wtf genes are prolific dual poison-antidote meiotic drivers

Abstract

Meiotic drivers are selfish genes that bias their transmission into gametes, defying Mendelian inheritance. Despite the significant impact of these genomic parasites on evolution and infertility, few meiotic drive loci have been identified or mechanistically characterized. Here, we demonstrate a complex landscape of meiotic drive genes on chromosome 3 of the fission yeasts Schizosaccharomyces kambucha and S. pombe. We identify S. kambucha wtf4 as one of these genes that acts to kill gametes (known as spores in yeast) that do not inherit the gene from heterozygotes. wtf4 utilizes dual, overlapping transcripts to encode both a gamete-killing poison and an antidote to the poison. To enact drive, all gametes are poisoned, whereas only those that inherit wtf4 are rescued by the antidote. Our work suggests that the wtf multigene family proliferated due to meiotic drive and highlights the power of selfish genes to shape genomes, even while imposing tremendous costs to fertility.

Keywords: Meiotic drive; S. pombe; chromosomes; evolution; evolutionary biology; genes; genetic conflict; genomics; infertility; meiosis; selfish genes.

Figure 1.. A complex meiotic drive landscape on Sk and Sp chromosome 3 is revealed by recombination mapping.

(A) A cross between Sk and Sp generates a heterozygote that has low fertility and preferentially transmits Sk alleles on all three chromosomes into viable gametes (Zanders et al., 2014). (B) Generation of chromosome 3 introgression diploids 1–8. Sk-derived DNA is shown in purple while Sp-derived DNA is shown in green. The origin of the Sp/Sk mosaic chromosome is depicted in Figure 1—figure supplement 1. (C) Phenotypes of rec12∆/rec12∆ introgression/Sk diploids. See Figure 1—source data 1 for breakpoints between Sk-derived DNA (purple) and Sp-derived DNA (green). Chromosome transmission was followed using the heterozygous markers at the ade6 locus: hph is short for the hphMX4 marker gene which confers resistance to hygromycin (Hyg^R). The percentage of gametes that inherit both markers (heterozygous disomes, likely aneuploids and diploids) and (after excluding the heterozygous disomes) the percent of gametes that inherit the marker from the pure Sk chromosome are shown. Over 100 viable gametes were tested for each diploid; raw data can be found in Figure 1—source data 2. * indicates p-value<0.01 (G-test) compared to rec12∆/rec12∆ Sk control (from Zanders et al. (2014)). (D) Fine-scale mapping of the drive locus starting with the introgression from diploid 1. Strains that were recombinant between the ura4 locus and an introduced kanMX4 marker gene were selected and their phenotypes were tested in crosses to Sk. The recombinant strain with the smallest amount of Sp DNA that retained the phenotype (sensitivity to drive by an Sk chromosome) is shown in detail. This introgression strain was mated to Sk to generate diploid 9. These analyses identified a ~30 kb candidate region (see blow up) containing a drive locus. In Sp, this region contains wtf4 and the wtf3 pseudogene. The syntenic region in Sk contains only one wtf gene, wtf4. DOI: http://dx.doi.org/10.7554/eLife.26033.003

Figure 1—figure supplement 1.. Generation of mosaic chromosome 3 used in Figure 1B.

The goal of these crosses was to generate a strain containing mostly Sp-derived DNA on chromosome 3 in an otherwise Sk background. This effort was complicated by the different karyotypes of Sp and Sk chromosomes 2 and 3 (Zanders et al., 2014). We used rec12∆ strains to limit recombination, but rare recombinants (e.g. SZY239 and SZY247) can still be obtained via selection. Markers derived from the Sk parent are shown in purple, while Sp-derived markers are green. We first isolated hybrids in which Sk and Sp markers on chromosomes 2 and 3 were uncoupled, suggesting rare recombination events had occurred between Sk and Sp chromosomes 2 and 3. Such events have the potential to generate chromosome 3 variants with mostly Sp DNA, but with an Sk karyotype, as occurred in SZY247. We then performed the illustrated crosses to move that chromosome into a different strain background with pure Sk chromosomes 1 and 2. We finally sequenced SZY558 and verified the strain has Sk chromosomes 1 and 2 and Sp DNA on chromosome 3 until between SNPs at positions 1,804,477 and 1,810,659. DOI: http://dx.doi.org/10.7554/eLife.26033.006

Figure 2.. Sk wtf4 is a self-sufficient meiotic driver that kills gametes that do not inherit the gene.

