLiM for substrate docking by Clb3, we have added the final missing piece to the specificity matrix of substrate targeting by major cell-cycle cyclins in S. cerevisiae: LP motif for G1-cyclin Cln2 (Bhaduri and Pryciak, 2011, Kõivomägi et al., 2011), RxL motif for S-cyclin Clb5 (Loog and Morgan, 2005, Wilmes et al., 2004), PxF motif for G2-cyclin Clb3 (this study), and LxF motif for M-cyclin Clb2 (Örd et al., 2019a). Our work also demonstrates that the hp, which was originally thought to bind only a single SLiM, the “RxL or Cy motif,” can accommodate a wide variety of motifs. This leaves us with the important conclusion that a SLiM and its binding pocket are not a fixed pair. Instead, by only minimal modifications in both SLiMs and the pockets, orthogonal specificity can be achieved. At the same time, diverged hp pockets can retain common specificity for certain SLiMs (Figure 6G). The common specificity could mediate the universal functions of Clb cyclins, such as triggering DNA replication (Donaldson et al., 1998), but the cyclin-specific motifs could allow greater temporal control of phosphorylation events, as there are differences in functions of even closely related cyclins. For example, Clb3 promotes timely spindle assembly, but is incapable of substituting Clb2 in triggering anaphase or inhibiting mitotic exit (Pecani and Cross, 2016, Richardson et al., 1992).

The PxF motif was found in an uncharacterized ORF Ypr174c, and after the search for similar motifs also in Tgl5, Sen1, and Boi1. As Clb3 exhibits a unique localization at the nuclear periphery, and it has the lowest nuclear abundance among the major B-type cyclins (Clb5, Clb3, and Clb2; Örd et al., 2019b), precise targeting of selected proteins could be the main function of the PxF motif, while the more abundant Clb5 and Clb2 could constitute the major accumulating CDK activity affecting the phosphorylation of the bulk of the targets (which is predicted to exceed 500 targets in yeast; Ubersax et al., 2003).

We found that the PxF motif controls Ypr174c phosphorylation in vivo and that phosphorylated Ypr174c recruits Cdc5 to SPBs in metaphase. Thus, we propose to name Ypr174c as Csa1 for “Cdc5 SPB anchor.” When also considering the finding that clb3(hpm) mutation was associated with a delay from G1/S to spindle formation, and also, a delay in the expression of Clb2, it can be suggested that the sharp Clb3 specificity profile could have a key role in triggering the processes that should not occur in early S phase, where the preceding Clb5 plays the major role. For example, spindle formation should be allocated to the time slot after SPB duplication, which takes place around the onset of G1 and S cyclins (Jaspersen et al., 2004).

Finally, we performed a study to determine if the known hp-docking SLiMs show diverse specificity when the core motif or the flanking residues are changed. This revealed that the specificity depends on both the core and flanking regions, suggesting that cyclins as docking hubs can provide a variety of specificities to temporally fine-tune the cell cycle. Such diverse cyclin specificity enables us to create different switching orders over the cell cycle. Furthermore, SLiMs specific for the remaining representatives of the nine cell-cycle cyclins, besides the three major ones studied here, are yet to be identified. For example, the observed functional divergence of cyclins Clb3 and Clb4 suggests that yet another level of cyclin-specific substrate targeting governs functions of G2 cyclins. We also studied the specificities of the three major human cyclin-CDK complexes, which predicted similar potential for fine-tuning the substrate-targeting specificity by varying the SLiMs. For example, positions immediately N-terminal to the RxL motif were very sensitive to both negatively and positively charged residues for cyclin B and E, but only to negatively charged residues for cyclin A. Also, we found that cyclin E has more limited RxL specificity, with the preference for R over K at the L-2 position and for F at the L+1 position. This finding confirms a prediction based on a large-scale screen showing that fewer proteins bind specifically to cyclin E (73) compared to cyclin A (227), which led to a hypothesis that cyclin E may have a more restricted substrate specificity (Pagliuca et al., 2011).

In conclusion, the ability of cyclins to function as signaling hubs of the cell cycle provides a system with diverse options for assembling signaling networks using SLiMs. Furthermore, the study reveals that conserved binding pockets can accommodate diverse classes of SLiMs. This finding further supports the prediction that the majority of biological signal interactions remain to be discovered and that the specificity of signal processing is ubiquitously mediated via SLiMs and their globular binding partner interactions (Iakoucheva et al., 2002). Combining the extended design space of possible cyclin docking sites with other linear parameters of the CDK multisite phosphorylation code (Örd et al., 2019b) opens up new ways to understand the temporal coordination of cell cycle processes, and potentially, also reveals new ways to use the same encoding rules of SLiMs in synthetic biology applications.

