Kar4, the Yeast Homolog of METTL14, is Required for mRNA m⁶A Methylation and Meiosis

Abstract

KAR4, the yeast homolog of the mammalian mRNA N⁶A-methyltransferase complex component METTL14, is required for two disparate developmental programs in Saccharomyces cerevisiae: mating and meiosis. To understand KAR4’s role in yeast mating and meiosis, we used a genetic screen to isolate 25 function-specific mutant alleles, which map to non-overlapping surfaces on a predicted structure of the Kar4 protein (Kar4p). Most of the mating-specific alleles (Mat⁻) abolish Kar4p’s interaction with the transcription factor Ste12p, indicating that Kar4p’s mating function is through Ste12p. In yeast, the mRNA methyltransferase complex was previously defined as comprising Ime4p (Kar4p’s paralog and the homolog of mammalian METTL3), Mum2p (homolog of mammalian WTAP), and Slz1p (MIS), but not Kar4p. During meiosis, Kar4p interacts with Ime4p, Mum2p, and Slz1p. Moreover, cells lacking Kar4p have highly reduced levels of mRNA methylation during meiosis indicating that Kar4p is a key member of the methyltransferase complex, as it is in humans. Analysis of kar4Δ/Δ and 7 meiosis-specific alleles (Mei⁻) revealed that Kar4p is required early in meiosis, before initiation of S-phase and meiotic recombination. High copy expression of the meiotic transcriptional activator IME1 rescued the defect of these Mei− alleles. Surprisingly, Kar4p was also found to be required at a second step for the completion of meiosis and sporulation. Over-expression of IME1 in kar4Δ/Δ permits pre-meiotic S-phase, but most cells remained arrested with a monopolar spindle. Analysis of the function-specific mutants revealed that roughly half became blocked after premeiotic DNA synthesis and did not sporulate (Spo⁻). Loss of Kar4p’s Spo function was suppressed by overexpression of RIM4, a meiotic translational regulator. Overexpression of IME1 and RIM4 together allowed sporulation of kar4Δ/Δ cells. Taken together, these data suggest that Kar4p regulates meiosis at multiple steps, presumably reflecting requirements for methylation in different stages of meiotic gene expression.

Author Summary

In yeast, KAR4 is required for mating and meiosis. A genetic screen for function-specific mutations identified 25 alleles that map to different surfaces on a predicted structure of the Kar4 protein (Kar4p). The mating-specific alleles interfere with Kar4p’s ability to interact with the transcription factor Ste12p, its known partner in mating. The meiosis-specific alleles revealed an independent function: Kar4p is required for entry into meiosis and initiation of S-phase. During meiosis, Kar4p interacts with all components of the mRNA methyltransferase complex and kar4Δ/Δ mutants have greatly reduced levels of mRNA methylation. Thus, Kar4p is a member of the yeast methyltransferase complex. Overexpression of the meiotic transcriptional activator IME1 rescued the meiotic entry defect but did not lead to sporulation, implying that Kar4p has more than one meiotic function. Suppression by Ime1p overexpression led to arrest after premeiotic DNA synthesis, but before sporulation. Loss of Kar4’s sporulation function can be suppressed by overexpression of a translation regulator, Rim4p. Overexpression of both IME1 and RIM4 allowed sporulation in kar4Δ/Δ cells.

Introduction

The budding yeast, Saccharomyces cerevisiae, undergoes cell differentiation when it enters either the mating pathway, resulting in the fusion of two haploid cells, or meiosis, leading to sporulation and generation of haploids. Each of these pathways is mediated by specific transcription factors that activate different large-scale programs, with little overlap between their transcriptional profiles. Kar4p was originally identified as a transcription factor required for efficient yeast mating (Kurihara, Beh et al. 1994, Kurihara, Stewart et al. 1996), but it is also required for meiosis (Kurihara, Stewart et al. 1996). This raises the question of whether there are overlapping functions in the two programs.

During mating, the pheromone response triggers a mitogen-activated protein kinase cascade that activates numerous proteins, including Ste12p, the major transcription factor required for mating (Reviewed in Bardwell 2005). Ste12p turns on many genes, including a mating-specific transcript of KAR4, and Kar4p then works with Ste12p to activate a subset of late-acting mating genes (Kurihara, Stewart et al. 1996, Lahav, Gammie et al. 2007). The set of genes that requires Kar4p is known, but why they require Kar4p for their expression is still not well understood. Kar4p-dependent genes seem to have promoters that contain weak Ste12p-binding sites (Pheromone Response Elements, PREs) (Lahav, Gammie et al. 2007, Su, Tamarkina et al. 2010). A potential model is that Kar4p acts to stabilize Ste12p at these weak promoters. The best studied KAR4-dependent genes are KAR3 and CIK1, which encode the two subunits of a kinesin-like motor protein responsible for nuclear congression in newly fused zygotes (Meluh and Rose 1990, Page, Satterwhite et al. 1994). When KAR4 is absent from both mating cells, the nuclei fail to fuse (Kurihara, Stewart et al. 1996); as a result, the zygote buds off haploid daughter cells. Kar4p’s activity is Ste12-dependent, and overexpression of STE12, or KAR3 together with CIK1, can largely suppress the kar4Δ mating defect (Kurihara, Stewart et al. 1996, Gammie, Stewart et al. 1999).

Similarly, meiosis is a complex process with a distinct transcriptional program; cells lacking Kar4p fail to complete the meiotic program. Entry into meiosis is regulated by a complex interplay of nutritional and cell state signals converging mainly on the promoter of IME1, the “early” meiotic transcription factor (Neiman 2011). Ime1p activates the expression of genes required early during meiosis for pre-meiotic DNA synthesis and the initiation of meiotic recombination. The middle meiotic transcription factor, Ndt80p, activates the expression of genes required to complete meiotic recombination and progress through the divisions and spore maturation (Neiman 2011, van Werven and Amon 2011, Winter 2012). Among the genes activated by Ime1p is IME2, encoding a protein kinase homologous to human CDK2 (Smith and Mitchell 1989, Kominami, Sakata et al. 1993, Guttmann-Raviv, Boger-Nadjar et al. 2001). Ime2p both down-regulates Ime1p activity (Guttmann-Raviv, Martin et al. 2002) and phosphorylates a variety of downstream meiotic regulators (Dirick, Goetsch et al. 1998, Clifford, Marinco et al. 2004, Sedgwick, Rawluk et al. 2006, Holt, Hutti et al. 2007, Ahmed, Bungard et al. 2009, Corbi, Sunder et al. 2014). Among other functions, Ime2p is responsible for the removal of the Cdk inhibitor Sic1p, and therefore cells without IME2 fail to initiate meiotic DNA replication (Dirick, Goetsch et al. 1998, Sedgwick, Rawluk et al. 2006).

Both the IME1- and IME2-dependent early meiotic pathways require the RNA-binding protein Rim4p (Soushko and Mitchell 2000, Deng and Saunders 2001). The early meiotic genes that are Rim4p-dependent are required for the initiation of G1 arrest, pre-meiotic DNA synthesis, and meiotic recombination (van Werven and Amon 2011). At later stages in meiosis, Rim4p blocks the translation of a set of mRNAs by sequestering them into an amyloid-like aggregate until their protein products are required (Berchowitz, Kabachinski et al. 2015, Jin, Zhang et al. 2015). Inhibition of translation is relieved by the phosphorylation of Rim4p by Ime2p (Carpenter, Bell et al. 2018) and subsequent degradation by autophagy (Wang, Zhang et al. 2020).

Although it is known that deletion of KAR4 affects meiosis, the mechanism describing Kar4p’s function in meiosis is not known (Kurihara, Stewart et al. 1996). While STE12 is expressed in diploids (Ellahi, Thurtle et al. 2015), it is not totally required for meiosis because cells lacking Ste12p can still form two spore asci (Marston, Tham et al. 2004). Because Kar4p’s activity in mating is Ste12p-dependent (Kurihara, Stewart et al. 1996, Gammie, Stewart et al. 1999, Lahav, Gammie et al. 2007), it is likely that Kar4p has a separate meiotic function. KAR4 is expressed in G1 cells prior to entry into meiosis, and cells without KAR4 fail to initiate meiosis, suggesting that it acts very early in the pathway (Kurihara, Stewart et al. 1996). Interestingly, KAR4 is induced again 8–12 hours into sporulation (Kurihara, Stewart et al. 1996), suggesting that it may be required at more than one stage. Clues to Kar4p’s function during meiosis can be found in its homology. Kar4p is a member of the MT-A70 family of N⁶-methyladenosine (m⁶A) methyltransferases (Bujnicki, Feder et al. 2002). However, Kar4p is missing several key residues important for S-adenosyl methionine binding and catalysis (Bujnicki, Feder et al. 2002, Wang, Doxtader et al. 2016, Wang, Feng et al. 2016) and is unlikely to have retained enzymatic activity. A second MT-A70 family member in yeast, IME4, is also required for meiosis, although unlike Kar4p, Ime4p has retained enzymatic activity (Clancy, Shambaugh et al. 2002, Agarwala, Blitzblau et al. 2012). Ime4p is part of the MIS (Mum2p, Ime4p, and Slz1p) complex that methylates adenines in mRNAs during meiosis. Mutations in the IME4 catalytic site or loss of members of the MIS complex result in reduced or delayed sporulation (Clancy, Shambaugh et al. 2002, Agarwala, Blitzblau et al. 2012). MIS complex-mediated mRNA methylation promotes expression of IME1 by negatively regulating the expression of the transcriptional repressor, Rme1p. Methylation of RME1 transcripts leads to their degradation, causing reduced Rme1p levels and increased expression of IME1 (Bushkin, Pincus et al. 2019). Recently, m⁶A has been found to be widespread and dynamic during yeast meiosis, occurring in over 1000 transcripts including IME1, IME2, and RIM4 (Bodi, Button et al. 2010, Schwartz, Agarwala et al. 2013, Scutenaire, Plassard et al. 2022, Varier, Sideri et al. 2022), which are enriched on polyribosomes (Bodi, Bottley et al. 2015).

Comparison of RNA methyltransferase homologs from different species showed that yeast KAR4 and IME4 diverged early in eukaryotic evolution, and both homologs are present in higher eukaryotes (Bujnicki, Feder et al. 2002). KAR4 and IME4 are homologous to human METTL14 and METTL3, respectively. Although mRNA methylation is performed in yeast by the MIS complex, RNA methylation in humans is performed by the METTL3/ METTL14 complex together with WTAP, the human homolog of Mum2p (Agarwala, Blitzblau et al. 2012, Liu, Yue et al. 2014). Recently published crystal structures showed S-adenosyl methionine binding to METTL3, with METTL14 playing a structural role in mRNA binding (Wang, Doxtader et al. 2016, Wang, Feng et al. 2016). In yeast, Kar4p and Ime4p have been reported to physically interact, as have Kar4p and Mum2p (Uetz, Giot et al. 2000, Ito, Chiba et al. 2001, Morgan 2006), although the functional significance of the interactions is not known. Mitotic expression of the MIS complex is sufficient to induce mRNA methylation (Agarwala, Blitzblau et al. 2012), suggesting that Kar4p may not be required for methylation. However, it is important to note that Kar4p is also present in mitotic cells, thus making it possible that Kar4p was involved in the methylation they observed during mitosis in those experiments.

To better understand how Kar4p performs its functions in different developmental pathways, we conducted a genetic screen for alleles of KAR4 that differentially affect its functions in mating and meiosis. Remarkably, Kar4p is required for at least two distinct functions in meiosis in addition to its function in mating. Kar4p’s mating function appears to be solely about its ability to interact with Ste12p, but the meiotic functions appear to be more complex. We found that kar4Δ/Δ mutants have serious deficiencies in mRNA m⁶A levels during meiosis and that Kar4p interacts with all members of the MIS complex. Taken together, these data imply that Kar4p is a member of the yeast methyltransferase complex. Interestingly, Kar4p, Ime4p, and Mum2p are required for another function during meiosis that appears to be independent of mRNA methylation.

Results

kar4Δ/Δ Cells Reversibly Arrest Prior to Meiotic DNA Replication

During mating, the nuclei of kar4Δ mutant zygotes fail to fuse, remaining separated due to the lack of induction of the kinesin motor protein Kar3p/Cik1p (Kurihara, Stewart et al. 1996). To determine where kar4Δ/Δ mutants are blocked in meiosis, strains expressing GFP-TUB1 and SPC42-mCherry were used to observe the microtubules and spindle pole bodies by fluorescence microscopy. The progression of both wild-type and kar4Δ/Δ cells was visualized at 12, 24, and 48 hours after induction of meiosis (Fig 1A). Wild type cells progressed through both meiotic divisions, reaching 34% sporulation after 48 hours. In contrast, 100% of kar4Δ/Δ cells remain arrested with a monopolar spindle after 48 hours (Fig 1A).

Fig 1. Kar4p is required early in meiosis.

(A) Fluorescence microscopy of the spindle pole body (Spc42-mCherry) and microtubules (GFP-Tub1) across a time course of meiosis (12, 24, and 48 hours post transfer into sporulation media) in wild type and kar4Δ/Δ. Graphs are the quantification of the number of cells in different meiotic stages (Monopolar Spindle, Meiosis I, Meiosis II, and Spores). Experiments were run in three biological replicates for each strain and at least 100 cells were counted for each replicate. Error bars represent standard deviation. (B) Flow cytometry analysis of DNA content in wild type and kar4Δ/Δ across the same meiotic time course of the microscopy. DNA was stained with propidium iodide.

To further characterize the meiotic block in kar4Δ/Δ cells, we used flow cytometry to measure DNA content at 12, 24, and 48 hours. A significant number of wild-type cells entered S-phase by 12 hours; by 48 hours the cultures comprised a mixture of 2N and 4N cells. In contrast, the kar4Δ/Δ population comprised cells with 2N DNA content throughout the time course. We conclude that the kar4Δ/Δ mutants had not initiated DNA synthesis (Fig 1B), indicating that these cells were arrested very early in the meiotic program. To determine whether the arrest was reversible, the viability of the kar4Δ/Δ cells was assessed. The kar4Δ/Δ diploids remained viable after several days in sporulation media and returned to mitotic growth with near 100% efficiency.

KAR4 Has Different Functions in Mating and Meiosis

To determine if Kar4p has distinct functions in mating and meiosis, we designed a genetic screen for mutant alleles that might differentially affect its function in the two pathways. If Kar4p’s functions in mating and meiosis were the same, we would expect that all alleles would affect both pathways. In contrast, if Kar4p performs distinct roles in the two pathways, we would expect to isolate two classes of mutations that differentially affect function: Mat⁺ Mei⁻ and Mat⁻ Mei⁺. It is also possible that Kar4p has only one function, but that the two pathways require different levels of Kar4p activity; if so, one pathway may be more sensitive to perturbation than the other. We would then expect to isolate two classes of mutant alleles, one affecting both pathways and the other affecting only the pathway with the more stringent requirement. For example, isolation of only Mat⁻ Mei⁻ and Mat⁺ Mei⁻ alleles would suggest that KAR4’s meiotic function is easier to perturb than the mating function.

KAR4 on a centromeric plasmid was mutagenized in vitro, and the plasmid was transformed into a MATa kar4Δ strain (MY 10128) (Table S1 and Table S2). To assess Kar4p’s function during mating, we used a time-limited plate mating assay (Gammie and Rose 2002), which is sensitive to defects in nuclear fusion. The transformants were replica-plated together with a MATα kar4Δ (MY11297) (Table S1) (Tong, Evangelista et al. 2001) tester lawn onto rich media and allowed to mate for three hours before being replica-plated onto synthetic media to select for diploids. Under these conditions, the MATa kar4Δ control strain showed a severe mating defect (Mat⁻) compared to the wild type (Fig 2A, Fig 2B).

Fig 2. Kar4p has distinct functions in mating and meiosis.

(A) Schematic of the screen used to identify separation of function mutants of KAR4. (B) Limited mating assay to determine the ability of the KAR4 alleles to facilitate mating. (Left) Resulting growth on media selective for diploids after a limited mating assay between a strain carrying the indicated KAR4 allele and kar4Δ. (Right) Growth of strains carrying the indicated KAR4 allele mated to kar4Δ on YPD (non-selective rich media) overnight. (C) Spot assay to assess the ability of the different KAR4 alleles to initiate meiotic recombination after 24 hours of exposure to meiosis inducing conditions. (Left) Selection for successful recombination events by selecting for His+ Can^R recombinants. (Right) Growth of all strains on YPD. Spots are 10-fold serial dilutions of 1 OD of each sample.

To assess Kar4p’s function in meiosis, we used a P_MFA1-HIS3 cassette integrated in place of the CAN1 gene in the MATα parent (Tong, Evangelista et al. 2001). The P_MFA1-HIS3 cassette is only expressed in cells that are phenotypically MATa. The MATa and MATα parents, as well as the MATa/α diploid, are His⁻. However, diploids competent for meiosis can produce MATa His⁺ Can^R haploid offspring by inducing meiotic recombination and undergoing sporulation. Alternatively, MATa His⁺ Can^R diploids can be generated if cells that have undergone meiotic recombination are returned to a rich growth medium and allowed to re-enter the mitotic pathway. After overnight incubation of the mating plate to allow multiple rounds of mating, the Mat⁻ mutants formed a sufficient number of diploids for further analysis (Gammie and Rose 2002). The diploids were induced to enter meiosis and then spotted onto media to select for His+ Can^R recombinants. Under these conditions, the KAR4/kar4Δ control strains showed robust growth (Mei⁺), whereas the kar4Δ/kar4Δ diploid showed no growth (Mei⁻). Interestingly, some alleles exhibited an intermediate phenotype (Fig 2B).

A total of 25,000 transformants were screened, and three classes of mutants were identified. As expected, many mutants were defective for both mating and meiosis. As these were likely to be null mutations, they were not studied further. After rescreening and retesting, we identified 36 independent mutants representing 25 unique alleles showing separation of function (5 sites were mutated resulting in the same amino acid substitution 2 or more times). Two alleles comprised double mutants in which one or both sites were recovered as single mutations. After further analysis, 9 unique alleles were competent for mating, but not meiosis (Mat⁺ Mei⁻) and 11 alleles were defective for mating, but apparently competent for meiosis (Mat⁻ Mei⁺). Because we identified all three possible classes of mutant alleles (null mutations and both types of separation of function alleles), we conclude that Kar4p has distinct, separable functions in mating and meiosis (Table 1).

Table 1.

Characterization of the mating/meiotic defects and interactions of the separation of function alleles of KAR4. Mat⁻ alleles in blue and Mei⁻ alleles in red.

Allele	Mating	Ste12 (2-hybrid)	Meiosis	Ime4 (2-hybrid)	Mum2 (Co-IP)	% Sporulation after Ime1 OE*
KAR4	+	+	++++	+	++++	75
kar4Δ	-	-	-	-	-	3
A157P	-	-	++++	++	++++	43
H213Y	-	++	++++	++++	++++	61
F186S	-	-	++++	+++	+	52
E183K	-	-	++++	++++	++++	52
L255F	-	-	++++	++++	++++	50
P273L	-	-	++++	+++	++++	48
W212R	-	++	++++	++	+++	32
C127Y	-	++	++++	++++	++	65
W156G	-	-	+++	-	++	47
P293L	-	-	+++	-	++++	43
G126A	-	-	++	-	++	7
T264P	+	++++	+	++++	+++	0
Q132K	+	+	+	-	++	0
E260K	+	+	-	++	+	0
G287D	+	+	-	-	-	0
E260G	+	++++	-	+++	++	0
A310Stop	+	++++	-	++	++	0
ΔP281V282	+	++++	-	-	-	0
M207R	+	++++	-	++	-	0
A85P	+	++++	-	++++	++	0

^*Sporulation was measured in three biological replicates after IME1 overexpression

^**“++++” ~ 100% WT, “+++” ~ 10% WT, “++” ~ 1% WT, and “+” ~ 0.1% WT

To identify the causal mutations in KAR4, the plasmids were recovered and sequenced. The alleles were interspersed throughout the primary sequence of KAR4, suggesting that Kar4p is not organized into distinct functional domains (Table 1, S Fig 1). To determine if the function-specific alleles map to different surfaces of the protein, we took advantage of the recent advances in protein folding algorithms to generate AlphaFold (Jumper, Evans et al. 2021) predicted structure of Kar4p and mapped the alleles to that structure (Fig 2C). The two classes of alleles clustered into distinct regions of the predicted protein. The Mat− alleles (blue) cluster tightly within one region, whereas the Mei− alleles (red) are found broadly distributed over the opposite surface. The finding that distinct regions of the predicted protein are affected by the two classes of mutations suggests that these regions mediate different functions in mating and meiosis.

Kar4p Engages in a Function Specific Interaction with Ste12p

One possible model for how the mutations differentially impact Kar4p’s functions is that they block the ability of Kar4p to interact with different protein partners. Kar4p’s activity during mating is dependent on Ste12, over-expression of Ste12p suppresses the loss of Kar4p (Kurihara, Stewart et al. 1996), and Kar4p interacts with Ste12p during the pheromone response (Lahav, Gammie et al. 2007). Therefore, it is likely that the kar4 Mat⁻ alleles interfere with mating by disrupting the Kar4p-Ste12p interaction. To test the ability of the KAR4 alleles to interact with Ste12p, we leveraged Ste12p’s function as a transcriptional activator to perform a one-hybrid assay between the Kar4p mutants and Ste12p. We moved the alleles onto a construct of KAR4 fused to the Gal4p DNA-binding domain (KAR4-GBD) previously used to show that Kar4p’s mating activity is Ste12p-dependent (Lahav, Gammie et al. 2007). In the presence of mating pheromone, the Kar4 hybrid protein recruits Ste12p to drive transcription of a HIS3 reporter gene, allowing cells to grow in the absence of histidine. In strains carrying alleles that disrupt the Kar4p-Ste12p interaction, the reporter will not be transcribed. Of the 20 mutants tested, 8 of the 11 Mat− alleles caused reduced interaction with Ste12p. All 9 of the Mat+ alleles allowed transcription of the reporter gene (Fig 3A, Table 1). Interestingly, three alleles were defective for mating but still showed some interaction with Ste12p by this assay. The presence of these alleles suggests that there may be some threshold of Kar4p function that is required for mating or that Kar4p may do more than just stabilize Ste12p at weak promoters during mating. We conclude that most alleles that interfere with mating function do so by disrupting the Kar4p-Ste12p interaction.

Fig 3. Kar4p engages in a function specific interaction with Ste12p and Ime4p.

(A) A one-hybrid assay to determine the ability of the different Kar4p alleles fused to the Gal Binding Domain to interact with Ste12p upon exposure to alpha-factor. (Left) Growth on non-selective media both with and without 3 μM alpha-factor. (Right) Growth on SC-His media to select for alleles that can maintain an interaction with Ste12p and drive expression of the HIS3 reporter gene. (B) A two-hybrid assay was used to determine the ability of the different Kar4p alleles to interact with Ime4p fused to the Gal Activating Domain after 24 hours of exposure to meiosis inducing conditions. (Left) Growth on non-selective media. (Right) Growth on both SC-HIS and SC-ADE to select for the alleles that can maintain an interaction with Ime4p and drive expression of the two reporter genes HIS3 and ADE2. Spots are 10-fold serial dilutions of 1 OD of each sample.

Kar4p’s Interaction with Ime4p is Correlated with the Mei Function

It seemed likely that Kar4p’s role in meiosis might also involve a function specific binding partner. Ime4p, the paralog of Kar4p, was a good candidate for several reasons: (1) the orthologs of Ime4p and Kar4p in most eukaryotes have been shown to form a heterodimer that functions to methylate mRNA, (2) both ime4Δ/Δ and kar4Δ/Δ arrest at similar points in meiosis, and (3) a physical interaction between Kar4p and Ime4p was previously detected by high-throughput yeast two-hybrid assays (Ito, Chiba et al. 2001). To confirm the interaction, we created a fusion of IME4 to the Gal4p activating domain (IME4-GAD) for use in the yeast two-hybrid system with the KAR4-GBD fusion described above. Our findings agree with the previous report that Kar4p and Ime4p physically interact (Fig 3B). We also detected the interaction when the fusions were swapped (KAR4-GAD with IME4-GBD) (data not shown).

To identify residues on Kar4p important for the interaction with Ime4p, we tested the alleles in the two-hybrid assay. All 8 alleles that are proficient for wild type levels of meiotic function were able to drive varying levels of expression of the reporter gene. Of the four alleles that had reduced meiotic function, three were not able to express the reporter gene. Of the 12 alleles that showed either reduced or total loss of meiotic function, 6 were not able to interact with Ime4p and the other 6 had varying levels of expression of the reporter gene (Fig 3B, Table 1). One interpretation of these data is that the alleles causing a defect in meiotic function, but not in the interaction with Ime4p, may nevertheless impact the function of Kar4p in meiosis. We conclude that Kar4p’s role in meiosis is correlated with its interaction with Ime4p, implying that Kar4p may also have a role in mRNA methylation.

Kar4p is Required for Efficient mRNA Methylation During Meiosis

To address whether Kar4p is required for mRNA methylation during meiosis, we measured m⁶A levels in mRNAs from sporulating cells. To increase the efficiency and synchrony of meiosis, we moved the kar4Δ/Δ, ime4Δ/Δ, and slz1Δ/Δ into the SK1 background. Cells were harvested after four hours in sporulation-inducing conditions, which we determined was when m⁶A levels peaked during meiosis in the strains we used for these experiments (data not shown). In agreement with previous findings, we were able to detect m⁶A levels at 25% of wild type in the slz1Δ/Δ strain. (Fig 4A). Levels of m⁶A were only 17% of wild type in kar4Δ/Δ indicating that Kar4p is required for wild type levels of mRNA methylation during meiosis. Interestingly, in kar4Δ/Δ slz1Δ/Δ double mutants we were unable to detect methylated mRNA. Slz1p is important for localization of the MIS complex to the nucleus (Schwartz, Agarwala et al. 2013), but some nuclear localization persisted in slz1Δ/Δ. Given that Kar4p localizes to the nucleus during mating, it is tempting to speculate that nuclear localization of the other members of the MIS complex persisted in the slz1Δ/Δ mutant in meiosis due to Kar4p.

Fig 4. Kar4p is required for mRNA m⁶A methylation.

(A) mRNA m⁶A levels measured using an ELISA like assay from EpiGenTek. The indicated mutations were made in the SK1 strain background and samples were harvested after four hours of exposure to meiosis inducing conditions. Experiments were run in three biological replicates for each strain and error bars represent standard deviation. (B) Western Blot of 3xFLAG-Rme1p in wild type and kar4Δ/Δ across a time course of meiosis. Kar2p is used as a loading control. (C) Western blot of 3xMYC-Ime4p after four hours in meiosis inducing media with either 100 μM cycloheximide or an equivalent amount of DMSO in both wild type and kar4Δ/Δ. (Top) 3xMYC-Ime4p levels with DMSO. (Middle) 3xMYC-Ime4p levels with cycloheximide. (Bottom) Quantification of three biological replicates of the cycloheximide chase experiment. The strains used are in the SK1 background. Kar2p is used as a loading control. “*” indicates a non-specific band.

Meiotic mRNA methylation reduces the levels of the transcriptional repressor Rme1p. In mutants lacking Ime4p (or catalytic mutants of Ime4p), Rme1p levels are increased causing lower levels of IME1 transcription (Bushkin, Pincus et al. 2019). Consistent with a role for Kar4p in mRNA methylation, the levels of Rme1p tagged with three FLAG repeats remained high in kar4Δ/Δ across meiosis, elevated at least 2-fold for the duration of the time course (Fig 4B).

Kar4p may be required for mRNA methylation either because it is required for the activity of the complex, or by stabilizing Ime4p. By using cycloheximide to block new protein synthesis during meiosis, we found that Ime4p does not turn over faster in the absence of Kar4p (Fig 4C). Indeed, Ime4p levels were slightly higher in kar4Δ/Δ compared to wild type, which could be due to the delay in sporulation in those mutants. We conclude that Kar4p is required for the activity of the methyltransferase complex and not for stabilizing Ime4p.

Kar4p Interacts with the MIS Complex Proteins Mum2p and Slz1p

If Kar4p is a member of the methyltransferase complex, it should also interact with the other known components, Mum2p and Slz1p. A high throughput study using the two-hybrid method suggested an interaction between Kar4p and Mum2p (Uetz, Giot et al. 2000). To confirm this, we used co-immunoprecipitation of Kar4p fused to the 3xHA epitope and Mum2p fused to the 4xMYC epitope. We were able to confirm an interaction between Kar4p and Mum2p (Fig 5A). In addition, we used the alleles of KAR4 and co-immunoprecipitations to ask whether Kar4p and Mum2p engage in a function specific interaction during meiosis. Interaction of the Kar4p mutant proteins with Mum2p was highly correlated with the interaction of Kar4p with Ime4p. Only 4 out of the 20 mutant Kar4p proteins tested showed an interaction with Mum2p that was different from that observed with Ime4p (Fig 5A, Table 1). Three mutant proteins (P293L, G126A, and Q132K) interacted with Mum2p and retained some meiotic functionality, but were strongly defective for interaction with Ime4p, by the two-hybrid assay. One possibility is that there is still an interaction between the Kar4p and Ime4p that is facilitated by the interaction with Mum2p. This may explain why these alleles retain some meiotic function. The fourth anomalous mutant protein, M207R, maintained an interaction with Ime4p, but was defective for both meiotic function and interaction with Mum2p. This suggests that interaction with Ime4p alone is not sufficient for facilitating Kar4p’s meiotic functions. Taken together, Kar4p’s meiotic function appears dependent upon interaction with two key members of the MIS complex, Ime4p and Mum2p.

Fig 5. Kar4p interacts with the MIS complex components Mum2p and Slz1p.

(A) Western blots of total protein and co-IPs between Mum2p-4MYC and the alleles of Kar4p-3HA. (Top) Total protein samples from the extracts that were used for the Co-IPs. Kar2p is used as a loading control. Alleles proficient for Kar4p’s meiotic function (Mei⁺) are in blue and alleles in red are not (Mei⁻). (Bottom) Co-IPs where Mum2p-4MYC was purified and the co-purification of Kar4p-3HA was assayed. (B) Western blots of total protein and Co-IPs between Slz1p-3HA and Kar4p-9MYC. (Left) Total protein samples from the extracts that were used for the co-IPs. “*” indicates a non-specific band. (Right) Co-IPs where Slz1p-3HA was purified and the co-purification of Kar4p-13MYC was assayed. “‡” indicates the heavy chain of IgG from the anti-HA magnetic beads used for the Co-IP. (C) Western blots of total protein and Co-IPs between Kar4p-9MYC and Slz1p-3HA. (Left) Total protein samples from the extracts that were used for the Co-IPs. (Right) Co-IPs where Kar4p-9MYC was purified and the co-purification of Slz1–3HA was assayed.

To determine if Kar4p also interacts with Slz1p we generated strains carrying Slz1p fused to the 3xHA epitope and Kar4p fused to the 13xMYC epitope and performed co-immunoprecipitation assays during meiosis. We detected interaction between the two proteins either by precipitating Slz1p or Kar4p (Fig 5B and Fig 5C). We did not examine whether the interaction is affected by the function specific KAR4 alleles. Thus, Kar4p interacts with all members of the MIS complex and is required for mRNA methylation. These findings point to Kar4p being a member of the mRNA methyltransferase complex.

Kar4p’s Meiotic Function Is Partially Suppressed by Overexpression of IME1

To gain insight into Kar4p’s meiotic function, we performed a high-copy suppressor screen. The strain used in the initial mutant screen (MY10128) was transformed with a YEp24 2μ plasmid-based genomic library (Carlson and Botstein 1982) and the transformants were mated to the kar4Δ MATα can1::P_MFA1HIS3 strain used to discover the Mei− alleles. Colonies that could now undergo meiotic recombination were identified after induction of sporulation by growth on appropriate selective media (SC-histidine, plus canavanine). The kar4Δ/Δ recombination defect was suppressed by 2μ plasmids carrying chromosomal fragments containing KAR4 (6x), IME1 (12x), and IME2 (1x). To confirm the identity of the suppressing genes, IME1 and IME2 were subcloned onto 2μ vectors and retested (Fig 6A). Suppression by IME1 was not as strong as KAR4, suggesting that Kar4p may have additional functions other than acting through Ime1p. Suppression by IME2 was much weaker than IME1 (Fig 6A).

Fig 6. IME1 overexpression partially suppresses the kar4Δ/Δ meiotic defect.

(A) Growth of kar4Δ/Δ cells on SC-HIS+canavanine to screen for recombination after exposure to meiosis-inducing conditions carrying either KAR4, IME1, or IME2 cloned onto high-copy number 2μ plasmids. (B) Fluorescence microscopy of the spindle pole body (Spc42-mcherry) and microtubules (GFP-Tub1) across a time course of meiosis (12, 24, and 48 hours post movement into sporulation media) in wild type and kar4Δ/Δ with IME1 overexpressed from an estradiol-inducible promoter. Graphs are the quantification of the number of cells in different meiotic stages (Monopolar Spindle, Meiosis I, Meiosis II, and Spores). Experiments were run in three biological replicates for each strain and at least 100 cells were counted for each replicate. 1 μM of estradiol was used to induce expression. Error bars represent standard deviation. (C) Flow cytometry analysis of DNA content in wild type and kar4Δ/Δ with IME1 overexpressed across the same meiotic time course of the microscopy. DNA was stained with propidium iodide. (D) Western blot showing GFP-Ime1p levels across a time course of meiosis in wild type, kar4Δ/Δ, and kar4Δrme1Δ/ kar4Δrme1Δ. Kar2p is used as a loading control. (E) qPCR measuring IME1 transcript levels in wild type, kar4Δ/Δ, and kar4Δrme1Δ/ kar4Δrme1Δ. Data were normalized to levels of PGK1 expression. (F) Flow cytometry analysis of DNA content in wild type and kar4Δ/Δ carrying either the wild type allele of RME1, the hypomorphic −308A allele of RME1, or a deletion of RME1. DNA was stained with propidium iodide.

To further investigate IME1’s suppression of the kar4Δ/Δ meiotic phenotype, we used a non-native, estradiol-inducible promoter to overexpress IME1 in cells containing GFP-TUB1 and SPC42-mCherry. In this system (McIsaac, Gibney et al. 2014), the promoter is recognized by an artificial transcription factor with four zinc fingers (Z₄EV), effectively eliminating any off-target effects. Addition of estradiol triggered rapid sporulation in wild-type diploid cells (P_Z4EV-IME1) resulting in 57% sporulation after 24 hours and 81% sporulation by 2 days. In contrast, induction of P_Z4EV -IME1 in kar4Δ/Δ cells resulted in no sporulation after 24 hours, only 7% sporulation after 48 hours, and reached a peak of 12% sporulation after 2 days. Most cells remained arrested with a monopolar spindle (Fig 6B), however almost all cells had completed premeiotic S-phase (Fig 6C). After induction of P_Z4EV-IME1, KAR4 cells retained 94% viability. In contrast, the P_Z4EV-IME1-induced kar4Δ/Δ strains retained only 27% viability (S Fig 2).

We conclude that IME1 overexpression allows premeiotic S-phase, consistent with the appearance of meiotic recombinants. However, only a small fraction of cells complete sporulation in the kar4Δ/Δ cells, indicating that there are remaining defects in meiosis that are not suppressed by IME1 overexpression. Given Kar4p’s role in RNA methylation and the role of RNA methylation in meiotic entry, we hypothesize that the requirement for Kar4p in meiotic DNA synthesis and recombination is through the expression of IME1. In support of this hypothesis, Ime1p levels were reduced at least 2-fold in kar4Δ/Δ compared to wild type over the first 12 hours of meiosis (Fig 6D). A much less severe impact was observed for the level of IME1 transcript (Fig 6E). Interestingly, there was no difference in the level of IME1 transcript at 4 hours between wild type and kar4Δ/Δ, although there was a significant defect in the level of Ime1p protein. Recent work confirmed that IME1 transcripts are methylated (Scutenaire, Plassard et al. 2022, Varier, Sideri et al. 2022), suggesting that the loss of methylation in kar4Δ/Δ may impact the translatability of IME1 mRNA. Together, these data suggest that Kar4p is required for the normal induction of IME1 expression.

Kar4p’s Meiotic Function is Partially Suppressed by Loss of Rme1p

Recent work showed that RNA methylation acts upstream of IME1 expression through regulating the levels of Rme1p (Bushkin, Pincus et al. 2019). RNA methylation acts to destabilize RME1 transcript and thereby reduce the levels of Rme1p. Reducing Rme1p levels either by hypomorphic allele (RME1-308A) found in the SK1 strain background or by deleting RME1 permits meiotic DNA replication in ime4Δ/Δ, but does not permit sporulation (Bushkin, Pincus et al. 2019). Given Kar4p’s role in RNA methylation and its impact on Ime1p levels, we sought to determine if loss of Rme1p could also permit meiotic DNA replication in kar4Δ/Δ. Wild type strains harboring either the hypomorphic allele or the deletion underwent meiotic replication faster than strains carrying the wild type RME1 allele (Fig 6F). In kar4Δ/Δ with these mutations, cells also underwent meiotic replication, although not to the same extent that occurs upon overexpression of IME1 (Fig 6F). Like IME1 overexpression, the RME1 mutations did not permit spore formation in kar4Δ/Δ (S Fig 3). A double mutant lacking both Kar4p and Rme1p expressed wild type levels of both IME1 transcript and protein (Fig 6D, Fig 6E). Again, we note that rather small changes in mRNA levels in the double mutant (compared to kar4Δ/Δ alone) resulted in much larger increases in protein level. Possibly IME1 is also regulated at the translational level, by a mechanism other than mRNA methylation, or the double mutant may have increased expression of IME4, and increased levels of mRNA methylation compared to kar4Δ/Δ. These findings, in conjunction with the ability of IME1 overexpression to rescue the early kar4Δ/Δ defect, further support the conclusion that Kar4p acts upstream of IME1 expression. Increasing Ime1p through overexpression of IME1 or by removing a negative regulator can bypass the early requirement for Kar4p and mRNA methylation.

Kar4p has Two Distinct Meiotic Functions

We next tested whether overexpression of IME1 would rescue the separation of function alleles described above. Plasmids bearing the alleles were transformed into a diploid kar4Δ/Δ strain carrying the P_Z4EV-IME1 construct. Transformant strains were incubated in sporulation media, induced with estradiol, and examined for spore formation after 48 hours. As expected, many of the Mei⁻ mutants were unable to form spores, like kar4Δ/Δ. Remarkably, two of the mutants (W156G and P293L) that show reduced meiotic function were able to form spores at wild type levels after overexpression of IME1. However, the three alleles that had even greater reduction in meiotic function (G126A, T264P, and Q132K) were not able to sporulate at wild type levels after IME1 overexpression (Table 1). Given the high degree of correlation between the alleles that are both Mei⁺ and Spo⁺, it is difficult to discern whether the Spo function is a distinct meiotic function of Kar4p or simply requires higher levels of methyltransferase activity.

To elucidate whether the Mei and Spo functions are distinct functions of Kar4p, we examined the ability of IME1 overexpression to suppress the meiotic defects caused by deletions of other members of the methyltransferase complex. If the meiotic defects can be rescued by IME1 overexpression alone, this would suggest that the Spo function may be a distinct meiotic function. In our strains, deletions of KAR4, IME4, and MUM2 completely fail to sporulate; the slz1Δ/Δ had low levels of sporulation (S Fig 4). Overexpression of IME1 did not rescue sporulation in ime4Δ/Δ or mum2Δ/Δ mutants (Fig 7, S Fig 4). However, the slz1Δ/Δ mutant was suppressed by overexpressing IME1. A catalytic mutant of Ime4p (ime4-D348A, W351A referred to as ime4-cat from this point forward) lacks Ime4p methyltransferase activity and has been shown to block meiosis (Clancy, Shambaugh et al. 2002). Interestingly, the catalytic mutant was rescued by overexpression of IME1 alone, suggesting that the lack of methylation is largely manifest as a defect in Ime1p function. This further suggests that Kar4p’s Spo function does not involve RNA methylation and represents a distinct meiotic function of Kar4p (Fig 7, S Fig 4). By extension, this finding implies that Slz1p’s function in meiosis is only through its role in mRNA methylation.

Fig 7. Co-overexpression of IME1 and RIM4 permits sporulation in kar4Δ/Δ.

Sporulation of wild type, kar4Δ/Δ, mum2Δ/Δ, ime4Δ/Δ, slz1Δ/Δ, and ime4-cat/ime4Δ with either endogenous expression of IME1 and RIM4, overexpression of RIM4, overexpression of IME1, or overexpression of both IME1 and RIM4. 1 μM of estradiol was used to induce expression. All dyads, triads, and tetrads were counted across three biological replicates for each strain. At least 100 cells were counted after 48 hours post addition of estradiol. Error bars represent standard error.

Taken together, these observations indicate that Kar4p is required at more than one step during meiosis. An early step that is upstream of meiotic DNA synthesis and recombination (Kar4p’s Mei function) appears to be associated with its role in RNA methylation. A later step acts either during or downstream of recombination but upstream of commitment to meiosis, given the ability of kar4Δ/Δ mutants to resume mitotic growth after IME1 overexpression. The later step (Kar4p’s Spo function) appears to be a distinct function of Kar4p and other members of the methyltransferase complex, other than Slz1p, that does not involve RNA methylation. Given the role of Slz1p in localizing the complex to the nucleus, the non-catalytic function may be specific to the cytoplasm.

Overexpression of RIM4 with IME1 leads to spore formation in kar4Δ/Δ

To better understand Kar4p’s “Spo” function, we designed an unbiased high-copy suppressor screen to look for genes that could rescue the sporulation defect of P_Z4EV-IME1 induced kar4Δ/Δ diploids. We transformed the same 2μ library used previously into the P_Z4EV-IME1 kar4Δ/Δ diploid strain. Transformants were washed off plates and induced to sporulate by addition of β-estradiol. After 3 days, the cultures were digested with zymolyase, which kills diploids that do not complete meiosis, but spores are resistant to this treatment. Survivors were plated on media selecting for the suppressing plasmid. Plasmids were extracted from haploid colonies and their inserts identified by DNA sequencing. From 13 independent transformant colonies, we recovered 5 different KAR4-containing genomic fragments. From 3 other independent transformants, we recovered the same 7.5 kb fragment from chromosome VIII. Sequence analysis determined that it contained the sequences for RIM4, SNF6, YHL026C, and a portion of the RIM101 gene. Subcloning of the genomic fragments onto new 2μ plasmids identified RIM4 as the suppressor gene. For all further experiments, we put RIM4 and IME1 under the control of an estradiol-inducible promoter (P_Z3EV). In confirmation P_Z3EV-IME1, P_Z3EV-RIM4 induced kar4Δ/Δ strains were able to form spores, while the uninduced strains did not (Fig 7, S Fig 4). Overexpression of RIM4 alone was able to permit sporulation in slz1Δ/Δ, but not any of the other mutants (Fig 7, S Fig 4).

Given the ability of IME1 and RIM4 co-overexpression to rescue the kar4Δ/Δ defect, we asked if it could also rescue the ime4Δ/Δ and mum2Δ/Δ meiotic defects. For both mutants, the co-overexpression of IME1 and RIM4 was able to permit sporulation at levels comparable to those seen in kar4Δ/Δ (Fig 7, S Fig 4). Taken together, these data suggest that Kar4p, Ime4p, and Mum2p may all participate in a second RNA methylation-independent meiotic function. The requirement for RIM4 overexpression suggests that this function may involve translational regulation of meiotic transcripts.

Discussion

We took advantage of the power of yeast genetics to understand how KAR4 is integrated into multiple differentiation pathways in yeast. The genetic screen for separation-of-function mutants identified multiple alleles that specifically disrupt either Kar4p’s mating or meiotic function. These alleles are interspersed on Kar4p’s primary sequence, suggesting that it does not have separate functional domains. However, the three-dimensional model threaded using the AlphaFold predicted structure of Kar4p (Jumper, Evans et al. 2021) shows that these alleles cluster in separate regions of the protein.

The mechanism of Kar4p’s mating function has been relatively well-characterized. It is known to interact with Ste12p to activate a subset of pheromone-induced genes, specifically those with “weak” PREs, which tend to be expressed later in mating and at higher pheromone concentration. Although Kar4p does not bind DNA directly, electrophoretic-mobility shift assays suggest that it facilitates the binding of Ste12p to Kar4p-dependent promoters. Evidence for direct interaction comes from the ability of Kar4p to recruit Ste12p and activate transcription in a one-hybrid transcription assay (Lahav, Gammie et al. 2007). All Mat⁺ mutants interact with Ste12p by this assay, and all but three of the Mat− mutants were fully defective for interaction. Thus, we conclude that the Mat⁻ mutations are specifically defective for mating because they fail to interact with Ste12p and that Mat⁻ alleles likely define a region on Kar4p that is required for binding Ste12p. Furthermore, these data in combination with the fact that loss of Ste12p does not fully block meiosis indicate that the Kar4p-Ste12p interaction is not required for meiosis.

In meiosis, Kar4p may have two different functions. It is first required prior to pre-meiotic S-phase and meiotic recombination. Kar4p is required for efficient mRNA methylation during meiosis and interacts with all members of the yeast methyltransferase complex, indicating that it is a functional part of the MIS complex. This finding indicates that the yeast mRNA methyltransferase complex is more like that of other eukaryotes than previously thought. These findings are supported by concurrent work from Ensinck et al. (2023) that also showed that Kar4p is part of the methyltransferase complex and is required for mRNA methylation (manuscript in preparation). Previous work as well as our data show mRNA methylation via the MIS complex functions to permit entry into meiosis by regulating the transcript levels of RME1, encoding a key transcriptional repressor of IME1 (Bushkin, Pincus et al. 2019). In kar4Δ/Δ, as in cells lacking Ime4p or expressing a catalytically dead mutant of Ime4p, we see elevated levels of Rme1p. Overexpression of IME1 suppresses the kar4Δ/Δ Mei⁻ defect but does not rescue the Spo⁻ defect. IME1 overexpression does however permit sporulation in both ime4-cat/Δ and slz1Δ/Δ. The ability of IME1 overexpression to rescue the catalytically dead version of Ime4p and the Mei⁻ defect indicates that Kar4p’s Mei function involves mRNA methylation. The additional finding that IME1 overexpression can suppress the defect associated with the loss of nuclear localization of the complex mediated by Slz1p indicates that Kar4p’s Mei function occurs in the nucleus. Together, these findings support a model in which Kar4p acts upstream of Ime1p and is necessary for efficient activation of IME1 expression through nuclear mRNA methylation. In accordance with this model, we see reduced levels of Ime1p in kar4Δ/Δ during meiosis, and reduced levels of Rme1p suppresses the early meiotic DNA replication defect.

When the Kar4p Mei⁻ function is bypassed by IME1 overexpression, cells progress through DNA replication and recombination, but become arrested prior to the meiotic divisions and spore formation suggesting that Kar4p has additional functions in meiosis. Work on Ime4p also suggested that the protein has additional non-catalytic functions in meiosis based on the finding that cells lacking both Rme1p and Ime4p progress through pre-meiotic S-phase, but not meiotic divisions. However, cells lacking Rme1p and expressing a catalytically dead mutant of Ime4p did progress through meiosis and form spores (Bushkin, Pincus et al. 2019). In mammalian systems, METTL3 (Ime4p) has also been shown to have non-catalytic functions that act to enhance the translation of bound transcripts (Lin, Choe et al. 2016, Wei, Huo et al. 2022). Most likely, the complex in yeast is also engaging in these additional functions.

RIM4 overexpression can rescue the kar4Δ/Δ Spo⁻ defect. Given that slz1Δ/Δ can be fully suppressed by Ime1p overexpression alone, it is surprising that RIM4 overexpression alone is sufficient to enhance sporulation in wild type cells and suppress slz1Δ/Δ. One interpretation is that suppression entails Rim4p’s early activating function rather than its later function in blocking translation. We speculate that Kar4p might work with Rim4p or other proteins to activate or support the translation of Ime2p or other regulators of genes required for the meiotic divisions and spore formation. For example, the “middle” meiotic transcription factor, Ndt80p, which regulates genes required for spore formation, is repressed by Sum1p and derepressed by Ime2p. Accordingly, the defect in spore formation may be due to mis-regulation of NDT80 or the mis-regulation of mRNAs within NDT80’s regulon. This model is supported by findings presented in a companion manuscript that show reduced Ime2p levels and NDT80 expression in kar4Δ/Δ with IME1 overexpressed. The finding that the meiotic defects of both ime4Δ/Δ and mum2Δ/Δ are also suppressed by co-overexpressing both IME1 and RIM4 suggests that Ime4p and Mum2p may also be involved in this other function.

In summary, the genetic analysis described herein implicates KAR4 as a major regulator of S. cerevisiae meiosis. It is required both early, before DNA replication and later, to facilitate entry into the meiotic divisions. These two meiotic functions are partially separable from each other, and from Kar4p’s previously characterized role in mating. This work also establishes that the yeast mRNA methylation machinery is more like that of other eukaryotes than previously thought. In a concurrent paper, we explore the impact of losing this member of the mRNA methyltransferase machinery on the meiotic transcriptome and present further evidence that the later function appears to involve mRNA methylation-independent post-transcriptional regulation during meiosis.

Materials and Methods

Media

In general, strains were grown in 2% glucose YPD media unless otherwise noted. Synthetic complete (SC) media lacking the necessary nutrients for selection were made with yeast nitrogen base supplemented with the necessary nutrients except for the nutrient synthesized by the selectable marker. Pre-sporulation media was yeast nitrogen base (1% w/v), peptone (2% w/v), and potassium acetate (1% w/v). Sporulation media was either 1% (w/v) potassium acetate supplemented with histidine, uracil, and leucine or complete sporulation media which consists of yeast extract (0.25% w/v), potassium acetate (1.5% w/v), and 0.1% glucose.

Hydroxylamine mutagenesis and genetic screen for separation-of-function alleles of KAR4

A centromere-based plasmid carrying URA3 and KAR4 with an internal 3xHA tag under the control of its native promoter (pMR2654) was mutagenized with hydroxylamine (see Table S2). The plasmid (10 μg) was incubated with 500 μl of 7% hydroxylamine solution for 20 hours at 37° (Rose, Winston et al. 1990). The reaction was stopped by adding 10 μl 5M NaCl and 50 μl 1mg/ml BSA, and the DNA was ethanol-precipitated and resuspended in TE. The mutagenized DNA was transformed into the yeast strain MY 10128 (see Table S1). Individual transformants were patched onto SC-URA media, 50 to a plate, with a positive and negative control (MY 11596 and 11597). The patches were replica-printed onto YEPD with a lawn of MATα MY11297 and allowed to mate at 30° C for 3 hours (limited mating) before replica-printing on SC-LEU-LYS media to detect mating defects. The mating plates were returned to the 30° C incubator overnight, then replica printed onto SC-LEU-LYS-URA media the next day to select for diploids from all patches (overnight mating). Diploids from the overnight mating were replica printed onto Complete Sporulation media and incubated at room temperature for one day, then at 30° for three days. The sporulation plates were finally replica printed on SC-HIS-ARG media with added canavanine to detect successful entry into meiosis. The limited mating plates were compared to the meiosis plates to detect potential separation of function mutants. Potential hits were repatched (nine to a plate) from the original transformation plate and rescreened. When transformants displayed a consistent phenotype, the plasmid was extracted using a yeast mini-prep kit (Zymo Research, Irvine, CA) and electroporated into bacteria. Bacterial colonies were grown up and the plasmids were extracted using a kit (Qiagen, Germantown, MD). The samples were sent for sequencing with KAR4-specific primers. After sequencing, the plasmids were retransformed into MY 10128 and assayed a final time for both mating and meiotic defects. To further address the severity of the meiotic defect, cells were sporulated as described above and spotted in 10-fold serial dilutions on the SC-HIS-ARG media. Alleles that supported wild type levels (formed colonies at the highest dilution in the series) of recombination were scored as “++++”. Alleles that only formed colonies on the 10⁻² serial dilution were scored as “+++” and so on.

Kar4p Structure

AlphaFold was used to generate the predicted structure of Kar4p (Jumper, Evans et al. 2021). Molecular graphics and analyses were performed with the UCSF Chimera package (http://www.rbvi.ucsf.edu/chimera)/, developed by the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco (supported by NIGMS P41-GM103311) (Pettersen, Goddard et al. 2004).

Site-directed Mutagenesis of KAR4-GBD Plasmid

To move the alleles from the screen onto a KAR4-GBD fusion for use in yeast one-hybrid assays, we employed PCR mutagenesis. A pair of mutagenic primers was designed for each allele, and they were used to amplify the entire pMR4997 plasmid. Immediately after the PCR was finished, DpnI restriction enzyme was added to the PCR reaction tube, and the mixture was incubated at 37° for an additional 2.5 hrs. to digest the methylated template DNA. The remaining mutagenized vectors were electroporated into E. coli, and several candidate colonies from each plate were grown up, miniprepped, and sequenced to find the desired clones.

Yeast One-Hybrid and Two-Hybrid

To assess whether the Kar4p alleles retained the ability to interact with Ste12p, a yeast one-hybrid assay was performed essentially as previously described (Lahav, Gammie et al. 2007). Briefly, the plasmids were transformed into the yeast two-hybrid reporter strain PJ69-4A, and dilutions of logarithmically growing cultures were normalized and spotted in 10-fold serial dilutions onto SC-TRP and SC-HIS plates with either 3 μM α-factor or an equivalent volume of methanol. The plates were incubated at 30° for several days, and interaction was assessed as ability to drive transcription of the selective marker (HIS3).

A two-hybrid assay was used to both verify the interaction between Kar4p and Ime4p as well as determine whether any of the Kar4p alleles impacted this interaction. IME4 was PCR-amplified from chromosomal DNA using the primers “IME4 PGAD fwd” (5’ AAAGAGATCGAATTCCCGGGGATCCATATGATTAACGATAAACTAGT 3’) and “IME4 PGAD rev” (5’ TACTACGATTCATAGATCTCTGCAGGTTTACTGAGCAAAATATAGGT 3’). The P_GAD C2 vector (pMR3659) was linearized with BamHI/PstI double restriction digest and transformed into yeast with the PCR product. Candidates growing on SC-LEU media had their plasmids extracted, sequence-confirmed (Genewiz, South Plainfield, NJ), and transformed into the PJ69-4A yeast two-hybrid reporter strain. This strain was mated to the yeast-two hybrid strain of the opposite mating type carrying the KAR4-GBD or KAR4^SOFA-GBD constructs. Matings were selected on SC-LEU, -TRP. Once diploids were isolated, they were induced to sporulate as described above for 24 hours before being plated in 10-fold serial dilutions on both SC-LEU, -TRP (growth), SC-HIS (selection), and SC-ADE (selection). A diploid containing the IME4-GAD and only an empty GBD was used as a negative control.

For both the one- and two-hybrid assays, alleles that facilitated growth on the selective media at every dilution were scored as “++++”. Every loss of a plus indicates that an allele only drove expression to support growth at 10-fold lower dilution. For example, an allele that could drive expression such that colonies formed in the 10⁻³ serial dilution would be scored as “++++” and an allele that could drive expression such that colonies formed in the 10⁻² serial dilution would be scored as “+++”.

Sporulation

Liquid cultures were sporulated as previously described (Elrod, Chen et al. 2009). Cultures were grown overnight, back diluted into YPA, grown for 16–18 hours, washed with water, resuspended at an OD of 0.5 in 1% potassium acetate supplemented with the amino acids necessary for each strain, and incubated at 26°C for varying amounts of time depending on the experiment being conducted. For sporulation assays involving the overexpression of IME1 and RIM4, the only difference was that these cultures were induced with 1 μM of β-estradiol or an equivalent volume of 100% ethanol upon addition of the cells to the 1% potassium acetate media. Experiments using strains from the SK1 background were sporulated in much the same way with the only differences being cells were resuspended in 1% potassium acetate at an OD of 2.0 and were sporulated at 30°C.

mRNA Methylation

SK1 cultures were sporulated for four hours as described above. Cells were lysed using the FastPrep (MP) with four rounds of bead beating at 4.0 m/s for 40 seconds with one minute on ice in between each round. Total RNA was then harvested using the Qiagen RNeasy kit. mRNA was purified from these total RNA samples using Oligod(T) magnetic beads (Thermofisher). mRNA methylation was measured using the EpiQuik fluorometric m⁶A RNA methylation quantification kit (EpiGenTek). Briefly, purified mRNA was bound to wells and then washed before the addition of an antibody to m⁶A. Wells were then washed again before the addition of a detection antibody followed by an enhancer/developer solution. The fluorescence was measured using a plate reader at 530_EX/590_EM nm and m⁶A levels were quantified for each sample using a standard curve generated from a positive control provided with the kit. Measurements from an ime4Δ/Δ strain were used to subtract background from the data collected from other mutants.