(A) Allele transmission and propidium iodide (PI) staining phenotypes of diploids 11–19. Sk-derived DNA is purple, Sp-derived DNA is green. The cartoons depict chromosome 3. Chromosomes 1 and 2 are derived from Sk in diploids 11–15, but are from Sp in diploids 16–19. For diploids 11–15, allele transmission was monitored by following heterozygous markers at the ura4 locus, which is tightly linked to wtf4 (estimated 7–17 cM based on physical distance [Young et al., 2002]). PI dye is excluded from living spores, but not dead spores that have lost membrane integrity, such as those destroyed by drive. The percent of spores that exclude PI is shown as a proxy of fertility (Figure 2—source data 1). The PI phenotypes and ura4 locus allele transmission for diploids 11, 12, 14 and 15 were compared to those of the wild-type Sk control (diploid 13). * indicates p-value<0.01 (G-test). For diploids 16–19, allele transmission was followed using markers at the ade6 locus, which is where the empty vector or wtf gene constructs are integrated. The integrations introduced a dominant drug resistance gene and mutated ade6⁺. Because these diploids all had codominant alleles at ade6, we could detect progeny that inherited both ade6 alleles (less than 10% of the total population). These progeny are excluded from the data presented above, but all the raw data are presented in Supplementary file 1. The PI phenotypes and allele transmission for diploids 17–19 were compared to the empty vector control (diploid 16) and * indicates p-value<0.01 (G-test). See Supplementary file 1 for the markers used for each diploid and the raw data for allele transmission and Supplementary file 2 for the PI staining raw data. Over 200 viable gametes were scored for allele transmission and over 200 spores (>50 4-spore asci) were assayed for PI staining. (B) Images of PI staining and transmitted light (TL) in an ascus with no drive containing all alive spores (top) and in an ascus with drive where two of the four spores are dead (bottom). Scale bar represents three microns. DOI: http://dx.doi.org/10.7554/eLife.26033.007

Figure 3.. Sk wtf4 has the capacity to make two proteins and Wtf4-GFP shows a dual localization pattern.

(A) Model for meiotic drive of Sk wtf4 via a poison-antidote mechanism. (B) wtf4 creates a long and an alternative short transcript. See Figure 3—figure supplement 1 for a depiction of the long-read RNA sequencing data on which this model is based (Kuang et al., 2017). (C) Sk Wtf4-GFP localization in diploids where drive does [right] or does not occur [left]. Cells were imaged prior to the first meiotic division [top] and as mature asci [bottom]. (D) Asci generated by diploids of the same genotypes as in (C) stained with PI to label dead cells (those lacking wtf4). DOI: http://dx.doi.org/10.7554/eLife.26033.009

Figure 3—figure supplement 1.. Sp wtf4 has alternate transcriptional start sites.

Our annotation of the wtf4 gene with alternate start sites predicted is shown at the top in the same format as Figures 3–5. The PomBase annotation for Sp wtf4 is shown below that in blue. The transcript locations from one replicate of the meiotic transcript time courses sequenced by Kuang et al. (2017) are shown below in red and orange. The IsoSeq consensus reads shown should represent full-length transcripts, and each represents a number of individual sequencing reads. Only transcripts represented by 11 or more reads are displayed. Many of the transcripts vary by only a few nucleotides at the 5’ or 3’ ends and appear identical in the image. The time the samples were taken after meiotic induction are shown on the left. No transcripts with 11 or more reads were observed at earlier time points. Introns are represented by thin lines with blue arrows and the coding sequences are represented by the thick boxes. There are two major transcriptional start sites and the splice sites of intron 5 are different from those in the PomBase annotation. We did not verify two possible additional transcript types observed only at 10 hr, or explore their possible functional relevance. The data were visualized using IGV (Thorvaldsdottir et al., 2013). DOI: http://dx.doi.org/10.7554/eLife.26033.010

Figure 4.. Sk wtf4 creates two proteins using alternate transcripts: an antidote and a gamete-killing poison.

(A) Separation of function wtf4 alleles. The red stars indicate start codon mutations. (B) Allele transmission and PI staining phenotypes of Sp diploids with the indicated Sk wtf4 alleles integrated at ade6 on chromosome 3, as in diploids 16–19 in Figure 2A. Spores that inherited both alleles at ade6 are eliminated from the data presented above, but the complete data are found in Supplementary file 1. * indicates p-value<0.01 (G-test) compared to empty vector (or wild-type control) for allele transmission and fertility as assayed by PI staining. See Supplementary file 1 for raw data and the markers used to monitor allele transmission for each diploid and Supplementary file 2 for the PI staining raw data. Over 200 viable gametes were scored for allele transmission for all diploids except diploid 24, from which we genotyped 50. Over 200 spores (>50 4-spore asci) were assayed for PI staining of each diploid. DOI: http://dx.doi.org/10.7554/eLife.26033.011

Figure 5.. Wtf4 antidote is spore-specific and Wtf4 poison spreads throughout the ascus.

(A) Constructs tagging either the Wtf4 antidote (top) or poison (bottom) proteins. The red stars indicate start codon mutations. (B) Allele transmission and PI staining phenotypes for tagged alleles, as in Figure 4B. See Supplementary file 1 for raw data and the markers used to monitor allele transmission for each diploid and Supplementary file 2 for the PI staining raw data. We could not reliably use PI to assay fertility of mCherry-tagged strains because of color similarity, but in viable spore yield assays the mCherry^antidote-wtf4 allele gave a similar phenotype to wtf4. * indicates p-value<0.01 (G-test) compared to empty vector (or wild-type control). Over 200 viable gametes were scored for allele transmission and over 200 spores (>50 4-spore asci) were assayed for PI staining. (C) Wtf4 poison (cyan) and antidote (magenta) protein localization prior to the first meiotic division (left) and in a mature ascus (right). Scale bar represents three microns. TL, transmitted light. DOI: http://dx.doi.org/10.7554/eLife.26033.012

Figure 5—figure supplement 1.. Spectral unmixing verifies true signal.

Wtf4 poison (cyan) and antidote (magenta) protein localization in a mature ascus processed using linear unmixing [top] and unprocessed [bottom]. Scale bar represents three microns. DOI: http://dx.doi.org/10.7554/eLife.26033.013