STAR★Methods

Key Resources Table

REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
anti-HA.11 (clone 16B12) Biolegend Cat. No. 901501; RRID: AB_2565006
anti-Clb2 Santa Cruz Biotechnology sc-9071; RRID: AB_667962
Goat anti-rabbit IgG, HRP-conjugated Labas AS N/A
Goat anti-mouse IgG, HRP-conjugated Labas AS N/A
Bacterial and Virus Strains
E. coli BL21-CodonPlus(DE3) -RP Agilent Technologies Catalog Code: 230255
Chemicals, Peptides, and Recombinant Proteins
HA peptide (CYPYDVPDYAGYP
YDVPDYAG)		 ProImmune N/A
PxF peptide (PPKGPNFYAK) Storkbio N/A
1-NM-PP1 Carbosynth FA10330
Phos-tag Acrylamide AAL-107 Wako Chemicals 304-93521
ATP, [γ-32P]-10mCi/ml Hartmann Analytical SRP-501
SuperSignal West Pico PLUS Chemiluminescent Substrate ThermoFisher Scientific 34577
Bio-Rad Protein Assay Dye Reagent Concentrate Bio-rad #5000006
cyclin E-Cdk2 Millipore 14-475
cyclin A-Cdk2 Millipore 14-448
cyclin B-Cdk1 Millipore 14-450
bovine histone H1 Sigma-Aldrich 14-155
Deposited Data
Unprocessed autoradiographs, western blot images, and confocal microscopy images of this study This study Mendeley Data: https://doi.org/10.17632/yxzzy26hbd.1
Experimental Models: Organisms/Strains
S. cerevisiae strains used in this study are listed in Table S1 N/A
Recombinant DNA
Plasmids used in this study are listed in Table S2 N/A
Software and Algorithms
MATLAB scripts for cell tracking and quantification of fluorescence signals Doncic et al., 2013 N/A
SLiMSearch4 Krystkowiak and Davey, 2017 http://slim.ucd.ie/slimsearch/
PSIPRED v3.3 Buchan et al., 2013 http://bioinf.cs.ucl.ac.uk/psipred/
UCSF Chimera Pettersen et al., 2004 https://www.cgl.ucsf.edu/chimera/
I-TASSER Yang et al., 2015 https://zhanglab.ccmb.med.umich.edu/I-TASSER/
MM-align Mukherjee and Zhang, 2009 https://zhanglab.ccmb.med.umich.edu/MM-align/
FlexPepDock Raveh et al., 2010 http://flexpepdock.furmanlab.cs.huji.ac.il/
Villalta et al.’s colocalization algorithm Villalta et al., 2011 https://se.mathworks.com/matlabcentral/fileexchange/30665-villalta-et-al-s-colocalization-algorithm
Open in a new tab

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Mart Loog (mart.loog@ut.ee).

Materials Availability

Yeast strains and plasmids used in this study are described in Tables S1 and S2 and are made available upon request from the Lead Contact, Mart Loog (mart.loog@ut.ee).

Data and Code Availability

Original data have been deposited to Mendeley Data: https://doi.org/10.17632/yxzzy26hbd.1.

For quantification of fluorescent proteins from time-lapse experiments, MATLAB scripts from Doncic et al. (2013) were adapted and used, and are available upon request.

Experimental Model and Subject Details

Yeast strains

S. cerevisiae strains were haploid MATa derivates of W303. All yeast strains used in the study are listed in Table S1 and are available upon request. Yeast cells were grown in YPD or Synthetic Complete media at 30°C. Gene deletions, promoter substitutions and tagging were performed using methods based on PCR and homologous recombination (Janke et al., 2004, Longtine et al., 1998). All gene modifications were confirmed by DNA sequencing. The Cdk1 activity sensor based on Mcm2-Mcm3 nuclear localization (Liku et al., 2005) and export signals was fused to EGFP and expressed under ADH1 promoter from a pRS304-based vector integrated to ADH1 promoter locus.

Method Details

Protein purification

The plasmids used for recombinant protein expression are listed in Table S2. Ypr174c, Tgl5(554-750), Sen1(920-1065) and Spc29 were fused N-terminally with GST, while Boi1, Ndd1 and Plm2 were tagged N-terminally with 6xHis and were expressed in BL21RP cells at 16°C using 0.3 mM IPTG. Sic1(1-33) constructs contained mutations T2A, T5S and R8A. Sic1(1-33) were fused with a linker sequence ELQGGGGG and Protein G B1 domain with C-terminal 6xHis. Sic1 constructs were expressed at 37°C. His-tagged proteins were purified using standard Chelating Sepharose (GE Healthcare) and were eluted using imidazole. For purification of GST-tagged proteins, Glutathione Sepharose (GE Healthcare) was used. 6xHis-Cdc14 was purified as described previously (Bremmer et al., 2012).

Yeast cyclin-Cdk1 complexes were purified from S. cerevisiae cultures where tagged cyclin was overexpressed from GAL1 promoter. Clb5, Clb3 and Clb2 were TAP-tagged and purified as described previously (Puig et al., 2001, Ubersax et al., 2003). Cln2 with N-terminal 3HA tag was purified as described in McCusker et al. (2007). All yeast lysates were prepared using Mixer Mill MM 400 (Retch). Cks1 was expressed in E. coli BL21 and purified as described in Reynard et al. (2000).

