The wtf meiotic driver gene family has unexpectedly persisted for over 100 million years

Abstract

Meiotic drivers are selfish elements that bias their own transmission into more than half of the viable progeny produced by a driver+/driver- heterozygote. Meiotic drivers are thought to exist for relatively short evolutionary timespans because a driver gene or gene family is often found in a single species or in a group of very closely related species. Additionally, drivers are generally considered doomed to extinction when they spread to fixation or when suppressors arise. In this study, we examine the evolutionary history of the wtf meiotic drivers first discovered in the fission yeast Schizosaccharomyces pombe. We identify homologous genes in three other fission yeast species, S. octosporus, S. osmophilus, and S. cryophilus, which are estimated to have diverged over 100 million years ago from the S. pombe lineage. Synteny evidence supports that wtf genes were present in the common ancestor of these four species. Moreover, the ancestral genes were likely drivers as wtf genes in S. octosporus cause meiotic drive. Our findings indicate that meiotic drive systems can be maintained for long evolutionary timespans.

Keywords: S. octosporus; S. pombe; evolutionary biology; genetics; genomics; meiosis; meiotic drive; selfish genes; spore killers; wtf.

Figure 1.. wtf homologs are found outside of Schizosaccharomyces pombe.

(A) Model for meiotic drive of wtf genes in S. pombe, modified from Nuckolls et al., 2017. All spores are exposed to the poison protein, but those that inherit the wtf driver are rescued by the antidote protein. (B) Schematic phylogeny of Schizosaccharomyces species based on published reports (Brysch-Herzberg et al., 2019; Rhind et al., 2011). Our analyses of percent identity between orthologs (Supplementary file 1a) agree with this tree topology. MYA represents million years ago. Annotations of all the identified S. osmophilus genes can be found in Figure 1—source data 2. To the right of the phylogeny, we list the numbers of wtf homologs found by position-specific iterated-basic local alignment search tool (PSI-BLAST) and BLASTn searches. *The S. osmophilus genome is not fully assembled, so the number represents the wtf homologs found within the assembled contigs.

Figure 1—figure supplement 1.. Maps of the wtf gene family members in Schizosaccharomyces octosporus, S. osmophilus, S. cryophilus, and S. pombe.

Genome maps of wtf genes from (A) S. octosporus, (B) S. osmophilus, (C) S. cryophilus, and (D) S. pombe. Genes on the forward strand are shown above each chromosome, whereas genes on the reverse strand are shown below chromosomes. Genes that contain an alternate translational start site near the beginning of exon 2 are shown in purple. Such an alternate translational start site is used to encode Wtf^poison proteins in S. pombe wtf drivers. Genes that lack the potential alternate start site are shown in green. Genes lacking the alternate translational start site in S. pombe can encode Wtf^antidote proteins and act as suppressors of drive. Predicted pseudogenes are indicated with an asterisk (*). The four S. pombe wtf genes with unknown functions are shown in light blue. The S. pombe map is modified from Eickbush et al., 2019. Annotations of the novel wtf genes can be found in Supplementary file 1b-1d.

Figure 2.. Schizosaccharomyces pombe wtf genes share features with other wtf genes outside of S. pombe.

(A) Schematic wtf loci of the Schizosaccharomyces species. Orange boxes correspond to exons (E1 indicates exon 1, E2 indicates exon 2, etc.), the red boxes represent 5S rDNA genes, the blue box represents a pseudogenized wag gene, and the yellow box is a long terminal repeat (LTR) from a Tf transposon. The predicted translational start sites for the antidote (ATG in exon 1) and poison (ATG in exon 2) proteins are indicated, as is the FLEX transcriptional regulatory motif (Supplementary file 1b-d). (B) Long-read RNA sequencing of mRNAs from meiotic S. octosporus cells revealed two main transcript isoforms of the wtf25 gene, presumably encoding an antidote and a poison protein, respectively. cDNA reads obtained using the Oxford Nanopore Technologies (ONT) platform are shown in pink. Blue lines indicate sequences missing in the reads due to splicing. The diagram at the top depicts the two main transcript isoforms. The 3' transcript ends shown in the diagram correspond to the major transcript end revealed by cDNA reads.

