Teosinte Pollen Drive guides maize diversification and domestication by RNAi

Benjamin Berube; Evan Ernst; Jonathan Cahn; Benjamin Roche; Cristiane de Santis Alves; Jason Lynn; Armin Scheben; Daniel Grimanelli; Adam Siepel; Jeffrey Ross-Ibarra; Jerry Kermicle; Robert A Martienssen

doi:10.1038/s41586-024-07788-0

. 2024 Aug 7;633(8029):380–388. doi: 10.1038/s41586-024-07788-0

Teosinte Pollen Drive guides maize diversification and domestication by RNAi

Benjamin Berube ¹, Evan Ernst ¹, Jonathan Cahn ¹, Benjamin Roche ¹, Cristiane de Santis Alves ¹, Jason Lynn ¹, Armin Scheben ², Daniel Grimanelli ³, Adam Siepel ², Jeffrey Ross-Ibarra ⁴, Jerry Kermicle ⁵, Robert A Martienssen ^1,^✉

PMCID: PMC11390486 PMID: 39112710

Abstract

Selfish genetic elements contribute to hybrid incompatibility and bias or ‘drive’ their own transmission^1,2. Chromosomal drive typically functions in asymmetric female meiosis, whereas gene drive is normally post-meiotic and typically found in males. Here, using single-molecule and single-pollen genome sequencing, we describe Teosinte Pollen Drive, an instance of gene drive in hybrids between maize (Zea mays ssp. mays) and teosinte mexicana (Z. mays ssp. mexicana) that depends on RNA interference (RNAi). 22-nucleotide small RNAs from a non-coding RNA hairpin in mexicana depend on Dicer-like 2 (Dcl2) and target Teosinte Drive Responder 1 (Tdr1), which encodes a lipase required for pollen viability. Dcl2, Tdr1 and the hairpin are in tight pseudolinkage on chromosome 5, but only when transmitted through the male. Introgression of mexicana into early cultivated maize is thought to have been critical to its geographical dispersal throughout the Americas³, and a tightly linked inversion in mexicana spans a major domestication sweep in modern maize⁴. A survey of maize traditional varieties and sympatric populations of teosinte mexicana reveals correlated patterns of admixture among unlinked genes required for RNAi on at least four chromosomes that are also subject to gene drive in pollen from synthetic hybrids. Teosinte Pollen Drive probably had a major role in maize domestication and diversification, and offers an explanation for the widespread abundance of ‘self’ small RNAs in the germ lines of plants and animals.

Subject terms: Genetic hybridization, Plant domestication, RNAi, Evolutionary genetics, Plant genetics

An example of hybrid incompatibility between maize and teosinte reveals a selfish toxin–antidote system mediated by small RNAs that may have contributed to the origin of maize.

Main

The introduction of novel genetic variation through hybridization is an important evolutionary catalyst⁵, as adaptive introgression in hybrid individuals can increase fitness under new environmental conditions and lead to geographical expansion and diversification⁶. Modern maize, for example, was first domesticated from a close relative of Z. mays ssp. parviglumis (teosinte parviglumis) in the lowlands of southwest Mexico approximately 9000 bp, but admixture from a second teosinte, Z. mays ssp. mexicana, 4,000 years later, appears to have catalysed rapid expansion across the Americas³. The combination of divergent genomes, however, can also result in hybrid sterility, inviability and necrosis^7–9. The Bateson–Dobzhansky–Muller (BDM) model accounts for such scenarios, via the interaction of deleterious mutations in distinct populations and at least some of these incompatibilities stem from intragenomic conflict triggered by selfish genetic elements^2,10.

Meiotic drive depends on selfish elements that actively manipulate reproductive development to facilitate their own preferential transmission¹¹. Chromosomal drive refers to the manipulation of chromosome segregation during asymmetric female meiosis, as centromeres, heterochromatic knobs and telomeres exert mechanical advantages that favour their inclusion in the egg cell^1,12–14. Examples include Abnormal 10 (Ab10) in both maize and teosinte populations^15,16. Conversely, gene drive occurs preferentially in males and is achieved via disruption of post-meiotic reproductive development resulting in segregation distortion^17,18. These systems tend to occur in sperm or haploid spores and involve toxin–antidote (or distorter–responder) pairs in close genetic linkage. Gametes that do not inherit the drive locus are selectively killed, resulting in overrepresentation of the driver¹¹. The mouse t-complex^19,20, Drosophila Segregation Distorter (SD) complex^21,22 and Schizosaccharomyces pombe/kombucha wtf spore killers^23,24 are all autosomal drivers that selectively kill competing wild-type gametes in heterozygotes.

Because selfish genetic elements often impose fitness and fertility penalties, tremendous selective pressure is placed on regions of the genome that can evolve suppressors²⁵. As a consequence, drive systems undergo recurrent cycles of suppression and counter-suppression; although drive is predicted to be widespread, most systems exist in a cryptic state, either through suppression or fixation^11,26. It is through hybridization with naive individuals that suppression is lost and drive is once again apparent²⁷, reinforcing species barriers and influencing patterns of introgression in hybrid individuals via genetic linkage^28,29.

Here we characterize a male-specific segregation distortion system in introgression lines between maize (Z. mays ssp. Mays) and teosinte mexicana (Z. mays ssp. mexicana), hereafter referred to as Teosinte Pollen Drive (TPD). We implicate small interfering RNAs (siRNAs) from a mexicana-specific long non-coding hairpin RNA in close genetic linkage with the centromere of chromosome 5 as the primary factor mediating pollen killing. Co-segregation of a genetically linked hypomorphic (partially functional) Dcl2 allele suppresses this effect via the reduction of secondary 22-nucleotide (nt) siRNAs and is reinforced by a second unlinked antidote (Tpd2) on chromosome 6. Survey sequencing of modern and traditional varieties of maize from Mexico and sympatric populations of teosinte implicate TPD in patterns of mexicana introgression, and in maize dispersal and domestication.

TPD in maize hybrids

Hybridization between maize and teosinte is subject to unilateral cross-incompatibility^30,31, but pollination of maize by mexicana pollen is frequent³². Consistently, genome-wide assessments of introgression in sympatric collections have provided evidence for asymmetric gene flow from mexicana to maize^32,33. To further explore the reproductive consequences of hybridization, multiple sympatric collections of mexicana were crossed to the Midwestern US dent inbred W22, resulting in variable rates of pollen abortion that typically decreased in subsequent generations. However, a subset of late backcross (BC) lines (hereafter TPD) displayed an unusually consistent rate of pollen abortion (75.5 ± 2.48%) relative to W22 (6.02 ± 2.95%; P < 0.0001, Welch’s t-test) despite normal vegetative and reproductive development (Fig. 1a–c and Extended Data Fig. 1a). The pollen abortion phenotype was absent after three rounds of selfing in TPD BC₈S₃ plants (6.40 ± 2.26%; P < 0.0001, Welch’s t-test), suggesting that heterozygosity was required (Fig. 1d). In reciprocal crosses, pollination of TPD ears with W22 pollen resulted in the independent assortment of fertile, semi-sterile and fully male sterile progeny in a 2:1:1 ratio (Fig. 1e and Supplementary Table 1). These results indicated the presence of two unlinked loci responsible for pollen survival that were transmitted to all individuals in the next generation, but only through pollen. Because this phenotype was observed only in heterozygotes, we reasoned that it stemmed from an incompatibility between the W22 genome and regions of mexicana introgression after meiosis, reminiscent of genic drivers that distort patterns of inheritance via selective gamete killing^20,24. Consistently, meiotic progression in TPD plants was normal until the tetrad stage, following the separation of each haploid complement (Fig. 1f). This phenotype, although strictly post-meiotic, appeared to progress gradually, ultimately resulting in arrested pollen grains with a heterogenous overall diameter and varying degrees of starch accumulation (Fig. 1c,g).

Fig. 1 — a, Anther florets (5 mm) from wild-type (WT; left) and *TPD* (right) plants. Scale bars, 1 mm. b, Mature pollen grains from WT (left) and *TPD* (right) plants. Arrowheads denote developmentally arrested pollen grains. Scale bars, 0.1 mm. c, Viable pollen grains are plump and darkly stained with iodine potassium iodide (I₂KI), whereas arrested pollen grains (arrowheads) exhibit reduced diameter and incomplete staining. Scale bars, 0.1 mm. d, Quantification of pollen abortion rates in *TPD* backcross (BC_11,12), WT and *TPD* self-fertilized (BC₈S₃) lines. Data are mean ± s.d. (n = 6–8). ****P < 0.0001 and not significant (NS; two-tailed Student’s t-test). e, Phenotypic segregation ratios in replicate reciprocal crosses. The numbers above the bar represent the sample size for each progeny population. The red dashed lines denote a perfect 2:1:1 phenotypic segregation ratio. f, Fluorescein diacetate (FDA) viability staining of tetrads from *TPD* plants. Pollen viability is progressively restricted to a single spore following meiosis. Panels show differential interference contrast (DIC), FDA and merged images. Scale bars, 50 µm. g, Viability scoring of *TPD* and WT tetrads shown in panel f. *TPD* spores exhibit significantly reduced viability at the tetrad stage. n = 3 biological replicates, 952 total tetrads assayed. Data are mean ± s.d. *P < 0.05 and **P < 0.01 (Welch’s t-test). h, Single-pollen grain genome sequencing. Imputed allele frequencies at *mexicana* markers in a population of 178 mature pollen grains collected from *TPD* plants. Chr. chromosome. i, Imputed *mexicana* marker density on chromosomes 5 and 6 for individual pollen grain genome sequences. Multiple *mexicana* haplotypes (blue) are selfishly inherited in viable *TPD* pollen grains (n = 178) but not in WT pollen grains (n = 32). Values shown are plotted using a 500-kb sliding window (h,i).

Extended Data Fig. 1 — a, Crossing scheme of the *TPD* phenotype. When back-crossed as male, all the progeny of semi-sterile *TPD* plants display the semi-sterile pollen phenotype instead of the expected 1:1 fertile:semi-sterile ratio. Graphics in panel a were created using BioRender (https://biorender.com). b, Representative ears from *Tpd1 Bt1/bt1*; *Tpd2 Y1/y1* reciprocal crosses with *bt1; y1* testers, demonstrating severe segregation distortion (“drive” of *Bt1*), but only through the male. *bt1 (brittle1*, collapsed kernels); y1 (*yellow1*, white kernels). c, Summary of molecular and morphological mapping of the *Tpd1* interval. Molecular mapping was performed using *Tpd/++* x W22 segregating progeny, whereas morphological mapping was performed by crossing *Tpd1 Bt1/tpd1 bt1* plants to *bt1* testers. d, Molecular mapping of the *Tpd2* interval. SNP markers are shown in blue with recombination frequencies in red.

Genetic mapping revealed that brittle endosperm 1 (bt1) on chromosome 5 and yellow endosperm 1 (y1) on chromosome 6 were linked with the pollen abortion phenotype (Extended Data Fig. 1b,c). Backcrosses to y1;bt1 yielded 100% Bt1 kernels instead of 50%, but only when TPD was used as a pollen parent (Extended Data Fig. 1b). The frequency of white kernels (y1) was in agreement with recombination estimates (21–22%). This bias was strongly indicative of gene drive resembling similar incompatibility systems in rice³⁴, although we could not formally exclude other forms of incompatibility that also result in segregation distortion. To exclude such possibilities, we sequenced the genomes of two homozygous TPD lines (BC₈S₃ and BC₅S₂) to define 408,031 high-confidence single-nucleotide polymorphisms (SNPs) corresponding to regions of mexicana introgression. Next, we sequenced the genomes of individual surviving pollen grains from TPD plants, rationalizing that if segregation distortion was occurring in pollen, the causative regions would be overrepresented. We found that several intervals occurred at much higher frequencies than expected after eight backcrosses (Fig. 1h). Of note, introgression intervals on chromosomes 5 and 6 were consistently observed in all surviving pollen (Fig. 1i), strongly indicative of post-meiotic gene drive. We designated these loci as Tpd1 and Tpd2, respectively.

A Dicer-Like 2 toxin–antidote complex

To determine the relative contributions of Tpd1 and Tpd2 to pollen abortion and survival, we separated the components by maternal transmission into fertile, semi-sterile (‘drive’) and fully sterile classes (Fig. 2a). Each progeny class had distinct rates of pollen abortion (Fig. 2b) and showed significant differences in flowering time (Fig. 2c). Fertile segregants were phenotypically wild type and showed no transmission defects, whereas drive plants recapitulated the canonical TPD pollen abortion phenotype. By contrast, male reproductive development in sterile plants was developmentally retarded, displaying severely delayed anthesis and reduced overall shed (Fig. 2a,c). Consequently, crosses performed with this pollen showed minimal seed set and often failed entirely. We collected pools of plants from the fertile and sterile phenotypic classes (Fig. 2d) for bulk segregant analysis, and found that Tpd1 was differentially enriched in sterile plants, whereas Tpd2 was enriched in fertile plants (Fig. 2e). This indicated that Tpd1 alone was sufficient to ‘poison’ the male germ line and that this most likely occurred pre-meiotically, as only a single copy of Tpd1 was required. Genetic mapping placed Tpd1 in a large interval surrounding the centromere of chromosome 5, whereas Tpd2 was placed in a 1.5-Mb interval on chromosome 6L (Extended Data Fig. 1c,d).

Fig. 2 — a, Representative tassels from fertile, semi-sterile and sterile plants in a maternally segregating population. Scale bar, 1 cm. b, I₂KI viability staining of pollen from the same genotypes as in panel a. Scale bar, 0.1 mm. c, Measurement of days to anthesis in fertile, semi-sterile and sterile phenotypic classes. n is given at the bottom of the plots. ****P < 0.0001 (two-tailed Mann–Whitney test). d, Genotypic segregation ratios in reciprocal crosses. The numbers at the top of the bars represent the sample size for each progeny population. The red dashed lines denote a perfect 1:1:1:1 genotypic segregation ratio. Normal segregation is only observed in maternal progeny. e, Bulk segregant analysis of fertile and sterile progeny pools indicates that *Tpd1* (red arrowhead) is necessary and sufficient for dominant male sterility (toxin), whereas *Tpd2* (blue arrowhead) is associated with fertility (antidote). FDR ≤ 0.01 (Benjamini–Hochberg method). f, Dot plots of chromosomes 5 and 6 showing multiple alignment between the *TPD* and W22 reference genomes. The blue lines and shaded regions correspond to five fully scaffolded intervals of *mexicana* introgression (indicated by arrowheads). As in panel e, the red and blue arrowheads mark the *Tpd1* and *Tpd2* intervals, respectively. The small purple arrowheads indicate breakpoints of an approximately 13-Mb paracentric inversion present within the *Tpd1* haplotype on chromosome 5L. g, Schematics summarizing the *Tpd1* and *Tpd2* intervals, as well as associated markers. The 13-Mb inversion is indicated as a reverse arrow.

The unusual transmission of TPD led us to liken it to previously described selfish genetic elements that operate via post-meiotic gamete killing^20,22,24. These systems generally encode a toxin (or distorter) that acts in trans to disrupt proper reproductive development. Only gametes containing a cell-autonomous antidote (or resistant responder allele) can suppress these effects in a gametophytic manner. Although the toxin was clearly encoded by Tpd1, the TPD system was unusual in that it featured a genetically unlinked antidote, namely, Tpd2. However, the absence of tpd1;Tpd2 recombinants in the progeny of W22 × TPD crosses argued that Tpd2 alone was insufficient for suppression of pollen abortion (Fig. 2d and Supplementary Table 2). We reasoned that this might reflect the additional requirement for another antidote, linked to Tpd1, that could explain the observed rate of pollen abortion (approximately 75%). Linked modifiers in drive systems are common and generally ascribed to the co-evolutionary struggle between distorters and rapidly accumulating suppressors^11,22.

SNP genotyping of the two homozygous lines identified 13 mexicana introgression intervals, 7 of which were shared between backcross generations (Extended Data Fig. 2a). As predicted from the single-pollen sequencing data, the highest regions of SNP density were present on chromosome 5 (Tpd1) and chromosome 6 (Tpd2), coinciding with Bt1 and close to Y1, respectively (Extended Data Fig. 2a). However, other regions strongly overrepresented in homozygous progeny were only partially overrepresented in TPD pollen, including additional peaks on chromosomes 5S, 6S and 6L (Extended Data Fig. 2b). This probably reflected the presence of recombinant pollen grains that competed poorly during pollination.

Extended Data Fig. 2 — a, Whole genome plots of homozygous *mexicana* SNP density present within *Tpd1; Tpd2* lines. The upper plot corresponds to data from bulked seedlings after 8 backcrosses and 3 self-pollinations (BC₈S₃) whereas the lower plot is from BC₅S₂ plants. SNP density is consolidated in 250 kb genomic bins. Physical locations for morphological markers *Bt1* and Y1, as well as *mexicana* derived RNAi genes, are labelled in red. 7/13 introgression intervals overlap in both independently maintained homozygous lines. b, Allele frequency at *mexicana* markers in 96 pollen grains from four different *TPD* plants subjected to single pollen grain sequencing. Regions highlighted in red were over-represented in viable pollen grains. c, Quantification of pollen viability in *Tpd1* + */−; Tpd2* + */−; rgd1/+* and *Tpd1* + */−; Tpd2* + */−; Rgd1* pollen demonstrating gametophytic suppression via germline segregation of the *rgd1* null allele. n ≥ 9 plants per genotype, ≥ 200 pollen grains per plant. **** p < 0.0001 (Welch’s t-test).