Kinase assays

The assay mixture consisted of 50 mM HEPES-KOH, pH 7.4, 150 mM NaCl, 5 mM MgCl2, 20 mM imidazole, 2% glycerol, 0.2 mg/ml BSA, 500 nM Cks1 and 500 μM ATP [(with added [γ-32P]-ATP (Hartmann Analytic)].

Substrate protein concentrations were in the range of 0.5-3 μM (in the linear [S] versus v0 range, several-fold below the estimated KM value). Bovine histone H1 (Sigma-Aldrich) was used in 80 μg/ml concentration. Cyclin-Cdk complexes were 0.5-2 nM in the assays. In assays with human cyclin-Cdk complexes, cyclin E-Cdk2 (Millipore, 14-475), cyclin A-Cdk2 (Millipore, 14-448) and cyclin B-Cdk1 (Millipore 14-450) were used. For Ypr174c dephosphorylation assay, 1 μM Ypr174c was phosphorylated with 1 nM Clb3 for 30 minutes prior to addition of Cdc14. PPKGPNFYAK competitor peptide was dissolved in 50 mM HEPES-KOH, pH 7.4.

The reactions were stopped using SDS-PAGE sample buffer at time points below 10% of initial substrate turnover. The proteins were separated using SDS-PAGE. For separation of differentially phosphorylated forms of Ypr174c, Phos-tag SDS-PAGE (8% acrylamide, 50 μM Phos-tag (Wako Chemicals), 100 μM MnCl2) was used. Incorporated 32P signals were detected using Amersham Typhoon 5 Biomolecular Imager (GE Healthcare Life Sciences) and were quantified using ImageQuant TL (Amersham Biosciences). All kinase assays were performed in at least two replicates.

Western blotting

For analysis of asynchronous cultures, Ypr174c-6HA expressing cells were grown in 5 mL YPD at 30°C to OD 0.8, collected by centrifugation and flash-frozen. For synchronized metaphase release, PGALS-CDC20 cells were grown in YP Raf+Gal to OD 0.3, then cells were collected and resuspended in YP Raf. Cells were grown for 3 hours, resulting in over 90% of cells being in metaphase, then released to anaphase by addition of 2% galactose. For lysis, cells were resuspended in urea lysis buffer and disrupted by bead beating. The lysates were separated using SDS-PAGE and proteins were transferred to nitrocellulose membranes using Pierce G2 Fast Blotter (Thermo Scientific). anti-HA.11 epitope tag antibody (1:500) (clone 16B12, BioLegend Cat. No. 901501) and HRP-conjugated anti-mouse IgG antibody (1:7,500) from Labas, Estonia were used to detect Ypr174c-6HA. For detection of Clb2, rabbit polyclonal Clb2 antibody (1:500) (y-180, Santa Cruz Biotechnology) and HRP-conjugated anti-rabbit antibody (1:7500) from Labas, Estonia were used. All western blottings were performed in at least two biological replicates.

Co-immunoprecipitation

Yeast cells expressing Cdc5-6HA were grown in YPD to OD 0.8, cells were collected and snap frozen. GST-tagged Ypr174c was expressed in BL21RP cells at 16°C using 0.3 mM IPTG for induction. GST-Ypr174c was bound to Glutathione Sepharose (GE Healthcare) and washed using lysis buffer (50 mM HEPES, pH 7, 100 mM NaCl, 1 mM MgCl2, 0.1% NP40). The Sepharose with GST-Ypr174c was resuspended in kinase buffer (50 mM HEPES, pH 7.4, 150 mM NaCl, 5 mM MgCl2, 0.5 mM ATP) and phosphorylated with Clb3-Cdk1 to saturation. GST-Ypr174c was then washed with IP buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP40, 1 mM DTT). The yeast cells were resuspended in IP buffer and disrupted by bead beating. The cleared lysate was mixed with GST-Ypr174c and the mixtures were incubated for 1 hour at 4°C. Next, the beads were washed thoroughly using IP buffer and the proteins were eluted using SDS-PAGE sample buffer.

Time-lapse fluorescence microscopy

Prior to experiment, cells were grown at 30 °C in synthetic complete media with 2% glucose (SC) to OD 0.2–0.6. Cells were pipetted onto 0.8-mm cover glass and covered with a 1-mm thick 1.5% SC/glucose-agarose pad (NuSieve™ GTG™ Agarose, Lonza). Cells were incubated under the agarose pad for 60 min before starting the experiment. Imaging was carried out using a Zeiss Observer Z1 microscope with a 63 × /1.4NA oil immersion objective and Axiocam 506 mono camera (Zeiss), using 3 × 3 binning. The temperature of the agarose pad was held at 30 °C using a Tempcontrol 37–2 digital (PeCon). Images were taken every 3 min. The experiments were 8 h long and up to 14 positions were imaged in one experiment using an automated stage and ZEN software (Zeiss). Definite Focus was used to keep the cells in focus during the experiment. The excitation of EGFP-tagged proteins was done using a Colibri 470 LED module with exposure times of 15 ms. Imaging of Cdc5-yeVenus and cyclins fused with yeCitrine was performed using a Colibri 505 LED module for 500 ms. Whi5-mCherry and NLS-NES-mCherry were imaged using a Colibri 540–580 LED module for 750 ms and 400 ms, respecively. All Colibri modules were used at 25% power. Image segmentation, cell tracking, and quantification of fluorescence signals was performed using MATLAB (The MathWorks, Inc.) as described in Doncic et al. (2013). All plots with microscopy data contain data from at least two experiments, the exact number of cells analyzed from each strain is presented in Table S1.