Figure 2—figure supplement 1.. Limited conservation of Wtf proteins.

The percent identity shared among all 113 Wtf predicted antidote proteins from Schizosaccharomyces octosporus, S. osmophilus, S. cryophilus, and S. pombe (isolate FY29033) aligned with MAFFT (L-INS-I; BLOSSUM62 scoring matrix/k=2; Gap open penalty of 2; offset of 0.123; Katoh et al., 2002; Katoh and Standley, 2013). Note that the antidote protein sequences are the same as the poison protein sequences except the antidotes contain an additional ~35–45 amino acids at the N terminus encoded by exon 1, which is not found in the poison proteins. The plot shows highest conservation at the N terminal sequences, which are found only in the antidote proteins. Pseudogenes were excluded from the analyses. The data are shown in 10 amino acid sliding windows. The alignment can be found in Figure 2—figure supplement 1—source data 1.

Figure 2—figure supplement 2.. Many wtf genes in Schizosaccharomyces octosporus harbor the FLEX motif in intron 1.

(A) The FLEX motif identified by the de novo motif discovery tool multiple em for motif elicitation (MEME). A total of 49 Mei4 target genes in S. pombe and their orthologs in S. octosporus, S. osmophilus, and S. cryophilus were used as input for MEME. MEME analyses were conducted for each species separately and for all species combined. (B) wtf genes containing the FLEX motif in intron 1. The motif scanning tool find individual motif occurrence (FIMO) was used to find the FLEX motif in the genomes of S. pombe, S. octosporus, S. cryophilus, and S. osmophilus. The 11-bp FLEX motif identified by the MEME analysis using 146 genes as input was provided to FIMO for motif scanning. All wtf genes containing a FIMO hit in intron 1 are shown with the p-value of the FIMO hit in intron 1 presented on a –log10 scale. We found the default p-value cutoff of FIMO (1E-4) being too permissive and applied a cutoff of 3E-6 to distinguish confident hits from unreliable hits.

Figure 2—figure supplement 3.. Transcription levels of predicted poison and antidote isoforms of intact wtf genes in Schizosaccharomyces octosporus.

Long-read (Oxford Nanopore) RNA sequencing was performed on mRNAs isolated from S. octosporus cells undergoing meiosis. All intact wtf genes are shown with the read count of the long transcript (encoding putative antidote) in gray and the read count of the short transcript (encoding putative poison) in black. The bold gene names indicate the genes with a confident FLEX motif hit in intron 1. The underlined gene names indicate the genes analyzed by deletion (Figure 8). The read counts of the two isoforms can be found in Supplementary file 1b.

Figure 3.. Genomic context of wtf genes.

The wtf genes are found in a limited number of genomic contexts. The wtf genes are represented as orange boxes, wag genes are in blue, and long terminal repeats (LTRs) are in yellow. NA indicates not applicable as wag genes are absent from Schizosaccharomyces pombe, and LTRs are absent from S. octosporus.

Figure 3—figure supplement 1.. Distance between 5S rDNA and wtf genes.

The distance in base pairs between 5S rDNA and the coding sequence of a wtf gene in (A) Schizosaccharomyces osmophilus and (B) S. octosporus. Only wtf genes with a flanking 5S rDNA were considered. The wtf gene is collapsed at 0, and the flanking sequences were considered in 100 base pair bins.

Figure 3—figure supplement 2.. Homology between distinct 5S rDNA-wtf and wag-wtf units.

The regions containing wtf genes with the indicated genomic contexts were aligned with MAFFT to find the percent sequences identity (Katoh et al., 2002; Katoh and Standley, 2013). The percent identity is shown in 50 base pair sliding windows. (A) The percent identity shared among 37 wtf-5S rDNA units from Schizosaccharomyces octosporus. (B) The percent identity shared among 17 wtf-wag units from S. octosporus.