To determine gene content in these and other introgression intervals, we performed de novo genome assembly from homozygous Tpd1;Tpd2 BC₈S₃ seedlings (see Methods; Supplementary Table 3) with fully scaffolded mexicana introgression intervals on chromosomes 5 and 6 (Fig. 2f,g). We noted the presence of a 1.9-Mb mexicana introgression interval on chromosome 5S linked to the Tpd1 haplotype and strongly overrepresented in both our bulk sequencing and single-pollen grain data (Figs. 1h,i and 2f). Within this interval, we identified ten genes with expression in pollen, one of which, Dcl2, had excess nonsynonymous substitutions within conserved domains (Fig. 3a), suggesting the possibility of adaptive evolutionary change³⁵. Absolute genetic linkage (n = 214) between this locus (hereafter dcl2^T) and Tpd1 was conditioned on passage through the male germ line from heterozygous TPD plants, whereas recombination between dcl2^T and Tpd1 occurred at the expected frequency (approximately 12%) when crossed as female (Fig. 3b). This was very strong evidence for a linked antidote and probably explained the maintenance of this interval across 13 backcross generations.

Fig. 3 — a, Genome-wide *mexicana* SNP density in bulk-sequenced *Tpd1;Tpd2* (BC₈S₃) plants. A subset of *mexicana* introgression intervals (in addition to *Tpd1* and *Tpd2*) are selectively maintained and include RNAi factors. A *mexicana*-derived allele of *Dcl2* (*dcl2*^T) with a high rate of nonsynonymous substitution is maintained in linkage to *Tpd1*. dsRBD, double-stranded RNA-binding domain. b, Rates of recombination between *dcl2*^T and *Tpd1* in replicate reciprocal crosses. *dcl2*^T exhibits tight pseudolinkage with *Tpd1* when propagated as male (0 cM), but not as female (18.7 ± 1.6 cM). The numbers above the bars represent the sample size for each progeny population. c, Measurements of pollen viability in *Tpd1/tpd1* and *tpd1* plants containing combinations of *Dcl2*, *dcl2*^T and *dcl2-mu1*. Addition of the *dcl2-mu1* hypomorphic allele is sufficient for suppression of *Tpd1*-mediated pollen abortion. Data points correspond to measurements from individual plants (n = 6–10). **P < 0.01 and ***P < 0.001 (two-tailed Mann–Whitney test). d, Volcano plots of 21-nt (n = 9), 22-nt (n = 212) and 24-nt (n = 6) small RNA (sRNA) clusters that are differentially expressed in WT and *TPD* pollen. The accumulation of ectopic 22-nt siRNAs occurs specifically in *TPD* pollen. log₂ fold change ≥ 2, FDR ≤ 0.01. e, Representative ears from replicate crosses containing WT *Dcl2* (W22 × *Tpd1/tpd1*) or *dcl2-mu1* (W22 × *dcl2-mu1* *Tpd1/dcl2-mu1* tpd1) in linkage to *Tpd1*. Pollen parents homozygous for *dcl2-mu1* restore the seed set. Scale bar, 4 cm.

Dcl2 encodes a Dicer-like protein responsible for the production of 22-nt siRNAs from hairpins, as well as secondary small RNAs from double-stranded RNA templates produced by the coordinated action of RNA-DEPENDENT RNA POLYMERASE 6 (RDR6) and SUPPRESSOR OF GENE SILENCING 3 (SGS3)³⁶. In Arabidopsis thaliana, DCL2 function is superseded by DCL4 and endogenous levels of 22-nt siRNAs are low³⁷. However, DCL2 can fulfill roles in silencing and antiviral immunity when DCL4 function is lost^37,38, sometimes resulting in ‘toxic’ pleiotropic defects associated with gene targets of 22-nt siRNAs^37,39,40. These observations stem from the unique biological properties of 22-nt siRNAs, which are responsible for propagation of systemic silencing signals that move between cells⁴¹ and transitive amplification of silencing in both cis and trans⁴². In dcl2^T, nonsynonymous changes were clustered within the DExD/H RNA helicase domain of Dicer (Fig. 3a), which has been shown to alter substrate preference and processing efficiency of double-stranded RNA, but not hairpin RNA, in both plants and invertebrates^43–45.

To explore the role of 22-nt siRNAs in the TPD phenotype, we tested mutants in 22-nt siRNA biogenesis for their ability to act as antidotes. We isolated maternal dcl2^T recombinants and compared them with the dcl2-mu1 allele in the W22 inbred background, which has a Mu transposon insertion in the 5′ untranslated region, 200 bp upstream of the start codon. In dcl2^T/dcl2-mu1 Tpd1, pollen abortion was partially suppressed, whereas pollen from dcl2-mu1/dcl2-mu1 Tpd1 plants were almost fully viable (Fig. 3c). This meant that stacking over the dcl2^T allele had a synergistic effect, strongly supporting its role as a partial antidote, and indicating that the sporophytic production of 22-nt siRNAs in diploid meiotic cells was responsible for the TPD phenotype. To test the idea that 22-nt siRNAs might be responsible for TPD, we sequenced pollen small RNAs from TPD and wild-type siblings and found that although small RNA composition was similar overall, the Tpd1 haplotype triggered a strong, 22-nt-specific response (Fig. 3d). Consistent with these 22-nt small RNAs being responsible for the TPD phenotype, we observed almost complete rescue of sterility in dcl2-mu1/dcl2-mu1 Tpd1/ + pollen parents (Fig. 3e). Several other introgression intervals observed in one or the other backcross individual also included genes encoding components of the small RNA biogenesis pathway, including ago1a, ago1b and rgd1, the homologue of SGS3 (Fig. 3a and Extended Data Fig. 2a). These intervals were also observed in single-pollen grain sequencing along with dcl2^T (Extended Data Fig. 2b). To determine whether these genes were also capable of acting as an antidote, we crossed mutants in rgd1 to TPD plants. Segregation of rgd1 in the germ line of heterozygotes resulted in close to 50% viable pollen (Extended Data Fig. 2c), suggesting that it functions as a cell-autonomous gametophytic suppressor in a manner similar to Tpd2. We concluded that mutants in primary 22-nt small RNA synthesis (dcl2-mu1) blocked production of the toxin, whereas mutants in secondary 22-nt small RNA synthesis (dcl2^T and rgd1), and potentially in small RNA function (ago1a and ago1b), acted as antidotes.

22-nt small RNAs target a pollen lipase

To identify the origin and the targets of DCL2-dependent small RNAs, we performed small RNA sequencing from wild-type, dcl2^T and dcl2-mu1 plants. Analysis revealed that 22-nt siRNAs were the dominant species in wild-type pollen (Extended Data Fig. 3a,b) and defined 804 high-confidence 22-nt siRNA pollen-specific clusters (log₂ fold change ≥ 2, false discovery rate (FDR) ≤ 0.01; Supplementary Table 4). As expected, these clusters depended on Dcl2 (P < 0.0001, determined by analysis of variance (ANOVA)) and there were even fewer 22-nt siRNAs in dcl2-mu1 than in dcl2^T (Extended Data Fig. 3c). Over half (54.6%) of all pollen-specific 22-nt species were derived from endogenous hairpin precursors (hpRNAs; Extended Data Fig. 3d,e,g). Hairpin short interfering RNAs (hp-siRNAs) were disproportionately 22 nt long, derived from a single strand (Extended Data Fig. 4a,b) with high thermodynamic stability (Extended Data Fig. 4c,d). On the basis of these criteria (and a minimum expression cut-off), we identified 28 hp-siRNA-producing loci in the genome, with at least one hairpin on every chromosome except chromosome 4 (average 2.1 ± 1.3 per chromosome). hp-siRNAs can serve as a powerful means to silence transposons⁴⁶, and 22-nt siRNAs targeting Gypsy and Copia LTR retrotransposons were abundant in pollen, as were those targeting Mutator and CACTA elements (Extended Data Fig. 3d). We also found evidence for pollen-specific silencing of at least 30 protein-coding genes (Extended Data Fig. 3d,f,g). Germline specificity is a common feature in SD systems, as such factors can avoid the evolutionary conflicts imposed by pleiotropic fitness defects in the diploid stage of the life cycle⁴⁷.

Extended Data Fig. 3 — a, Distribution and relative abundances of small RNA size classes in WT pollen libraries. Bars indicate mean ± SD. n = 3 biological replicates. b, Comparison of relative abundances for 21-nt, 22-nt, and 24-nt sRNA size classes in WT leaf and pollen samples. Both 22-nt and 24-nt sRNAs show significant increases in pollen. Bars indicate mean ± SD. n = 3 biological replicates. **** p < 0.0001 (Welch’s t-test). c, Comparison of 22-nt sRNA levels in *Dcl2*, *dcl2*^T, and *dcl2-mu1* pollen at 804 pollen-specific loci. Values shown are log₂ transformed counts per million (CPM) averaged across replicates. n = 3 replicates per genotype. **** p < 0.0001 (ANOVA test). d, Summary of relative contributions for 22-nt sRNA producing loci in WT pollen. Hairpin/inverted repeat (IR) hp-siRNAs represent the largest fraction of 22-nt species. e, Heatmap showing 22-nt hp-siRNA levels at hpRNA loci in leaf and pollen. f, Heatmap showing 22-nt siRNA levels at protein-coding genes in leaf and pollen. g, Browser shots showing 22-nt hp-siRNA accumulation at a hpRNA locus on chromosome 1 (left) and 22 nt siRNA silencing at a representative protein-coding gene. Scale is CPM.

Extended Data Fig. 4 — a, hp-siRNAs are expected to show strand bias. Measurement of strand score (min[plus, minus]/max[plus, minus]) at 28 putative hairpin precursors and randomly selected siRNA clusters from wild-type (W22) maize. A value of zero indicates complete strand bias, whereas a value of 1 indicates unbiased accumulation from both strands. n = 28. **** p < 0.0001 (Welch’s t-test). b, log2 read count at hairpin precursors indicate 22 nt size bias. n = 28. **** p < 0.0001 (ANOVA). c, Example of a 73 nt stretch from a 4,480 bp hairpin precursor demonstrating near-complete reverse complementarity. d, Mountain plots measuring thermodynamic stability of the *Tpd1* hairpin from *mexicana* and another randomly selected hairpin structure.

In TPD pollen, we observed the accumulation of 158 ectopic 22-nt siRNA clusters across the genome (log₂ fold change ≥ 2, FDR ≤ 0.01; Supplementary Table 5), and a general upregulation of genes associated with 22-nt siRNA biogenesis and function (Extended Data Fig. 5a). Nearly 60% of all ectopic 22-nt siRNAs in TPD pollen targeted transposable elements of the P Instability Factor (PIF)/Harbinger superfamily (Extended Data Fig. 5b), whose expression was TPD specific (Extended Data Fig. 5c–e). This superfamily is also activated following intraspecific hybridization and anther culture in rice⁴⁸. However, a subset of protein-coding genes was also targeted in TPD pollen specifically (Extended Data Fig. 5b). Given that a reduction in 22-nt siRNAs suppressed the TPD phenotype, we hypothesized that inappropriate silencing of these genes might disrupt male reproductive development. In total, we identified four genes that gained ectopic 22-nt siRNAs in TPD pollen, approximately 62% of which came from a single gene (Zm00004b012122) that is also located on chromosome 5S (Extended Data Fig. 6a). Relative to other targets, this gene exhibited highly specific expression in pollen (Extended Data Fig. 6b,c). Zm00004b012122 encodes a GDSL triacylglycerol lipase/esterase, defined by a core catalytic sequence motif (GDSxxDxG), with roles in lipid metabolism, host immunity and reproductive development⁴⁹. In maize, both male sterile 30 (ms30) and irregular pollen exine 1 (ipe1) mutants disrupt genes encoding a GDSL lipase and are completely male sterile^50,51. Similar functions have been reported in rice⁵² and Arabidopsis⁵³.

Extended Data Fig. 5 — a, RNAi genes (*Sgs3*/*Rgd1*, *Rdr6*, *Ago1e* and *Dcl2*) associated with 22 nt biogenesis and function are upregulated in *TPD* pollen. Expression is shown in TMM normalized counts. Bars show mean ± SD. n = 5 replicates per condition. **** p < 0.0001, *** p < 0.001, ** p < 0.01 (FDR). b, Relative abundances of *TPD*-dependent 22 nt siRNAs mapping to annotated elements. Pie chart inset shows proportions of 22 nt siRNAs targeting genes in CPM. c, Browser shot showing transcriptional activation at PIF/Harbinger elements in *TPD* pollen as well as 22 nt siRNA accumulation. d, Quantification of mRNA expression at 258 PIF/Harbinger superfamily elements in WT and *TPD* pollen. **** p < 0.0001 (Mann-Whitney test). e, 22 nt siRNA levels at 42 PIF/Harbinger elements in WT and *TPD* pollen. **** p < 0.0001 (Mann-Whitney test).

Extended Data Fig. 6 — a, Browser shots showing ectopic accumulation of 22-nt siRNAs at protein-coding genes in *TPD* pollen. Scale in counts per million (CPM). b, RNA-seq tissue expression of 22-nt siRNA targets specific to *TPD* pollen, data from ref. ¹²⁵. Bars show mean ± SD. n = 3 replicates per tissue. c, RNA-seq expression of 22-nt siRNA targets in WT and *TPD* pollen. Bars show mean ± SD. n = 5 replicates per condition. d, Western blot comparing TDR1 protein levels in WT and *TPD* pollen, anthers, and leaf. Protein levels were normalized using Heat Shock Protein 90 (HSP90). e, p-nitrophenyl butyrate esterase activity assay in 5 mm anthers and pollen from WT, *TPD*, and *Tpd1* + /− plants. **f, g**, GO term biological processes up-regulated in f, *TPD* and g, WT pollen (FDR ≤ 0.001). Upregulated genes in *TPD* pollen were associated with RNA metabolism, ribosome assembly, and cytoplasmic translation as well as G2 mitotic arrest. This could reflect translational repression via 22-nt siRNAs. Interestingly, a subset of genes associated with endoplasmic reticulum (ER)-nucleus signalling was also up-regulated¹²⁶, while genes associated with glycerol metabolism, the primary backbone for TAG synthesis, were downregulated. In pollen, the accumulation of TAGs in lipid droplets (LDs) is critical for proper membrane expansion and pollen tube growth¹²⁷.

DCL2-dependent 22-nt siRNAs engage primarily in translational repression of their targets⁵⁴, and consistently all four target genes had similar or higher levels of mRNA in TPD pollen (Extended Data Fig. 6c). We raised antiserum to the GDSL lipase for immunoblotting, choosing a surface-exposed peptide located between putative pro-peptide-processing sites reflecting endoplasmic reticulum localization⁵¹. The GDSL lipase protein accumulated strongly in both 5-mm anthers and mature pollen from wild-type plants, but was absent from leaf and from TPD anthers and pollen, supporting the conclusion that 22-nt siRNAs mediate translational repression (Extended Data Fig. 6d). Furthermore, whole-protein extracts from TPD anthers had reduced esterase activity, which was ameliorated in pollen containing Tpd2 but not in pollen with Tpd1 alone (Extended Data Fig. 6e). Gene ontology analysis of genes upregulated in wild-type and TPD pollen strongly supported translational suppression of the GDSL lipase as the primary cause of developmental arrest and abortion of pollen in TPD plants (Extended Data Fig. 6f,g). Finally, mRNA expression began post-meiotically at the 3-mm (tetrad) stage, peaking in 5-mm anthers and mature pollen (Extended Data Fig. 7a). This expression pattern was conspicuously similar to the developmental window in which TPD pollen abortion begins (Fig. 1f), suggesting that this gene might act as a ‘responder’ to Tpd1-driven distortion. On the basis of all these observations, we defined Zm00004b012122 as the primary candidate for targeting by Tpd1 toxin activity, renaming it Teosinte drive responder 1 (Tdr1).

Extended Data Fig. 7 — a, RT-qPCR of the *Tdr1* transcript throughout anther development and in mature pollen. Bars show mean ± SD. n = 2 replicates per condition. b, RT-qPCR of the *Tpd1* transcript during anther development and in mature pollen in WT and *Tpd*. Bars show mean ± SD. n = 2 replicates per condition. **c, d**, Single-cell expression at different stages of meiosis of c, *Tdr1* and d, *Dcl2*, using single cell RNAseq data⁵⁹. Early and late expression of *Dcl2* coincides with *Tpd1* and *Tdr1*, respectively.

Hairpin siRNAs trigger pollen abortion

As ectopic silencing at protein-coding genes only occurred in the presence of the Tpd1 haplotype, we reasoned that the distorter must generate small RNAs capable of triggering silencing in trans. In plants, microRNAs, secondary siRNAs and hp-siRNAs all have this capacity⁵⁵. Processed small RNA duplexes are loaded into ARGONAUTE (AGO) proteins, passenger strands are released and RNase H-like slicing activity is targeted by guide strand homology, as is translational repression⁵⁶. Silencing can be amplified via the coordinated action of RDR6 and SGS3 (ref. ⁴²). RNase H-mediated slicing results in an exposed 5′-phosphate that allows for ligation of 3′ cleavage products. Using an improved degradome sequencing technique in TPD pollen, iPARE-seq (see Methods), we identified putative cleavage sites responsible for triggering silencing at the Tdr1 locus (Fig. 4a,b). We simultaneously searched for non-coding RNA within the Tpd1 haplotype that produced 22-nt sRNAs capable of triggering silencing. This approach yielded only one candidate: a large hpRNA similar to those identified previously in wild-type pollen (Fig. 4c). This hairpin was uninterrupted in the mexicana-derived Tpd1 interval and produced high levels of TPD-specific 22-nt hp-siRNAs (Fig. 4d,e). In the W22 genome, we identified two large transposon insertions that interrupted this locus, which produced no small RNA, indicating that it was non-functional in maize, consistent with being responsible for TPD (Fig. 4c). By comparison with centromere placement in other maize inbreds⁴, the hairpin is on the short arm of chromosome 5, 5 Mb from the centromere.