Confocal microscopy

The imaging of Ypr174c-EGFP/mCherry, EGFP-NES-Clb3 and mCherry-Tgl5 presented in Figure S3 was performed using LSM 710 confocal laser scanning microscope (Zeiss) with AlphaPlnAPO 63x/1.46 objective (Zeiss) and Orca-R2 C10600-10B camera (Hamamatsu). RMC 7812 Z2 module was used for exitation of EGFP-tagged proteins and DPSS 561-10 module was used for mCherry-tagged proteins. Prior to induction of Clb3 and Tgl5 from GAL1 promoter, cells were grown in SC supplemented with 2% raffinose, then, 2% galactose was added. Cells were covered with 1.5% SC/galactose-agarose as described above. Images were taken 2 hours after galactose induction. For colocalization quantification, images were first subjected to mean filtering and background subtraction. Pearson’s correlation coefficient was then calculated for every image using an automated algorithm (Villalta et al., 2011).

Bioinformatics

Secondary structure prediction of Ypr174c was performed using PSIPred 4.0 (Buchan et al., 2013). The search for PxF motifs in the disordered regions of S. cerevisiae proteome was carried out using SlimSearch4 (Krystkowiak and Davey, 2017). The motif [PILVM]PxxPxFxx[KR] was searched from protein regions with IUPRED disorder score over 0.3.

Clb3-PxF interaction model was created on the basis of the crystal structure of cyclin A-RxL (2CCI (Cheng et al., 2006)). First, the predicted structural model of Clb3 was obtained using I-TASSER (Yang et al., 2015). Clb3-RxL structure was then received by aligning cyclin A-RxL and Clb3 structures with MM-align (Mukherjee and Zhang, 2009). Then, based on the alignment in Figure 6A, the RxL peptide was swapped with Ypr174c PxF motif in UCSF Chimera (Pettersen et al., 2004). The interaction was refined locally using FlexPepDock (Raveh et al., 2010).

Quantification and Statistical Analysis

The data from time-lapse microscopy experiments is from at least two independent replicate experiments. All replicate experiments are included in the data. The statistical details of the experiments can be found in the figure legends, the exact number of cells used in microscopy data is presented in Table S1. Mann-Whitney U test was used to test the difference of medians. For comparisons of mean values, unpaired two-tailed t test was used.

Acknowledgments

We thank D. Morgan and P. Pryciak for valuable comments on the manuscript and S. Rubin for assistance with modeling the docking interaction. We are grateful to J. Mihhejev for excellent technical assistance. The work was funded by European Research Council grant 649124, Archimedes foundation grant "Centre of Excellence for Molecular Cell Technologies" TK143, and the Estonian Science Agency grant PRG550 to M.L.

Author Contributions

M.Ö. and M.L. directed the study; M.Ö., K.K.P., R.K., K.M., T.O., I.B., I.F., E.V., and M.K. cloned the constructs, made the yeast strains, and purified the proteins; M.Ö., K.K.P., R.K., K.M., T.O., I.B., and M.K. performed the experiments; and M.Ö. and M.L. wrote the manuscript.

Declaration of Interests

The authors declare no competing interests.

Published: June 16, 2020

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.celrep.2020.107757.

Supplemental Information

Document S1. Figures S1–S7 and Tables S1 and S2
mmc1.pdf (7.8MB, pdf)
Document S2. Article plus Supplemental Information
mmc2.pdf (11.8MB, pdf)

