Cell cycle regulated expression of the WHI7 Start repressor gene

ABSTRACT

Periodic transcriptional waves along the cell cycle ensure the accurate progression of the different cell cycle phases through the timely regulated expression of cell cycle proteins. The G1/S transition (Start) consists in the activation of a transcriptional program by G1 CDKs through the inactivation of Start transcriptional repressors, Whi5 and Whi7 in yeast or Rb in mammals. Here, we provide a comprehensive characterization of the transcriptional regulation of the Start repressor Whi7 in budding yeast. We found that WHI7 is a cell cycle regulated gene that shows periodic expression peaking in G1. Our results demonstrate that the three cell cycle transcriptional programs related to G1 and their corresponding transcription factors are involved in the transcriptional control of WHI7. Specifically, we identified that the transcriptional regulators Swi5 and Mcm1-Yox1, which are involved in late M and early G1 expression, and the transcription factors MBF and SBF, which are responsible for G1/S expression, are able to associate and regulate the WHI7 gene. In summary, in this work, we provide new mechanisms for the regulation of the Start repressor Whi7, which highlights the precise and complex control of the cell cycle machinery governing the G1/S transition.

Introduction

The cell cycle consists of an orderly and timely succession of processes by which cells proliferate. A sophisticated regulatory network with multiple levels of control, governs cell cycle progression, ensuring that cell division occurs accurately. Cell division requires synthesis of numerous proteins. Not surprisingly, a crucial strategy in cell cycle regulation involves the control of gene transcription. Indeed, during the cell cycle, there is a dramatic reorganization of gene expression, known as cell cycle regulated transcription or periodic transcription. Transcript levels of many genes fluctuate significantly throughout the cell cycle, with a peak at particular periods to optimize the production of proteins when their functions are needed. The regulation of gene transcription through the cell cycle has been extensively studied in budding yeast, revealing that hundreds of genes (up to 20% of the genome) encoding not only critical cell cycle regulators but also proteins with specific roles in the cell cycle, are periodically expressed along the cell cycle [1–3].

Gene expression during the cell cycle is tightly regulated, with the existence of a number of specific transcriptional programs responsible for the coregulated expression of different gene clusters. Thus, waves of periodic expression occur at particular cell cycle stages, such as G1/S, S phase, G2/M and M/G1, with the activation and repression of specific transcriptional regulatory factors. An interesting hallmark of these transcriptional programs is that they are intertwined, since one wave of gene expression periodically expresses transcription factors that regulate the next wave [4]. On the other hand, these different cell cycle transcriptional programs are intimately linked to the activity of the cyclin-dependent protein kinases (CDKs) that drive cell cycle events: the transcriptional programs affect CDK activity by providing the timely expression of specific cyclin genes or CDK-inhibitors (CKI) genes, and in turn, CDKs phosphorylate transcription factors to regulate their activity [5,6]. Thus, transcription factors play crucial roles in cell proliferation and control of their activity is integrated at the heart of the cell cycle control system. Importantly, the regulatory principles underlying cell cycle control are conserved among eukaryotes, and deregulated cell cycle gene expression can lead to uncontrolled cell divisions that cause cell death and diseases [7].

The G1/S transition (Start in budding yeast), in which cells irreversibly commit to initiate a new round of division, is the main step in cell cycle control. The Start regulatory network, which is highly conserved across eukaryotes, consists in the activation of a specific transcriptional program that triggers the rapid expression of numerous genes encoding cell cycle regulators and proteins essential for early cellular events [8,9]. In Saccharomyces cerevisiae, SBF and MBF transcription factors drive this Start transcriptional wave. SBF and MBF are heterodimers that share the regulatory subunit Swi6 but present a different sequence-specific DNA-binding protein, Swi4 (SBF) or Mbp1 (MBF) [2]. Counteracting SBF and MBF, the Whi5 and Whi7 transcriptional repressors (functional homologs of Retinoblastoma family proteins in mammalian cells) inhibit the activation of the Start transcriptional program. Thus, although SBF is bound to the promoters of Start genes in early G1, Start transcription is turned off due to the association of Whi5 and Whi7 repressors to target promoters [10–12]. In late G1, CDK Cdc28 associated with cyclin Cln3 initially promotes the basal expression of SBF-dependent genes, including the CLN1 and CLN2 Start cyclin genes [13]. As a consequence, the accumulation of Cdc28-Cln1 and Cdc28-Cln2 activities leads to the full induction of the Start transcriptional program through the inactivation of the repressors and the activation of SBF and MBF [8], creating a positive feedback loop that provides coherence and irreversibility to the G1/S transition [14,15]

Although Whi5 and Whi7 are related proteins, notable functional distinctions between these repressors exist [16]. Interestingly, the expression of WHI7, but not WHI5, is significantly upregulated under various stress conditions [17,18], including cell wall stress [16,19] where Whi7, rather than Whi5, becomes the primary Start repressor [16]. In addition, Pho85, a multifunctional CDK that signals to the cell when environmental conditions are favorable [20,21], promotes Start through the specific downregulation of the Whi7 repressor at multiple levels [22]. Finally, a novel role of Whi7 in quiescence has also been proposed [23]. All these observations point to a functional specialization of Whi7 compared to Whi5 to connect the response to adverse conditions with the cell cycle control network.