Target site prediction uncovered four abundant hp-siRNA species predicted to target the Tdr1 transcript in trans (Fig. 4f) resembling ‘proto-microRNA’⁵⁷. Three of these began with 5′-C, indicating loading into Ago5, and had iPARE-seq support, indicating cleavage of Tdr1 (Fig. 4g). However, the most abundant hp-siRNA, Tpd1-siRNAb, was 22 nt in length and began with 5′-A, indicating loading into Ago2 (Fig. 4g). Tpd1-siRNAb has an asymmetric bulge predicted to enhance silencing transitivity and systemic spread between cells⁵⁸, and had only limited iPARE-seq support, indicating translational repression (Fig. 4b). To confirm that silencing of Tdr1 was responsible for the TPD phenotype, we generated two independent frameshift alleles within the catalytic domain using CRISPR–Cas9 (Fig. 4h). Homozygotes for tdr1-1 and tdr1-2 had identical male sterile phenotypes, with extensive pollen abortion that phenocopied Tpd1 (Fig. 4i–k).

Expression of the Tpd1 hairpin was observed pre-meiotically in 1–3-mm anthers, as well as in microspores (4-mm anthers) where expression of Tdr1 was first detected, but not in mature pollen (Extended Data Fig. 7b,c). According to published single-cell RNA sequencing data from developing maize pollen⁵⁹, Dcl2 is also expressed pre-meiotically, consistent with its role in generating 22-nt hp-siRNA from Tpd1 (Extended Data Fig. 7d). Dcl2 is not expressed in bicellular microspores, but is expressed in mature pollen consistent with an additional function in production of secondary small RNAs from Tdr1 (Extended Data Fig. 7d). These results indicate a sequential order of events, in which expression of Tpd1 pre-meiotically deposits small RNAs in microspores where they target Tdr1. Subsequent expression of Dcl2 in mature pollen then promotes secondary small RNA production and translational suppression. Identification of Tdr1 provided insight into the function of Tpd2. Tpd1 hp-siRNAs were unaffected by Tpd2, which was instead required to suppress secondary small RNA biogenesis from Tdr1, along with the mexicana allele of Dcl2, namely, dcl2^T (Extended Data Fig. 8a). This indicates that Tpd2 and dcl2^T have additive effects on suppressing secondary small RNAs, consistent with their role as partial antidotes (Extended Data Fig. 8b). Although the molecular identity of Tpd2 remains unknown, the 1.5-Mb Tpd2 interval contains six pollen-expressed genes in W22 (Extended Data Fig. 8c). One of these genes encodes the maize homologue of Arabidopsis RNA-DIRECTED DNA METHYLATION (RDM1), a critical component of the RNA-directed DNA methylation pathway⁶⁰. This gene is significantly overexpressed in TPD pollen (Extended Data Fig. 8c), and it is possible that increased activity of RNA-directed DNA methylation might compete with the production of secondary small RNAs^61,62, although further experimentation is required to support this idea.

Extended Data Fig. 8 — a, Browser shots showing ectopic accumulation of 22nt siRNAs at *Tdr1* (left) and *Tpd1* hairpin (right) in fertile (*tpd1; tpd2*, grey), drive (*Dcl2*^T *Tpd1/Dcl2 tpd1; Tpd2/tpd2*, blue) and sterile (*Dcl2 Tpd1/Dcl2 tpd1; tpd2*, red) pollen from maternal segregants. Scale in counts per million (CPM). *Tpd2* and *Dcl2*^T reduce small RNAs from *Tdr1* (left) but not from the *Tpd1* hairpin (right), consistent with a cell autonomous role in secondary small RNA biogenesis and silencing. b, Table summarizing genotypes transmitted when *TPD* is backcrossed to W22 as female (left column) or male (right column). Only the combination of *dcl2*^T *Tpd1* (linked) and *Tpd2* is transmitted through pollen. Recombinants between *dcl2*^T and *Tpd1* occur at the expected frequency but are not transmitted through pollen (Fig. 3b), presumably because of higher siRNA production during and after meiosis. c, *Rdm1* (Zm00004b029511) is one of six genes in the *Tpd2* interval expressed in pollen, and is overexpressed in *TPD* pollen.

TPD, RNAi and the origin of modern maize

Population-level studies of maize traditional varieties identified an uninterrupted mexicana-derived haplotype surrounding the centromere of chromosome 5 (refs. ^32,63) with high rates of linkage disequilibrium⁶³. Consistent with reduced recombination, fine-mapping of Tpd1 yielded very few informative recombinants (21 of 7,549) and none proximal to the hairpin (Extended Data Fig. 1c). Comparative analysis of the TPD and W22 genomes revealed three megabase-scale inversions, one of which corresponded to a 13-Mb event within the Tpd1 haplotype and including Bt1 on chromosome 5L (Fig. 2f,g). The presence of this inversion, along with its physical proximity to the centromere, explained our mapping data (Extended Data Fig. 1c) and strongly suggested that the Tpd1 haplotype behaves as a single genetic unit.

The 13-Mb paracentric inversion in the Tpd1 haplotype (W22 chromosome 5: 115,316,812–124,884,039) almost entirely encompasses ‘region D’ adjacent to centromere 5 (W22 chromosome 5: 118,213,716–126,309,970), which has undergone a dramatic domestication sweep in all maize inbreds relative to teosinte⁴. This region includes Bt1, which undergoes drive in the TPD system (Extended Data Fig. 1c). Our synthetic hybrids with maize inbred W22 retained approximately 13 intervals of the mexicana genome that persisted in serial backcrosses (Extended Data Fig. 2a,b). Four of these intervals are tightly linked to genes encoding AGO proteins, specifically Ago1a, Ago1b, Ago2b and Ago5b, all of which are expressed in the male germ line (Extended Data Fig. 2). According to 5′ nucleotide analysis, these AGO proteins are predicted to bind to Tpd1-hp-siRNAa–d (Ago2 and Ago5), as well as secondary Tdr1 22-nt siRNAs (Ago1), and it is conceivable that hypomorphic alleles could also act as partial antidotes in combination with Tpd2. In addition to intervals encoding Dcl2, Rdm1 and Rgd1/Sgs3, this means that 7 of the 13 intervals are tightly linked to genes required for RNAi. These correlations suggest that there has been strong selection on all of these modifiers to ameliorate the toxic effects of Tpd1, resulting in apparent gene drive.

In traditional maize varieties, but not in sympatric mexicana, significant correlations were observed in mexicana ancestry between 11 of the 13 intervals (Extended Data Fig. 9a, Supplementary Tables 6 and 7 and Supplementary Discussion). By contrast, variation at Tdr1 displays no such correlation with the co-inherited intervals in traditional maize varieties (Extended Data Fig. 9a). In fact, Tdr1 is strongly monomorphic in traditional maize varieties, whereas in mexicana, Tdr1 displays extreme polymorphism (Extended Data Fig. 9b). We considered the possibility that this locus has evolved to become immune to silencing in modern maize, a predicted outcome of selfish genetic systems¹¹. A recent survey of maize and teosinte genome sequences⁶⁴ has revealed that three of the four Tpd1-hp-siRNA target sites in Tdr1 exhibit extensive polymorphism in maize and teosinte, including an in-frame deletion of the target site seed region for Tpd1-hp-siRNAa and a SNP at position 11 in target sites for Tpd1-hp-siRNAb, which are predicted to reduce or abolish cleavage and translational inhibition, respectively (Fig. 5a). TPD pollinations of the temperate inbred B73, which carries the deletion haplotype, resulted in 50% partially sterile (44 of 83) and fully fertile (35 of 83) offspring in advanced backcrosses, as well as rare fully sterile presumptive recombinants (4 of 83), consistent with these expectations. Surveys of the frequency of the deletion haplotype across Zea found it widespread, suggesting an origin before speciation of Z. mays from Zea luxurians and Zea diploperennis (Fig. 5b), whereas it is absent from Zea nicaraguagensis and Tripsacum dactyloides. The frequency of the deletion haplotype is relatively low in mexicana (12%) compared with parviglumis (46%), and increases in tropical maize, traditional maize varieties, popcorn and inbreds, where it is nearly fixed in several modern inbred groups (98%), suggesting a trajectory of spread to North and South America.

Extended Data Fig. 9 — a, Frequency of mexicana-derived alleles were calculated for 1 Mb intervals associated with *TPD* on chromosomes 1,2,3,4,5,6 and 10. Correlations are shown between population means from each of 14 maize traditional varieties (left) and sympatric *mexicana* populations (right). Intervals on chromosomes 5, 6 and 10 include *Dcl2* (5.19), *Tdr1* (5.40), *Tpd1* (5.79), *Rgd1/Sgs3* (6.3) and candidate genes *Ago1a* (6.3), *Tpd2/Rdm1* (6.98), *Ago1b* and *Ago2b* (10.134). Correlations were observed for most of the intervals in maize traditional varieties, except for *Tdr1* (green arrow), but only for intervals including *Tpd1*, *Rgd1* and *Tpd2* in *mexicana*. Spearman correlation coefficients are displayed as a heatmap. b, *mexicana*-derived ancestry in each of 14 maize traditional varieties (above) and sympatric *mexicana* populations (below) in *Dcl2*, *Tdr1, Tpd1* and *Tpd2* intervals. The *Tdr1* interval (green) is monomorphic in most of the maize traditional varieties, but shows extreme dimorphism in 7 out of 14 sympatric *mexicana* populations.

Fig. 5 — a, Sequence complement of the *Tpd1*-hp-siRNAa and *Tpd1*-hp-siRNAb target sites in *Tdr1*, indicating a 27-bp in-frame deletion found in modern maize, maize traditional varieties and in teosinte that removes the *Tpd1*-hp-siRNAa seed sequence, and a SNP on the eleventh nucleotide of *Tpd1*-hp-siRNAb that is predicted to reduce binding. b, Pie charts indicating the frequency of the deletion in 1,483 resequenced genomes from maize and teosinte, aligned with the B73 reference genome (GATK 3.0). The deletion allele (blue) arose in teosinte and quickly spread through maize traditional varieties in Central and South America, before fixation in modern stiff stalk, but not in tropical maize inbred lines. High frequencies of heterozygosity in *mexicana* and *parviglumis* are consistent with recent or ongoing pollen drive.

Discussion

TPD is a toxin–antidote system that defies Mendelian inheritance and may have a history of selfish evolution, like other hybrid incompatibilities that cause gamete killing. Unlike teosinte crossing barriers tcb-1, Ga-1 and Ga-2 (ref. ⁶⁵), which prevent fertilization, TPD resembles BDM incompatibility (also known as Dobzhansky–Muller incompatibility or DMI) in that it acts post-zygotically, resulting in sterile progeny. In canonical BDM, however, hybrid sterility is due to the unmasking of deleterious alleles, so that fertility eventually recovers in recurrent backcrosses to either parent. In TPD, backcrosses to maize result in pollen abortion no matter how many backcross generations are observed. This is because TPD is a special case of BDM that is consistent with meiotic drive. For gamete killers to spread via meiotic drive, they must compensate somehow for loss of fertility⁶⁶. Loss of fertility may have posed a challenge for the spread of TPD in populations of teosinte. Therefore, establishing the evolutionary origin of TPD by meiotic drive will require additional population-level data and modelling, so that other explanations for gamete killing can be excluded⁶⁷. In practice, maize–teosinte hybrids are extremely vigorous with numerous tassels, so that these wind-pollinated species may be less sensitive to reductions in male fertility. This is especially true during domestication, when early domesticates are typically less prolific than wild relatives, and at lower population size. In such circumstances, segregation distortion in hybrids could affect patterns of introgression between maize and teosinte.

Tpd1 encodes a long non-coding hpRNA that produces specific 22-nt hp-siRNAs in the male germline and kill pollen grains by targeting the genetically linked responder gene Tdr1 (Extended Data Fig. 10a,b). This effect is countered by at least two gametophytic antidotes: a linked hypomorphic allele of Dcl2 and the unlinked Tpd2 locus on chromosome 6 (Extended Data Fig. 10c). The genetic architecture of this system, consisting of multiple linked and unlinked loci, deviates from previously established toxin–antidote systems. In rice, for instance, the qHMS7 quantitative trait locus is a selfish genetic element composed of two tightly linked open reading frames³⁴. Similarly, the wtf4 driver in S. pombe features two alternatively spliced transcripts derived from the same locus²⁴. By contrast, the Tpd1 haplotype results from tight pseudolinkage between Tpd1, Tdr1 and dcl2^T on chromosome 5, but only when transmitted through the male (Extended Data Fig. 8b). Although recombinants occur in single-pollen grains, they are not transmitted to the next generation (Fig. 1), and maternal recombinants between dcl2^T and Tpd1 are completely male sterile (Fig. 2). These recombinants produce far more secondary 22-nt small RNAs at Tdr1 (Extended Data Fig. 8a), providing an explanation for the failure to transmit recombinants through pollen. Tpd2 is unlinked but acts cell autonomously, so that independent assortment of Tpd1 and Tpd2 occurs in female gametes, but never in male, implying that gametophytic suppression of pollen killing requires co-segregation of Tpd2 with Tpd1. Although unlinked suppressors are relatively rare, a similar system has been reported in fission yeast⁶⁸. In both cases, the selective suppression of drive can be interpreted as selfish behaviour on the part of the antidote. Ultimately, cycles of suppression and counter-suppression can be expected to result in complex, polygenic drivers that exist in a continuum of cryptic states (Extended Data Fig. 11), and the conspicuous maintenance of mexicana introgression intervals containing RNAi factors supports this idea (Extended Data Figs. 2a and 11).

Extended Data Fig. 10 — a, The *TPD* system is defined by *mexicana* introgression intervals on chromosomes 5 and 6. *Tpd1* encodes a pre-meiotically expressed *mexicana*-specific hairpin that produces abundant 22nt hp-siRNAs. b, These hp-siRNAs trigger secondary siRNAs amplification by RDR6 and SGS3/RGD1 at the *Tdr1* gene when it starts being transcribed at the late tetrad stage, which in turn target *Tdr1* for translational repression (red ribosomes). In surviving microspores (dark yellow background) *Tpd2* and *dcl2*^T repress secondary siRNAs processing, restoring translation and fertility (green ribosome). c, Only pollen grains of the genotype *dcl2*^T *Tpd1*; *Tpd2* are viable, and all other competing gametes are eliminated. Other RNAi genes (*Sgs3*/*Rgd1*, *Ago1*, *Ago2*, *Ago5*) can act as partial suppressors by affecting levels of siRNAs.

Extended Data Fig. 11 — After the antidotes arise in an ancestral teosinte population, the *Tpd1* toxin arises and gains a transmission advantage when linked to the antidote genes. In extant populations of *Z. mexicana* and *Z. mays*, some antidotes are fixed, while others are polymorphic or lost. The demographic model was based on ref. ¹²⁸ and the conceptual framework of selfish evolution was adapted from ref. ⁶⁷. Graphics were created using BioRender (https://biorender.com).

Genome scans of sympatric maize and mexicana have identified multiple regions of introgression associated with adaptive variation, some of which overlap with the genomic interval corresponding to the Tpd1 haplotype³² and other intervals undergoing distortion⁶⁹, and we found that intervals associated with drive in pollen are significantly correlated with each other in maize traditional varieties, but not in sympatric mexicana populations (Extended Data Fig. 9). We postulated that the most powerful suppressor of all would be an ‘immune’ target gene, in which hp-siRNA target sites in Tdr1 had been mutated. Such in-frame immune haplotypes were found in wild taxa in Zea and have been progressively fixed from tropical to temperate stiff-stalk maize inbreds (Fig. 5), suggesting that TPD may be an ancient system that has influenced admixture throughout the history of the genus, reaching fixation in modern maize. TPD complements the hypothesized role of Ab10, a chromosomal driver of female meiosis that simulations suggest may have been responsible for the redistribution of heterochromatic knobs in maize, parviglumis and mexicana^15,70, potentially along with thousands of linked genes¹⁶.

Our results suggest that DCL2-dependent 22-nt small RNAs stemming from long hpRNAs function as selfish genetic elements in pollen. In Arabidopsis, 22-nt siRNA biogenesis is carefully regulated due to ectopic silencing of host genes^37,40,42,54, and 21–22-nt siRNAs from pollen mediate triploid seed abortion^71,72 and can block self-fertilization⁷³. In Drosophila melanogaster^74,75, silencing of protein-coding genes by recently evolved hairpins is important for male reproductive development⁷⁵, whereas in Drosophila simulans, the Winters sex-ratio distortion system is actually suppressed by two hpRNAs, Not much yang (Nmy) and Too much yin (Tmy), which act as antidotes and are essential for male fertility and sex balance^76,77. In mammals, endo-siRNAs in the oocyte are generated from hairpin and antisense precursors by an oocyte-specific Dicer isoform (Dcr-O) and have an essential function in global translational suppression^78–80. The remarkable parallels between all of these systems, and between Dcr-O and dcl2^T, which both have potential defects in the helicase domain, invites speculation that selection for selfish behaviour is an efficient means by which germline small RNAs can propagate within a population. Such propagation provides a plausible origin for ‘self’-targeting small RNAs in the germlines of plants and animals.