References

  1. Athenstaedt K., Daum G. Tgl4p and Tgl5p, two triacylglycerol lipases of the yeast Saccharomyces cerevisiae are localized to lipid particles. J. Biol. Chem. 2005;280:37301–37309. doi: 10.1074/jbc.M507261200. [DOI] [PubMed] [Google Scholar]
  2. Bhaduri S., Pryciak P.M. Cyclin-specific docking motifs promote phosphorylation of yeast signaling proteins by G1/S Cdk complexes. Curr. Biol. 2011;21:1615–1623. doi: 10.1016/j.cub.2011.08.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bhaduri S., Valk E., Winters M.J., Gruessner B., Loog M., Pryciak P.M. A docking interface in the cyclin Cln2 promotes multi-site phosphorylation of substrates and timely cell-cycle entry. Curr. Biol. 2015;25:316–325. doi: 10.1016/j.cub.2014.11.069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bloom J., Cross F.R. Multiple levels of cyclin specificity in cell-cycle control. Nat. Rev. Mol. Cell Biol. 2007;8:149–160. doi: 10.1038/nrm2105. [DOI] [PubMed] [Google Scholar]
  5. Botchkarev V.V., Jr., Rossio V., Yoshida S. The budding yeast Polo-like kinase Cdc5 is released from the nucleus during anaphase for timely mitotic exit. Cell Cycle. 2014;13:3260–3270. doi: 10.4161/15384101.2014.953882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Botchkarev V.V., Jr., Garabedian M.V., Lemos B., Paulissen E., Haber J.E. The budding yeast Polo-like kinase localizes to distinct populations at centrosomes during mitosis. Mol. Biol. Cell. 2017;28:1011–1020. doi: 10.1091/mbc.E16-05-0324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brautigan D.L., Shenolikar S. Protein serine/threonine phosphatases: keys to unlocking regulators and substrates. Annu. Rev. Biochem. 2018;87:921–964. doi: 10.1146/annurev-biochem-062917-012332. [DOI] [PubMed] [Google Scholar]
  8. Bremmer S.C., Hall H., Martinez J.S., Eissler C.L., Hinrichsen T.H., Rossie S., Parker L.L., Hall M.C., Charbonneau H. Cdc14 phosphatases preferentially dephosphorylate a subset of cyclin-dependent kinase (Cdk) sites containing phosphoserine. J. Biol. Chem. 2012;287:1662–1669. doi: 10.1074/jbc.M111.281105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brown N.R., Noble M.E.M., Endicott J.A., Johnson L.N. The structural basis for specificity of substrate and recruitment peptides for cyclin-dependent kinases. Nat. Cell Biol. 1999;1:438–443. doi: 10.1038/15674. [DOI] [PubMed] [Google Scholar]
  10. Brown N.R., Lowe E.D., Petri E., Skamnaki V., Antrobus R., Johnson L.N. Cell cycle cyclin B and cyclin A confer different substrate recognition properties on CDK2. Cell Cycle. 2007;6:1350–1359. doi: 10.4161/cc.6.11.4278. [DOI] [PubMed] [Google Scholar]
  11. Buchan D.W.A., Minneci F., Nugent T.C.O., Bryson K., Jones D.T. Scalable web services for the PSIPRED Protein Analysis Workbench. Nucleic Acids Res. 2013;41:W349–W357. doi: 10.1093/nar/gkt381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chen J., Saha P., Kornbluth S., Dynlacht B.D., Dutta A. Cyclin-binding motifs are essential for the function of p21CIP1. Mol. Cell. Biol. 1996;16:4673–4682. doi: 10.1128/mcb.16.9.4673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cheng K.-Y., Noble M.E.M., Skamnaki V., Brown N.R., Lowe E.D., Kontogiannis L., Shen K., Cole P.A., Siligardi G., Johnson L.N. The role of the phospho-CDK2/cyclin A recruitment site in substrate recognition. J. Biol. Chem. 2006;281:23167–23179. doi: 10.1074/jbc.M600480200. [DOI] [PubMed] [Google Scholar]
  14. Davey N.E., Van Roey K., Weatheritt R.J., Toedt G., Uyar B., Altenberg B., Budd A., Diella F., Dinkel H., Gibson T.J. Attributes of short linear motifs. Mol. Biosyst. 2012;8:268–281. doi: 10.1039/c1mb05231d. [DOI] [PubMed] [Google Scholar]
  15. Donaldson A.D., Raghuraman M.K., Friedman K.L., Cross F.R., Brewer B.J., Fangman W.L. CLB5-dependent activation of late replication origins in S. cerevisiae. Mol. Cell. 1998;2:173–182. doi: 10.1016/s1097-2765(00)80127-6. [DOI] [PubMed] [Google Scholar]
  16. Doncic A., Eser U., Atay O., Skotheim J.M. An algorithm to automate yeast segmentation and tracking. PLoS ONE. 2013;8:e57970. doi: 10.1371/journal.pone.0057970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Dowdy S.F., Hinds P.W., Louie K., Reed S.I., Arnold A., Weinberg R.A. Physical interaction of the retinoblastoma protein with human D cyclins. Cell. 1993;73:499–511. doi: 10.1016/0092-8674(93)90137-f. [DOI] [PubMed] [Google Scholar]