Protein Extraction

Proteins were extracted using alkaline lysis followed by TCA precipitation. Briefly, samples were incubated with 150 μl of 1.85 M NaOH on ice for 10 minutes. An equal volume of 50% TCA was then added to the samples and they were left to incubate at 4°C for 10 minutes. Samples were then washed with acetone before being resuspended in 100 μl of 2x sample buffer. Samples were then boiled for five minutes and then kept on ice for five minutes before being run on an SDS-PAGE gel or stored for later use at −80°C.

Western Blotting

Protein samples were resolved on an appropriate percentage SDS-PAGE gel before being transferred to a PVDF membrane using a semi-dry transfer apparatus (TransBlot SD BioRad) at 16 volts for 36 minutes. After transfer, membranes were blocked for 30 minutes using 10% milk in TBS before being incubated with primary antibody (anti-HA (12CA5) 1:1,000, anti-HA (rabbit: Cell Signaling C29F4) 1:2,500, anti-MYC (rabbit: Novus NB600-336SS) 1:1,000, anti-FLAG (Sigma M2) 1:1000, anti-GFP (BD Living Colors 632377) 1:1000) for one hour at room temperature. Membranes were then washed three times for ten minutes each with 0.1% TBS-Tween20 (TBST) before being incubated with a secondary antibody in 1% milk in TBST for 30 minutes at room temperature. Incubation with secondary antibody (Donkey anti-mouse IgG (Jackson ImmunoResearch) 1:10,000 or Donkey anti-rabbit IgG (ImmunoResearch) 1:20,000) was then followed by another three washes for ten minutes each with TBST before being incubated with Immobilon Western HRP substrate (Millipore) for 5 minutes and imaged using the G-Box from SynGene. Densitometry was conducted using ImageJ.

Co-Immunoprecipitation

Cells were sporulated for four hours as described above and a volume of cells equivalent to 50 OD units was harvested for each strain. Cells were resuspended in a lysis buffer (1% Triton-X 100, 0.2 M Tris-HCl pH 7.4, 0.3 M NaCl, 20% glycerol, 0.002 M EDTA) supplemented with 10x protease inhibitor (Pierce) and lysed using the FastPrep (MP) as described above. Protein extracts were then incubated with 15 μL of anti-MYC or HA magnetic beads (Pierce) for one hour at room temperature. Beads were washed three times with lysis buffer before being resuspended in 2x sample buffer and boiled for 5 minutes to elute proteins off the beads. Samples were then resolved on 8% SDS-PAGE gels and western blotting was conducted as described above.

Microscopy

For microscopy, 100 μl of a sporulated culture expressing GFP-TUB1 integrated at TUB1 (Straight, Marshall et al. 1997) and SPC42-mCherry was spun down and resuspended in 10 μl of 1% potassium acetate media. Cells were imaged on a DeltaVision deconvolution microscope (Applied Precision, Issaquah, WA) based on a Nikon TE200 (Melville, NY) using a 100x/numerical aperture 1.4 objective, a 50W mercury lamp, and a Photometrics Cool Snap HQ charge-coupled device camera (Photometrics, Tucson, AZ). All images were deconvolved using Applied Precision SoftWoRx imaging software. At least one hundred cells were counted in all cases, and the percentage of cells in each meiotic stage including mature spores were determined using the morphology of the spindle and number of spindle pole bodies.

Flow Cytometry

Cells were induced to sporulate as described above. At each time point measured, ~5×10⁶ cells were pelleted and washed in dH₂O. Cells were fixed in 70% ethanol and stored at −20 °C for one hour up to several days. After fixation, cells were washed in 500 μl 50mM sodium citrate (pH 7.2) and incubated in 0.5 ml sodium citrate containing 0.25 mg/ml RNaseA for two hours at 37°C. After RNaseA treatment, 500 μl of 8 μg/mL propidium iodide in sodium citrate buffer was added to each sample and incubated overnight in the dark at 4 °C. The cells were then sonicated at 50% duty cycle, output setting 3, for 4×12 pulses on ice before being transferred to Falcon 2054 tubes. FACS was conducted at the Flow Cytometry and Cell Sorting Shared Resource core at Georgetown University and data was analyzed using FCS Express.

High-copy suppression assays and screen

To find suppressors of KAR4’s “Mei” function, MY10128 was transformed with a YEp24 2μ plasmid library (Carlson and Botstein 1982) and the colonies were mated overnight against a lawn of MY11297. Diploids from this cross were selected with SC-LEU-LYS-URA media and then replica-printed onto complete sporulation media (described above) and incubated at room temperature for one day, then at 30°C for three days. The sporulation plates were finally replica-printed on SC-His-Arg media with added canavanine to detect successful recombination/suppression. When colonies grew after meiosis, plasmids were extracted from them and sent for sequencing. IME1 and IME2 were subcloned into pRS426 to confirm that they were the causal genes on the isolated fragments.

For the screen to find genes that rescued the sporulation defect of Z₄EV-IME1 induced kar4Δ diploids, the same 2μ library was transformed into the Z₄EV-IME1 kar4Δ diploid strain. Transformants were washed off plates and induced to sporulate by addition of 1 μM estradiol. After 3 days at 30° C, the cultures were digested with zymolyase to kill diploid cells, and dilutions were plated on SC –URA media to retain the plasmids in any surviving cells or spores. Candidate colonies were replica-plated onto a sst2Δ lawn. Cells that formed halos on the sst2Δ lawn were further tested for papillation on media containing canavanine to eliminate any diploids that survived zymolyase digestion. Plasmids were extracted from haploid cells and sent for sequencing, and any plasmids containing fragments other than KAR4 were retransformed and reinduced to confirm that they led to spore formation. RIM4 was then subcloned into pRS426.

P_Z4EV-IME1 and P_Z3EV-RIM4 Construction

The β-estradiol-inducible promoter (P_Z4EV) tagged with KanMX was PCR-amplified from plasmid pRSM29 (gift of R. Scott McIsaac) using IME1-specific promoters. The PCR product was transformed into MS3215, and transformants were selected based on G418 resistance and confirmed by colony PCR. A successful integrant was mated against strain DBY12416 (gift of the Botstein laboratory), which contains the gene for the Z₄EV artificial transcription factor under the ACT1 promoter. Spores from this cross were dissected to yield a haploid strain that contained kar4Δ, P_Z4EV-IME1, ura3Δ and Z₄EV. The P_Z3EV-RIM4 strain was made basically the same way except the promoter was amplified from vector RB3483 (gift of Patrick Gibney) and made in a haploid strain containing the appropriate Z₃EV artificial transcription factor.

qPCR

Approximately 10 OD units of cells were harvested at the indicated time points. RNA was extracted from cells using the Qiagen RNeasy kit. Briefly, cells were lysed, and the supernatant was cleared and mixed with an equal volume of 70% ethanol. The mixture was then loaded on the column provided with the kit and the wash steps were followed as detailed in the kit’s protocol. The on-column DNase treatment (Qiagen) was used to remove any DNA from the RNA sample. Total RNA was eluted in 100 μl of RNase free water. cDNA libraries were constructed using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems) with 10 μl of the total RNA sample. The concentration of the resulting cDNA was measured using a nanodrop. qPCR reactions were set up using Power SYBR Green PCR Master Mix (Applied Biosystems) with a volume of the total RNA sample equivalent to 50 ng. The reactions were run on a CFX96 Real-Time System (BioRad) with reaction settings exactly as described in the master mix instructions with the only change being the addition of a melt curve at the end of the program. Results were analyzed using CFX Maestro. Primer sequences are as follows: PGK1 Forward 5’-CTCACTCTTCTATGGTCGCTTTC-3’, PGK1 Reverse 5’-AATGGTCTGGTTGGGTTCTC-3’, IME1 Forward 5’-ATGGCAACTGGTCCTGAAAG-3’, and IME1 Reverse 5’-GGAACGTAGATGCGGATTCAT-3’.