Figure 3—figure supplement 3.. Schizosaccharomyces octosporus wtf gene units supported by maximum likelihood phylogeny.

The regions flanking the wtf genes in S. octosporus were sorted into the color-coded groups shown based on maximum phylogenies shown in Figure 3—figure supplement 4 and Figure 3—figure supplement 5. Orange boxes correspond to wtf genes, the red boxes represent 5S rDNA genes, and the blue boxes represent wag genes. Genomic contexts without wag genes and with wag genes are shown separately in (A) and (B).

Figure 3—figure supplement 4.. Maximum likelihood phylogeny of the regions between Schizosaccharomyces octosporus wtf genes and a downstream flanking 5S rDNA gene.

The regions downstream of 67 S. octosporus wtf genes with a downstream 5S rDNA gene were aligned with MAFFT (Katoh et al., 2002), and a maximum likelihood phylogeny was built with PhyML (Guindon et al., 2010). Branch support values shown at the nodes (0–1) are SH-like aLRT values. The shaded clades and letter designations correspond to the colors and letters shown in Figure 3—figure supplement 3.

Figure 3—figure supplement 5.. Maximum likelihood phylogeny of the regions between Schizosaccharomyces octosporus wtf genes and an upstream flanking 5S rDNA gene.

The regions upstream of 40 S. octosporus wtf genes with an upstream 5S rDNA gene were aligned with MAFFT, and a maximum likelihood phylogeny was built with PhyML. Branch support values shown at the nodes (0–1) are SH-like aLRT values. The shaded clades and letter designations correspond to the colors and letters shown in Figure 3—figure supplement 3.

Figure 3—figure supplement 6.. Maximum likelihood phylogeny of Schizosaccharomyces octosporus wtf genes.

The sequences of 83 S. octosporus wtf genes were aligned using MAFFT, and a maximum likelihood phylogeny was constructed using PhyML. Branch support values shown at the nodes (0–1) are SH-like aLRT values. The color-coded letter designations to the right of the gene names indicate the phylogenetic groupings of the sequences flanking the wtf genes from Figure 3—figure supplement 3.

Figure 4.. Shared wtf locus in three fission yeast species.

(A) The syntenic region between clr4 and met17 in Schizosaccharomyces octosporus, S. osmophilus, S. cryophilus, and S. pombe is shown. The S. pombe locus shown is from the S. kambucha isolate. The orange boxes represent wtf genes, the blue boxes represent wag genes, the red arrows represent 5S rDNA, the green arrow represents tRNA-his, the gray boxes represent genes without a homolog in this region in the species shown, and the black boxes represent genes that are syntenic between the species. The phylogenetic relationship between species is shown to the left of the DNA representation. The orthologs of clr4 (B) and met17 (C) were aligned and used to build neighbor-joining trees that were midpoint rooted. Branch support (0–100) was calculated using bootstrap.

Figure 4—figure supplement 1.. Synteny between Schizosaccharomyces cryophilus wtf4 and S. pombe wtf6.

(A) The syntenic region containing cyp9 and ago1 is shown for all fission yeast species. An inversion in the S. pombe lineage separated cyp9 and ago1. There is a wtf gene upstream of ago1 in both S. pombe and S. cryophilus. The orange boxes represent the wtf genes. Five genes are numbered and shown in green to illustrate that the ancestor of S. pombe and S. cryophilus likely had a wtf gene between cyp9 and ago1. The black boxes represent additional orthologous genes in synteny. The orthologs of cyp9 (B) and ago1 (C) were aligned and used to build neighbor-joining trees that were midpoint rooted. Branch support (0–100) was calculated using bootstrap.

Figure 5.. Gene duplication and non-allelic gene conversion within wtf gene family.

All the predicted intact Wtf antidote amino acid sequences were aligned using MAFFT from Figure 2—figure supplement 1—source data 1 and used to build a maximum likelihood tree using PhyML. The Schizosaccharomyces pombe sequences were from the FY29033 isolate as it has more wtf genes than the reference genome. The S. pombe genes are shown in black, S. octosporus genes are in magenta, S. osmophilus genes are dark blue, and the S. cryophilus genes are cyan. The triangles represent multiple genes with the precise number indicated on the right. The branch support values (0–1) are SH-like aLRT values and are shown at each node.

Figure 5—figure supplement 1.. Gene duplication and non-allelic gene conversion within wtf gene family.

This depicts the same tree as Figure 5 but with all the tip labels displayed.

Figure 5—figure supplement 2.. Genetic algorithm recombination detection (GARD) analysis consistent with non-allelic gene conversion within wtf genes.

We used GARD (Kosakovsky Pond et al., 2006b) analysis to look for evidence of gene conversion within the wtf genes of (A) Schizosaccharomyces octosporus, (B) S. osmophilus, and (C) S. pombe. We considered only genes predicted to be meiotic drivers or suppressors. This analysis found that a hypothesis allowing multiple trees for different segments of the alignment is >100 times more likely than a hypothesis allowing only a single tree, supporting that non-allelic recombination has occurred within wtf genes. The analysis identified two likely breakpoints in each species. For S. pombe, the analysis is from Eickbush et al., 2019.

Figure 5—figure supplement 3.. Contraction and expansion of repeat sequences in wtf genes.

The wtf genes of Schizosaccharomyces octosporus (A), S. osmophilus (C), and S. pombe (E) can contain the indicated repetitive sequences. The DNA (top) and amino acid (bottom) sequence logos representing the repeat regions are shown for each species. The size distribution of the repeat regions for all S. octosporus (A), S. osmophilus (C), and S. pombe (E) wtf genes is shown. The sizes are presented in base pairs instead of repeat units because the terminal repeats are not always full length. The S. pombe data are from Eickbush et al., 2019 and includes wtf genes from four different isolates. The repeat count in exon 4 of S. octosporus wtf genes, and the repeat count in exon 4 of S. osmophilus wtf genes is shown in Supplementary file 1r.

Figure 6.. wtf genes duplicated into pre-existing 5S rDNA.

Testing if lineage restricted wtf genes occur at sites where the ancestral species is inferred to have had a 5S rDNA gene. An example of this situation is illustrated in (A) where species A has a 5S-rDNA-flanked wtf gene, and species B has a 5S rDNA gene at the syntenic locus. (B) Number of wtf + 5S rDNA loci in species A (any of the gene layouts illustrated in (A)) with 5S rDNA at the syntenic locus in species B. This analysis only considers loci that contain 5S-rDNA-flanked wtf gene in species A but contain no wtf genes in species B. Data in Supplementary file 1p and q were used to test this hypothesis.

Figure 6—figure supplement 1.. wtf gene duplication models.

(A) Model of duplication via non-allelic gene conversion: (1) double strand of DNA with 5S rDNA depicted in red. (2) A double-strand DNA break (DSB) within the 5S rDNA and (3) 5’ end resection. (4) Strand invasion of an ectopic locus with a wtf gene flanked by 5S rDNA genes. (5) The repair template containing the wtf gene is copied to the site of the initiating DSB. (6) Strand displacement and annealing of the broken DNA. (7) Synthesis of DNA with wtf gene in the other strand and ligation to finalize repair. (8) wtf gene duplicated in a new locus (B) (1) 5S rDNA-wtf-5S rDNA unit. (2) Crossing-over between 5S rDNA repeats flanking a wtf gene can generate an extrachromosomal circular DNA. (3) This circle can recombine with an ectopic locus containing a 5S rDNA. (4) Generation of a new wtf locus.

Figure 7.. wtf genes can encode for poison and antidote proteins.

Spot assay of serial dilutions of Saccharomyces cerevisiae cells on non-inducing (SC -His -Trp -Ura) and inducing (SC -His -Trp -Ura+500 nM β-estradiol) media. Each strain contains [TRP1] and [URA3] ARS CEN plasmids that are either empty (EV) or carry the indicated β-estradiol inducible wtf alleles. (A) Schizosaccharomyces octosporus wtf25^poison-GFP and wtf25^antiddote-mCherry (B) S. osmophilus wtf41^poison and wtf41^antidote, and (C) S. cryophilus wtf1^poison and wtf1^antidote. The dilution factor is 0.2 starting at OD = 1. (D) A representative cell carrying a [URA3] plasmid with β-estradiol inducible S. octosporus wtf25^poison-GFP (cyan). (E) A representative cell carrying a [TRP1] plasmid with β-estradiol inducible S. octosporus wtf25^antidote-mCherry (magenta). (F) A representative S. cerevisiae cell carrying a [URA3] plasmid with β-estradiol inducible S. octosporus wtf25^poison-GFP (cyan) and [TRP1] plasmid with β-estradiol inducible S. octosporus wtf25^antidote-mCherry (magenta). In all the experiments, the cells were imaged approximately 4 hr after induction with 500 nM β-estradiol. TL = transmitted light. Scale bar represents 2 µm.

Figure 7—figure supplement 1.. Some wtf genes outside of Schizosaccharomyces pombe encode for poison and antidote proteins.

Spot assay of serial dilutions of Saccharomyces cerevisiae cells on non-inducing (SC -His -Trp -Ura) and inducing (SC -His -Trp -Ura+500 nM β-estradiol) media. Each strain contains [TRP1] and [URA3] ARS CEN plasmids that are either empty (EV) or carry the indicated β-estradiol inducible alleles. (A) S. osmophilus wtf19^poison and wtf19^antidote, (B) S. octosporus wtf61^poison and wtf61^antidote. The dilution factor is 0.1 for (A) 0.2 for (B) with a starting OD = 1 for both panels.

Figure 7—figure supplement 2.. Non-cognate Wtf^antidotes fail to rescue cells from Wtf^poisons.

Spot assay of serial dilutions of Saccharomyces cerevisiae cells on non-inducing (SC -His -Trp -Ura) and inducing (SC -His -Trp -Ura+500 nM β-estradiol) media. Each strain contains [TRP1] and [URA3] ARS CEN plasmids that are either empty (EV) or carry the indicated β-estradiol inducible Wtf^poison and wtf^antidote alleles. (A) Schizosaccharomyces octosporus wtf61 and S. osmophilus wtf41, (B) S. cryophilus wtf1 and S. osmophilus wtf41, (C) S. octosporus wtf61 and S. octosporus wtf25, (D) S. cryophilus wtf1 and S. octosporus wtf25, and (E) S. pombe wtf4 and S. octosporus wtf25. In (C–E), the Wtf25^poison protein was tagged with GFP, and the Wtf25^antidote protein was tagged with mCherry. The percent identity between the coding sequences of the wtf^poison alleles and the percent amino acid identity shared by the Wtf^poison proteins are shown at the top of each panel. The dilution factor for all plates is 0.2 starting at OD = 1.

Figure 7—figure supplement 3.. The distribution of Schizosaccharomyces octosporus Wtf25 proteins is similar to S. pombe Wtf4 proteins.

(A) Images of cells carrying a [URA3] plasmid with β-estradiol inducible S. octosporus wtf25^poison-GFP. Wtf25^poison-GFP signal is distributed in the cytoplasm, with potential endoplasmic reticulum localization (yellow arrows). (B) Images of cells carrying a [TRP1] plasmid with β-estradiol inducible S. octosporus wtf25^antidote-mCherry (magenta). Wtf25^antidote-mCherry signal can be observed within vacuoles (white arrows) and as large aggregates (yellow arrows). (C) Images of cells carrying a [URA3] plasmid with β-estradiol inducible S. octosporus wtf25^poison-GFP (cyan) and [TRP1] plasmid with β-estradiol inducible S. octosporus wtf25^antidote-mCherry (magenta). There is colocalization of Wtf25^poison-GFP and Wtf25^antidote-mCherry signal within vacuoles (black arrows). In all the experiments, the cells were imaged approximately 4 hr after induction with 500 nM β-estradiol. TL = transmitted light. Scale bars represent 2 µm. All images were captured with the same settings. Images in panel A are not shown at the same brightness and contrast as panels B and C to better visualize the signal.

Figure 8.. Three Schizosaccharomyces octosporus wtf genes, when individually deleted, caused spore viability loss in heterozygous crosses but not in homozygous crosses.

Deletion mutants of seven S. octosporus wtf genes were obtained, and crosses were performed. Heterozygous deletion cross but not homozygous deletion cross of wtf25, wtf68, or wtf33 resulted in significant spore viability loss. Spore viability was measured using octad dissection analysis (see Materials and methods). Representative octads are shown in Figure 9, Figure 9—figure supplements 1–6 and Figure 8 and Figure 9—source data 2. Numerical data are provided in Supplementary file 2b. p-Values (Fisher’s exact test) for crosses with >5% spore viability reduction compared to the wild-type control are shown and calculated in Figure 8—source data 1.

Figure 9.. Some Schizosaccharomyces octosporus wtf genes cause meiotic drive.

(A) Representative octads dissected from asci produced from a wtf25 heterozygous deletion cross. The labels A–H indicate the 8 spores dissected from each ascus, and the labels 1–11 indicate the 11 asci analyzed. The genotypes of clones were determined by replica plating onto antibiotic-containing plates. Raw data of all octads can be found in Figure 9—source data 2. (B) The percentages of spores that were viable and with indicated genotypes produced by wtf25⁺/wtf25Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—source data 1. (C) Classification of octads derived from wtf25⁺/wtf25Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as p-values cannot reach the significance threshold if the total number of octads ≤5. (D) Correlation between transmission distortion ratio and poison isoform expression level. The transmission distortion ratio represents the proportion of wtf containing spores in total viable spores produced by a wtf⁺/wtfΔ heterozygote, and the read counts are those shown in Figure 2—figure supplement 3. Numerical data of transmission distortion ratio of each wtf gene can be found in Supplementary files 2b-h.

Figure 9—figure supplement 1.. Octad dissection analysis of wtf68 heterozygous deletion cross.

(A) Representative octads dissected from asci produced from a wtf68 heterozygous deletion cross. The labels A–H indicate the 8 spores dissected from each ascus, and the labels 1–11 indicate the 11 asci analyzed. The genotypes of clones were determined by replica plating. Raw data of all octads can be found in Figure 9—figure supplement 1—source data 2. (B) The percentages of spores that were viable and with indicated genotypes in wtf68⁺/wtf68Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—figure supplement 1—source data 1. (C) Classification of octads derived from wtf68⁺/wtf68Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as p-values cannot reach the significance threshold if the total number of octads ≤5.

Figure 9—figure supplement 2.. Octad dissection analysis of wtf33 heterozygous deletion cross.

(A) Representative octads dissected from asci produced from a wtf33 heterozygous deletion cross. The labels A–H indicate the 8 spores dissected from each ascus, and the labels 1–11 indicate the 11 asci analyzed. The genotypes of clones were determined by replica plating. Raw data of all octads can be found in Figure 9—figure supplement 2—source data 2. (B) The percentages of spores that were viable and with indicated genotypes in wtf33⁺/wtf33Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—figure supplement 2—source data 1. (C) Classification of octads derived from wtf33⁺/wtf33Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as p-values cannot reach the significance threshold if the total number of octads ≤5.

Figure 9—figure supplement 3.. Octad dissection analysis of wtf60 heterozygous deletion cross.

(A) Representative octads dissected from asci produced from a wtf60 heterozygous deletion cross. The labels A–H indicate the 8 spores dissected from each ascus, and the labels 1–11 indicate the 11 asci analyzed. The genotypes of clones were determined by replica plating. Raw data of all octads can be found in Figure 9—figure supplement 4—source data 2. (B) The percentage of spores that were viable and with indicated genotypes in wtf60⁺/wtf60Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—figure supplement 4—source data 1. (C) Classification of octads derived from wtf60⁺/wtf60Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as P values cannot reach the significance threshold if the total number of octads ≤5.

Figure 9—figure supplement 4.. Octad dissection analysis of wtf46 heterozygous deletion cross.

(A) Representative octads dissected from asci produced from a wtf46 heterozygous deletion cross. The labels A–H indicate the 8 spores dissected from each ascus, and the labels 1–11 indicate the 11 asci analyzed. The genotypes of clones were determined by replica plating. Raw data of all octads can be found in Figure 9—figure supplement 3—source data 2. (B) The percentage of spores that were viable and with indicated genotypes in wtf46⁺/wtf46Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—figure supplement 3—source data 1. (C) Classification of octads derived from wtf46⁺/wtf46Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as p-values cannot reach the significance threshold if the total number of octads ≤5.

Figure 9—figure supplement 5.. Octad dissection analysis of wtf21 heterozygous deletion cross.

(A) Representative octads dissected from asci produced from a wtf21 heterozygous diploid. The coordinates A–H stand for 8 spores dissected from 1 ascus, and rows 1–11 represent 11 octad asci analyzed. The genotypes of clones were determined by replica plating. Raw data of all octads can be found in Figure 9—figure supplement 6—source data 1. (B) The percentage of spores that were viable and with indicated genotypes in wtf21⁺/wtf21Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—figure supplement 6—source data 1. (C) Classification of octads derived from wtf21⁺/wtf21Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as p-values cannot reach the significance threshold if the total number of octads ≤5.

Figure 9—figure supplement 6.. Octad dissection analysis of wtf62 heterozygous deletion cross.

(A) Representative octads dissected from asci produced from a wtf62 heterozygous deletion cross. The labels A–H indicate the 8 spores dissected from each ascus, and the labels 1–11 indicate the 11 asci analyzed. The genotypes of clones were determined by replica plating. Raw data of all octads can be found in Figure 9—figure supplement 6—source data 2. (B) The percentage of spores that were viable and with indicated genotypes in wtf62⁺/wtf62Δ cross. The p-value was calculated using exact binomial test, and numerical data are provided in Figure 9—figure supplement 6—source data 1. (C) Classification of octads derived from wtf62⁺/wtf62Δ cross according to the number of viable spores with and without a wtf gene deletion. The p-values were calculated using the exact binomial test. The p-values are only displayed if a pair of octad types have more than five octads in total, as p-values cannot reach the significance threshold if the total number of octads ≤5.

Figure 10.. Schizosaccharomyces octosporus wtf25 is a poison-and-antidote killer meiotic driver.

(A) Schematic of the wtf25 alleles integrated at the leu1 (SOCG_02003) locus. Black asterisks indicate start codon mutations. The start codon for the putative wtf25^poison coding sequence is mutated in the wtf25^{antidote-only} allele, and the start codon for the putative wtf25^antidote coding sequence is mutated in the wtf25^poison-only allele. (B) The wild-type wtf25 allele integrated at the leu1 locus can act as a meiotic driver by killing spores not inheriting it in a heterozygous cross, while wtf25^{antidote-only} mutant allele integrated at the same locus was unable to kill spores not inheriting it in a heterozygous cross. p-Value calculations using a binomial test of goodness-of-fit are shown in Figure 10—source data 1 and 2. (C) The wtf25^poison-only allele integrated at leu1 can cause self-killing in spores that do not inherit wild-type wtf25 at the endogenous locus. The effects of the wtf25^poison-only allele were compared to a control cross in which an empty vector was integrated at leu1. Numerical data are provided in Supplementary file 2i, and the p-value calculation is shown in Figure 10—source data 3.