  18. Dunn K.W., Kamocka M.M., McDonald J.H. A practical guide to evaluating colocalization in biological microscopy. Am. J. Physiol. Cell Physiol. 2011;300:C723–C742. doi: 10.1152/ajpcell.00462.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ear P.H., Booth M.J., Abd-Rabbo D., Kowarzyk Moreno J., Hall C., Chen D., Vogel J., Michnick S.W. Dissection of Cdk1-cyclin complexes in vivo. Proc. Natl. Acad. Sci. USA. 2013;110:15716–15721. doi: 10.1073/pnas.1305420110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Elia A.E.H., Rellos P., Haire L.F., Chao J.W., Ivins F.J., Hoepker K., Mohammad D., Cantley L.C., Smerdon S.J., Yaffe M.B. The molecular basis for phosphodependent substrate targeting and regulation of Plks by the Polo-box domain. Cell. 2003;115:83–95. doi: 10.1016/s0092-8674(03)00725-6. [DOI] [PubMed] [Google Scholar]
  21. Fitch I., Dahmann C., Surana U., Amon A., Nasmyth K., Goetsch L., Byers B., Futcher B. Characterization of four B-type cyclin genes of the budding yeast Saccharomyces cerevisiae. Mol. Biol. Cell. 1992;3:805–818. doi: 10.1091/mbc.3.7.805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hertz E.P.T., Kruse T., Davey N.E., López-Méndez B., Sigurðsson J.O., Montoya G., Olsen J.V., Nilsson J. A conserved motif provides binding specificity to the PP2A-B56 phosphatase. Mol. Cell. 2016;63:686–695. doi: 10.1016/j.molcel.2016.06.024. [DOI] [PubMed] [Google Scholar]
  23. Hirschi A., Cecchini M., Steinhardt R.C., Schamber M.R., Dick F.A., Rubin S.M. An overlapping kinase and phosphatase docking site regulates activity of the retinoblastoma protein. Nat. Struct. Mol. Biol. 2010;17:1051–1057. doi: 10.1038/nsmb.1868. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Iakoucheva L.M., Brown C.J., Lawson J.D., Obradović Z., Dunker A.K. Intrinsic disorder in cell-signaling and cancer-associated proteins. J. Mol. Biol. 2002;323:573–584. doi: 10.1016/s0022-2836(02)00969-5. [DOI] [PubMed] [Google Scholar]
  25. Janke C., Magiera M.M., Rathfelder N., Taxis C., Reber S., Maekawa H., Moreno-Borchart A., Doenges G., Schwob E., Schiebel E., Knop M. A versatile toolbox for PCR-based tagging of yeast genes: new fluorescent proteins, more markers and promoter substitution cassettes. Yeast. 2004;21:947–962. doi: 10.1002/yea.1142. [DOI] [PubMed] [Google Scholar]
  26. Jaspersen S.L., Huneycutt B.J., Giddings T.H., Jr., Resing K.A., Ahn N.G., Winey M. Cdc28/Cdk1 regulates spindle pole body duplication through phosphorylation of Spc42 and Mps1. Dev. Cell. 2004;7:263–274. doi: 10.1016/j.devcel.2004.07.006. [DOI] [PubMed] [Google Scholar]
  27. Kataria M., Mouilleron S., Seo M.-H., Corbi-Verge C., Kim P.M., Uhlmann F. A PxL motif promotes timely cell cycle substrate dephosphorylation by the Cdc14 phosphatase. Nat. Struct. Mol. Biol. 2018;25:1093–1102. doi: 10.1038/s41594-018-0152-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kõivomägi M., Valk E., Venta R., Iofik A., Lepiku M., Morgan D.O., Loog M. Dynamics of Cdk1 substrate specificity during the cell cycle. Mol. Cell. 2011;42:610–623. doi: 10.1016/j.molcel.2011.05.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Krystkowiak I., Davey N.E. SLiMSearch: a framework for proteome-wide discovery and annotation of functional modules in intrinsically disordered regions. Nucleic Acids Res. 2017;45(W1):W464–W469. doi: 10.1093/nar/gkx238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liku M.E., Nguyen V.Q., Rosales A.W., Irie K., Li J.J. CDK phosphorylation of a novel NLS-NES module distributed between two subunits of the Mcm2-7 complex prevents chromosomal rereplication. Mol. Biol. Cell. 2005;16:5026–5039. doi: 10.1091/mbc.E05-05-0412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Linke C., Chasapi A., González-Novo A., Al Sawad I., Tognetti S., Klipp E., Loog M., Krobitsch S., Posas F., Xenarios I., Barberis M. A Clb/Cdk1-mediated regulation of Fkh2 synchronizes CLB expression in the budding yeast cell cycle. NPJ Syst. Biol. Appl. 2017;3:7. doi: 10.1038/s41540-017-0008-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Longtine M.S., McKenzie A., 3rd, Demarini D.J., Shah N.G., Wach A., Brachat A., Philippsen P., Pringle J.R. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast. 1998;14:953–961. doi: 10.1002/(SICI)1097-0061(199807)14:10<953::AID-YEA293>3.0.CO;2-U. [DOI] [PubMed] [Google Scholar]
  33. Loog M., Morgan D.O. Cyclin specificity in the phosphorylation of cyclin-dependent kinase substrates. Nature. 2005;434:104–108. doi: 10.1038/nature03329. [DOI] [PubMed] [Google Scholar]