Methods

Plant material and growth conditions

The TPD lineage traces to teosinte mexicana collected near Copándaro, Michoacán, Mexico in December 1993. Gamete a, plant 4 of collection 107 was used in an initial outcross to the Midwestern US dent inbred W22 and subsequently backcrossed. Tpd1;Tpd2 (BC₈S₃) homozygous lines were used for whole-genome sequencing and de novo genome assembly. All additional experiments were performed using Tpd1/tpd1; Tpd2/tpd2 (BC₁₁–BC₁₃) plants or populations derived from maternal segregation of these lines. The lbl1-rgd1 and dcl2-mu1 alleles were backcrossed to W22 four or more times. dcl2-mu1 was isolated from the Uniform-Mu line UFMu-12288. All genetic experiments used segregating wild-type progeny as experimental controls. Plants were grown under greenhouse and field conditions.

Phenotyping and microscopy

All pollen phenotyping was performed using mature 5-mm anthers before anthesis. Individual anthers were suspended in PBS and dissected using forceps and an insulin syringe. Starch viability staining was performed using Lugol solution (L6146-1L, Sigma). Measurements for days to anthesis were taken for three replicate crosses (Tpd1/tpd1;Tpd2/tpd2 × W22) with staggered planting dates in three different field positions. The leaf collar method⁸¹ was combined with routine manual palpation of the topmost internode to track reproductive stages. Meiotic anthers were dissected, fixed in 4% paraformaldehyde plus MBA buffer⁸², and stained with DAPI for visualization. For tetrad viability assays, anthers from the upper floret of an individual spikelet were dissected and stored in MBA. One anther was used for staging and the others were dissected to release the tetrads. FDA viability staining was performed as previously described⁸³. To control for artefacts associated with sample handling, only intact tetrads (four physically attached spores) were considered.

Genotyping and marker design

For routine genotyping, tissue discs were collected with a leaf punch and stored in 96-well plates. To extract genomic DNA, 20 μl of extraction solution (0.1 M NaOH) was added to each well and samples were heated to 95 °C for 10 min and then placed immediately on ice. To neutralize this solution, 90 μl of dilution solution (10 mM Tris + 1 mM EDTA, pH to 1.5 with HCl) was added. PCRs, using 1–2 μl of this solution as template, were performed using GoTaq G2 Green Master Mix (M7822, Promega). Secondary validation of genotyping reactions was performed as needed using the Quick-DNA Plant/Seed Miniprep kit (D6020, Zymo Research). Bulk Illumina and Nanopore data from Tpd1;Tpd2 seedlings was used for co-dominant molecular marker design (Supplementary Table 8). When possible, markers based on simple sequence length polymorphisms were prioritized, but a number of restriction fragment length polymorphisms were also designed. W22, Tpd1/tpd1;Tpd2/tpd2 and Tpd1;Tpd2 genomic DNA was used to validate marker segregation before use. The dcl2-mu1 insertion was amplified by combining gene-specific forward and reverse primers with a degenerate terminal inverted repeat primer cocktail. The insertion was subsequently validated by Sanger sequencing.

High-molecular-weight genomic DNA extraction

High-molecular-weight (HMW) genomic DNA was used as input for all Nanopore and bulk Illumina sequencing experiments. For extraction, bulked seedlings were dark treated for 1 week before tissue collection. Four grams of frozen tissue was ground under liquid N₂ and pre-washed twice with 1.0 M sorbital. The tissue was then transferred to 20 ml pre-warmed lysis buffer (100 mM Tris-HCl (pH 9.0), 2% w/v CTAB, 1.4 M NaCl, 20 mM EDTA, 2% PVP-10, 1% 2-mercaptoethanol, 0.1% sarkosyl and 100 μg ml⁻¹ proteinase K), mixed gently and incubated for 1 h at 65 °C. Organic extraction in phase-lock tubes was performed using 1 vol phenol:chloroform:isoamyl alcohol (25:24:1) followed by 1 vol chloroform:isoamyl alcohol. DNA was precipitated by adding 0.1 vol 3 M NaOAc (pH 5.2) followed by 0.7 vol isopropanol. HMW DNA was hooked out with a pasteur pipette and washed with 70% EtOH, air dried for 2 min and resuspended in 200 μl Tris-HCl (pH 8.5; EB). The solution was treated with 2 μl 20 mg ml⁻¹ RNase A at 37 °C for 20 min followed by 2 µl 50 mg ml⁻¹ proteinase K at 50 °C for 30 min. 194 μl EB, 100 µl NaCl and 2 μl 0.5 M EDTA were added, and organic extractions were performed as before. DNA was precipitated with 1.7 vol EtOH, hooked out of solution with a pasteur pipette, washed with 70% EtOH and resuspended in 50 μl EB.

Nanopore and Hi-C sequencing, TPD genome assembly and annotation

HMW DNA from Tpd1;Tpd2 BC₈S₃ was gently sheared by passage through a P1000 pipette 20 times before library preparation with the Oxford Nanopore Technologies Ligation Sequencing gDNA (SQK-LSK109) protocol with the following modifications: (1) DNA repair, end-prep and ligation incubation times were extended to 20 min each; (2) 0.8× vol of a custom SPRI bead solution was used for reaction cleanups^84,85; and (3) bead elutions were carried out at 50 °C for 5 min. Libraries were sequenced on the MinION device with R9.4.1 flow cells. Offline base calling of Oxford Nanopore Technologies reads was performed with Guppy 5.0.7 and the R9.4.1 450-bp super accuracy model. Reads longer than 1 kb were assembled into contigs using Flye 2.9-b1768 (ref. ⁸⁶) with options ‘--extra-params max_bubble_length=2000000 -m 20000 -t 48 --nano-raw’. The same long reads were aligned to the Flye contigs (filtered to keep only the longest alternatives) using minimap2 2.22-r1101 (ref. ⁸⁷), and these alignments were passed to the PEPPER-Margin-DeepVariant 0.4 pipeline⁸⁸ to polish the initial consensus. To correct remaining single-nucleotide variants and small indels, two Illumina PCR-free genomic DNA PE150 libraries were mapped to the long read polished consensus with bwa-mem2 2.2.1 (ref. ⁸⁹) for further polishing with NextPolish 1.3.1 (ref. ⁹⁰) followed by Hapo-G 1.2 (ref. ⁹¹), both with default options. Two biological replicate samples of BC₈S₃ leaf tissue were used to prepare Dovetail Omni-C Kit libraries following the manufacturer’s protocol, and sequenced as a PE150 run on a NextSeq500. These Hi-C reads were mapped to the polished contigs with the Juicer pipeline release 1.6 UGER scripts with options ‘enzyme=none’⁹². The resulting ‘merged_nodups.txt’ alignments were passed to the 3D DNA pipeline to iteratively order and orient the input contigs and correct misjoins⁹³. This initial automatic scaffolding resulted in 11 superscaffolds longer than 10 Mb. Correcting a single centromeric break during manual review with JBAT⁹⁴ resulted in the expected 10 pseudomolecules. One 6-Mb contig was identified as bacterial with no contacts and was discarded. The remaining unscaffolded contigs were of organelle origin (n = 9, 625 kb) or aligned to the pseudomolecules (n = 116, 12 Mb). Coding gene predictions from the NRGene 2.0 W22 (ref. ⁹⁵) were projected onto the TPD genome assembly using Liftoff 1.6.2 (ref. ⁹⁶) with options ‘-polish -copies -chroms <chrom_map>’. An average Phred quality value (QV) score for the assembly was estimated from a 20-mer database of the Illumina reads using merqury 1.4.1 (ref. ⁹⁷) with default options. Assembly completeness was also assessed with BUSCO 5.5.0 (ref. ⁹⁸) with options ‘-m genome --miniprot’. See Supplementary Table 3 for assembly metrics.

RNA extraction

Tissue was collected, snap frozen in liquid nitrogen and stored at −80 °C. Samples were ground into a fine powder using a mortar and pestle on liquid nitrogen. Of pre-extraction buffer (100 mM Tris-HCl (pH 8.0), 150 mM LiCl, 50 mM EDTA (pH 8.0), 1.5% v/v SDS and 1.5% 2-mercaptoethanol), 800 µl was added and mixed by vortexing. Of acid phenol:chloroform (pH 4.7–5.0), 500 µl was added and samples were mixed then spun down at 13,000g for 15 min at 4 °C. The aqueous layer was extracted, and 1 ml TRIzol per 200 mg input tissue was added. Samples were mixed by vortex and incubated at room temperature for 10 min. Chloroform (200 µl) per 1 ml TRIzol was added and samples were mixed by vortexing and then incubated at room temperature for 2 min. Samples were then spun down at 13,000g for 15 min at 4 °C. The aqueous phase was extracted and cleaned up using the Zymo RNA Clean and Concentrator-5 kit (R1013, Zymo Research). Only samples with RNA integrity scores of 9 or more were used for quantitative PCR (qPCR) and sequencing.

Reverse transcription and RT–qPCR

For reverse transcription, 1 µg of total RNA was treated with ezDNase (11766051, Thermo Fisher) according to the manufacturer’s instructions. Reverse transcription was performed with SuperScript IV VILO Master Mix (11756050, Thermo Fisher). Following reverse transcription, complementary DNA (cDNA) was diluted 1:20 in dH₂0 to be used as template in qPCR with reverse transcription (RT–qPCR).

All RT–qPCR experiments were performed on an Applied Biosystems QuantStudio 6 system in 96-well plate format using PowerUp SYBR Green Master Mix (A25741, Thermo Fisher). Before use in experiments, primer efficiency was tested for each primer set using a standard curve generated from serial dilutions of cDNA template. Only primer sets with efficiencies between 90% and 110% were used (Supplementary Table 9). For experiments, three or more biological replicates (independent cDNA samples from discrete plants) were assayed per genotype, and two or more technical replicates were set up for each reaction condition. Raw Ct (cycle threshold) from technical replicates were averaged, and ^∆Ct (mean Ct^exp – mean Ct^ref) was calculated using Elfa9 as a housekeeping reference. ^∆∆Ct values (^∆Ct^cond1 – ^∆Ct^cond2) were calculated between genotypes and converted to fold change (2⁽^–^∆∆Ct)).

Whole-genome sequencing and SNP calling

For HMW DNA from separately maintained Tpd1;Tpd2 lineages (BC₈S₃ and BC₅S₂) and from bulk segregation analysis maternal pools, extractions were as detailed above. Libraries were prepared using the Illumina TruSeq DNA PCR-Free kit (20015962, Illumina) with 2 μg of DNA input. Samples were sequenced on a NextSeq500 platform using 2 × 150-bp high-output run. Adapter trimming was performed with Cutadapt (v3.1)⁹⁹. Paired-end reads were aligned to the W22 reference genome⁹⁵ with BWA-MEM (v0.7.17)¹⁰⁰. Alignments were filtered by mapping quality (mapQ ≥ 30), and PCR duplicates were removed using SAMtools (v1.10)¹⁰¹. SNP calling was performed using Freebayes (v1.3.2)¹⁰². Putative SNP calls were filtered by quality, depth and allele frequency (allele frequency = 1) to obtain a high-confidence mexicana marker set that was subsequently validated against the TPD assembly. For bulk segregation analysis¹⁰³, SNP calls were filtered against the gold-standard TPD marker set. Reference and alternate allele frequencies at each marker were calculated and the average signal was consolidated into 100-kb bins. The ∆SNP index was then calculated for each bin in a sliding window.

Single-pollen grain sequencing

Pollen grains from Tpd1/tpd1;Tpd2/tpd2 plants were suspended in ice-cold PBS on a microscope slide under a dissecting scope. Individual plump, viable pollen grains were deposited into the 0.2-ml wells of a 96-well plate using a p20 pipette. Lysis and whole-genome amplification were performed using the REPLI-g single-cell kit (150345, Qiagen) with the following modifications: one-fourth of the specified volume of amplification mix was deposited in each well and isothermal amplification was limited to 5 h. All steps before amplification were performed in a UV-decontaminated PCR hood. Whole-genome analysis products were cleaned up using a Genomic DNA Clean & Concentrator kit (D4067, Zymo Research), and yields were quantified using with the QuantiFluor dsDNA system (E2670, Promega) in a 96-well microplate format.

Libraries were prepared using the TruSeq Nano DNA High Throughput kit (20015965, Illumina) with 200 ng input. Samples were sequenced on a NextSeq500 platform using 2 × 101-bp high-output runs. Quality control, adapter trimming, alignment and SNP calling were performed as above. BCFtools 1.14 (ref. ¹⁰⁴) was used to derive genotype calls from single-pollen grains at the predefined marker positions and then passed to GLIMPSE 1.1.1 (ref. ¹⁰⁵) for imputation. All calls at validated marker sites were extracted and encoded in a sparse matrix format (rows = markers, columns = samples) and encoded (1 = alt allele, −1 = ref allele, 0 = missing). To assess mexicana introgression in individual pollen grains, mean SNP signal was calculated in 100-kb bins across the genome. A sliding window (1-Mb window, 200-kb step) was applied to smooth the data and identify regions with mexicana SNP density. To identify genomic intervals overrepresented in surviving TPD pollen grains, aggregate allele frequency was calculated across all pollen grains at each marker site.

RNA sequencing and analysis

Five biological replicates were prepared for each biological condition (Tpd1/tpd1;Tpd2/tpd2 and tpd1;tpd2 siblings). Of total RNA, 5 µg was ribosome depleted using the RiboMinus Plant Kit (A1083808, Thermo Fisher), and libraries were prepared using the NEXTFLEX Rapid Directional RNA-seq kit (NOVA-5138-08, PerkinElmer). The size distribution of completed libraries was assessed using an Agilent Bioanalyzer, and quantification was performed using a KAPA Library Quantification kit (KK4824, Roche). Libraries were sequenced on a NextSeq500 platform using a 2 × 150-bp high-output run. Trimmed reads were aligned to the W22 reference with STAR in two-pass alignment mode¹⁰⁶. Read counts were assigned to annotated features using featureCounts¹⁰⁷. For transposable element expression, multi-mapping reads were assigned fractional counts based on the number of identical alignments. Differential expression analysis was performed using edgeR¹⁰⁸. To avoid false positives, a stringent cut-off (log₂ fold change ≥ 2, FDR ≤ 0.001) was used to call differentially expressed genes. Gene ontology analysis (Fisher’s exact test, P < 0.01) was performed using topGO¹⁰⁹, and the results were visualized using rrvgo¹¹⁰. For data visualization, alignment files were converted to a strand-specific bigwig format using deepTools¹¹¹.

Small RNA sequencing and analysis

For comparisons between Tpd1/tpd1;Tpd2/tpd2 and tpd1;tpd2 pollen, three biological replicates were used. Two biological replicates were used for dcl2^T−/− and dcl2-mu1^−/− pollen samples. Libraries were constructed with the NEXTFLEX Small RNA-Seq V3 kit (NOVA-5132-06, PerkinElmer) using 2 μg of total RNA input per library and the gel-free size selection protocol. The size distribution of completed libraries was assessed using an Agilent Bioanalyzer, and quantification was performed using a KAPA Library Quantification kit (KK4824, Roche). Libraries were sequenced on a NextSeq500 platform using a 1 × 76-bp run. Adapters were trimmed using cutadapt⁹⁹, and the 4-bp unique molecular identifier sequences on either side of each read were removed.

Reads were filtered using pre-alignment to a maize structural RNA consensus database using bowtie2 (ref. ¹¹²). Alignment and de novo identification of small RNA loci were performed with ShortStack¹¹³, using a minimum CPM cut-off of 5, and only clusters with clear size bias (21, 22 or 24 nt) were retained in downstream analysis. Differential sRNA accumulation was performed with edgeR¹⁰⁸ (log₂ fold change ≥ 2, FDR ≤ 0.01). The accumulation of size and strand-biased hp-siRNAs was used to identify hairpin loci throughout the genome. For each locus, the underlying primary sequence was tested for reverse complementarity, and RNA secondary structure prediction was performed using RNAfold¹¹⁴. Non-hp-siRNA targets were only retained if they showed negligible strand bias (that is, evidence of a double-stranded RNA template for processing by a Dicer-like enzyme).

iPARE-seq and analysis

iPARE-seq is an improvement on degradome sequencing by PARE-seq¹¹⁵. For iPARE-seq libraries, 40 μg of total RNA was poly(A) selected using a Dynabeads mRNA Purification Kit (61006, Thermo Fisher). Of poly(A) RNA, 1 µg was ligated to the 5′ PARE adapter (100 pmol) in 10% DMSO, 1 mM ATP, 1X T4 RNA ligase 1 buffer (B0216L, New England Biolabs), 25% PEG8000 with 1 μl (40U) of RNaseOUT (10777019, Thermo Fisher) and 1 μl T4 RNA ligase 1 (M0204S, New England Biolabs) in a reaction volume of 100 μl. Ligation reactions were performed for 2 h at 25 °C followed by overnight incubation at 16 °C. Samples were then purified using RNA Clean XP beads (A63987, Beckman Coulter) and eluted in 18 μl dH₂0. Chemical fragmentation of ligated RNA to 200 nt or fewer was performed using the Magnesium RNA fragmentation kit (E6150S, New England Biolabs). Of RNA fragmentation buffer, 2 µl was added and samples were incubated at 94 °C for 5 min followed by a transfer to ice and the addition of 2 μl of RNA Stop solution. Samples were purified using the RNA Clean & Concentrator-5 kit (R1013, Zymo Research) and eluted in 11 μl H₂0. Reverse transcription was performed as follows: 10 μl of RNA, 1 μl of 10 mM dNTP and 2 μl of random primer mix (S1330S New England Biolabs) were mixed and incubated for 10 min at 23 °C, and then put on ice for 1 min. The following was then added: 4 μl of 5X SuperScript IV buffer, 1 μl of 100 mM DTT, 1 μl of RNaseOUT and 1 μl of Superscript IV (200U). The reaction was incubated for 10 min at 23 °C, followed by 10 min at 50 °C. Of Tris-EDTA, 80 µl was then added to this mixture.