A large number of genes required for G1 functions are expressed during the transition from mitosis to early G1. These genes were grouped into two clusters. The first class of M/G1 expressed genes is called the SIC1 gene cluster and is regulated by the paralogous transcription factors Swi5 and Ace2. Although they bind to the same DNA sequences [24], there are genes specifically regulated by one of these transcription factors such as the endonuclease HO and chitinase CTS1 that are specifically regulated by Swi5 and Ace2, respectively, and genes, such as SIC1 that are regulated by both factors [25,26]. The second class of M/G1 expressed genes is the Mcm1 gene cluster. It includes, among others, the genes encoding cyclin Cln3 that initially activate Start and the transcription factor of the Start transcriptional program Swi4. The expression of these genes is mediated by the MADS box transcription factor Mcm1, which binds to the Early Cell-cycle Box (ECB) located in the promoter of target genes [27,28]. Importantly, most of the genes in the Mcm1 gene cluster contain, adjacent to the Mcm1 binding site, a binding site for the paralogous homeodomain repressors Yox1 or Yhp1 [29]. However, the interaction between Yox1 and Yhp1 homeodomain proteins with Mcm1 is potent enough to anchor these suppressors to the ECB-Mcm1 complex even in the absence of a specific DNA-binding site. As Mcm1 is constitutively bound to promoter genes throughout the cell cycle, it is the presence of these repressors that blocks Mcm1-mediated transcription throughout most of the cell cycle [29].

As commented above, tight regulation of transcriptional activators and repressors constitutes a key mechanism of cell cycle control. Our group has previously described new mechanisms of control of the Start transcriptional repressor Whi7, including the cell cycle dependent regulation of Whi7 function, stability, phosphorylation and subcellular localization [12,16,22]. Here, we add a new layer in the cell cycle regulation of Whi7 showing that WHI7 gene is periodically expressed peaking in G1 and that this periodic expression depends on a precise and complex control in the M/G1 and G1/S transitions.

Materials and methods

Strains, plasmids and growing conditions

The yeast strains used in this study are listed in Supplementary Table 1. Gene deletions, promoter replacements and insertions of tags were generated by standard one-step PCR-based methods [30,31].

The centromeric plasmid pWHI7 expressing C-terminal HA-tagged wild-type Whi7 was a gift from Dr. M. Aldea [32]. Plasmids containing mutations in the WHI7 promoter either at the MCB site (pWHI7^MCBmut), Swi5 binding site (pWHI7^Swi5BSmut), ECB site (pWHI7^ECBmut), Yox1 binding site (pWHI7^Yox1BSmut) or a combination of sites (pWHI7^{Swi5BSmut,MCBmut}, pWHI7^{Yox1BSmut,ECBmut}, pWHI7^{MCBmut,ECBmut}) were obtained by site-directed mutagenesis with the highly processive Pfu polymerase using pWHI7 as template. Briefly, mutagenesis was achieved using a PCR program consisting of 20 cycles, with 1 min of denaturation at 95°C, 1 min 30 s of annealing at 52°C, and 10 min of extension at 68°C, using primers containing the desired changes in the sequence indicated in Table 1. Following the PCR reaction, the mixture was incubated with DpnI for 3 h at 37°C. This step specifically digested the methylated DNA strands originated in E. coli (which served as a template for the reaction), leaving the newly synthesized (non-methylated) strands intact. The product obtained from this digestion was used to transform competent DH5α cells.

Table 1.

Regulatory sequences responsible for periodic gene expression identified by sequence analysis of WHI7 promoter.

Site	Transcription factor	Sequence	Position	Mutated sequence
Mcm1 BS (ECB)	Mcm1	TCCCTTTTTAG	−138*	TGGGTTTTAG
Yox1 BS	Yox1	CATTAAA	−164	CATCGAA
Mbp1 BS (MCB)	Mbp1	ACGCGT	−217	ACATGT
Swi5 BS	Swi5	CCAGCC	−380	CACGCC

Asterisk indicate reverse strand.

The plasmid pWHI7:sfGFP-CLN2^PEST expresses under the control of the WHI7 promoter an unstable version of GFP due to the fusion with the Cln2-PEST degron. This plasmid was constructed in two steps. First, the WHI7 promoter (550 bp) obtained by PCR from genomic DNA using forward and reverse oligonucleotides containing HindIII and BamHI restriction sites, respectively, was cloned into HindIII/BamHI digested pRS413 vector to obtain pWHI7pr. Second, sfGFP-CLN2^PEST obtained by PCR from the plasmid pGAL1:sfGFP-CLN2^PEST [33] using forward and reverse oligonucleotides containing NotI and SacII restriction sites, respectively, was cloned into NotI/SacII digested pWHI7pr. pWHI7^MCBmut:sfGFP-CLN2^PEST was obtained by directed mutagenesis of the MCB site of the WHI7 promoter in pWHI7pr:sfGFP-CLN2^PEST, as described above.

The plasmid pWHI7:lacZ was constructed by cloning the WHI7 promoter (550 bp) into the plasmid YEp356R-lacZ, which was kindly provided by Dr. F. Estruch [34]. For this, the WHI7 promoter obtained by PCR from genomic DNA using forward and reverse oligonucleotides containing PstI and BamHI restriction sites, respectively, was cloned into PstI/BamHI digested YEp356R-lacZ. pWHI7^MCBmut:lacZ, pWHI7^Yox1BSmut:lacZ, pWHI7^ECBmut:lacZ and pWHI7^{Yox1BSmut,ECBmut}:lacZ, were obtained by directed mutagenesis of the corresponding binding sites of the WHI7 promoter in pWHI7:lacZ, as described above. Plasmid pSCB:lacZ expresses the lacZ reporter under the control of a fragment from −507 to −367 from the HO promoter bearing three SCB elements. Plasmid pMCB:lacZ expresses the lacZ reporter under the control of three MCB elements.

Cells were grown to exponential phase (below 1 × 10⁷ cells mL⁻¹) at 25°C in standard yeast extract-peptone-dextrose (YPD) medium or synthetic complete (SC) medium with 2% glucose. Plasmid-carrying strains were grown in minimal SC medium supplemented as required to select auxotrophic markers. For G1 arrest and release, exponentially growing MATa cells were arrested with 15 μg mL⁻¹ α-factor (GenScript Cat. No: RP01002) for 3 h, washed three times and resuspended in fresh medium in the absence of pheromone. For telophase arrest and release, exponential cdc15^ts cells were incubated for 3 h at 37°C and released by transferring the culture to 25°C. For metaphase arrest and release, MET3:CDC20 growing cells were incubated in the presence of 3 mM L-methionine for 3 h, washed three times with SC medium and released into SC without methionine.