  34. Lowe E.D., Tews I., Cheng K.Y., Brown N.R., Gul S., Noble M.E.M., Gamblin S.J., Johnson L.N. Specificity determinants of recruitment peptides bound to phospho-CDK2/cyclin A. Biochemistry. 2002;41:15625–15634. doi: 10.1021/bi0268910. [DOI] [PubMed] [Google Scholar]
  35. McCusker D., Denison C., Anderson S., Egelhofer T.A., Yates J.R., 3rd, Gygi S.P., Kellogg D.R. Cdk1 coordinates cell-surface growth with the cell cycle. Nat. Cell Biol. 2007;9:506–515. doi: 10.1038/ncb1568. [DOI] [PubMed] [Google Scholar]
  36. Miller C.J., Turk B.E. Homing in: mechanisms of substrate targeting by protein kinases. Trends Biochem. Sci. 2018;43:380–394. doi: 10.1016/j.tibs.2018.02.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Morgan D.O. New Science Press; 2007. The Cell Cycle: Principles of Control. [Google Scholar]
  38. Mukherjee S., Zhang Y. MM-align: a quick algorithm for aligning multiple-chain protein complex structures using iterative dynamic programming. Nucleic Acids Res. 2009;37:e83. doi: 10.1093/nar/gkp318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Örd M., Venta R., Möll K., Valk E., Loog M. Cyclin-specific docking mechanisms reveal the complexity of M-CDK function in the cell cycle. Mol. Cell. 2019;75:76–89.e3. doi: 10.1016/j.molcel.2019.04.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Örd M., Möll K., Agerova A., Kivi R., Faustova I., Venta R., Valk E., Loog M. Multisite phosphorylation code of CDK. Nat. Struct. Mol. Biol. 2019;26:649–658. doi: 10.1038/s41594-019-0256-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pagliuca F.W., Collins M.O., Lichawska A., Zegerman P., Choudhary J.S., Pines J. Quantitative proteomics reveals the basis for the biochemical specificity of the cell-cycle machinery. Mol. Cell. 2011;43:406–417. doi: 10.1016/j.molcel.2011.05.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pearson K. VII. Mathematical contributions to the theory of evolution. III. Regression, heredity, and panmixia. Philos. Trans. R. Soc. London. Ser. A. 1896;187:253–318. [Google Scholar]
  43. Pecani K., Cross F.R. Degradation of the mitotic cyclin Clb3 is not required for mitotic exit but is necessary for G1 cyclin control of the succeeding cell cycle. Genetics. 2016;204:1479–1494. doi: 10.1534/genetics.116.194837. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Pettersen E.F., Goddard T.D., Huang C.C., Couch G.S., Greenblatt D.M., Meng E.C., Ferrin T.E. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 2004;25:1605–1612. doi: 10.1002/jcc.20084. [DOI] [PubMed] [Google Scholar]
  45. Puig O., Caspary F., Rigaut G., Rutz B., Bouveret E., Bragado-Nilsson E., Wilm M., Séraphin B. The tandem affinity purification (TAP) method: a general procedure of protein complex purification. Methods. 2001;24:218–229. doi: 10.1006/meth.2001.1183. [DOI] [PubMed] [Google Scholar]
  46. Raveh B., London N., Schueler-Furman O. Sub-angstrom modeling of complexes between flexible peptides and globular proteins. Proteins. 2010;78:2029–2040. doi: 10.1002/prot.22716. [DOI] [PubMed] [Google Scholar]
  47. Ravid T., Hochstrasser M. Diversity of degradation signals in the ubiquitin-proteasome system. Nat. Rev. Mol. Cell Biol. 2008;9:679–690. doi: 10.1038/nrm2468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Reynard G.J., Reynolds W., Verma R., Deshaies R.J. Cks1 is required for G(1) cyclin-cyclin-dependent kinase activity in budding yeast. Mol. Cell. Biol. 2000;20:5858–5864. doi: 10.1128/mcb.20.16.5858-5864.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Reynolds D., Shi B.J., McLean C., Katsis F., Kemp B., Dalton S. Recruitment of Thr 319-phosphorylated Ndd1p to the FHA domain of Fkh2p requires Clb kinase activity: a mechanism for CLB cluster gene activation. Genes Dev. 2003;17:1789–1802. doi: 10.1101/gad.1074103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Richardson H., Lew D.J., Henze M., Sugimoto K., Reed S.I. Cyclin-B homologs in Saccharomyces cerevisiae function in S phase and in G2. Genes Dev. 1992;6:2021–2034. doi: 10.1101/gad.6.11.2021. [DOI] [PubMed] [Google Scholar]
  51. Russo A.A., Jeffrey P.D., Patten A.K., Massagué J., Pavletich N.P. Crystal structure of the p27Kip1 cyclin-dependent-kinase inhibitor bound to the cyclin A-Cdk2 complex. Nature. 1996;382:325–331. doi: 10.1038/382325a0. [DOI] [PubMed] [Google Scholar]
  52. Schulman B.A., Lindstrom D.L., Harlow E.D. Substrate recruitment to cyclin-dependent kinase 2 by a multipurpose docking site on cyclin A. Proc. Natl. Acad. Sci. USA. 1998;95:10453–10458. doi: 10.1073/pnas.95.18.10453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Schwob E., Nasmyth K. CLB5 and CLB6, a new pair of B cyclins involved in DNA replication in Saccharomyces cerevisiae. Genes Dev. 1993;7:1160–1175. doi: 10.1101/gad.7.7a.1160. [DOI] [PubMed] [Google Scholar]
  54. Segal M., Clarke D.J., Maddox P., Salmon E.D., Bloom K., Reed S.I. Coordinated spindle assembly and orientation requires Clb5p-dependent kinase in budding yeast. J. Cell Biol. 2000;148:441–452. doi: 10.1083/jcb.148.3.441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Song S., Grenfell T.Z., Garfield S., Erikson R.L., Lee K.S. Essential function of the polo box of Cdc5 in subcellular localization and induction of cytokinetic structures. Mol. Cell. Biol. 2000;20:286–298. doi: 10.1128/mcb.20.1.286-298.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sundin B.A., Chiu C.-H., Riffle M., Davis T.N., Muller E.G.D. Localization of proteins that are coordinately expressed with Cln2 during the cell cycle. Yeast. 2004;21:793–800. doi: 10.1002/yea.1133. [DOI] [PubMed] [Google Scholar]
  57. Swaffer M.P., Jones A.W., Flynn H.R., Snijders A.P., Nurse P. CDK substrate phosphorylation and ordering the cell cycle. Cell. 2016;167:1750–1761.e16. doi: 10.1016/j.cell.2016.11.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Takeda D.Y., Wohlschlegel J.A., Dutta A. A bipartite substrate recognition motif for cyclin-dependent kinases. J. Biol. Chem. 2001;276:1993–1997. doi: 10.1074/jbc.M005719200. [DOI] [PubMed] [Google Scholar]
  59. Tompa P., Davey N.E., Gibson T.J., Babu M.M. A million peptide motifs for the molecular biologist. Mol. Cell. 2014;55:161–169. doi: 10.1016/j.molcel.2014.05.032. [DOI] [PubMed] [Google Scholar]
  60. Topacio B.R., Zatulovskiy E., Cristea S., Xie S., Tambo C.S., Rubin S.M., Sage J., Kõivomägi M., Skotheim J.M. Cyclin D-Cdk4,6 drives cell-cycle progression via the retinoblastoma protein’s C-terminal helix. Mol. Cell. 2019;74:758–770. doi: 10.1016/j.molcel.2019.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Ubersax J.A., Woodbury E.L., Quang P.N., Paraz M., Blethrow J.D., Shah K., Shokat K.M., Morgan D.O. Targets of the cyclin-dependent kinase Cdk1. Nature. 2003;425:859–864. doi: 10.1038/nature02062. [DOI] [PubMed] [Google Scholar]
  62. Villalta J.I., Galli S., Iacaruso M.F., Antico Arciuch V.G., Poderoso J.J., Jares-Erijman E.A., Pietrasanta L.I. New algorithm to determine true colocalization in combination with image restoration and time-lapse confocal microscopy to MAP kinases in mitochondria. PLoS ONE. 2011;6:e19031. doi: 10.1371/journal.pone.0019031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Wilmes G.M., Archambault V., Austin R.J., Jacobson M.D., Bell S.P., Cross F.R. Interaction of the S-phase cyclin Clb5 with an “RXL” docking sequence in the initiator protein Orc6 provides an origin-localized replication control switch. Genes Dev. 2004;18:981–991. doi: 10.1101/gad.1202304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Wohlschlegel J.A., Dwyer B.T., Takeda D.Y., Dutta A. Mutational analysis of the Cy motif from p21 reveals sequence degeneracy and specificity for different cyclin-dependent kinases. Mol. Cell. Biol. 2001;21:4868–4874. doi: 10.1128/MCB.21.15.4868-4874.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Yang J., Yan R., Roy A., Xu D., Poisson J., Zhang Y. The I-TASSER suite: protein structure and function prediction. Nat. Methods. 2015;12:7–8. doi: 10.1038/nmeth.3213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Yu H., Braun P., Yildirim M.A., Lemmens I., Venkatesan K., Sahalie J., Hirozane-Kishikawa T., Gebreab F., Li N., Simonis N. High-quality binary protein interaction map of the yeast interactome network. Science. 2008;322:104–110. doi: 10.1126/science.1158684. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Figures S1–S7 and Tables S1 and S2
mmc1.pdf (7.8MB, pdf)
Document S2. Article plus Supplemental Information
mmc2.pdf (11.8MB, pdf)

Data Availability Statement

Original data have been deposited to Mendeley Data: https://doi.org/10.17632/yxzzy26hbd.1.

For quantification of fluorescent proteins from time-lapse experiments, MATLAB scripts from Doncic et al. (2013) were adapted and used, and are available upon request.

RESOURCES