Target indirect capture was performed with 100 μl Dynabeads MyOne Streptavidin T1 beads (65601, Thermo Fisher) as per the manufacturer's instructions. Of the reverse transcription reaction, 100 µl was used as input, and captured cDNA molecules were eluted in 50 μl. Second-strand synthesis was performed using 5U Klenow fragment (M0210S, New England Biolabs) with 100 µM dNTPs and 1 μM of iPARE adapter primer (5′-NNNNTCTAGAATGCATGGGCCCTCCAAG-3′) for 1 h at 37 °C and incubation at 75 °C for 20 min. Samples were purified with a 1:1 ratio of AMPure XP SPRI beads (A63880, Beckman Coulter) and resuspended in 51 μl EB. Of sample, 50 µl was used for library preparation with the NEB Ultra DNA library kit (E7370S, New England Biolabs). Barcoded samples were sequenced with a NextSeq500 2 × 150-bp high-output run. Use of the directional iPARE adapter allows for the retention of directionality even when using a non-directional DNA-seq kit. Cutadapt⁹⁹ was used to search and recover the adapter sequence in both 5′ and 3′ orientation (forward in read1 or read2, respectively). Read1 adapter reads were trimmed for the 3′ adapter if present, and the 5′ iPARE adapter was subsequently removed. Potential polyA tails were also removed, and only reads of 20 nt or more were retained. Read2 adapter reads were processed in an identical manner. Filtered reads were mapped using Bowtie2 (ref. ¹¹²) and the 5′ position of each read (the cloned 5′-monophosphate corresponding to the position of AGO-mediated cleavage) was extracted using BEDtools¹¹⁶ with CPM normalization. Small RNA target prediction was performed using psRNATarget¹¹⁷.

Protein extraction and western blotting

Fresh anthers or pollen were collected and snap frozen in liquid nitrogen. Samples were then ground to a fine powder in a mortar and pestle over liquid nitrogen and resuspended in freshly prepared extraction buffer (2 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% v/v NP-40, 5% v/v glycerol, 1 mM PMSF and 1 ml Roche protease inhibitor cocktail per 30 g input tissue) and vortexed thoroughly. Samples were then centrifuged at 14,000 rpm at 4 °C for 5 min to pellet cellular debris, and the aqueous fraction was transferred to another tube. This step was then repeated twice more. Protein extracts were quantified using the Pierce Detergent Compatible Bradford Assay Kit (23246, Thermo Fisher) on a Promega Glomax-Multi+ plate reader.

To assess the role of 22-nt siRNAs in translational repression, antiserum was raised to a peptide (SRKGAPPSSPPLSPPKLGA) from the Zm00004b012122 protein in collaboration with PhytoAB. Specificity was determined as follows: (1) blots using pollen protein extracts showed a single band at roughly the expected size, and (2) blots using leaf protein extracts showed no band in concordance with expected pollen/anther specificity. A rabbit polyclonal HSP90-2 antibody (AS11 1629, Agrisera), a constitutive isoform with high expression, was used as loading control in all western blot experiments. For comparisons of protein abundance between wild-type and TPD pollen/anthers, 2 µg of protein was denatured at 95 °C for 5 min in an appropriate volume of 2X Laemmli buffer (120 nM Tris-Cl (pH 6.8), 4% v/v SDS, 0.004% bromophenol blue, 20% v/v glycerol, 0.02% w/v bromophenol blue and 350 mM DTT). Samples were run on a 4–20% Mini-PROTEAN TGX Precast Gel (4561094, Bio-Rad) with a Precision Plus Protein Dual Xtra Prestained standard (1610377, Bio-Rad).

Transfer to a PVDF membrane was performed using a Bio-Rad Trans-Blot Turbo Transfer system. Membranes were blocked using 5% w/v powdered milk in 1X TBS-T (20 mM Tris, 150 mM NaCl and 0.1% Tween-20) for 1 h at room temperature. Subsequently, the membrane was cut and incubated with primary antibody (1:3,000 dilution in blocking solution) at 4 °C overnight with gentle agitation. Three 15-min membrane washes were performed with 1X TBS-T at room temperature. Membranes were then incubated with a 1:3,000 goat anti-rabbit IgG H&L (PHY6000, PhytoAB) secondary antibody for 1 h at room temperature. Following three more washes with 1X TBS-T, membranes were incubated for 5 min with ECL Prime detection reagent (RPN2236, Amersham) and visualized using a Bio-Rad ChemiDoc Touch Imaging System.

Esterase enzymatic activity assay

Esterase activity assays were performed using the colorimetric substrate p-nitrophenyl butyrate (N9876, Sigma) at a final concentration of 1 mM in 0.5 M HEPES (pH 6.5). For assays using whole 5-mm anthers, 100 μg of total protein was used as input for each sample, whereas 50 μg was used for pollen. Individual samples were prepared in cuvettes at a volume of 1.5 ml. Upon addition of the total protein extract, samples were gently mixed, and an initial 410-nm absorbance reading was taken to serve as a per sample baseline. Samples were then incubated at 30 °C, and absorbance readings were taken every 5 min for a total of 12 timepoints. This experiment was replicated three times for each genotype. All absorbance readings were taken using a Thermo Scientific Genesys 20 spectrophotometer.

Detection of selective sweeps in candidate regions associated with TPD

We investigated signals of selection in genomic regions associated with TPD using selscan (v1.2.0a)¹¹⁸ to calculate the genome-wide normalized absolute integrated haplotype score (iHS) statistics for individual SNPs and in 10-kb windows. iHS is suitable for identifying selection in a single population and relies on the presence of ongoing sweeps and a signal of selection from unusually long-range linkage disequilibrium. We also used VCFtools (v0.1.16)¹¹⁹ to calculate Weir and Cockerham’s F_ST in 10-kb windows to assess signals of selection based on changes in allele frequency between populations. Phased SNPs for modern temperate maize lines, teosinte and T. dactyloides were obtained from Grzybowski et al.¹²⁰, and SNPs for 265 CIMMYT traditional varieties were obtained from Yang et al.¹²¹ and phased with Beagle (v5.4)¹²². A phased and imputed set of 42,387,706 genome-wide concatenated SNPs was used for the analysis of selection. The T. dactyloides allele was set to be the ancestral allele. A consensus genetic map curated by Ed Coe was obtained from MaizeGDB¹²³, and SNP positions were interpolated to genetic positions. Weighted F_ST was calculated for each unique population pair. For iHS, 10-kb windows were binned into 10 quantiles based on the number of SNPs they contained, and empirical P values for each window were calculated within each quantile. The statistic calculated was the number of extreme (top 5%) |iHS| scores per window. Empirical P values for iHS and F_ST were then calculated from the rank of each window based on the respective statistics. We adjusted these P values for multiple testing of different populations using the Bonferroni method. TPD-linked regions (dcl2, rdm1, tdr1 and hairpin region) and their 1-kb upstream and downstream regions were intersected with the 10-kb windows using bedtools (v2.30)¹¹⁶ and assigned the lowest P value of all intersecting windows. To validate our selection scan, we also investigated windows intersecting with a set of four known domestication genes¹²⁴.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Online content

Any methods, additional references, Nature Portfolio reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41586-024-07788-0.

Supplementary information

Supplementary Information^{(653KB, pdf)}

Supplementary Discussion, Supplementary Tables 1–3 and Supplementary Tables 6–9

Reporting Summary^{(73.5KB, pdf)}

Supplementary Tables^{(114.3KB, pdf)}

Supplementary Table 4^{(196.5KB, xlsx)}

22nt clusters enriched in WT pollen vs WT leaf

Supplementary Table 5^{(56KB, xlsx)}

22nt clusters enriched in TPD pollen vs WT pollen

Acknowledgements

We thank K. Dawe and J. Birchler for discussions and for sharing cytogenetic data. Research in the Martienssen laboratory is supported by the US National Institutes of Health (NIH) grant R35GM144206, the National Science Foundation Plant Genome Research Program and the Howard Hughes Medical Institute. The authors acknowledge assistance from the Cold Spring Harbor Laboratory Shared Resources, which are funded in part by a Cancer Center Support grant (5PP30CA045508). B.B. was supported by a predoctoral fellowship from the National Science Foundation.

Extended data figures and tables

Author contributions

B.B., J.K. and R.A.M. designed the study. B.B., E.E., B.R., C.d.S.A. and J.L. performed the experiments with advice from D.G. B.B., E.E., J.C., A.Scheben, J.R.-I. and R.A.M. analysed the data and/or its significance. B.B. and R.A.M. wrote the manuscript with contributions from J.C., J.R.-I. and A.Scheben, A.Siepel and R.A.M. acquired funding.

Peer review

Peer review information

Nature thanks the anonymous reviewers for their contribution to the peer review of this work.

Data availability

Sequencing datasets generated during the current study are available at the NCBI (Gene Expression Omnibus SuperSeries: GSE234925). Datasets used for genome assembly are available at the Sequence Read Archive (BioProject: PRJNA937229). This Whole-Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession JARBIH000000000. The version described in this paper is version JARBIH010000000. All materials are available on request.

Code availability

All code is available on Github (https://github.com/martienssenlab/TPD-manuscript).

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

is available for this paper at 10.1038/s41586-024-07788-0.

Supplementary information

The online version contains supplementary material available at 10.1038/s41586-024-07788-0.