Western blot analysis

Protein extraction, SDS-PAGE and western blot analysis were performed as previously described [22]. Briefly, approximately 1 × 10⁸ cells were collected, resuspended in 100 μL water, and incubated for 5 min at room temperature after adding 100 μL 0.2 M NaOH. Cells were collected by centrifugation, resuspended in 50 μL Laemmli Sample Buffer, and incubated for 5 min at 95°C. Extracts were clarified by centrifugation, and equivalent amounts of protein were resolved in an SDS-PAGE gel and transferred onto a nitrocellulose membrane (GE Healthcare). The primary antibodies used were monoclonal anti-HA conjugated with peroxidase 3F10 antibody (Roche Diagnostics, Cat. No: 12013819001) diluted 1:5000, monoclonal anti-Cdc2 p34 (PSTAIRE) sc-53 (Santa Cruz Biotechnology Inc.) diluted 1:2000 and polyclonal anti-G6PDH (Thermo Scientific, Cat. No: BS-6989 R) diluted 1:30000. The secondary antibody (when required) was an anti-rabbit IgG Horseradish Peroxidase conjugate (Thermo Fisher Scientific, Cat. No: 170–6516 or 31,460 respectively) diluted 1:20000. The uncropped western blots are shown in Figure S2.

Gene expression analysis

mRNA levels were analyzed by RT-qPCR as previously described [22]. Briefly, approximately 2 × 10⁸ cells were lysed in a FastPrep Precellys24 (Bertin Technologies). Total RNA was extracted using the Qiagen RNeasy kit according to the manufacturer’s protocol. RNA was quantified using a NanoDrop 2000 Spectrophotometer (Thermo Scientific). 10 μg of RNA were incubated with DNase (Roche), and after DNase inactivation and incubation with oligo dT, cDNA was obtained with reverse transcriptase Maxima Reverse Transcriptase (Thermo Scientific). cDNA was analyzed in triplicate by quantitative RT-PCR with a DNA Engine Peltier Thermal Cycler (Bio-Rad) using the NZYSpeedy qPCR Green Master Mix. The mRNA levels of the genes of interest were quantified relative to ACT1 mRNA using the ΔCt method. For β-galactosidase activity assays, approximately 10⁸ cells were harvested, washed with water, and resuspended in 200 µL of Z buffer (60 mM Na₂HPO₄, 40 mM NaH₂PO₄·H₂O, 10 mM KCl, 1 mM MgSO₄, 0.1% β-mercaptoethanol, pH 7.0) containing 1 mg mL⁻¹ of zymolyase-20T. After 20 min of incubation at 30°C, the extracts were centrifuged and the supernatants were preserved. Different quantities of extracts (20–180 µL) were taken, and 900 µL of Z buffer was added. The reaction was initiated by the addition of 180 µL of 4 mg mL⁻¹ o-nitrophenyl-β-d-galactopiranoside (ONPG). The samples were incubated at 30°C until they become yellow. The reaction was stopped with 450 µL of Na₂CO₃ 1 M and A₄₂₀ was determined. Activity is expressed in U µg⁻¹ of protein where 1 U is defined as A₄₂₀ ×10³ × (min of incubation)⁻¹. The protein concentration was determined using the Bradford assay.

Fluorescence microscopy

Conventional fluorescence microscopy was performed using an Axioskop 2 microscope (Zeiss). Images were captured with an AxioCam MRm camera (Zeiss) and AxioVision v4.7 software (Zeiss). Image analysis and fluorescence intensity quantification were done using the ImageJ software (https://imagej.net/software/fiji/). The imaged cells were segmented using DIC acquisition with the aid of YeastMate [35] and ImageJ. After background subtraction, the mean GFP fluorescence intensity was quantified for the whole cell.

Chromatin immunoprecipitation analysis

Chromatin immunoprecipitation analysis were carried out as previously described [22]. Dynabeads Protein G (Invitrogen) previously bound to an HA-probe (F-7) antibody (Santa Cruz Biotechnology Inc. Cat. No: SC-7392), monoclonal GFP antibody (Roche Diagnostics, Cat. No: 11814460001) or monoclonal anti-c-myc 9E10 antibody (Roche Diagnostics, Cat. No: 1667149) were used.

Statistical methods and reproducibility

Graphs and statistical analyses (two-tailed unpaired t-test and one-sample t-test) were performed using GraphPad Prism software. For all the strains used in this study, at least two independent clones were tested, with similar results. All data were obtained from multiple independent biological replicates (ranging from 2 to 19), as indicated in the figure legends.

Results

The WHI7 gene shows periodic expression peaking in G1

Many cell cycle regulators are encoded by genes whose expression level fluctuates throughout the cell cycle. Our group has described that the Start repressor Whi7 is an unstable protein whose level and phosphorylation are cell cycle regulated [12]. To further characterize the molecular basis of the cell cycle regulation of Whi7, we investigate whether WHI7 gene is expressed periodically. Therefore, we conducted an analysis of WHI7 mRNA levels in synchronized cultures. Cells were arrested in G1 by incubation with α-factor and, after release from the arrest, the levels of WHI7 mRNA, as well as SIC1 and CLN2 mRNA (as examples of well-known cell cycle regulated genes expressed in the M/G1 and G1/S transition, respectively) were analyzed by quantitative RT-PCR (Figure 1). Cell cycle progression was also monitored by bud emergence as an indication of Start execution. Two cycles of division are shown, and it has to be noted that, unlike the case of typical G1/S genes like CLN2, which are barely expressed in the presence of the pheromone, the first G1/S transition could not be analyzed for WHI7 because the gene is induced by α-factor [36]. As it can be observed, SIC1 mRNA accumulation reflected mitotic exit and G1 entry at approximately 80 min after the release; later, the CLN2 mRNA peak marked the execution of Start at around 100 min. In the case of WHI7, mRNA levels indeed fluctuate along the cell cycle, although the observed 3-fold oscillation is milder than those of CLN2 and SIC1 mRNAs. Concerning its periodicity, maximum expression of the WHI7 gene occurs during the G1 phase spanning from SIC1 to CLN2 peaks.