References

1.Sandler, L. & Novitski, E. Meiotic drive as an evolutionary force. Am. Nat.91, 105–110 (1957). 10.1086/281969 [DOI] [Google Scholar]
2.Presgraves, D. C. The molecular evolutionary basis of species formation. Nat. Rev. Genet.11, 175–180 (2010). 10.1038/nrg2718 [DOI] [PubMed] [Google Scholar]
3.Kistler, L. et al. Multiproxy evidence highlights a complex evolutionary legacy of maize in South America. Science362, 1309–1313 (2018). 10.1126/science.aav0207 [DOI] [PubMed] [Google Scholar]
4.Schneider, K. L., Xie, Z., Wolfgruber, T. K. & Presting, G. G. Inbreeding drives maize centromere evolution. Proc. Natl Acad. Sci. USA113, E987–E996 (2016). 10.1073/pnas.1522008113 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Anderson, E. & Stebbins, G. L. Hybridization as an evolutionary stimulus. Evolution8, 378–388 (1954). 10.2307/2405784 [DOI] [Google Scholar]
6.Arnold, M. L. Transfer and origin of adaptations through natural hybridization: were Anderson and Stebbins right? Plant Cell16, 562–570 (2004). 10.1105/tpc.160370 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Bayes, J. J. & Malik, H. S. Altered heterochromatin binding by a hybrid sterility protein in Drosophila sibling species. Science326, 1538–1541 (2009). 10.1126/science.1181756 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Tang, S. & Presgraves, D. C. Evolution of the Drosophila nuclear pore complex results in multiple hybrid incompatibilities. Science323, 779–782 (2009). 10.1126/science.1169123 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Bomblies, K. et al. Autoimmune response as a mechanism for a Dobzhansky–Muller-type incompatibility syndrome in plants. PLoS Biol.5, e236 (2007). 10.1371/journal.pbio.0050236 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.McLaughlin, R. N. Jr & Malik, H. S. Genetic conflicts: the usual suspects and beyond. J. Exp. Biol.220, 6–17 (2017). 10.1242/jeb.148148 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Lindholm, A. K. et al. The ecology and evolutionary dynamics of meiotic drive. Trends Ecol. Evol.31, 315–326 (2016). 10.1016/j.tree.2016.02.001 [DOI] [PubMed] [Google Scholar]
12.Fishman, L. & Saunders, A. Centromere-associated female meiotic drive entails male fitness costs in monkeyflowers. Science322, 1559–1562 (2008). 10.1126/science.1161406 [DOI] [PubMed] [Google Scholar]
13.Chmátal, L. et al. Centromere strength provides the cell biological basis for meiotic drive and karyotype evolution in mice. Curr. Biol.24, 2295–2300 (2014). 10.1016/j.cub.2014.08.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Fishman, L. & McIntosh, M. Standard deviations: the biological bases of transmission ratio distortion. Annu. Rev. Genet.53, 347–372 (2019). 10.1146/annurev-genet-112618-043905 [DOI] [PubMed] [Google Scholar]
15.Buckler, E. S. 4th et al. Meiotic drive of chromosomal knobs reshaped the maize genome. Genetics153, 415–426 (1999). 10.1093/genetics/153.1.415 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Dawe, R. K. et al. A kinesin-14 motor activates neocentromeres to promote meiotic drive in maize. Cell173, 839–850.e18 (2018). 10.1016/j.cell.2018.03.009 [DOI] [PubMed] [Google Scholar]
17.Lyon, M. F. Transmission ratio distortion in mice. Annu. Rev. Genet.37, 393–408 (2003). 10.1146/annurev.genet.37.110801.143030 [DOI] [PubMed] [Google Scholar]
18.McDermott, S. R. & Noor, M. A. F. The role of meiotic drive in hybrid male sterility. Phil. Trans. R. Soc. B365, 1265–1272 (2010). 10.1098/rstb.2009.0264 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Herrmann, B. G., Koschorz, B., Wertz, K., McLaughlin, K. J. & Kispert, A. A protein kinase encoded by the t complex responder gene causes non-Mendelian inheritance. Nature402, 141–146 (1999). 10.1038/45970 [DOI] [PubMed] [Google Scholar]
20.Bauer, H., Willert, J., Koschorz, B. & Herrmann, B. G. The t complex-encoded GTPase-activating protein Tagap1 acts as a transmission ratio distorter in mice. Nat. Genet.37, 969–973 (2005). 10.1038/ng1617 [DOI] [PubMed] [Google Scholar]
21.Hartl, D. L. Genetic dissection of segregation distortion. I. Suicide combinations of SD genes. Genetics76, 477–486 (1974). 10.1093/genetics/76.3.477 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Larracuente, A. M. & Presgraves, D. C. The selfish segregation distorter gene complex of Drosophila melanogaster. Genetics192, 33–53 (2012). 10.1534/genetics.112.141390 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Zanders, S. E. et al. Genome rearrangements and pervasive meiotic drive cause hybrid infertility in fission yeast. eLife3, e02630 (2014). 10.7554/eLife.02630 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Nuckolls, N. L. et al. wtf Genes are prolific dual poison–antidote meiotic drivers. eLife6, e26033 (2017). 10.7554/eLife.26033 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Lewontin, R. C. & Dunn, L. C. The evolutionary dynamics of a polymorphism in the house mouse. Genetics45, 705–722 (1960). 10.1093/genetics/45.6.705 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Hurst, L. D. & Pomiankowski, A. Causes of sex ratio bias may account for unisexual sterility in hybrids: a new explanation of Haldane’s rule and related phenomena. Genetics128, 841–858 (1991). 10.1093/genetics/128.4.841 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Coughlan, J. M. The role of conflict in shaping plant biodiversity. New Phytol.10.1111/nph.19233 (2023). [DOI] [PMC free article] [PubMed]
28.Phadnis, N. & Orr, H. A. A single gene causes both male sterility and segregation distortion in Drosophila hybrids. Science323, 376–379 (2009). 10.1126/science.1163934 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Zhang, L., Sun, T., Woldesellassie, F., Xiao, H. & Tao, Y. Sex ratio meiotic drive as a plausible evolutionary mechanism for hybrid male sterility. PLoS Genet.11, e1005073 (2015). 10.1371/journal.pgen.1005073 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Kermicle, J. L. & Allen, J. P. Cross-incompatibility between maize and teosinte. Maydica35, 399–408 (1990). [Google Scholar]
31.Lu, Y., Hokin, S. A., Kermicle, J. L., Hartwig, T. & Evans, M. M. S. A pistil-expressed pectin methylesterase confers cross-incompatibility between strains of Zea mays. Nat. Commun.10, 2304 (2019). 10.1038/s41467-019-10259-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Hufford, M. B. et al. The genomic signature of crop-wild introgression in maize. PLoS Genet.9, e1003477 (2013). 10.1371/journal.pgen.1003477 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Rojas-Barrera, I. C. et al. Contemporary evolution of maize landraces and their wild relatives influenced by gene flow with modern maize varieties. Proc. Natl Acad. Sci. USA116, 21302–21311 (2019). 10.1073/pnas.1817664116 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Wang, C. et al. A natural gene drive system confers reproductive isolation in rice. Cell186, 3577–3592.e18 (2023). 10.1016/j.cell.2023.06.023 [DOI] [PubMed] [Google Scholar]
35.Yang, Z. & Bielawski, J. P. Statistical methods for detecting molecular adaptation. Trends Ecol. Evol.15, 496–503 (2000). 10.1016/S0169-5347(00)01994-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Yoshikawa, M., Peragine, A., Park, M. Y. & Poethig, R. S. A pathway for the biogenesis of trans-acting siRNAs in Arabidopsis. Genes Dev19, 2164–2175 (2005). 10.1101/gad.1352605 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Parent, J.-S., Bouteiller, N., Elmayan, T. & Vaucheret, H. Respective contributions of Arabidopsis DCL2 and DCL4 to RNA silencing. Plant J.81, 223–232 (2015). 10.1111/tpj.12720 [DOI] [PubMed] [Google Scholar]
38.Deleris, A. et al. Hierarchical action and inhibition of plant Dicer-like proteins in antiviral defense. Science313, 68–71 (2006). 10.1126/science.1128214 [DOI] [PubMed] [Google Scholar]
39.Bouché, N., Lauressergues, D., Gasciolli, V. & Vaucheret, H. An antagonistic function for Arabidopsis DCL2 in development and a new function for DCL4 in generating viral siRNAs. EMBO J.25, 3347–3356 (2006). 10.1038/sj.emboj.7601217 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Wu, Y.-Y. et al. DCL2- and RDR6-dependent transitive silencing of SMXL4 and SMXL5 in Arabidopsis dcl4 mutants causes defective phloem transport and carbohydrate over-accumulation. Plant J.90, 1064–1078 (2017). 10.1111/tpj.13528 [DOI] [PubMed] [Google Scholar]
41.Taochy, C. et al. A genetic screen for impaired systemic RNAi highlights the crucial role of DICER-LIKE 2. Plant Physiol.175, 1424–1437 (2017). 10.1104/pp.17.01181 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Mlotshwa, S. et al. DICER-LIKE2 plays a primary role in transitive silencing of transgenes in Arabidopsis. PLoS ONE3, e1755 (2008). 10.1371/journal.pone.0001755 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Tagami, Y., Motose, H. & Watanabe, Y. A dominant mutation in DCL1 suppresses the hyl1 mutant phenotype by promoting the processing of miRNA. RNA15, 450–458 (2009). 10.1261/rna.1297109 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Welker, N. C. et al. Dicer’s helicase domain discriminates dsRNA termini to promote an altered reaction mode. Mol. Cell41, 589–599 (2011). 10.1016/j.molcel.2011.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Aderounmu, A. M., Aruscavage, P. J., Kolaczkowski, B. & Bass, B. L. Ancestral protein reconstruction reveals evolutionary events governing variation in Dicer helicase function. eLife12, e85120 (2023). 10.7554/eLife.85120 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Slotkin, R. K., Freeling, M. & Lisch, D. Heritable transposon silencing initiated by a naturally occurring transposon inverted duplication. Nat. Genet.37, 641–644 (2005). 10.1038/ng1576 [DOI] [PubMed] [Google Scholar]
47.Bhutani, K. et al. Widespread haploid-biased gene expression enables sperm-level natural selection. Science371, eabb1723 (2021). 10.1126/science.abb1723 [DOI] [PubMed] [Google Scholar]
48.Shan, X. et al. Mobilization of the active MITE transposons mPing and Pong in rice by introgression from wild rice (Zizania latifolia Griseb.). Mol. Biol. Evol.22, 976–990 (2005). 10.1093/molbev/msi082 [DOI] [PubMed] [Google Scholar]
49.Ding, L.-N. et al. Advances in plant GDSL lipases: from sequences to functional mechanisms. Acta Physiol. Plant41, 151 (2019). 10.1007/s11738-019-2944-4 [DOI] [Google Scholar]
50.An, X. et al. ZmMs30 encoding a novel GDSL lipase is essential for male fertility and valuable for hybrid breeding in maize. Mol. Plant12, 343–359 (2019). 10.1016/j.molp.2019.01.011 [DOI] [PubMed] [Google Scholar]
51.Huo, Y. et al. IRREGULAR POLLEN EXINE2 encodes a GDSL lipase essential for male fertility in maize. Plant Physiol.184, 1438–1454 (2020). 10.1104/pp.20.00105 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Zhao, J. et al. RMS2 encoding a GDSL lipase mediates lipid homeostasis in anthers to determine rice male fertility. Plant Physiol.182, 2047–2064 (2020). 10.1104/pp.19.01487 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Tsugama, D., Fujino, K., Liu, S. & Takano, T. A GDSL-type esterase/lipase gene, GELP77, is necessary for pollen dissociation and fertility in Arabidopsis. Biochem. Biophys. Res. Commun.526, 1036–1041 (2020). 10.1016/j.bbrc.2020.03.179 [DOI] [PubMed] [Google Scholar]
54.Wu, H. et al. Plant 22-nt siRNAs mediate translational repression and stress adaptation. Nature581, 89–93 (2020). 10.1038/s41586-020-2231-y [DOI] [PubMed] [Google Scholar]
55.Borges, F. & Martienssen, R. A. The expanding world of small RNAs in plants. Nat. Rev. Mol. Cell Biol.16, 727–741 (2015). 10.1038/nrm4085 [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Fang, X. & Qi, Y. RNAi in plants: an Argonaute-centered view. Plant Cell28, 272–285 (2016). 10.1105/tpc.15.00920 [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Axtell, M. J., Westholm, J. O. & Lai, E. C. Vive la différence: biogenesis and evolution of microRNAs in plants and animals. Genome Biol.12, 221 (2011). 10.1186/gb-2011-12-4-221 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Manavella, P. A., Koenig, D. & Weigel, D. Plant secondary siRNA production determined by microRNA-duplex structure. Proc. Natl Acad. Sci. USA109, 2461–2466 (2012). 10.1073/pnas.1200169109 [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Nelms, B. & Walbot, V. Gametophyte genome activation occurs at pollen mitosis I in maize. Science375, 424–429 (2022). 10.1126/science.abl7392 [DOI] [PubMed] [Google Scholar]
60.Wongpalee, S. P. et al. CryoEM structures of Arabidopsis DDR complexes involved in RNA-directed DNA methylation. Nat. Commun.10, 3916 (2019). 10.1038/s41467-019-11759-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Jauvion, V., Rivard, M., Bouteiller, N., Elmayan, T. & Vaucheret, H. RDR2 partially antagonizes the production of RDR6-dependent siRNA in sense transgene-mediated PTGS. PLoS ONE7, e29785 (2012). 10.1371/journal.pone.0029785 [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Creasey, K. M. et al. miRNAs trigger widespread epigenetically activated siRNAs from transposons in Arabidopsis. Nature508, 411–415 (2014). 10.1038/nature13069 [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Romero Navarro, J. A. et al. A study of allelic diversity underlying flowering-time adaptation in maize landraces. Nat. Genet.49, 476–480 (2017). 10.1038/ng.3784 [DOI] [PubMed] [Google Scholar]
64.Chen, L. et al. Genome sequencing reveals evidence of adaptive variation in the genus Zea. Nat. Genet.54, 1736–1745 (2022). 10.1038/s41588-022-01184-y [DOI] [PubMed] [Google Scholar]
65.Lu, Y., Kermicle, J. L. & Evans, M. M. S. Genetic and cellular analysis of cross-incompatibility in Zea mays. Plant Reprod.27, 19–29 (2014). 10.1007/s00497-013-0236-5 [DOI] [PubMed] [Google Scholar]
66.Hartl, D. L. Population dynamics of sperm and pollen killers. Theor. Appl. Genet.42, 81–88 (1972). 10.1007/BF00277948 [DOI] [PubMed] [Google Scholar]
67.Sweigart, A. L., Brandvain, Y. & Fishman, L. Making a murderer: the evolutionary framing of hybrid gamete-killers. Trends Genet.35, 245–252 (2019). 10.1016/j.tig.2019.01.004 [DOI] [PubMed] [Google Scholar]
68.Bravo Núñez, M. A., Lange, J. J. & Zanders, S. E. A suppressor of a wtf poison–antidote meiotic driver acts via mimicry of the driver’s antidote. PLoS Genet.14, e1007836 (2018). 10.1371/journal.pgen.1007836 [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Barnes, A. C. et al. An adaptive teosinte mexicana introgression modulates phosphatidylcholine levels and is associated with maize flowering time. Proc. Natl Acad. Sci. USA119, e2100036119 (2022). 10.1073/pnas.2100036119 [DOI] [PMC free article] [PubMed] [Google Scholar]
70.McClintock, B., Kato Yamakake, T. A., Blumenschein, A. & Escuela Nacional de Agricultura (Mexico). Chromosome Constitution of Races of Maize: Its Significance in the Interpretation of Relationships between Races and Varieties in the Americas (Colegio de Postgraduados, 1981).
71.Borges, F. et al. Transposon-derived small RNAs triggered by miR845 mediate genome dosage response in Arabidopsis. Nat. Genet.50, 186–192 (2018). 10.1038/s41588-017-0032-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Martinez, G. et al. Paternal easiRNAs regulate parental genome dosage in Arabidopsis. Nat. Genet.50, 193–198 (2018). 10.1038/s41588-017-0033-4 [DOI] [PubMed] [Google Scholar]
73.Durand, E. et al. Dominance hierarchy arising from the evolution of a complex small RNA regulatory network. Science346, 1200–1205 (2014). 10.1126/science.1259442 [DOI] [PubMed] [Google Scholar]
74.Czech, B. et al. An endogenous small interfering RNA pathway in Drosophila. Nature453, 798–802 (2008). 10.1038/nature07007 [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Wen, J. et al. Adaptive regulation of testis gene expression and control of male fertility by the Drosophila hairpin RNA pathway. Mol. Cell57, 165–178 (2015). 10.1016/j.molcel.2014.11.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Tao, Y. et al. A sex-ratio meiotic drive system in Drosophila simulans. II: an X-linked distorter. PLoS Biol.5, e293 (2007). 10.1371/journal.pbio.0050293 [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Lin, C.-J. et al. The hpRNA/RNAi pathway is essential to resolve intragenomic conflict in the Drosophila male germline. Dev. Cell46, 316–326.e5 (2018). 10.1016/j.devcel.2018.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Flemr, M. et al. A retrotransposon-driven Dicer isoform directs endogenous small interfering RNA production in mouse oocytes. Cell155, 807–816 (2013). 10.1016/j.cell.2013.10.001 [DOI] [PubMed] [Google Scholar]
79.Tam, O. H. et al. Pseudogene-derived small interfering RNAs regulate gene expression in mouse oocytes. Nature453, 534–538 (2008). 10.1038/nature06904 [DOI] [PMC free article] [PubMed] [Google Scholar]
80.Su, R. et al. Global profiling of RNA-binding protein target sites by LACE-seq. Nat. Cell Biol.23, 664–675 (2021). 10.1038/s41556-021-00696-9 [DOI] [PubMed] [Google Scholar]
81.Begcy, K. & Dresselhaus, T. Tracking maize pollen development by the leaf collar method. Plant Reprod.30, 171–178 (2017). 10.1007/s00497-017-0311-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Bass, H. W. et al. A maize root tip system to study DNA replication programmes in somatic and endocycling nuclei during plant development. J. Exp. Bot.65, 2747–2756 (2014). 10.1093/jxb/ert470 [DOI] [PubMed] [Google Scholar]
83.Kalkar, S. A. & Neha, K. Evaluation of FDA staining technique in stored maize pollen. Middle East J. Sci. Res.12, 560–562 (2012).
84.Nagar, R. & Schwessinger, B. DNA size selection (>3–4 kb) and purification of DNA using an improved homemade SPRIbeads solution. Protocols.io10.17504/protocols.io.n7hdhj6 (2018).
85.Schalamun, M., Nagar, R. & Kainer, D. Harnessing the MinION: an example of how to establish long‐read sequencing in a laboratory using challenging plant tissue from Eucalyptus pauciflora. Mol. Ecol.10.1111/1755-0998.12938 (2018). [DOI] [PMC free article] [PubMed]
86.Kolmogorov, M., Yuan, J., Lin, Y. & Pevzner, P. A. Assembly of long, error-prone reads using repeat graphs. Nat. Biotechnol.37, 540–546 (2019). 10.1038/s41587-019-0072-8 [DOI] [PubMed] [Google Scholar]
87.Li, H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics34, 3094–3100 (2018). 10.1093/bioinformatics/bty191 [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Shafin, K. et al. Haplotype-aware variant calling with PEPPER-Margin-DeepVariant enables high accuracy in nanopore long-reads. Nat. Methods18, 1322–1332 (2021). 10.1038/s41592-021-01299-w [DOI] [PMC free article] [PubMed] [Google Scholar]
89.Vasimuddin, M., Misra, S., Li, H. & Aluru, S. Efficient architecture-aware acceleration of BWA-MEM for multicore systems. In 2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS) 314–324 (IEEE, 2019).
90.Hu, J., Fan, J., Sun, Z. & Liu, S. NextPolish: a fast and efficient genome polishing tool for long read assembly. Bioinformatics10.1093/bioinformatics/btz891 (2019). [DOI] [PubMed]
91.Aury, J.-M. & Istace, B. Hapo-G, haplotype-aware polishing of genome assemblies with accurate reads. NAR Genom. Bioinform.3, lqab034 (2021). 10.1093/nargab/lqab034 [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Durand, N. C. et al. Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst.3, 95–98 (2016). 10.1016/j.cels.2016.07.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
93.Dudchenko, O. et al. De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds. Science356, 92–95 (2017). 10.1126/science.aal3327 [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Durand, N. C. et al. Juicebox provides a visualization system for Hi-C contact maps with unlimited zoom. Cell Syst.3, 99–101 (2016). 10.1016/j.cels.2015.07.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
95.Springer, N. M. et al. The maize W22 genome provides a foundation for functional genomics and transposon biology. Nat. Genet.50, 1282–1288 (2018). 10.1038/s41588-018-0158-0 [DOI] [PubMed] [Google Scholar]
96.Shumate, A. & Salzberg, S. L. Liftoff: accurate mapping of gene annotations. Bioinformatics37, 1639–1643 (2021). 10.1093/bioinformatics/btaa1016 [DOI] [PMC free article] [PubMed] [Google Scholar]
97.Rhie, A., Walenz, B. P., Koren, S. & Phillippy, A. M. Merqury: reference-free quality, completeness, and phasing assessment for genome assemblies. Genome Biol.21, 245 (2020). 10.1186/s13059-020-02134-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
98.Manni, M., Berkeley, M. R., Seppey, M., Simão, F. A. & Zdobnov, E. M. BUSCO update: novel and streamlined workflows along with broader and deeper phylogenetic coverage for scoring of eukaryotic, prokaryotic, and viral genomes. Mol. Biol. Evol.38, 4647–4654 (2021). 10.1093/molbev/msab199 [DOI] [PMC free article] [PubMed] [Google Scholar]
99.Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal17, 10 (2011). 10.14806/ej.17.1.200 [DOI] [Google Scholar]
100.Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. Preprint at https://arxiv.org/abs/1303.3997 (2013).
101.Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics25, 2078–2079 (2009). 10.1093/bioinformatics/btp352 [DOI] [PMC free article] [PubMed] [Google Scholar]
102.Garrison, E. & Marth, G. Haplotype-based variant detection from short-read sequencing. Preprint at https://arxiv.org/abs/1207.3907 (2012).
103.Takagi, H. et al. QTL-seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant J.74, 174–183 (2013). 10.1111/tpj.12105 [DOI] [PubMed] [Google Scholar]
104.Danecek, P. et al. Twelve years of SAMtools and BCFtools. Gigascience10, giab008 (2021). 10.1093/gigascience/giab008 [DOI] [PMC free article] [PubMed] [Google Scholar]
105.Rubinacci, S., Ribeiro, D. M., Hofmeister, R. J. & Delaneau, O. Efficient phasing and imputation of low-coverage sequencing data using large reference panels. Nat. Genet.53, 120–126 (2021). 10.1038/s41588-020-00756-0 [DOI] [PubMed] [Google Scholar]
106.Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics29, 15–21 (2013). 10.1093/bioinformatics/bts635 [DOI] [PMC free article] [PubMed] [Google Scholar]
107.Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics30, 923–930 (2014). 10.1093/bioinformatics/btt656 [DOI] [PubMed] [Google Scholar]
108.Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics26, 139–140 (2009). 10.1093/bioinformatics/btp616 [DOI] [PMC free article] [PubMed] [Google Scholar]
109.Alexa, A. & Rahnenfuhrer, J. topGO: enrichment analysis for gene ontology. R package version 2.42.0 (2023).
110.Sayols, S. rrvgo: a Bioconductor package for interpreting lists of Gene Ontology terms. MicroPubl. Biol.10.17912/micropub.biology.000811 (2023). [DOI] [PMC free article] [PubMed]
111.Ramirez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res.44, 160–165 (2016). 10.1093/nar/gkw257 [DOI] [PMC free article] [PubMed] [Google Scholar]
112.Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods9, 357–359 (2012). 10.1038/nmeth.1923 [DOI] [PMC free article] [PubMed] [Google Scholar]
113.Axtell, M. J. ShortStack: comprehensive annotation and quantification of small RNA genes. RNA19, 740–751 (2013). 10.1261/rna.035279.112 [DOI] [PMC free article] [PubMed] [Google Scholar]
114.Gruber, A. R., Lorenz, R., Bernhart, S. H., Neuböck, R. & Hofacker, I. L. The Vienna RNA websuite. Nucleic Acids Res.36, W70–W74 (2008). 10.1093/nar/gkn188 [DOI] [PMC free article] [PubMed] [Google Scholar]
115.German, M. A., Luo, S., Schroth, G., Meyers, B. C. & Green, P. J. Construction of parallel analysis of RNA ends (PARE) libraries for the study of cleaved miRNA targets and the RNA degradome. Nat. Protoc.4, 356–362 (2009). 10.1038/nprot.2009.8 [DOI] [PubMed] [Google Scholar]
116.Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics26, 841–842 (2010). 10.1093/bioinformatics/btq033 [DOI] [PMC free article] [PubMed] [Google Scholar]
117.Dai, X., Zhuang, Z. & Zhao, P. X. psRNATarget: a plant small RNA target analysis server (2017 release). Nucleic Acids Res.46, W49–W54 (2018). 10.1093/nar/gky316 [DOI] [PMC free article] [PubMed] [Google Scholar]
118.Szpiech, Z. A. selscan 2.0: scanning for sweeps in unphased data. Bioinformatics40, btae006 (2024). [DOI] [PMC free article] [PubMed]
119.Danecek, P. et al. The variant call format and VCFtools. Bioinformatics27, 2156–2158 (2011). 10.1093/bioinformatics/btr330 [DOI] [PMC free article] [PubMed] [Google Scholar]
120.Grzybowski, M. W. et al. A common resequencing-based genetic marker data set for global maize diversity. Plant J.113, 1109–1121 (2023). 10.1111/tpj.16123 [DOI] [PubMed] [Google Scholar]
121.Yang, N. et al. Two teosintes made modern maize. Science382, eadg8940 (2023). [DOI] [PubMed]
122.Browning, B. L., Tian, X., Zhou, Y. & Browning, S. R. Fast two-stage phasing of large-scale sequence data. Am. J. Hum. Genet.108, 1880–1890 (2021). 10.1016/j.ajhg.2021.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
123.Portwood, J. L. II et al. MaizeGDB 2018: the maize multi-genome genetics and genomics database. Nucleic Acids Res.47, D1146–D1154 (2019). 10.1093/nar/gky1046 [DOI] [PMC free article] [PubMed] [Google Scholar]
124.Stitzer, M. C. & Ross-Ibarra, J. Maize domestication and gene interaction. New Phytol.220, 395–408 (2018). 10.1111/nph.15350 [DOI] [PubMed] [Google Scholar]
125.Walley, J. W. et al. Integration of omic networks in a developmental atlas of maize. Science353, 814–818 (2016). 10.1126/science.aag1125 [DOI] [PMC free article] [PubMed] [Google Scholar]
126.Liu, L. & Li, J. Communications between the endoplasmic reticulum and other organelles during abiotic stress response in plants. Front. Plant Sci.10, 749 (2019). 10.3389/fpls.2019.00749 [DOI] [PMC free article] [PubMed] [Google Scholar]
127.Taurino, M. et al. SEIPIN proteins mediate lipid droplet biogenesis to promote pollen transmission and reduce seed dormancy. Plant Physiol.176, 1531–1546 (2018). 10.1104/pp.17.01430 [DOI] [PMC free article] [PubMed] [Google Scholar]
128.Beissinger, T. M. et al. Recent demography drives changes in linked selection across the maize genome. Nat. Plants2, 16084 (2016). 10.1038/nplants.2016.84 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Information^{(653KB, pdf)}

Supplementary Discussion, Supplementary Tables 1–3 and Supplementary Tables 6–9

Reporting Summary^{(73.5KB, pdf)}

Supplementary Tables^{(114.3KB, pdf)}

Supplementary Table 4^{(196.5KB, xlsx)}

22nt clusters enriched in WT pollen vs WT leaf

Supplementary Table 5^{(56KB, xlsx)}

22nt clusters enriched in TPD pollen vs WT pollen

Data Availability Statement

All code is available on Github (https://github.com/martienssenlab/TPD-manuscript).

[CR1] 1.Sandler, L. & Novitski, E. Meiotic drive as an evolutionary force. Am. Nat.91, 105–110 (1957). 10.1086/281969 [DOI] [Google Scholar]

[CR2] 2.Presgraves, D. C. The molecular evolutionary basis of species formation. Nat. Rev. Genet.11, 175–180 (2010). 10.1038/nrg2718 [DOI] [PubMed] [Google Scholar]

[CR3] 3.Kistler, L. et al. Multiproxy evidence highlights a complex evolutionary legacy of maize in South America. Science362, 1309–1313 (2018). 10.1126/science.aav0207 [DOI] [PubMed] [Google Scholar]

[CR4] 4.Schneider, K. L., Xie, Z., Wolfgruber, T. K. & Presting, G. G. Inbreeding drives maize centromere evolution. Proc. Natl Acad. Sci. USA113, E987–E996 (2016). 10.1073/pnas.1522008113 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Anderson, E. & Stebbins, G. L. Hybridization as an evolutionary stimulus. Evolution8, 378–388 (1954). 10.2307/2405784 [DOI] [Google Scholar]

[CR6] 6.Arnold, M. L. Transfer and origin of adaptations through natural hybridization: were Anderson and Stebbins right? Plant Cell16, 562–570 (2004). 10.1105/tpc.160370 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR7] 7.Bayes, J. J. & Malik, H. S. Altered heterochromatin binding by a hybrid sterility protein in Drosophila sibling species. Science326, 1538–1541 (2009). 10.1126/science.1181756 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Tang, S. & Presgraves, D. C. Evolution of the Drosophila nuclear pore complex results in multiple hybrid incompatibilities. Science323, 779–782 (2009). 10.1126/science.1169123 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Bomblies, K. et al. Autoimmune response as a mechanism for a Dobzhansky–Muller-type incompatibility syndrome in plants. PLoS Biol.5, e236 (2007). 10.1371/journal.pbio.0050236 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR10] 10.McLaughlin, R. N. Jr & Malik, H. S. Genetic conflicts: the usual suspects and beyond. J. Exp. Biol.220, 6–17 (2017). 10.1242/jeb.148148 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Lindholm, A. K. et al. The ecology and evolutionary dynamics of meiotic drive. Trends Ecol. Evol.31, 315–326 (2016). 10.1016/j.tree.2016.02.001 [DOI] [PubMed] [Google Scholar]

[CR12] 12.Fishman, L. & Saunders, A. Centromere-associated female meiotic drive entails male fitness costs in monkeyflowers. Science322, 1559–1562 (2008). 10.1126/science.1161406 [DOI] [PubMed] [Google Scholar]

[CR13] 13.Chmátal, L. et al. Centromere strength provides the cell biological basis for meiotic drive and karyotype evolution in mice. Curr. Biol.24, 2295–2300 (2014). 10.1016/j.cub.2014.08.017 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Fishman, L. & McIntosh, M. Standard deviations: the biological bases of transmission ratio distortion. Annu. Rev. Genet.53, 347–372 (2019). 10.1146/annurev-genet-112618-043905 [DOI] [PubMed] [Google Scholar]

[CR15] 15.Buckler, E. S. 4th et al. Meiotic drive of chromosomal knobs reshaped the maize genome. Genetics153, 415–426 (1999). 10.1093/genetics/153.1.415 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR16] 16.Dawe, R. K. et al. A kinesin-14 motor activates neocentromeres to promote meiotic drive in maize. Cell173, 839–850.e18 (2018). 10.1016/j.cell.2018.03.009 [DOI] [PubMed] [Google Scholar]

[CR17] 17.Lyon, M. F. Transmission ratio distortion in mice. Annu. Rev. Genet.37, 393–408 (2003). 10.1146/annurev.genet.37.110801.143030 [DOI] [PubMed] [Google Scholar]

[CR18] 18.McDermott, S. R. & Noor, M. A. F. The role of meiotic drive in hybrid male sterility. Phil. Trans. R. Soc. B365, 1265–1272 (2010). 10.1098/rstb.2009.0264 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Herrmann, B. G., Koschorz, B., Wertz, K., McLaughlin, K. J. & Kispert, A. A protein kinase encoded by the t complex responder gene causes non-Mendelian inheritance. Nature402, 141–146 (1999). 10.1038/45970 [DOI] [PubMed] [Google Scholar]

[CR20] 20.Bauer, H., Willert, J., Koschorz, B. & Herrmann, B. G. The t complex-encoded GTPase-activating protein Tagap1 acts as a transmission ratio distorter in mice. Nat. Genet.37, 969–973 (2005). 10.1038/ng1617 [DOI] [PubMed] [Google Scholar]

[CR21] 21.Hartl, D. L. Genetic dissection of segregation distortion. I. Suicide combinations of SD genes. Genetics76, 477–486 (1974). 10.1093/genetics/76.3.477 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Larracuente, A. M. & Presgraves, D. C. The selfish segregation distorter gene complex of Drosophila melanogaster. Genetics192, 33–53 (2012). 10.1534/genetics.112.141390 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Zanders, S. E. et al. Genome rearrangements and pervasive meiotic drive cause hybrid infertility in fission yeast. eLife3, e02630 (2014). 10.7554/eLife.02630 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Nuckolls, N. L. et al. wtf Genes are prolific dual poison–antidote meiotic drivers. eLife6, e26033 (2017). 10.7554/eLife.26033 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Lewontin, R. C. & Dunn, L. C. The evolutionary dynamics of a polymorphism in the house mouse. Genetics45, 705–722 (1960). 10.1093/genetics/45.6.705 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Hurst, L. D. & Pomiankowski, A. Causes of sex ratio bias may account for unisexual sterility in hybrids: a new explanation of Haldane’s rule and related phenomena. Genetics128, 841–858 (1991). 10.1093/genetics/128.4.841 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Coughlan, J. M. The role of conflict in shaping plant biodiversity. New Phytol.10.1111/nph.19233 (2023). [DOI] [PMC free article] [PubMed]

[CR28] 28.Phadnis, N. & Orr, H. A. A single gene causes both male sterility and segregation distortion in Drosophila hybrids. Science323, 376–379 (2009). 10.1126/science.1163934 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Zhang, L., Sun, T., Woldesellassie, F., Xiao, H. & Tao, Y. Sex ratio meiotic drive as a plausible evolutionary mechanism for hybrid male sterility. PLoS Genet.11, e1005073 (2015). 10.1371/journal.pgen.1005073 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Kermicle, J. L. & Allen, J. P. Cross-incompatibility between maize and teosinte. Maydica35, 399–408 (1990). [Google Scholar]

[CR31] 31.Lu, Y., Hokin, S. A., Kermicle, J. L., Hartwig, T. & Evans, M. M. S. A pistil-expressed pectin methylesterase confers cross-incompatibility between strains of Zea mays. Nat. Commun.10, 2304 (2019). 10.1038/s41467-019-10259-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Hufford, M. B. et al. The genomic signature of crop-wild introgression in maize. PLoS Genet.9, e1003477 (2013). 10.1371/journal.pgen.1003477 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Rojas-Barrera, I. C. et al. Contemporary evolution of maize landraces and their wild relatives influenced by gene flow with modern maize varieties. Proc. Natl Acad. Sci. USA116, 21302–21311 (2019). 10.1073/pnas.1817664116 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR34] 34.Wang, C. et al. A natural gene drive system confers reproductive isolation in rice. Cell186, 3577–3592.e18 (2023). 10.1016/j.cell.2023.06.023 [DOI] [PubMed] [Google Scholar]

[CR35] 35.Yang, Z. & Bielawski, J. P. Statistical methods for detecting molecular adaptation. Trends Ecol. Evol.15, 496–503 (2000). 10.1016/S0169-5347(00)01994-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Yoshikawa, M., Peragine, A., Park, M. Y. & Poethig, R. S. A pathway for the biogenesis of trans-acting siRNAs in Arabidopsis. Genes Dev19, 2164–2175 (2005). 10.1101/gad.1352605 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Parent, J.-S., Bouteiller, N., Elmayan, T. & Vaucheret, H. Respective contributions of Arabidopsis DCL2 and DCL4 to RNA silencing. Plant J.81, 223–232 (2015). 10.1111/tpj.12720 [DOI] [PubMed] [Google Scholar]

[CR38] 38.Deleris, A. et al. Hierarchical action and inhibition of plant Dicer-like proteins in antiviral defense. Science313, 68–71 (2006). 10.1126/science.1128214 [DOI] [PubMed] [Google Scholar]

[CR39] 39.Bouché, N., Lauressergues, D., Gasciolli, V. & Vaucheret, H. An antagonistic function for Arabidopsis DCL2 in development and a new function for DCL4 in generating viral siRNAs. EMBO J.25, 3347–3356 (2006). 10.1038/sj.emboj.7601217 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Wu, Y.-Y. et al. DCL2- and RDR6-dependent transitive silencing of SMXL4 and SMXL5 in Arabidopsis dcl4 mutants causes defective phloem transport and carbohydrate over-accumulation. Plant J.90, 1064–1078 (2017). 10.1111/tpj.13528 [DOI] [PubMed] [Google Scholar]

[CR41] 41.Taochy, C. et al. A genetic screen for impaired systemic RNAi highlights the crucial role of DICER-LIKE 2. Plant Physiol.175, 1424–1437 (2017). 10.1104/pp.17.01181 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Mlotshwa, S. et al. DICER-LIKE2 plays a primary role in transitive silencing of transgenes in Arabidopsis. PLoS ONE3, e1755 (2008). 10.1371/journal.pone.0001755 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Tagami, Y., Motose, H. & Watanabe, Y. A dominant mutation in DCL1 suppresses the hyl1 mutant phenotype by promoting the processing of miRNA. RNA15, 450–458 (2009). 10.1261/rna.1297109 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Welker, N. C. et al. Dicer’s helicase domain discriminates dsRNA termini to promote an altered reaction mode. Mol. Cell41, 589–599 (2011). 10.1016/j.molcel.2011.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Aderounmu, A. M., Aruscavage, P. J., Kolaczkowski, B. & Bass, B. L. Ancestral protein reconstruction reveals evolutionary events governing variation in Dicer helicase function. eLife12, e85120 (2023). 10.7554/eLife.85120 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Slotkin, R. K., Freeling, M. & Lisch, D. Heritable transposon silencing initiated by a naturally occurring transposon inverted duplication. Nat. Genet.37, 641–644 (2005). 10.1038/ng1576 [DOI] [PubMed] [Google Scholar]

[CR47] 47.Bhutani, K. et al. Widespread haploid-biased gene expression enables sperm-level natural selection. Science371, eabb1723 (2021). 10.1126/science.abb1723 [DOI] [PubMed] [Google Scholar]

[CR48] 48.Shan, X. et al. Mobilization of the active MITE transposons mPing and Pong in rice by introgression from wild rice (Zizania latifolia Griseb.). Mol. Biol. Evol.22, 976–990 (2005). 10.1093/molbev/msi082 [DOI] [PubMed] [Google Scholar]

[CR49] 49.Ding, L.-N. et al. Advances in plant GDSL lipases: from sequences to functional mechanisms. Acta Physiol. Plant41, 151 (2019). 10.1007/s11738-019-2944-4 [DOI] [Google Scholar]

[CR50] 50.An, X. et al. ZmMs30 encoding a novel GDSL lipase is essential for male fertility and valuable for hybrid breeding in maize. Mol. Plant12, 343–359 (2019). 10.1016/j.molp.2019.01.011 [DOI] [PubMed] [Google Scholar]

[CR51] 51.Huo, Y. et al. IRREGULAR POLLEN EXINE2 encodes a GDSL lipase essential for male fertility in maize. Plant Physiol.184, 1438–1454 (2020). 10.1104/pp.20.00105 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Zhao, J. et al. RMS2 encoding a GDSL lipase mediates lipid homeostasis in anthers to determine rice male fertility. Plant Physiol.182, 2047–2064 (2020). 10.1104/pp.19.01487 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Tsugama, D., Fujino, K., Liu, S. & Takano, T. A GDSL-type esterase/lipase gene, GELP77, is necessary for pollen dissociation and fertility in Arabidopsis. Biochem. Biophys. Res. Commun.526, 1036–1041 (2020). 10.1016/j.bbrc.2020.03.179 [DOI] [PubMed] [Google Scholar]

[CR54] 54.Wu, H. et al. Plant 22-nt siRNAs mediate translational repression and stress adaptation. Nature581, 89–93 (2020). 10.1038/s41586-020-2231-y [DOI] [PubMed] [Google Scholar]

[CR55] 55.Borges, F. & Martienssen, R. A. The expanding world of small RNAs in plants. Nat. Rev. Mol. Cell Biol.16, 727–741 (2015). 10.1038/nrm4085 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR56] 56.Fang, X. & Qi, Y. RNAi in plants: an Argonaute-centered view. Plant Cell28, 272–285 (2016). 10.1105/tpc.15.00920 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Axtell, M. J., Westholm, J. O. & Lai, E. C. Vive la différence: biogenesis and evolution of microRNAs in plants and animals. Genome Biol.12, 221 (2011). 10.1186/gb-2011-12-4-221 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Manavella, P. A., Koenig, D. & Weigel, D. Plant secondary siRNA production determined by microRNA-duplex structure. Proc. Natl Acad. Sci. USA109, 2461–2466 (2012). 10.1073/pnas.1200169109 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Nelms, B. & Walbot, V. Gametophyte genome activation occurs at pollen mitosis I in maize. Science375, 424–429 (2022). 10.1126/science.abl7392 [DOI] [PubMed] [Google Scholar]