Figure 1.

The WHI7 gene is periodically expressed peaking in G1. (a) Exponentially growing cultures of wild type (W303-1a) cells were arrested in G1 by incubation with α-factor for 3 h (>95% of cells arrested). After release from the arrest, cell cycle progression was monitored by bud emergence (indicative of Start activation) and the % of budded cells was scored at the indicated times. At least 200 cells were counted for each time. (b) SIC1 (M/G1 gene), CLN2 (Start gene) and WHI7 mRNA levels relative to ACT1 mRNA as a control were analyzed by quantitative RT-PCR. The maximum value of mRNA expression for each gene was normalized to 1. One representative experiment is shown.

The analysis of the WHI7 promoter sequence identified several known consensus regulatory elements responsible for periodic gene expression [29,37,38] (Table 1). These include the MCB (MluI Cell cycle Box) site responsible for cell cycle expression at Start, the Swi5-binding site responsible for cell cycle expression at mitotic exit and the ECB (Early Cell cycle Box) and Yox1 binding site involved in regulated expression at early G1. Interestingly, all these putative regulatory elements are consistent with the observed pattern of cell cycle regulated expression of the WHI7 gene in G1.

WHI7 gene expression depends on the G1/S transcription factors Mbp1 and Swi4

The Start transcriptional program is mediated by two related transcription factors, SBF (Swi4/Swi6) and MBF (Mbp1/Swi6). Although SBF and MBF preferentially bind to SCB and MCB regulatory elements, respectively, there is a significant functional overlap between them [39]. Because the WHI7 promoter contains a MCB element, we hypothesize that MBF (or/and SBF) could play a role in the expression of the gene. To test this, we initially examined the expression of the lacZ reporter gene driven by the WHI7 promoter (WHI7:lacZ gene fusion) in mutant strains lacking Mbp1, Swi4, or both. As controls, we also analyzed the expression of reporter genes SCB:lacZ and MCB:lacZ. As expected, due to the redundancy between SBF and MBF, the expression of SCB:lacZ and MCB:lacZ was reduced in the absence of either Swi4 or Mbp1 transcription factors (Figure 2(a)). Notably, the expression of WHI7:lacZ was also decreased in the mbp1 and swi4 mutant cells, with the double mutant strain showing an additive effect (Figure 2(a)). These findings support the role of Mbp1 and Swi4 as transcription factors in activating WHI7 expression during the cell cycle.

Figure 2.

WHI7 expression depends on Mbp1 and Swi4. (a) Wild-type (W303-1a), mbp1 (JCY2194), GAL1:SWI4 (JCY2268) and GAL1:SWI4 mbp1 (JCY2304) strains were transformed with either the plasmid pSCB:lacZ, pMCB:lacZ or pWHI7:lacZ. Cells were grown in galactose and switched overnight to glucose to inactivate SWI4 expression and β-galactosidase activity was measured. Data from the independent biological replicates are represented as mean ± s.D. The mean of the wt is referred to as 1. Two-tailed unpaired t-test was used; ** p ≤ 0.01 and * p ≤ 0.05. (b) Mbp1 and Swi4 association with CLN2, RNR1 and WHI7 promoters was analyzed by ChIP assays in W303-1a (no-tag), MBP1-HA (JCY2084) and SWI4-HA (JCY1954) strains. Data from the independent biological replicates are represented as mean ± s.D. The mean of No-tag is referred to as 1. For WHI7 promoter, one-sample t-test was used; * p ≤ 0.05.

To further explore the involvement of Mbp1 and Swi4 in WHI7 gene expression, we performed Chromatin immunoprecipitation (ChIP) experiments to analyze the binding of these transcription factors to the WHI7 promoter. As controls, we tested the association of Swi4 or Mbp1 with either CLN2 (a known Swi4-regulated gene) or RNR1 (a known Mbp1-regulated gene), respectively. Importantly, Mbp1, and to a lesser extent Swi4, were found to be recruited to the WHI7 promoter (Figure 2(b)). This result further confirms that WHI7 gene expression is regulated by the Start transcription factors Mbp1 and Swi4.

Finally, we aimed to directly investigate the role of the MCB element present in the WHI7 promoter in the regulation of gene expression. To achieve this, we mutated the MCB element via directed mutagenesis and analyzed the impact on Whi7 protein and WHI7 mRNA levels. As it is observed in Figure 3(a,b), there was a substantial decrease in both WHI7 mRNA levels and, consistently, in Whi7 protein levels, when the MCB was altered. In agreement with this, mutating the MCB sequence in the WHI7:lacZ reporter also led to reduced β-galactosidase activity compared to the wild-type version (Figure 3(c)). Additionally, the expression of an unstable version of the GFP protein under the control of the WHI7 promoter (WHI7:sfGFP-CLN2^PEST) indicated a significant decrease in the GFP fluorescence signal when the MCB site was removed (Figure 3(d)).

Figure 3.

The MCB site in WHI7 promoter is required for WHI7 full expression. (a) whi7 mutant strain (JCY1819) was transformed with a plasmid containing either the wild type (pWHI7) or the MCB mutated version (pWHI7^MCBmut) of the WHI7-HA gene. Whi7 protein levels were measured and normalized to Cdc28 (PSTAIRE) as loading control. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (b) In the same cells used in panel a, the level of WHI7 mRNA relative to ACT1 mRNA was analyzed by quantitative RT-PCR. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (c) Wild-type strain (W303-1a) was transformed with the plasmid pWhi7:lacZ or pWHI7^MCBmut:lacZ and β-galactosidase activity was measured. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (d) Whi7 promoter-driven expression was analyzed by fluorescence microscopy in whi7 mutant cells transformed with an empty vector or a plasmid expressing an unstable version of GFP (sfGFP-CLN2^PEST) from the wild type WHI7 promoter or its MCB mutated version. GFP mean fluorescence of the whole cell segmentation was quantified. The mean of the vector value is referred to as 1. n indicates the number of cells analyzed. Microscopy images of DIC and GFP are shown. Scale bar, 4 μm. Two-tailed unpaired t-test was used; ****p ≤ 0.0001 and *p ≤ 0.05.

In summary, our findings demonstrate that WHI7 expression is highly dependent on Mbp1 and Swi4, with the MCB regulatory element on the WHI7 promoter being essential for its full expression.

WHI7 gene expression depends on the M/G1 transcription factor Swi5

As previously mentioned, the WHI7 promoter contains a consensus binding site for the cell cycle transcription factor Swi5, and the timing of the WHI7 periodic expression is consistent with the gene being regulated by Swi5. Therefore, we next investigated the involvement of Swi5 in WHI7 gene expression. Initially, we observed a significant decrease in β-galactosidase activity in swi5 mutant cells carrying the WHI7:lacZ fusion gene compared to wild-type cells (Figure 4(a)). Further supporting a role for Swi5 in WHI7 regulation, when gene expression was examined in synchronized cultures after release from a metaphase arrest, whereas the expression of CLN2 gene (no regulated by Swi5) was not affected by the inactivation of Swi5, decreased mRNA levels were observed not only in the control Swi5-regulated gene SIC1 but also in the case of the WHI7 gene, although WHI7 expression still remains cell cycle regulated (Figure S1). Thus, our results indicate that Swi5 contributes to activating WHI7 gene expression in G1.

Figure 4.

Swi5 transcription factor activates WHI7 expression through the Swi5 binding site on its promoter. (a) Wild type (W303-1a) and swi5 mutant strain (JCY2620) were transformed with the plasmid pWhi7:lacZ and β-galactosidase activity was measured. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (b) Swi5 association with WHI7 and PCL2 promoters was analyzed by ChIP assays in W303-1a (no-tag) and SWI5-GFP (JCY2631) strains. Graph represents the mean ± s.D. derived from independent biological replicates. The No-tag value is referred to as 1. (c) cdc15^ts SWI5-GFP (JCY2629) mutant cells were arrested in telophase by incubation at 37°C for 3 hours and after released from the arrest Swi5 association with WHI7 and PCL2 promoters was analyzed by ChIP assays. The time 0 value was referred to as 1. (d) whi7 mutant strain (JCY1819) was transformed with plasmid pWHI7 (wt) or its derivatives pWHI7^MCBmut, pWHI7^Swi5BSmut or pWHI7^{Swi5BSmut,MCBmut}. The level of WHI7 mRNA relative to ACT1 mRNA was analyzed by quantitative RT-PCR. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (e) Ace2 association with WHI7 and CTS1 promoters was analyzed by ChIP assays in W303-1a (no-tag) and ACE2-HA (JCY0707) strains. Graph represents the mean ± s.D. derived from independent biological replicates. The No-tag value is referred to as 1. Two-tailed unpaired t-test was used; ****p ≤ 0.0001, *p ≤ 0.05 and ns p > 0.05.

Next, we conducted a ChIP analysis to explore Swi5 binding to the WHI7 promoter, using the PCL2 cyclin promoter (an early G1 Swi5-regulated gene) as a positive control. The results revealed a direct association of Swi5 with the WHI7 promoter in both asynchronous cultures (Figure 4(b)) and synchronized cultures after release from a telophase arrest (Figure 4(c)). Note that the kinetics of Swi5 association with the WHI7 promoter in the synchronized experiment mirrored that of the PCL2 cyclin promoter. These findings further support that Swi5 is indeed regulating WHI7 gene expression.

To further investigate the connection of Swi5 with WHI7 expression, we studied the functionality of the Swi5 binding site within the WHI7 promoter. As it can be observed in Figure 4(d), the mutation of the Swi5 binding site resulted in a significant decrease in WHI7 mRNA levels, similarly to the impact of altering the MCB. Noteworthy, the mutation of both the Swi5 binding site and the MCB exhibited an additive effect, further reducing WHI7 mRNA levels. This result reveals that both the Swi5 binding site and MCB regulatory elements independently contribute to WHI7 expression.

In conjunction with Swi5, another transcription factor, Ace2, is responsible for the regulation of the transcriptional wave at the end of mitosis. While certain genes are individually regulated by either Swi5 or Ace2, many others are controlled by both factors, as they recognize the same DNA sequence. Because of that, we investigated the binding of Ace2 to the WHI7 promoter. In contrast to the control Ace2-regulated gene CTS1, no association of Ace2 with the WHI7 gene was detected (Figure 4(e)). This result strongly suggests that WHI7 is regulated by Swi5 and not by Ace2.

The WHI7 periodic gene expression is perturbed when MCB and Swi5 binding sites are mutated

Once it was confirmed that the MCB and Swi5 binding sites are functional regulatory elements of WHI7 gene expression, we aimed to further evaluate their impact on the periodic expression of the gene. To do this, cultures of whi7 cells transformed with a centromeric plasmid containing either a wild type WHI7 gene, a WHI7 gene mutated at the MCB binding site, the Swi5 binding site, or both MCB and Swi5 binding sites were arrested in G1 using α-factor, and WHI7 expression was analyzed after the release from the arrest. The expression of CLN2 and SIC1 genes served as controls for cell cycle progression. When either the MCB site or the Swi5 binding site was individually mutated, a pattern of periodic expression with a peak in G1, similar to that of the wild-type WHI7 gene, was still observed (Figure 5). However, when both the MCB and Swi5 binding sites were simultaneously mutated, the maximum G1 periodic expression of WHI7 was abrogated. In all cases, the regulated expression of CLN2 and SIC1, as well as the budding index (data not shown), confirmed a similar degree of culture synchronization. These results clearly indicate that the cell cycle regulated expression of the WHI7 gene is mediated by the MCB and the Swi5 binding site regulatory elements, with each element capable of supporting gene periodic expression in G1.

Figure 5.

MCB and Swi5 binding sites are required for periodic WHI7 expression. exponentially growing cultures of whi7 mutant cells (JCY1819) transformed with either pWHI7, pWHI7^MCBmut, pWHI7^Swi5BSmut or pWHI7^{Swi5BSmut,MCBmut}, were arrested in G1 by incubation with α-factor for 3 h (>95% of cells arrested). After release from the arrest, WHI7, CLN2 (Start gene) and SIC1 (M/G1 gene) mRNA levels relative to ACT1 mRNA as a control were analyzed by quantitative RT-PCR. In all the panels, the maximum value of mRNA expression for each gene was normalized to 1. One representative experiment is shown.

Pho85 regulates WHI7 transcription through the Swi5 transcription factor

We have previously reported that Pho85-Pho80 inhibits WHI7 expression [22]. It is known that the CDK Pho85 regulates the Swi5 transcription factor. Specifically, Pho85-Pho80 interacts with Swi5 and pho85 mutant cells shows increased expression of some Swi5-regulated genes, such as ASH1, supporting the notion that Pho85 represses Swi5 function [40]. Having established that Swi5 promotes WHI7 expression, Swi5 emerges as a strong candidate for mediating the regulation of WHI7 gene expression by the Pho85 pathway. To elucidate this point, we analyzed whether the increased WHI7 mRNA and Whi7 protein levels found in pho85 mutant cells occur in the absence of Swi5. The results showed that the increase in both WHI7 mRNA and Whi7 protein levels in pho85 mutant cells was completely dependent on the presence of Swi5, as the pho85swi5 double mutant exhibited levels similar to those of the wild-type strain (Figure 6(a,b)). These results indicate that the CDK Pho85 represses WHI7 transcription through the regulation of the Swi5 transcription factor.

Figure 6.

Pho85 promotes WHI7 transcription through Swi5. (a) The level of WHI7 mRNA relative to ACT1 mRNA was analyzed by quantitative RT-PCR in wild type (W303-1a), pho85 (JCY2221) and pho85swi5 (JCY2622) strains. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (b) Whi7 protein level was measured in the same cells used in panel a by western blot. α-G6PDH is shown as loading control. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. Two-tailed unpaired t-test was used; *** p ≤ 0.001, ** p ≤ 0.01 and ns p > 0.05.

Mcm1 and Yox1 regulate WHI7 gene expression

A third putative regulatory element present in the WHI7 promoter that may be compatible with the observed periodic expression is the ECB (Early Cell cycle Box) sequence, which mediates expression in the early G1 phase. The MADS transcription factor Mcm1 binds to the ECB sequence to regulate this transcriptional wave, with a peak in early G1. To test the involvement of Mcm1 and the putative ECB element in regulating WHI7 gene expression, we first assayed the β-galactosidase activity in cells containing the WHI7:lacZ reporter gene with a mutated ECB element. Surprisingly, the results indicated an increase in WHI7 expression when the ECB sequence was inactivated (Figure 7(a)). In agreement with this, we observed an up-regulation of Whi7 protein and WHI7 mRNA levels when the ECB element was mutated (Figure 7(b,c)). This up-regulation was also observed when the ECB mutation was introduced in a promoter already mutated at the MCB site (Figure 7(b,c)). These observations demonstrate that the ECB element in WHI7 promoter is functional but unexpectedly plays a negative role in regulating WHI7 gene expression.

Figure 7.

The ECB element and Mcm1-Yox1 repress WHI7 expression. (a) β-galactosidase activity was measured in exponentially growing whi7 cells (JCY1819) transformed with either the plasmid pWHI7:lacZ, pWHI7^Yox1BSmut:lacZ, pWHI7^ECBmut:lacZ or pWHI7^{Yox1BSmut, ECBmut}:lacZ. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (b) The level of WHI7 mRNA relative to ACT1 mRNA was analyzed by quantitative RT-PCR in exponentially growing whi7 cells (JCY1819) transformed with either the plasmid pWHI7, pWHI7^MCBmut, pWHI7^ECBmut or pWHI7^{MCBmut, ECBmut}. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (c) Whi7 protein level was measured in the same cells as in panel b. Cdc28 (PSTAIRE) was used as loading control. Graph represents the mean ± s.D. derived from independent biological replicates. The wild-type value is referred to as 1. (d) Mcm1 and Yox1 association with CLN3, CDC6 and WHI7 promoters was analyzed by ChIP assays in W303-1a (no-tag), MCM1-myc (JCY2491) and YOX1-HA (JCY2482) strains. Graphs represent the mean ± s.D. derived from independent biological replicates. The No-tag value is referred to as 1. (e) Yhp1 association with WHI7 promoter was analyzed by ChIP assays in W303-1a (no-tag) and YHP1-HA (JCY2731) strain. Graph represents the mean ± s.D. derived from independent biological replicates. The No-tag value is referred to as 1. Two-tailed unpaired (a-c) or one-sample (d, e) t-test was used; ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05 and ns p > 0.05.

Mcm1 has a MADS box domain that interacts with homeodomain proteins and confers DNA binding specificity to regulate transcription. Thus, it has been described that Yox1 and Yhp1 are associated with Mcm1 to regulate target promoters, specifically acting as Mcm1 repressors. Mcm1 binds constitutively to promoters throughout the cell cycle, and the binding of these repressors helps to restrict transcriptional activity in several genes at the early G1 phase [29]. These repressors directly associate with DNA at binding sites adjacent to the Mcm1-ECB complex. Interestingly, this also seems to be the case for WHI7, since there is a putative Yox1 binding site next to the ECB site (Table 1). To study if Yox1 acts within the Mcm1-ECB context in the regulation of the WHI7 gene, we assayed β-galactosidase activity in cells containing the WHI7:lacZ reporter with a mutated Yox1 binding site. As shown in Figure 7(a), inactivation of the Yox1 binding site causes a mild but significant increase in WHI7 expression, suggesting that Yox1 acts as a repressor of the WHI7 gene. It should be noted that the inactivation of the Yox1 binding site has a milder effect than the ECB mutation. This is likely due to Yox1 ability to bind to the ECB through Mcm1 independently of the presence of a Yox1 binding site [29]. Supporting this, mutation of the Yox1 binding site had no effect in the absence of the ECB site (Figure 7(a)).

To further characterize the roles of Mcm1 and Yox1 in WHI7 expression, we analyzed the association of these transcriptional factors with the WHI7 promoter using ChIP assays (Figure 7(d)). We observed binding to positive control genes CDC6 and CLN3, known targets for Mcm1 and Yox1 proteins. Importantly, the association of both Mcm1 and Yox1 was also detected at the WHI7 promoter (Figure 7(d)). No binding to the WHI7 promoter was detected for the redundant Yhp1 repressor (Figure 7(e)), suggesting that WHI7 is a specific target for the Yox1 repressor.

In summary, these results support that the ECB element and the adjacent Yox1 binding site in the WHI7 promoter mediate repression of expression through the association of the Mcm1 and Yox1 transcription factors.

Discussion

In this work, we show that the levels of WHI7 mRNA oscillate within the cell cycle, peaking in G1, coinciding with M/G1 and G1/S transcriptional waves. We have unveiled the molecular mechanism behind this WHI7 cell cycle regulated expression characterizing the responsible transcription factors: Swi5 and Mcm1-Yox1 for M/G1 and Mbp1/Swi4 for G1/S (Figure 8). In S. cerevisiae there are three independent transcriptional programs that drive gene expression during G1. Remarkably, all these three programs are acting on WHI7 gene, highlighting the need for a robust control of WHI7 expression in M/G1 and Start.

Figure 8.

Model of WHI7 cell cycle regulated expression. at the end of mitosis, the Swi5 transcription factor activates WHI7 expression. Also at the M/G1 transition, Mcm1 and Yox1 repressor regulate WHI7 expression. During G1, Whi7 acts as a repressor of the Start transcriptional program. Later on in Start, Mbp1 and Swi4 activate WHI7 expression, which could function as a negative feedback mechanism on the Start transcriptional program (see discussion).

Functional redundancy between proteins is a common trait in cell cycle regulation. We have characterized that Swi5 activates WHI7 periodic expression through the Swi5 binding site on WHI7 promoter. Interestingly, Swi5 and its paralog Ace2 bind to the same DNA sequences [24]. Although both drive the M/G1 periodic expression, there is a limited overlapping group of genes, such as SIC1, as well as genes that are either Swi5-specific or Ace2-specific [25,26]. Our results have revealed that Swi5, but not Ace2, binds to the WHI7 promoter, suggesting that WHI7 is specifically regulated by Swi5. Interestingly, Ace2, like Swi5, is negatively regulated by Pho85 [41]. In this sense, we have observed that the increased expression of the WHI7 gene caused by Pho85 inactivation totally depends on the Swi5 protein. This observation further supports an exclusive role for Swi5 in WHI7 gene expression.

Regarding the Mcm1-ECB transcriptional wave, the key mechanism in this program is the repression of ECB-bound Mcm1 action by Yox1 and Yhp1 transcriptional repressors [29]. We detect binding of Yox1, but not of Yhp1, to the WHI7 promoter. Thus, although Yox1 and Yhp1 are redundant transcriptional repressors recognizing the same sequence, in the case of the WHI7 gene Yox1 must play the predominant role. The mutation of the ECB site usually diminishes gene transcription due to the inability of Mcm1 to bind to promoters, as happens with the CLN3 and SWI4 genes [42]. However, this is not the case for the WHI7 gene, since the ECB mutation increases gene expression. This distinct behavior could be explained by the combined existence of positive (Mcm1) and negative (Yox1/Yhp1) mechanisms acting through the ECB site. It is possible that, as due to a particular promoter context, one or the other mechanisms manifests. For instance, ECB mutation in CLN3 and SWI4 could impede Mcm1 from properly activating transcription, whereas in the case of WHI7 gene, ECB mutation could alleviate Yox1-mediated repression, allowing other adjacent transcriptional factors (Swi5, MBF, SBF?) to better activate expression. On the other hand, Yox1 binding site mutation increased WHI7 expression, which is expected considering the repressive function of Yox1. However, this up-regulation is milder than that observed when the ECB was mutated. As mentioned, Yox1 (and Yhp1) repress transcription through interaction with Mcm1, so it would be expected that mutation of the Yox1 binding site would be equivalent to the mutation of the ECB site. Rather, this result suggests that Yox1 could be acting as a repressor not only through the Yox1 binding site itself but also directly on the Mcm1-ECB complex. Supporting this, there is considerable evidence indicating that the important aspect is Yox1 and/or Yhp1 gaining access to the ECB-bound Mcm1 proteins, which also occurs in the absence of an adjacent Yox1/Yhp1 binding site [29].

We have also demonstrated that the WHI7 gene is controlled by the Start transcriptional program. Although G1/S genes appear to be regulated primarily either by the transcription factor SBF or MBF, many of them exhibit some degree of redundancy of these two factors [43]. For instance, SBF usually associated with SCB sites but can also bind to MCB sites [44] and vice-versa, they can substitute for the other (although not as efficiently) in the expression of several genes [39], or even genes can switch from one to the other factor depending on the circumstances [45]. In the case of the WHI7 gene, we have determined that both SBF and MBF bind to the WHI7 promoter in wild-type cells, and that inactivation of either Mbp1 or Swi4 causes unpaired expression of the WHI7:lacZ reporter gene, with the effect being very similar for both factors. These observations indicate that WHI7 is a bona fide gene whose expression is controlled to a similar extent by both MBF and SBF transcriptional factors.

A crucial question is to understand the biological significance behind the complex regulation of WHI7 transcription. Cell cycle regulators are usually expressed just before the moment in which they act. This is especially relevant for the transcription factors responsible for the different cell cycle regulated transcriptional waves, which are often periodically expressed in the previous wave [4]. Thus, it is reasonable that WHI7 expression is upregulated by Swi5 in the M/G1 wave, since Whi7 is a transcriptional repressor of the following transition, G1/S [12,16]. Under normal conditions, the Start transcriptional program is repressed from early G1 mainly by Whi5 [10,11] Interestingly, unlike the Whi5 repressor, which is specific to SBF promoters, Whi7 can bind not only to SBF but also to MBF and Swi4 monomeric, which broadens the possibilities to repress Start transcription [16]. In addition to the positive regulation of WHI7 expression in the M/G1 transition, we found that the Mcm1 transcription factor and Yox1 repressor, which also mediates gene expression at this transition, are also able to regulate WHI7 expression, although in this case negatively. The interplay between these positive and negative regulatory mechanisms acting in the early G1 may highlight the need of a versatile control of WHI7 expression at this cell cycle point to respond to different inputs.

The maximum expression peaking in early G1 of negative regulators of Start is not restricted to Whi7, as the CDK-G1 inhibitor Cip1 is also periodically expressed via Mcm1 [46]. The upregulation of Start repressors in early G1 could certainly help to prevent Start activation until late G1. Interestingly, both Whi7 and Cip1 are activated by environmental stress [16,46]. Thus, this periodic expression could reflect the convenience for cellular physiology of the existence of regulatory mechanisms to stop the initiation of a new cell cycle in adverse conditions.

We also described here that WHI7 is expressed in the G1/S transition, in an MBF and SBF-dependent manner. Considering that when the Start transcriptional wave is launched, Whi7 is being inactivated and exported out of the nucleus [12,16], the idea of increasing the expression of the WHI7 gene at this timepoint may seem puzzling. Strikingly, levels of the protein also begin to increase as cells progress through S-phase [12]. This suggests that Whi7 could have a function after Start. In this sense, a remarkable case is the CDC6 gene, whose expression, like WHI7, is cell cycle regulated by Swi5 and Mcm1 at the M/G1 boundary, and by SBF/MBF at the G1/S transition [27,47]. Cdc6 is an essential component of the pre-replicative complex (pre-RC), which forms during G1 to license DNA-replication origins to be fired later in S-phase [48]. Like the Whi7 protein, Cdc6 is an unstable protein. The first wave of expression provides de novo synthesis of Cdc6 protein required in G1 to assemble the pre-RC complex; the second burst of transcription in G1/S is relevant in cells after a long G1 period to resynthesize Cdc6 to guarantee the formation of the pre-RC complex [47]. What could be Whi7 post-Start function? One possible explanation for the G1/S-regulated transcription of WHI7 might be the existence of a Whi7-mediated negative feedback on Start transcription as cells exit G1 phase, similar to the case of the MBF repressor Nrm1. The NRM1 gene is expressed in the Start transcriptional wave in an MBF-dependent manner. As a consequence, MBF promoters are normally repressed as cells progress into S phase via negative feedback [49]. Unlike the Nrm1 repressor, which is specific to MBF, Whi7 can bind both, SBF and MBF, as well as monomeric Swi4 protein. Thus, the existence of a putative Whi7-mediated negative feedback might function to reinforce the switch-off of the Start transcriptional program, providing new possibilities to the feedback. Alternatively, it cannot be ruled out that Whi7 could play yet unknown functions during S/G2/M phases. Future work will be required to clarify this point.

We have previously described that Whi7 protein levels oscillate through the cell cycle, increasing from Start to mitosis when it begins to decline, and that Whi7 degradation is cell cycle regulated, with the protein being stable only in G1 [12]. This raised an intriguing situation: Whi7 protein levels are lower in G1 that in any other phase, despite G1 being the phase when the protein is stable. This result necessarily pointed to the existence of additional mechanisms controlling Whi7 synthesis. Here in this work, we rule out that transcriptional down-regulation could explain this puzzling situation. Rather, the periodic WHI7 expression described further contributes to this intriguing scenery, as the WHI7 gene is maximally expressed just in G1. Overall, all of our observations firmly support the existence of additional post-transcriptional regulatory mechanisms, such as regulation of mRNA export or translation, acting in the control of Whi7 cellular levels. This could open new clues in the complex regulation of Start transcriptional repressors, not only in yeast but also in mammalian cells.

Funding Statement

This study was supported by the Spanish Government and co-financed by ERDF from the European Union, grant number [PID2020-119793GB-I00] to JCI, and by the Generalitat Valenciana, grant number [CIGE2021-093 to MGA]. CRC was a recipient of a Predoctoral Fellowship from the Spanish Government [PRE2018-083178].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available within the article and its Supplementary Information, and from the corresponding author (JCI) on reasonable request.

Supplementary materials

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15384101.2024.2402192.