[CR60] 60.Wongpalee, S. P. et al. CryoEM structures of Arabidopsis DDR complexes involved in RNA-directed DNA methylation. Nat. Commun.10, 3916 (2019). 10.1038/s41467-019-11759-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Jauvion, V., Rivard, M., Bouteiller, N., Elmayan, T. & Vaucheret, H. RDR2 partially antagonizes the production of RDR6-dependent siRNA in sense transgene-mediated PTGS. PLoS ONE7, e29785 (2012). 10.1371/journal.pone.0029785 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR62] 62.Creasey, K. M. et al. miRNAs trigger widespread epigenetically activated siRNAs from transposons in Arabidopsis. Nature508, 411–415 (2014). 10.1038/nature13069 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR63] 63.Romero Navarro, J. A. et al. A study of allelic diversity underlying flowering-time adaptation in maize landraces. Nat. Genet.49, 476–480 (2017). 10.1038/ng.3784 [DOI] [PubMed] [Google Scholar]

[CR64] 64.Chen, L. et al. Genome sequencing reveals evidence of adaptive variation in the genus Zea. Nat. Genet.54, 1736–1745 (2022). 10.1038/s41588-022-01184-y [DOI] [PubMed] [Google Scholar]

[CR65] 65.Lu, Y., Kermicle, J. L. & Evans, M. M. S. Genetic and cellular analysis of cross-incompatibility in Zea mays. Plant Reprod.27, 19–29 (2014). 10.1007/s00497-013-0236-5 [DOI] [PubMed] [Google Scholar]

[CR66] 66.Hartl, D. L. Population dynamics of sperm and pollen killers. Theor. Appl. Genet.42, 81–88 (1972). 10.1007/BF00277948 [DOI] [PubMed] [Google Scholar]

[CR67] 67.Sweigart, A. L., Brandvain, Y. & Fishman, L. Making a murderer: the evolutionary framing of hybrid gamete-killers. Trends Genet.35, 245–252 (2019). 10.1016/j.tig.2019.01.004 [DOI] [PubMed] [Google Scholar]

[CR68] 68.Bravo Núñez, M. A., Lange, J. J. & Zanders, S. E. A suppressor of a wtf poison–antidote meiotic driver acts via mimicry of the driver’s antidote. PLoS Genet.14, e1007836 (2018). 10.1371/journal.pgen.1007836 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR69] 69.Barnes, A. C. et al. An adaptive teosinte mexicana introgression modulates phosphatidylcholine levels and is associated with maize flowering time. Proc. Natl Acad. Sci. USA119, e2100036119 (2022). 10.1073/pnas.2100036119 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR70] 70.McClintock, B., Kato Yamakake, T. A., Blumenschein, A. & Escuela Nacional de Agricultura (Mexico). Chromosome Constitution of Races of Maize: Its Significance in the Interpretation of Relationships between Races and Varieties in the Americas (Colegio de Postgraduados, 1981).

[CR71] 71.Borges, F. et al. Transposon-derived small RNAs triggered by miR845 mediate genome dosage response in Arabidopsis. Nat. Genet.50, 186–192 (2018). 10.1038/s41588-017-0032-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR72] 72.Martinez, G. et al. Paternal easiRNAs regulate parental genome dosage in Arabidopsis. Nat. Genet.50, 193–198 (2018). 10.1038/s41588-017-0033-4 [DOI] [PubMed] [Google Scholar]

[CR73] 73.Durand, E. et al. Dominance hierarchy arising from the evolution of a complex small RNA regulatory network. Science346, 1200–1205 (2014). 10.1126/science.1259442 [DOI] [PubMed] [Google Scholar]

[CR74] 74.Czech, B. et al. An endogenous small interfering RNA pathway in Drosophila. Nature453, 798–802 (2008). 10.1038/nature07007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR75] 75.Wen, J. et al. Adaptive regulation of testis gene expression and control of male fertility by the Drosophila hairpin RNA pathway. Mol. Cell57, 165–178 (2015). 10.1016/j.molcel.2014.11.025 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR76] 76.Tao, Y. et al. A sex-ratio meiotic drive system in Drosophila simulans. II: an X-linked distorter. PLoS Biol.5, e293 (2007). 10.1371/journal.pbio.0050293 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Lin, C.-J. et al. The hpRNA/RNAi pathway is essential to resolve intragenomic conflict in the Drosophila male germline. Dev. Cell46, 316–326.e5 (2018). 10.1016/j.devcel.2018.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR78] 78.Flemr, M. et al. A retrotransposon-driven Dicer isoform directs endogenous small interfering RNA production in mouse oocytes. Cell155, 807–816 (2013). 10.1016/j.cell.2013.10.001 [DOI] [PubMed] [Google Scholar]

[CR79] 79.Tam, O. H. et al. Pseudogene-derived small interfering RNAs regulate gene expression in mouse oocytes. Nature453, 534–538 (2008). 10.1038/nature06904 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR80] 80.Su, R. et al. Global profiling of RNA-binding protein target sites by LACE-seq. Nat. Cell Biol.23, 664–675 (2021). 10.1038/s41556-021-00696-9 [DOI] [PubMed] [Google Scholar]

[CR81] 81.Begcy, K. & Dresselhaus, T. Tracking maize pollen development by the leaf collar method. Plant Reprod.30, 171–178 (2017). 10.1007/s00497-017-0311-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR82] 82.Bass, H. W. et al. A maize root tip system to study DNA replication programmes in somatic and endocycling nuclei during plant development. J. Exp. Bot.65, 2747–2756 (2014). 10.1093/jxb/ert470 [DOI] [PubMed] [Google Scholar]

[CR83] 83.Kalkar, S. A. & Neha, K. Evaluation of FDA staining technique in stored maize pollen. Middle East J. Sci. Res.12, 560–562 (2012).

[CR84] 84.Nagar, R. & Schwessinger, B. DNA size selection (>3–4 kb) and purification of DNA using an improved homemade SPRIbeads solution. Protocols.io10.17504/protocols.io.n7hdhj6 (2018).

[CR85] 85.Schalamun, M., Nagar, R. & Kainer, D. Harnessing the MinION: an example of how to establish long‐read sequencing in a laboratory using challenging plant tissue from Eucalyptus pauciflora. Mol. Ecol.10.1111/1755-0998.12938 (2018). [DOI] [PMC free article] [PubMed]

[CR86] 86.Kolmogorov, M., Yuan, J., Lin, Y. & Pevzner, P. A. Assembly of long, error-prone reads using repeat graphs. Nat. Biotechnol.37, 540–546 (2019). 10.1038/s41587-019-0072-8 [DOI] [PubMed] [Google Scholar]

[CR87] 87.Li, H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics34, 3094–3100 (2018). 10.1093/bioinformatics/bty191 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR88] 88.Shafin, K. et al. Haplotype-aware variant calling with PEPPER-Margin-DeepVariant enables high accuracy in nanopore long-reads. Nat. Methods18, 1322–1332 (2021). 10.1038/s41592-021-01299-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR89] 89.Vasimuddin, M., Misra, S., Li, H. & Aluru, S. Efficient architecture-aware acceleration of BWA-MEM for multicore systems. In 2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS) 314–324 (IEEE, 2019).

[CR90] 90.Hu, J., Fan, J., Sun, Z. & Liu, S. NextPolish: a fast and efficient genome polishing tool for long read assembly. Bioinformatics10.1093/bioinformatics/btz891 (2019). [DOI] [PubMed]

[CR91] 91.Aury, J.-M. & Istace, B. Hapo-G, haplotype-aware polishing of genome assemblies with accurate reads. NAR Genom. Bioinform.3, lqab034 (2021). 10.1093/nargab/lqab034 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR92] 92.Durand, N. C. et al. Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst.3, 95–98 (2016). 10.1016/j.cels.2016.07.002 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR93] 93.Dudchenko, O. et al. De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds. Science356, 92–95 (2017). 10.1126/science.aal3327 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR94] 94.Durand, N. C. et al. Juicebox provides a visualization system for Hi-C contact maps with unlimited zoom. Cell Syst.3, 99–101 (2016). 10.1016/j.cels.2015.07.012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR95] 95.Springer, N. M. et al. The maize W22 genome provides a foundation for functional genomics and transposon biology. Nat. Genet.50, 1282–1288 (2018). 10.1038/s41588-018-0158-0 [DOI] [PubMed] [Google Scholar]

[CR96] 96.Shumate, A. & Salzberg, S. L. Liftoff: accurate mapping of gene annotations. Bioinformatics37, 1639–1643 (2021). 10.1093/bioinformatics/btaa1016 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR97] 97.Rhie, A., Walenz, B. P., Koren, S. & Phillippy, A. M. Merqury: reference-free quality, completeness, and phasing assessment for genome assemblies. Genome Biol.21, 245 (2020). 10.1186/s13059-020-02134-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR98] 98.Manni, M., Berkeley, M. R., Seppey, M., Simão, F. A. & Zdobnov, E. M. BUSCO update: novel and streamlined workflows along with broader and deeper phylogenetic coverage for scoring of eukaryotic, prokaryotic, and viral genomes. Mol. Biol. Evol.38, 4647–4654 (2021). 10.1093/molbev/msab199 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR99] 99.Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal17, 10 (2011). 10.14806/ej.17.1.200 [DOI] [Google Scholar]

[CR100] 100.Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. Preprint at https://arxiv.org/abs/1303.3997 (2013).

[CR101] 101.Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics25, 2078–2079 (2009). 10.1093/bioinformatics/btp352 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR102] 102.Garrison, E. & Marth, G. Haplotype-based variant detection from short-read sequencing. Preprint at https://arxiv.org/abs/1207.3907 (2012).

[CR103] 103.Takagi, H. et al. QTL-seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant J.74, 174–183 (2013). 10.1111/tpj.12105 [DOI] [PubMed] [Google Scholar]

[CR104] 104.Danecek, P. et al. Twelve years of SAMtools and BCFtools. Gigascience10, giab008 (2021). 10.1093/gigascience/giab008 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR105] 105.Rubinacci, S., Ribeiro, D. M., Hofmeister, R. J. & Delaneau, O. Efficient phasing and imputation of low-coverage sequencing data using large reference panels. Nat. Genet.53, 120–126 (2021). 10.1038/s41588-020-00756-0 [DOI] [PubMed] [Google Scholar]

[CR106] 106.Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics29, 15–21 (2013). 10.1093/bioinformatics/bts635 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR107] 107.Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics30, 923–930 (2014). 10.1093/bioinformatics/btt656 [DOI] [PubMed] [Google Scholar]

[CR108] 108.Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics26, 139–140 (2009). 10.1093/bioinformatics/btp616 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR109] 109.Alexa, A. & Rahnenfuhrer, J. topGO: enrichment analysis for gene ontology. R package version 2.42.0 (2023).

[CR110] 110.Sayols, S. rrvgo: a Bioconductor package for interpreting lists of Gene Ontology terms. MicroPubl. Biol.10.17912/micropub.biology.000811 (2023). [DOI] [PMC free article] [PubMed]

[CR111] 111.Ramirez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res.44, 160–165 (2016). 10.1093/nar/gkw257 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR112] 112.Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods9, 357–359 (2012). 10.1038/nmeth.1923 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR113] 113.Axtell, M. J. ShortStack: comprehensive annotation and quantification of small RNA genes. RNA19, 740–751 (2013). 10.1261/rna.035279.112 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR114] 114.Gruber, A. R., Lorenz, R., Bernhart, S. H., Neuböck, R. & Hofacker, I. L. The Vienna RNA websuite. Nucleic Acids Res.36, W70–W74 (2008). 10.1093/nar/gkn188 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR115] 115.German, M. A., Luo, S., Schroth, G., Meyers, B. C. & Green, P. J. Construction of parallel analysis of RNA ends (PARE) libraries for the study of cleaved miRNA targets and the RNA degradome. Nat. Protoc.4, 356–362 (2009). 10.1038/nprot.2009.8 [DOI] [PubMed] [Google Scholar]

[CR116] 116.Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics26, 841–842 (2010). 10.1093/bioinformatics/btq033 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR117] 117.Dai, X., Zhuang, Z. & Zhao, P. X. psRNATarget: a plant small RNA target analysis server (2017 release). Nucleic Acids Res.46, W49–W54 (2018). 10.1093/nar/gky316 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR118] 118.Szpiech, Z. A. selscan 2.0: scanning for sweeps in unphased data. Bioinformatics40, btae006 (2024). [DOI] [PMC free article] [PubMed]

[CR119] 119.Danecek, P. et al. The variant call format and VCFtools. Bioinformatics27, 2156–2158 (2011). 10.1093/bioinformatics/btr330 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR120] 120.Grzybowski, M. W. et al. A common resequencing-based genetic marker data set for global maize diversity. Plant J.113, 1109–1121 (2023). 10.1111/tpj.16123 [DOI] [PubMed] [Google Scholar]

[CR121] 121.Yang, N. et al. Two teosintes made modern maize. Science382, eadg8940 (2023). [DOI] [PubMed]

[CR122] 122.Browning, B. L., Tian, X., Zhou, Y. & Browning, S. R. Fast two-stage phasing of large-scale sequence data. Am. J. Hum. Genet.108, 1880–1890 (2021). 10.1016/j.ajhg.2021.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR123] 123.Portwood, J. L. II et al. MaizeGDB 2018: the maize multi-genome genetics and genomics database. Nucleic Acids Res.47, D1146–D1154 (2019). 10.1093/nar/gky1046 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR124] 124.Stitzer, M. C. & Ross-Ibarra, J. Maize domestication and gene interaction. New Phytol.220, 395–408 (2018). 10.1111/nph.15350 [DOI] [PubMed] [Google Scholar]

[CR125] 125.Walley, J. W. et al. Integration of omic networks in a developmental atlas of maize. Science353, 814–818 (2016). 10.1126/science.aag1125 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR126] 126.Liu, L. & Li, J. Communications between the endoplasmic reticulum and other organelles during abiotic stress response in plants. Front. Plant Sci.10, 749 (2019). 10.3389/fpls.2019.00749 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR127] 127.Taurino, M. et al. SEIPIN proteins mediate lipid droplet biogenesis to promote pollen transmission and reduce seed dormancy. Plant Physiol.176, 1531–1546 (2018). 10.1104/pp.17.01430 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR128] 128.Beissinger, T. M. et al. Recent demography drives changes in linked selection across the maize genome. Nat. Plants2, 16084 (2016). 10.1038/nplants.2016.84 [DOI] [PubMed] [Google Scholar]

PERMALINK

Teosinte Pollen Drive guides maize diversification and domestication by RNAi

Benjamin Berube

Evan Ernst

Jonathan Cahn

Benjamin Roche

Cristiane de Santis Alves

Jason Lynn

Armin Scheben

Daniel Grimanelli

Adam Siepel

Jeffrey Ross-Ibarra

Jerry Kermicle

Robert A Martienssen

Abstract

Main

TPD in maize hybrids

Fig. 1. Single-pollen sequencing reveals selfish inheritance in TPD.

Extended Data Fig. 1. Teosinte Pollen Drive and genetic mapping of Tpd1 and Tpd2.

A Dicer-Like 2 toxin–antidote complex

Fig. 2. A toxin–antidote system introduced from mexicana on chromosomes 5 and 6.

Extended Data Fig. 2. mexicana intervals introgressed into maize carry RNAi genes.

Fig. 3. Dcl2 from teosinte is a linked antidote for toxic 22-nt siRNA.

22-nt small RNAs target a pollen lipase

Extended Data Fig. 3. DCL2-dependent 22nt siRNAs from hairpins are prevalent in maize pollen.

Extended Data Fig. 4. Validation of highly abundant pollen hairpin precursors.

Extended Data Fig. 5. Origins and targets of 22 nt small RNAs in TPD Pollen.

Extended Data Fig. 6. TPD-dependent silencing of a GDSL lipase disrupts lipid metabolism.

Extended Data Fig. 7. Tpd1 and Dcl2 are expressed pre-meiotically, whereas Tdr1 is expressed in microspore and pollen.

Hairpin siRNAs trigger pollen abortion

Fig. 4. 22-nt siRNAs from a mexicana-derived hairpin (Tpd1) target Tdr1, an essential pollen gene.

Extended Data Fig. 8. Tpd2 suppresses 22nt secondary small RNAs.

TPD, RNAi and the origin of modern maize

Extended Data Fig. 9. Signatures of Teosinte Pollen Drive in modern maize, maize traditional varieties and sympatric mexicana populations.

Fig. 5. Tpd1 hp-siRNA target site deletion in tdr1 has spread to modern maize from teosinte.

Discussion

Extended Data Fig. 10. Mechanistic model of Teosinte Pollen Drive.

Extended Data Fig. 11. Evolutionary model of Teosinte Pollen Drive.

Methods

Plant material and growth conditions

Phenotyping and microscopy

Genotyping and marker design

High-molecular-weight genomic DNA extraction

Nanopore and Hi-C sequencing, TPD genome assembly and annotation

RNA extraction

Reverse transcription and RT–qPCR

Whole-genome sequencing and SNP calling

Single-pollen grain sequencing

RNA sequencing and analysis

Small RNA sequencing and analysis

iPARE-seq and analysis

Protein extraction and western blotting

Esterase enzymatic activity assay

Detection of selective sweeps in candidate regions associated with TPD

Reporting summary

Online content

Supplementary information

Acknowledgements

Extended data figures and tables

Author contributions

Peer review

Peer review information

Data availability

Code availability

Competing interests

Footnotes

Extended data

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases