Skip to main content
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2026 May 13;123(20):e2535316123. doi: 10.1073/pnas.2535316123

PAM1 regulates meiosis by coupling RNA processing to the chromosome axis

Qian Du a,1,2, Minghui Wang a,b,3, Choon-Lin Tiang a,4, Moira J Sheehan a,5, Paul Altendorf c,6, Ju-Kyung Yu c,7, Otto Hudecz d, Elisabeth Roitinger d, Chung-Ju Rachel Wang e, Robert Bukowski a,b, Robert B Meeley f, Clint Ko a,8, Inna N Golubovskaya g, Wojciech P Pawlowski a,2
PMCID: PMC13187813  PMID: 42127116

Significance

Meiosis is an evolutionarily conserved process, in which chromosomes interact with each other and exchange parts. This behavior is distinct during organism’s development and requires activity of many specialized genes that must be turned on during meiosis. We identified a gene called Pam1, which controls the activity of meiotic genes in maize in an unexpected and unique way. The protein encoded by Pam1 attaches to a meiosis-specific component of the chromosome scaffold, binds to RNAs transcribed from a large number of genes, and directs their processing. RNA-binding proteins have also been found to control meiosis in mammals and yeast, but, interestingly, they do not share the same evolutionary origin. Instead, they evolved independently to fulfill very similar functions.

Keywords: maize, meiosis, chromosomes, RNA-binidng proteins

Abstract

Meiosis is a fundamental process responsible for sexual reproduction and generating genetic diversity in the progeny. Its successful completion requires fine-tuning of expression programs of many genes: promoting expression of genes involved in meiotic processes and suppressing genes whose expression may interfere with meiosis. Molecular mechanisms involved in meiotic transcriptome regulation and controlling meiosis progression vary between plants, animals, and fungi and remain elusive. We found that the Plural abnormalities of meiosis1 (Pam1) gene in maize controls meiosis progression by tethering transcriptome processing to the meiosis-specific chromosome axis. Pam1 encodes an RNA binding protein that becomes associated with chromosomes during early meiotic prophase I, binds transcripts of a large number of meiosis-related genes, and affects their splicing by interacting with the CCR4-NOT RNA processing protein complex. Disrupting Pam1 function results in a wide array of severe meiosis defects affecting chromosome condensation and dynamics, nuclear envelope and cytoskeleton organization, as well as the overall meiosis progression. Pam1 controls only a subset of meiotic genes and processes, indicating that several programs directing transcriptome architecture collectively regulate meiosis. RNA-binding proteins have been found to control meiosis progression in fungi and animals, and it is now shown to be also the case in plants. Interestingly, these proteins all exhibit distinct modes of action and evolutionary origins, presenting a remarkable case of convergent evolution. Uncovering mechanisms controlling meiosis progression should enable engineering meiosis to benefit crop improvement efforts.


Meiosis involves a large number of cellular processes that are unique during organism’s development (1). In particular, specialized events occurring during meiotic prophase I are crucial for meiosis outcomes in creating genetic diversity in offspring while maintaining genome stability. These events, including programmed DNA double-strand break (DSB) formation, meiotic recombination, as well as homologous chromosome pairing and synapsis, are unlike anything cells encounter at other points in development (2). Thus, the transition to execute meiotic processes requires highly orchestrated changes in transcriptome architecture to first allow induction of meiotic genes that might be repressed in vegetative tissues (3) and then turning off these genes after meiosis is complete (4–6). In addition, meiotic cells must suppress the expression of genes required for premeiotic development but whose continuous expression during meiosis could interfere with meiosis progression. Evidence from several species suggests a tight and intricate regulation of meiotic progression by transcriptional, posttranscriptional, and epigenetic mechanisms (5–8).

Although the core processes of meiosis exhibit strong evolutionary conservation, the regulatory mechanisms controlling meiosis initiation and progression are different among the major groups of eukaryotes (1, 3, 6, 9, 10). The differences encompass both distinct regulatory proteins as well as divergent genetic pathways that facilitate meiosis initiation and progression (8, 11). Nonetheless, a common theme in the regulatory mechanism found in animals and fungi is the involvement of RNA binding proteins (RBP) (8, 12–16). RBPs control the dynamic regulation of the large number of RNAs involved in meiosis through mRNA degradation, stabilization, and processing (17). Interestingly, the meiosis-controlling RBPs in animals and fungi do not show a close evolutionary relationship and have distinct modes of action even though they share the similar function of controlling transcript abundance. In fission yeast, the Mmi1 protein selectively targets a subset of meiosis-specific transcripts for degradation during vegetative growth (12). To enable the initiation and progression of meiosis, the transcript of Mmi1 is sequestered by another RNA binding protein, Mei2 (13). In mammals and Drosophila, the mRNA stability factor MARF1 regulates meiosis progression by reducing transcript levels of a diverse group of genes in oocytes (14, 15, 18).

In plants, several genes, including Am1 in maize, as well as MEL1 and MEL2 in rice, have been implicated in meiosis initiation and progression, as mutants defective in these genes fail to enter meiosis (19–23). AM1 is a poorly studied protein with a central coiled-coil domain (19) while MEL1 is an Argonaute homolog involved in the biogenesis of phased small RNAs (phasiRNAs) (24). Another plant meiosis regulator MEL2 has been shown to directly bind RNAs (22) and premeiotically associate with cytoplasmic ribonucleoprotein granules involved in posttranscriptional gene regulation (23). Other genes, SWI/DYAD in Arabidopsis (25) and LEPTO1 in rice (26) have been found to facilitate proper progression of early meiosis but the exact effects of these genes remain poorly understood. Altogether, no molecular mechanisms controlling initiation and progression in plants have been uncovered.

To close this knowledge gap, we studied meiotic prophase I regulation in maize. We found that the maize homolog of MEL2, encoded by the Plural abnormalities of meiosis1 (Pam1) gene (27), regulates meiosis progression through a unique mechanism. The PAM1 protein is recruited to the meiotic chromosome axis and controls splicing of a large number of meiotic genes via the CCR4-NOT regulatory complex. The Pam1 gene has been reported to impact several meiotic processes, including early prophase chromosome dynamics, synapsis, and the synchrony of meiotic prophase progression among meiocytes within an anther (27). It is also thought to affect the formation of telomere bouquet (27), a cytological structure produced by telomeres attaching to the nuclear envelope in early zygotene (28). We found that role of Pam1 extends to events beyond the nucleus, including the establishment of meiosis-specific cytoskeleton. Altogether, our observations show that even though PAM1 plays a role similar to this of known meiosis-controlling RNA binding proteins of animals and fungi, it does it by engaging in a unique molecular mechanism, presenting a remarkable case of convergent evolution.

Results

Meiosis Progression Defects in the pam1 Mutant.

To fully elucidate the impact of the Pam1 gene on meiotic prophase I, we performed a comprehensive cytological analysis of the pam1 mutant. We first examined chromosome morphology by measuring chromosome width in Structured Illumination Microscopy. We found chromosome the width increasing during the transition from zygotene to pachytene in both wild type and mutant, (Fig. 1 A and B), reflecting the progressing chromosome condensation. However, chromosomes of pam1 were 30% thinner than wild-type chromosomes at both zygotene and pachytene, suggesting presence of significant abnormalities in chromosome condensation.

Fig. 1.

A three-panel figure with micrographs in A and C and a box plot in B comparing wild type and pam1 cells.

Meiotic prophase defects in the pam1 mutant. (A) Chromosome condensation defects in the pam1 mutant observed during prophase I using Structured Illumination-Microscopy. Chromosomes were stained with DAPI. (Scale bar, 5 μm.) (B) Quantification of chromatin condensation by measuring chromosome width. *** = significantly different according to Student’s t test (P = 0.01). Boxplot whiskers represent SE. The numbers of measurements and the numbers of cells analyzed were 220 and 15 for wild type in zygotene, 221 and 14 for wild type pachytene, 177 and 12 for pam1 in zygotene, and 400 and 20 for pam1 in pachytene, respectively. (C) Immunolocalization of tubulin in wild type and pam1 showing cytoskeleton organization differences. (Scale bar, 5 μm.)

In addition, we examined chromosome dynamics during meiotic prophase I, by following the patterns of chromosome pairing using fluorescence in situ hybridization (FiSH) with probes detecting the centromere of chromosome 4 (Cent4) and the subtelomerically located 5S rRNA locus on chromosome 2 (SI Appendix, Fig. S1A). We found that in wild-type meiocytes, the completion of Cent4 pairing followed 5S rRNA locus pairing (SI Appendix, Fig. S1A). This pattern was consistent with pairing being initiated in the subtelomeric region and progressing toward the centromere, a pattern thought to be facilitated by the presence of the telomere bouquet (28). In the pam1 mutant, the overall telomere-to-centromere chromosome pairing pattern was preserved. However, pairing of both Cent4 and 5S loci was delayed compared to wild type and remained incomplete at the end of pachytene (SI Appendix, Fig. S1A). These observations indicated that progression of chromosome pairing was disrupted in the pam1 mutant.

We also examined the progression of chromosome synapsis by monitoring the installation of ZYP1 (29), the central element protein of the synaptonemal complex. We found that synapsis was incomplete in the pam1 mutant, as many discontinuous ZYP1 fragments were present at the end of zygotene (SI Appendix, Fig. S1B).

We further investigated whether the role of Pam1 extends beyond chromosomes by analyzing the nuclear envelope. To do this, we first immunolocalized nuclear pores during meiotic prophase to track their behavior. We found that while in wild-type meiocytes nuclear pores clustered opposite the telomere bouquet, consistent with prior observations in rye (30), in pam1 meiocytes, the clustering was incomplete (SI Appendix, Fig. S1B). We also examined cytoskeleton installation during prophase I by applying an antibody against beta-tubulin. In wild-type meiocytes, beta-tubulin signals appeared fairly homogeneous throughout the cytoplasm, being organized into a reticular network (Fig. 1C). In contrast, pam1 mutant meiocytes exhibited a distinct tubulin organization, with pronounced tubulin bands around the nucleus stretching into the cytoplasm and accompanied by a discontinuous cortical meshwork at the cytoplasmic periphery (Fig. 1C). These patterns indicate that the Pam1 gene is essential for proper spatial organization of the meiotic cytoskeleton.

Disrupting meiocyte cytoskeleton in maize has been shown to disturb prophase I chromosome movements (31). These movements, described in a variety of species, are thought to facilitate homologous chromosome pairing, synaptonemal complex formation, and resolution of chromosome entanglements (interlocks) (32). Our live imaging observations conducted using MultiPhoton Excitation microscopy revealed a general decrease of chromosome mobility in pam1 mutant meiocytes compared to wild type in zygotene and pachytene, including the absence of rotational motions of the entire chromatin and a significant slowdown of the movement of individual chromosome segments (SI Appendix, Fig. S2 A and B). Moreover, we noticed that live pam1 meiocytes showed diffused chromatin appearance, corroborating our observations of a chromosome condensation defect in the mutant.

To assess how the collective effects of pam1 mutant defects impact the overall meiosis outcome, we examined pollen viability and found that it was reduced in the mutant to about 30% of the wild-type pollen viability (SI Appendix, Fig. S3A). On the other hand, there were no significant differences in the number of pollen grains between the mutant and wild type (SI Appendix, Fig. S3B), indicating that the numbers of cells entering meiosis in the mutant and wild type were not different. We further used 5-ethynyl-2’-deoxyuridine (EdU) staining to (33) examine whether premeiotic S phase progression was altered in the mutant but found no obvious signs of abnormalities (SI Appendix, Fig. S4) compared to the DNA replication patterns observed in wild-type maize anthers (33). These data suggest that Pam1 has a major role in meiosis but, in contrast, its impact on premeiotic processes is none or relatively minor.

Taken together, our observations of meiocyte development in the pam1 mutant suggest a broad role of Pam1 in controlling meiosis progression beyond its previously described role in controlling chromosome behavior during bouquet formation.

Localization of PAM1 to Meiotic Chromosomes.

To elucidate the mode of Pam1 action, we cloned the gene using a combination of a map-based approach and bulk-segregant sequencing (see Methods for details). We found that Pam1 corresponded to a gene annotated as Zm00001d030741. A guanine to adenine transition at the 3’ splice acceptor site of intron 3 of the gene was responsible for the pam1 mutation (SI Appendix, Fig. S5). To confirm Pam1 cloning, a second mutant allele, pam1-2, carrying an insertion of a Mutator transposon in the Pam1 coding region was identified and found to display similar meiosis defects as the original pam1-1 allele (SI Appendix, Fig. S6, see Methods for details).

We found that Pam1 encodes a putative RNA binding protein containing two ankyrin repeats at its N terminus, two nucleotide-binding domains in the middle part of the protein, including a putative RNA binding domain harboring an RNA recognition motif (RRM) and a LOTUS/OST-HTH domain, and two RING finger domains at the C-terminus (SI Appendix, Fig. S7). Phylogeny reconstruction showed that PAM1 belonged to a protein family with members present throughout the plant kingdom but exhibiting divergent domain architecture (SI Appendix, Fig. S8). The only previously studied member of the family has been rice MEL2 (21–23).

To study the subcellular localization of PAM1, we developed an antibody against the full-length PAM1 protein. In immunolocalization experiments using three-dimensional deconvolution microscopy, we observed PAM1 staining in premeiotic interphase and early prophase meiocytes (Fig. 2A and SI Appendix, Fig. S9) but not in somatic cells. In meiocytes, the PAM1 signal was present in the nucleus from leptotene to pachytene, although in leptotene some presence of PAM1 was also detected in the cytoplasm (Fig. 2A). There was no PAM1 signal in pam1 mutant meiocytes (SI Appendix, Fig. S10). Altogether, these observations suggested that PAM1 is a putative multidomain RNA binding protein present on chromosomes during the leptotene-pachytene period of meiotic prophase I.

Fig. 2.

A four-panel figure labeled A to D with micrographs in A and D, a bar graph in B, and a yeast two-hybrid assay results in C.

PAM1 uses its N-terminal ANK domain to interact with chromosome axis component ASY1. (A) Immunodetection of PAM1 and ASY1 showing their colocalization in wild-type meiocytes. (Scale bar, 5 μm.) (B) Quantification of PAM1 protein abundance with mass spectrometry in immunoprecipitation products obtained using the anti-ASY1 antibody. G = rabbit IgG (negative control), L = leptotene, Z = zygotene, P = pachytene, D = diplotene. Values for each stage are means of two to four biological replicates. Bars marked with different letters indicate significant differences according to the Newman–Keuls test (P = 0.05). (C) Verification of PAM1 interaction with ASY1 using yeast two-hybrid assays. Interaction is indicated by yeast cell growth on the TDO/SD (SD-Trp-Leu-His) medium of undiluted cultures and cultures at the 1:10 dilution compared to controls grown on the DDO/SD (SD-Trp-Leu) medium. (D) Verification of PAM1 interaction with ASY1 using the Bimolecular Fluorescence Complementation assay. The experiments were repeated three times. (Scale bar, 20 μm.)

Interaction of PAM1 with the Chromosome Axis.

To investigate the localization of PAM1 in more detail, we performed double immunostaining with antibodies against PAM1 and ASY1, a chromosome axis-associated HORMA domain protein (34). We found that PAM1 signals colocalized with ASY1 whenever PAM1 was present on meiotic chromosomes (Fig. 2A). Furthermore, the first appearance of PAM1 on chromosomes during leptotene was coincident with the loading of ASY1 onto chromosomes. ASY1 staining in pam1 mutant meiocytes was identical to that in the wild type, indicating that ASY1 loading was not dependent on the presence of PAM1 (SI Appendix, Fig. S10). In contrast, PAM1 staining was absent from asy1 mutant meiocytes (SI Appendix, Fig. S11), indicating that ASY1 was required for the presence of PAM1 on meiotic chromosomes.

To confirm the association between PAM1 and ASY1, we used Liquid Chromatography/Tandem Mass Spectrometry (LC/MSMS) to analyze proteins coimmunoprecipitated with ASY1 from extracts of wild-type anthers of the B73 inbred at different meiotic prophase substages. Based on peptide spectrum match (PSM) analyses, we found that at the leptotene stage PAM1 was significantly enriched compared to the rabbit IgG control (Fig. 2B), indicating that it is recruited to the chromosome at that time.

To further validate the interaction between PAM1 and ASY1, and identify the specific region of PAM1 that binds to ASY1, we used Bimolecular Fluorescence Complementation (BiFC) in Nicotiana benthamiana leaves and yeast two-hybrid assays (Fig. 2 C and D). These analyses revealed that ASY1 exhibited interaction with the full-length PAM1 protein but not with a truncated PAM1 missing the ankyrin repeat domain (PAM1ΔANK). This result was consistent with ankyrin domains being known to act in protein–protein interactions (35).

Taken together, these data indicated that PAM1 loads onto the chromosome axis in early meiotic prophase I by associating with ASY1 via its N-terminal ankyrin domain.

Role of PAM1 in the Control of Splicing of Meiotic Genes.

The presence of the RRM domain in PAM1 indicated that the protein is likely to bind RNAs and, in fact, RNA binding capacity has been identified in the rice homolog of PAM1, MEL2 (22). To confirm that PAM1 indeed binds RNA in vivo and identify the RNA species associating with PAM1, we conducted the native RNA immunoprecipitation (RNAIP) analysis (see Methods for details). To do it, we incubated the anti-PAM1 antibody with extracts from anthers containing meiocytes at the stages from leptotene to pachytene from the A344 inbred, the isogenic background of the pam1 mutant. In comparison to the negative control of rabbit IgG, we found that the PAM1 protein bound transcripts of 1582 genes at the log2-fold change ≥ 1 and the false discovery rate (FDR) < 0.05 (Dataset S1). These genes were located in different parts of the maize genome, exhibiting an overall distribution generally resembling local gene density patterns (SI Appendix, Fig. S12A). Gene Ontology (GO) analysis showed the terms most significantly enriched among the PAM1-associated genes to be “RNA processing”, “meiosis”, “meiosis cell cycle”, and “double-strand break repair” (Fig. 3A), in line with the pam1 mutant phenotype. We also found 14 cytoskeleton-related genes (Dataset S1), which could explain the cytoskeleton defects in the mutant. The transcript sites associated with the PAM1 protein were predominantly exons (58%) and UTRs (22%) (SI Appendix, Fig. S12B). Several nucleotide sequence motifs were enriched among the PAM1-transcripts (SI Appendix, Fig. S12C), likely representing PAM1 binding motifs.

Fig. 3.

Four panels show A a horizontal bar graph, B a volcano plot, C fluorescence microscopy images, and D genome viewer tracks.

PAM1 binds to multiple meiotic gene transcripts and facilitates their correct splicing. (A) Gene Ontology analysis of genes whose transcript were bound to PAM1 in the RNA immunoprecipitation (RNAIP) analysis showing enrichment in RNA processing and meiosis-related genes. (B) Volcano plot of transcripts identified in the RNAIP analysis of meiotic anthers with the anti-PAM1 antibody. Each dot represents a gene. Selected meiosis-related genes identified by RNAIP are listed. Enriched = transcripts overrepresented in PAM1 RNAIP compared to IgG control, Depleted = transcripts underrepresented in PAM1 RNAIP compared to IgG control, Nonspecific = transcripts without statistically significant change in abundance (FDR > 0.05). (C) Immunolocalization of AFD1 and DSY2 in wild-type and pam1 meiocytes. Both proteins showed substantially decreased signal in pam1 meiocytes (N = 20). (Scale bar, 5 μm.) (D) The effect of PAM1 on transcript splicing of seven meiosis-related genes. A344 (red) = RNA-seq read density in wild-type anthers. pam1 (blue) = RNA-seq read density in pam1 anthers. RNAIP (green) = read density from RNAIP experiments showing PAM1 transcript binding patterns. Ref-seq = gene structure. Exons exhibiting aberrant splicing patterns in the pam1 mutant are in black boxes.

Among the RNAIP-identified genes were eleven previously known meiotic genes, including seven core genes, Afd1, Dsy2, Mlh1, Msh4, and Mus81, Prd1, and Zyp1 (Fig. 3B). To investigate whether Pam1 indeed affected these genes, we conducted immunostaining of pam1 mutant meiocytes with antibodies against the AFD1, DSY2, and ZYP1 proteins. We found that all three proteins showed abnormal localization in the pam1 mutant, exhibiting weaker staining and discontinuous stretches, as opposed to strong continuous staining along the entire chromosome length found in wild type (Fig. 3C). These findings support the RNAIP results indicating that PAM1 affects specific meiotic proteins.

To examine whether the action of Pam1 affected the overall steady-state transcript levels, we conducted a global transcriptome analysis of pam1 mutant anthers at the stages between leptotene and pachytene. Compared to wild-type anthers of the A344 inbred, the pam1 mutant showed 478 up-regulated and 202 down-regulated differentially expressed genes (DEGs) at the log2-fold change cutoff of 2 and FDR < 0.05 (Datasets S2 and S3). Among the down-regulated DEGs, one of the most significantly enriched GO terms was “mRNA destabilization” (SI Appendix, Fig. S13). Among the up-regulated DEGs, the most significantly enriched GO term was “SCAR complex”, which is known to be involved in actin cytoskeleton assembly (36). Altogether, there were six up- and down-regulated cytoskeleton-related DEGs (Datasets S2 and S3).

Only 61 of the down-regulated genes and eight of the up-regulated genes were among the genes whose transcripts were identified in the RNAIP analysis (SI Appendix, Fig. S14A and Dataset S4). Furthermore, none of the elevn known meiotic genes detected by RNAIP showed differential expression. These findings suggested that PAM1 may be primarily involved in transcript processing rather than regulating transcription levels, transcript stabilization or degradation. Previous studies have indicated that some RRM-motif proteins bind to gene transcripts to affect a multitude of aspects of RNA metabolism, including alternative splicing (37–39). Alternative splicing may not necessarily alter transcript stability, resulting in no differences in steady-state transcript levels measured by RNA-seq. To examine this possibility, we examined RNAIP and RNA-seq read distribution in the seven core meiotic genes, Afd1, Dsy2, Mlh1, Msh4, and Mus81, Prd1, and Zyp1, using the Integrative Genomics Viewer (IGV) (Fig. 3D). The analysis revealed enrichment of RNAIP reads in exons of the meiotic genes demonstrating specific binding by PAM1. It also indicated reduction of specific exons in the genes in the pam1 mutant transcriptome, suggesting that in the pam1 mutant exons are skipped during transcript splicing. We designed primers targeting these exons and conducted validation by semiquantitative RT-PCR, which further confirmed these findings (SI Appendix, Fig. S14B). We then investigated if transcripts of these seven genes were correctly spliced in maize leaves, where the Pam1 gene is not expressed (40). We identified instances of altered splicing in transcripts of all seven genes, but in most cases these defects were not the same as those found in the pam1 mutant (SI Appendix, Fig. S15). These observations suggested the incorrect splicing of meiotic gene transcripts in somatic tissues was unlikely to be due to the absence of PAM1 in these cells.

We also performed a global analysis of alternative splicing with the rMATS software (41). Using FDR < 0.05, we identified in the pam1 mutant 446 alternative splicing events in 343 genes, encompassing five categories: skipped exons, alternative 5’ splice sites, alternative 3’ splice sites, retained introns, and mutually exclusive exons (SI Appendix, Fig. S16A and Dataset S5). Of the alternatively spliced genes, 86 were found in the anti-PAM1 RNAIP results, suggesting that PAM1 directly affects their splicing (SI Appendix, Fig. S16B and Dataset S6). However, only four of the 86 genes showed differential expression in the pam1 mutant (SI Appendix, Fig. S16C), which further supports the conclusion that PAM1 can modulate alternative splicing without affecting transcript abundance. Afd1, Dsy2, Mlh1, Msh4, Mus81, Prd1, and Zyp1 were found among the genes displaying alternative splicing in the rMATS analysis, although they did not exhibit low enough FDR due to high variability across the three biological replicates used. We hypothesize that it was because not all transcripts of these genes were misspliced and/or because the transcript abundance was too low for all splice variants to be fully represented in all replicates. Several RNA splicing-related genes were on the list of the 86 alternatively spliced genes detected in the RNAIP analysis (Dataset S6), suggesting that some alternative splicing events may be affected by PAM1 indirectly.

Taken together, the RNA analyses showed that PAM1 is a meiosis-specific splicing regulator affecting meiosis by posttranscriptionally altering splicing of meiosis-related genes, thereby modulating the expression of their protein products.

Interaction of PAM1 with the CCR4-NOT Complex.

Based on the presence of alternative splicing in the pam1 mutant, we hypothesized that PAM1 interacts with other proteins that facilitate mRNA processing. The PAM1 protein harbors in its central part an OST-HTH/LOTUS domain. LOTUS domains have been recently recognized as specialized RNA binding domains in plants and animals (42). They are known to interact with the CCR4-NOT protein complex, which, among other activities, functions in transcription initiation and elongation as well as splicing, mRNA quality control, and targeted transcript decay (43, 44). To determine whether PAM1 can indeed bind the CCR4-NOT complex, we tested the interaction between PAM1 and several CCR4-NOT complex proteins, NOT3, NOT10, and CCR4 using BiFC assays in N. benthamiana leaves. All three proteins showed interactions with the full-length PAM1 protein (Fig. 4A). In contrast, no interactions were detected when the LOTUS domain was deleted from PAM1. The interactions were confirmed using yeast two-hybrid assays (Fig. 4B). In addition, we found CCR4-NOT components NOT1 and NOT4 identified as target proteins in the ASY1 immunoprecipitation–LC/MSMS experiments (SI Appendix, Fig. S17). Taken together, these results indicate that PAM1 interacts with the CCR4-NOT complex through its LOTUS domain.

Fig. 4.

Two panel figure. Part A: micrographs of cells. Part B: yeast two hybrid assays results with columns 1 and 1/10 for S D /-L T H and S D /-L T media.

PAM1 interacts with the CCR4-NOT complex via its C terminal domains. (A) Interactions between key components of the CCR4-NOT complex and the full-length PAM1 protein (Left) as well as a truncated PAM1 missing the C terminus containing the LOTUS and RING finger domains (Right) identified using the BiFC assay. The experiments were repeated three times. C = chlorophyll, BF = brightfield. (Scale bar, 20 μm.) (B) Interactions between key components of the CCR4-NOT complex and the full-length PAM1 protein (Left) as well as a truncated PAM1 missing the C terminus containing the LOTUS and RING finger domains (Right) identified using yeast two-hybrid assays. Interaction were detected by yeast cell growth on the TDO/SD (SD-Trp-Leu-His) medium of undiluted cultures and cultures at the 1:10 dilution compared to controls grown on the DDO/SD (SD-Trp-Leu) medium. The interactions disappeared when C-terminal domains of PAM1 protein were removed. BD = bait vector, AD = prey vector.

Discussion

We have uncovered a unique mechanism controlling the expression of meiotic genes, in which the PAM1 protein associates with the meiotic chromosome axis and facilitates transcript splicing of key meiotic genes (Fig. 5). PAM1 is recruited to chromosomes in early leptotene and remains associated with them until pachytene, when most chromosome pairing- and recombination-related processes are completed (2, 45). This timing is consistent with the appearance of the mutant phenotypes of pam1.

Fig. 5.

Diagram of PAM1 complex binding meiotic transcripts to produce mRNAs for AFD1, ZYP1, and MLH1 to control meiotic processes.

A schematic showing the mode of action of PAM1.

Our findings revealed that correct splicing of meiotic gene transcripts required the CCR4-NOT complex to be anchored to the transcripts and to the meiotic chromosome axis. We show that PAM1 is recruited to the chromosome axis via its ankyrin domains and interacts with the CCR4-NOT complex via the LOTUS domain to regulate splicing of transcripts that it binds via the RRM domain. This model is consistent with the well-documented role of the CCR4-NOT complex in interacting with RNA binding proteins, such as AGO1, DDX6, and Pumilio, to mediate mRNA processing (46–48). CCR4-NOT functions in cotranscriptional mRNA processing as it is known to interact with RNA Polymerase II (49). Thus, we propose that the PAM1-CCR4-NOT complex is present on meiotic chromosomes to facilitate cotranscriptional splicing (50). We hypothesize that the PAM1-mediated anchoring of CCR4-NOT to the chromosome axis is needed to shield splicing from the challenges of nuclear environment present in meiocytes but not in somatic cells. For example, the very vigorous chromosome motility of early meiotic prophase I (31) has likely the capacity to disrupt protein–protein and protein–RNA interactions, which would interfere with transcription and RNA processing.

The role of PAM1 as a meiosis-controlling RNA binding protein is analogous to that of the previously described meiotic entry/progression regulators Mmi1 (13) in yeast and MARF1 in the mouse (14). However, the three proteins do not share common ancestry (SI Appendix, Fig. S18), suggesting that they represent an extraordinary case of convergent evolution. Furthermore, PAM1 is the only of the three proteins that exhibits ubiquitous chromosome association. Mei2 in fission yeast has been shown to be located on meiotic chromosomes but the protein is confined to a single location at the sme 2 locus (51).

PAM1 does not appear to be involved in meiosis initiation, as meiocyte failure to enter meiosis is not observed in either pam1-1 point mutant nor in the pam1-2 mutant, which caused by a transposon insertion. However, the presence of PAM1 protein in the cytoplasm during leptotene may be consistent with the observed association of the rice PAM1 homolog MEL2 with cytoplasmic ribonucleoprotein granules (23). Such behavior would be analogous to this of yeast Mei2, which is shuttled to the nucleus at the onset of prophase I (16).

The vast majority of PAM1-bound transcripts do not exhibit changes in steady-state mRNA levels, indicating that the main activity of the CCR4-NOT complex in association with PAM1 is facilitating splicing. Consequently, most DEGs identified in the pam1 mutant likely represent downstream effects of PAM1 function. Misregulation of these genes may be responsible for the phenotypic defects observed in pam1 mutants. In particular, the defects in prophase chromosome movement and cytoskeleton organization may stem from misregulation of the actin-controlling SCAR complex genes. The actin cytoskeleton is essential for chromosome movements in meiotic prophase I (31) and for chromosome segregation in anaphase (52, 53). Defects in both these processes are observed in pam1 mutants.

Genes identified in the RNAIP experiments and in the transcriptome analyses of pam1 mutant anthers do not represent the whole complement of meiotically expressed genes in maize (54, 55). Pam1-affected genes show only some overlap with genes whose transcripts are affected in am1 mutants (19, 56). Furthermore, despite the misregulation of a large number of genes resulting in major meiosis defects, some meiosis-specific processes proceed relatively normally in pam1 mutants, including DSB formation and early stages of their repair, as well as telomere attachment to the nuclear envelope (27). These data indicate that multiple programs of transcriptome control operate during meiotic prophase. These programs are controlled by distinct meiosis progression regulators, which include PAM1 and AM1. The mode of action of PAM1, where the protein binds to the meiotic chromosome axis to facilitate splicing, suggests that meiosis progression regulators have unique features that are distinct from those of transcriptome regulators acting in somatic cells.

Uncovering fundamental mechanisms controlling meiosis progression opens doors to efforts to engineer meiosis to benefit crop improvement methods, in particular developing ways to induce meiosis in vitro. The latter has enormous potential in many agronomically important crops that have extremely long generation times. Being able to induce meiosis would lead to shortening generation times and allow much more efficient breeding.

Methods

Cloning the Pam1 Gene.

The pam1 mutation was found in a population of maize plants of the A344 inbred line mutagenized with N-nitroso-N-methylurea (NMU) (27). Previous studies used chromosome translocations to show that the Pam1 gene was located close to the centromere of chromosome 1 (27). To clone the gene, a mapping population containing 1250 mutant plants was generated by crossing the pam1 mutant into several maize inbreds (A188, B73, F2, F7, H99, Mo17, W22, and W23) and selfing for two to three generations. Using single nucleotide polymorphism (SNP) markers, we narrowed down the interval around the Pam1 locus to a ca. 20 Mb segment in the pericentromeric region of chromosome 1. To identify the pam1 mutant lesion, two pools of pam1 mutant plants, one containing 32 individuals and the other one containing 27 individuals, were sequenced using the Illumina platform to a total coverage of about 16x. The resulting sequence was compared to the sequence of the A344 inbred, searching for genes harboring NMU-induced SNPs in their exon, promoter, and/or terminator regions. With this approach, we found that the only potential candidate in the Pam1 gene region was the predicted gene GRMZM2G061704. The gene contained only a single NMU-induced SNP capable of resulting in a mutant phenotype.

Analysis of the GRMZM2G061704 gene mRNA sequence obtained from de novo transcriptome assembly from RNA-seq data revealed that it was composed of 14 exons and 13 introns, with the open reading frame (ORF) of 3288 bp. The guanine to adenine transition at the 3’ splice acceptor site of intron 3 of the gene in the pam1 mutant was predicted to result in incorrect splicing of the intron (SI Appendix, Fig. S5), leading to a premature stop codon. Presence of incorrectly spliced Pam1 transcripts was confirmed by analyzing the pam1 mutant transcriptome. The Pam1 gene is predited to encode a 1095-amino acid-long relatively hydrophilic protein (hydrophobicity index=-0.29). The PAM1 protein contains two ankyrin domains (ANK; Pfam12796) near its N terminus (amino acids 26-136 and 323-416; SI Appendix, Fig. S7), two nucleotide binding domains: the RNA recognition motif (RRM, Pfam00076) and the LOTUS domain (Pfam12872), in the middle part of the protein (amino acids 532-612 and 668-746 respectively), and two C3HC4-type RING-finger domains (Pfam13920) near the C-terminus (amino acids 974-1020 and amino acids 1046-1090).

To confirm cloning the Pam1 gene, we used the cloned gene sequence to find an additional mutant pam1 allele in the TUSC (Trait Utility System for Corn) transposon mutagenesis program of Pioneer Hi-Bred (57, 58). Screening batches of Mu-mutagenized lines, with PCR primers specific to the Pam1 gene and the Mu transposon, we identified a second mutant allele of pam1, which we named pam1-2, carrying a Mu insertion in the 12th exon of the Pam1 gene. As expected, plants homozygous for the pam1-2 mutation showed male sterility phenotypes. Cytological observations of pam1-2 meiocytes using Structured Illumination Microscopy (SIM) revealed defective chromosome condensation as well as impaired tubulin organization (SI Appendix, Fig. S6). In addition, meiocyte development in pam1-2 was asynchronous within anthers, in contrast to wild type. These defects were similar to those observed in the pam1-1 mutant. Based on these results, we concluded that we cloned the Pam1 gene.

Anti-PAM1 Antibody Generation.

The anti-PAM1 antibody was developed against the full-length PAM1 protein, which was expressed using the Champion pET100 Directional TOPO Expression Kit (Invitrogen). The recombinant PAM1 protein contained the 6 × his tag, which was used for protein purification. The anti-PAM1 antibody was produced in rabbits by Cocalico Biologicals, Inc. (Reamstown, PA).

Slides Preparation for Immunolocalization Analyses.

Anthers at desired meiosis stages were dissected from developing tassels and fixed for 45 min at room temperature in 4% paraformaldehyde solution in Buffer A (59). Immunolocalization experiments were performed as previously described (59) using the following antibodies: rabbit anti-PAM1, guinea pig anti-ASY1 (60), rabbit anti-ASY1 (29), rat anti-AFD1 (61), rabbit anti-DSY2 (62), mouse anti-Nuclear Pore Complex (30), rabbit anti-tubulin (Sigma-Aldrich), and guinea pig anti-ZYP1 (29).

FiSH Slide Preparation.

Slides for FISH analyses were prepared as previously described (27) using probes detecting the centromere of maize chromosome 4 (Cent4) and the 5S rRNA locus (63).

Immuno-FiSH Slide Preparation.

Slides for Immuno-FiSH analyses prepared as previously described (64) using FiSH probe detecting the telomere repeats (27) as well as the anti-Nuclear Pore Complex (30) and the anti-ZYP1 (29) antibodies.

Three-Dimensional Deconvolution Microscopy.

Slides were examined using DeltaVision imaging workstation (Applied Precision). Three-dimensional stacks of images were collected across the entire thickness of the specimen with optical sections 150 nm apart. The image stacks were subjected to constrained iterative deconvolution and analyzed with the softWoRx software (Applied Precision). Colocalization analyses were conducted in three-dimensional space to distinguish between actual colocalization and accidental overlap.

Measurement of Chromosome Condensation with Structured Illumination Microscopy (SIM).

Chromosome preparation for SIM was performed as previously described with minor changes (62). In brief, 1 μL meiocytes suspension were placed in the center of an 18×18 mm No.1.5 coverslip. 0.5 μL of a polymerizing acrylamide solution was added to the meiocyte drop and mixed for 3 s. A second coverslip was placed on top of the meiocyte suspension and polymerization continued for 40 min at room temperature. Chromosomes were stained with 10 μg/mL DAPI (10 μg/mL). Slides were examined under an Elyra Super-Resolution Microscope (Zeiss). Chromosome width was measured using Image J (NIH).

Live Imaging of Chromosome Motility.

For live imaging observations, anthers were placed in an eight-chamber culture slide (Nunc-155411, Thermo Fisher Scientific) in artificial pond water (APW) buffer (31). The medium was supplemented with 50 μg/mL DAPI, 20 μM of a vital mitochondrial stain Rhodamine 123, and DMSO at the concentration of up to 1%. The solution was removed after 1.5 h of incubation at room temperature and replaced with APW buffer to remove unincorporated dyes before imaging. Images were captured using the LSM880 confocal/multiphoton inverted microscope (Zeiss). Experiments using pam1 and control wild-type A344 inbred anthers were performed side-by-side four times under identical conditions.

Chromosome dynamics was tracked based on DAPI staining, using the TrackMate plugin on Image J (NIH) using settings as previously described (31). The Blob diameter was set to 1 μM with an appropriate threshold, and the linking distance was set to 4 μM with Gap-closing maximum distance of 1 μM and Gap-closing maximum frame gap of 2. The identified tracks were manually examined and selected if they were observed at at least 5 consecutive time points. To visualize the dynamic movement, tracks were set to “show tracks backward in time” and recorded using the “Capture Overlay” function.

Pollen Viability Analysis.

Pollen viability was examined using Alexander stain as previously described (65) using six pam1 mutant and wild-type anthers.

S Phase Progression Analysis.

DNA replication progression during premeiotic S phase was analyzed using 5-ethynyl-2′-deoxyuridine (EdU) as previously described (33).

BiFC Assays.

Full-length coding regions of the maize Asy1, Not3, and Pam1 genes, as well as truncated versions of Pam1, were cloned into Gateway-compatible vectors pSPYNE (n-eYFP) and pSPYCE (c-eYFP) (66). The resulting constructs, together with empty N-eYFP and C-eYFP vectors and the p19 silencing plasmid, were transformed into Agrobacterium tumefaciens strain GV3101. Agrobacterium suspensions were infiltrated into leaves of five-week-old Nicotiana benthamiana plants as described previously (67). After four days of incubation in a growth chamber, the leaves were examined under the LSM710 Confocal microscope (Zeiss). Each experiment was repeated three times.

Yeast Two-Hybrid Assays.

Yeast two-hybrid assays were performed using the Matchmaker Gold Yeast Two-Hybrid system (Clontech). The full-length and truncated coding regions of the Pam1 gene were cloned into the pGADT7 vector and the full-length coding regions of Asy1 and Not3 were cloned into the pGBKT7 vector. The vectors were cotransformed into the yeast Y2H Gold strain, with empty pGADT7 and pGBKT7 vectors as negative controls. After culturing on the DDO (SD/-Trp/-Leu) medium at 30 °C, positive clones were picked and transferred to the TDO (SD/-His/-Trp/-Leu) medium with the appropriate positive and negative controls to detect interactions.

RNA-Seq Analyses.

Total RNA was extracted from prophase I anthers using the NucleoSpin RNA purification kit (Macherey-Nagel). The integrity and concentration of RNA were examined with gel electrophoresis and the Qubit 3.0 Fluorometer (Thermo Fisher Scientific). 100 ng of RNA was used for cDNA library preparation with the NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs) following the manufacturer’s protocol. Library quality was examined with the Agilent Bioanalyzer (Agilent Technologies). The cDNA libraries were sequenced using the Illumina platform.

RNA Immunoprecipitation (RNAIP) Analysis.

To perform the RNAIP analysis (68), 50 μL of Protein A beads (Invitrogen) were incubated overnight with 30 μg anti-PAM1 antibody or rabbit IgG. The resin was washed three times with 500 μL of TBST buffer. About 0.2 g of prophase I anthers collected from plants of the A344 inbred line were ground in liquid nitrogen and total protein extraction was performed by adding 1 mL RNAIP buffer (20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM DTT, 2 mM EDTA, 1 mM PMSF, 0.1% NP-40, 400 units/mL RnaseOUT (Invitrogen), 1 μg/mL protease inhibitor cocktail (Sigma-Aldrich). The extract was clarified by centrifuging at 12,000×g for 30 min and incubated with the beads with gentle rotation for 4 h. The beads were washed with RNAIP buffer three times. RNA was purified with the NucleoSpin RNA purification kit (Macherey-Nagel) and resuspended in 15 μL DEPC-treated water. All experiments were performed at 4 °C. Sequencing libraries were produced using the NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs). RNAIP was conducted on three biological replicates of anti-PAM1 antibody treatment and two biological replicates of the rabbit IgG control that were processed separately but side-by-side throughout the experiment. Principal component analysis (PCA) of RNAIP data indicated separation of the anti-PAM1 antibody treated samples from the IgG controls (SI Appendix, Fig. S19).

Semiquantitative RT-PCR.

200 ng of total RNA was used for reverse transcription PCR using the Luna Universal One-Step RT-qPCR Kit (New England Biolabs), following the manufacturer’s protocol, except that the PCR cycle number was decreased to 30. Primers used in the experiment are listed in SI Appendix, Table S1. 10 μL of PCR was separated in an agarose gel. Gel image was captured using the ChemiDoc Imaging System (Bio-Rad) and the quantitative analysis was performed with the Image Lab Software (Bio-Rad).

Bioinformatics Analysis.

For RNA-seq analysis, raw Illumina sequence reads were aligned to the maize reference genome (version 4) (69) with the STAR software (version 2.7.10a) (70) using the following settings: --outFilterMultimapNmax 1 --outFilterMismatchNmax 2 --outSAMtype BAM SortedByCoordinate. To generate a count table, the --quantMode GeneCounts command in STAR was used. Differentially expressed genes (DEGs) were defined using DESeq2 (71).

For RNAIP analysis, raw sequence reads were aligned to the maize reference genome (69) using Bowtie 2 (version 2.4.5) (72) with default settings. Reads with mapping quality of less than 20 were removed. The MACS2 (version 2.2.7.1) (73) software was used for peak calling with the following parameters: --shift 0 -g 2.0e9, --nomode, l --bw 40 --keep-dup=1 -q 0.01 -m 2 5. The bamcompare command of Deeptools (74) was used for producing bigwig files, with parameters --binSize 20 --operation subtract --normalizeUsing CPM -p 5 --smoothLength 60. The bigwig files were imported to the Integrated Genome Browser (IGV) to view read density along chromosomes. Peaks overlapping genome features were categorized based on distance: The 2 kb regions upstream from the transcription starting sites (TSS) were considered 5’ UTR, the 2 kb regions downstream from transcription termination sites (TTS) were considered 3’ UTR, and the regions between TSS and TTS, including exons and introns, were considered gene bodies. Peaks further away than 2 kb upstream from TSS and 2 kb downstream from TTS were considered intergenic. The relationship between RNAIP targets and gene density were investigated in 1 Mb nonoverlapping sliding windows. The meme command of the MEME Suite (version 5.4.1) (75) was used to search for enriched motifs in the RNAIP target genes.

Gene Ontology (GO) enrichment was performed using agriGO (version 2.0) (76). GO terms with FDR of 0.05 or less were considered significant.

Alternative splicing was analyzed with rMATS 4.2.0 (41) using the following settings: rmats.py --b1 A344.txt --b2 paml.txt --gtf maize_v4.gtf --od A344_paml_rmats --tmp tmp -t single --readLength 85 --nthread 2.

Supplementary Material

Appendix 01 (PDF)

pnas.2535316123.sapp.pdf (26.5MB, pdf)

Dataset S01 (XLSX)

pnas.2535316123.sd01.xlsx (163.8KB, xlsx)

Dataset S02 (XLSX)

pnas.2535316123.sd02.xlsx (25.4KB, xlsx)

Dataset S03 (XLSX)

pnas.2535316123.sd03.xlsx (46.2KB, xlsx)

Dataset S04 (XLSX)

pnas.2535316123.sd04.xlsx (14.2KB, xlsx)

Dataset S05 (XLSX)

Dataset S06 (XLSX)

pnas.2535316123.sd06.xlsx (12.3KB, xlsx)

Acknowledgments

We would like to thank Changbin Chen and Chris Franklin for generously sharing antibodies, Yan He for providing the maize asy1 mutant, Jian Hua and Margaret Frank for help with designing protein–protein interaction experiments, Yadanar Htike for help with immunolocalization experiments, Daniel Mead for help with pollen viability analysis, Nick Van Eck for help with growing maize plants, and Jian Hua, Changbin Chen, and Quinn Johnson for comments on the manuscript. Live imaging and SIM observations were conducted at the BRC Imaging Facility at the Cornell Institute of Biotechnology with the help of Johanna Dela Cruz and Rebecca Williams. This work was supported by a USDA-NIFA postdoctoral fellowship 2006-35304-17279 to M.J.S., a USDA-AFRI Grant NYC-149562 to W.P.P., and NSF Grant IOS-1841696 to W.P.P. W.P.P. is a Cornell Institute for Food Systems Faculty Fellow.

Author contributions

Q.D., C.-L.T., M.J.S., and W.P.P. designed research; Q.D., M.W., C.-L.T., M.J.S., O.H., E.R., C.-J.R.W., and C.K. performed research; P.A., J.-K.Y., O.H., E.R., R.B.M., and I.N.G. contributed new reagents/analytic tools; Q.D., M.W., and R.B. analyzed data; and Q.D. and W.P.P. wrote the paper.

Competing interests

The authors declare no competing interest.

Footnotes

This article is a PNAS Direct Submission.

Contributor Information

Qian Du, Email: qd36@cornell.edu.

Wojciech P. Pawlowski, Email: wp45@cornell.edu.

Data, Materials, and Software Availability

RNA-seq and RNAIP sequence data were deposited at NCBI under accession number GSE216990 (77).

Supporting Information

References

  • 1.Loidl J., Conservation and variability of meiosis across the Eukaryotes. Annu. Rev. Genet. 50, 293–316 (2016). [DOI] [PubMed] [Google Scholar]
  • 2.Grey C., de Massy B., Coupling crossover and synaptonemal complex in meiosis. Genes Dev. 36, 4–6 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Bower F., Schnittger A., How to switch from mitosis to meiosis: Regulation of germline entry in plants. Annu. Rev. Genet. 55, 427–452 (2021). [DOI] [PubMed] [Google Scholar]
  • 4.Nelms B., Walbot V., Gametophyte genome activation occurs at pollen mitosis I in maize. Science 375, 424–429 (2022). [DOI] [PubMed] [Google Scholar]
  • 5.Cairo A., et al. , Meiotic exit in Arabidopsis is driven by P-body-mediated inhibition of translation. Science 377, 629–634 (2022). [DOI] [PubMed] [Google Scholar]
  • 6.Sun S., et al. , Znhit1 controls meiotic initiation in male germ cells by coordinating with Stra8 to activate meiotic gene expression. Dev. Cell 57, e904 (2022). [DOI] [PubMed] [Google Scholar]
  • 7.Berger F., Meiosis as a mechanism for epigenetic reprogramming and cellular rejuvenation. Development 151, dev203046 (2024). [DOI] [PubMed] [Google Scholar]
  • 8.Li Y., et al. , The landscape of RNA binding proteins in mammalian spermatogenesis.. Science 386, eadj8172 (2024). [DOI] [PubMed] [Google Scholar]
  • 9.Pawlowski W. P., Sheehan M. J., Ronceret A., In the beginning: The initiation of meiosis. Bioessays 29, 511–514 (2007). [DOI] [PubMed] [Google Scholar]
  • 10.Zhou A., Pawlowski W. P., Regulation of meiotic gene expression in plants. Front. Plant Sci. 5, 413 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Ishiguro K. I., Mechanisms of meiosis initiation and meiotic prophase progression during spermatogenesis.. Mol. Aspects Med. 97, 101282 (2024). [DOI] [PubMed] [Google Scholar]
  • 12.Harigaya Y., et al. , Selective elimination of messenger RNA prevents an incidence of untimely meiosis. Nature 442, 45–50 (2006). [DOI] [PubMed] [Google Scholar]
  • 13.Chen H. M., Futcher B., Leatherwood J., The fission yeast RNA binding protein Mmi1 regulates meiotic genes by controlling intron specific splicing and polyadenylation coupled RNA turnover. PLoS One. 6, e26804 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Su Y. Q., et al. , MARF1 regulates essential oogenic processes in mice. Science 335, 1496–1499 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Nishimura T., et al. , Human MARF1 is an endoribonuclease that interacts with the DCP1:2 decapping complex and degrades target mRNAs. Nucleic Acids Res. 46, 12008–12021 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Watanabe Y., Shinozaki-Yabana S., Chikashige Y., Hiraoka Y., Yamamoto M., Phosphorylation of RNA-binding protein controls cell cycle switch from mitotic to meiotic in fission yeast. Nature 386, 187–190 (1997). [DOI] [PubMed] [Google Scholar]
  • 17.Xiao R., et al. , Pervasive chromatin-RNA binding protein interactions enable RNA-based regulation of transcription. Cell 178, e118 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Zhu L., Kandasamy S. K., Liao S. E., Fukunaga R., LOTUS domain protein MARF1 binds CCR4-NOT deadenylase complex to post-transcriptionally regulate gene expression in oocytes. Nat. Commun. 9, 4031 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Pawlowski W. P., et al. , Maize AMEIOTIC1 is essential for multiple early meiotic processes and likely required for the initiation of meiosis. Proc. Natl. Acad. Sci. U.S.A. 106, 3603–3608 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Nonomura K., et al. , A germ cell specific gene of the ARGONAUTE family is essential for the progression of premeiotic mitosis and meiosis during sporogenesis in rice. Plant Cell 19, 2583–2594 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Nonomura K., et al. , A novel RNA-recognition-motif protein is required for premeiotic G1/S-phase transition in rice (Oryza sativa L.).. PLoS Genet. 7, e1001265 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Miyazaki S., Sato Y., Asano T., Nagamura Y., Nonomura K., Rice MEL2, the RNA recognition motif (RRM) protein, binds in vitro to meiosis-expressed genes containing U-rich RNA consensus sequences in the 3’-UTR. Plant Mol. Biol. 89, 293–307 (2015). [DOI] [PubMed] [Google Scholar]
  • 23.Mimura M., Ono S., Somashekar H., Nonomura K. I., Impact of protein domains on the MEL2 granule, a cytoplasmic ribonucleoprotein complex maintaining faithful meiosis progression in rice. New Phytol. 243, 2235–2250 (2024). [DOI] [PubMed] [Google Scholar]
  • 24.Capilla-Perez L., et al. , The synaptonemal complex imposes crossover interference and heterochiasmy in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A. 118, e2023613118 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yang C., et al. , SWITCH 1/DYAD is a WINGS APART-LIKE antagonist that maintains sister chromatid cohesion in meiosis. Nat. Commun. 10, 1755 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Zhao T., et al. , The OsRR24/LEPTO1 type-B response regulator is essential for the organization of leptotene chromosomes in rice meiosis. Plant Cell 30, 3024–3037 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Golubovskaya I. N., Harper L. C., Pawlowski W. P., Schichnes D., Cande W. Z., The pam1 gene is required for meiotic bouquet formation and efficient homologous synapsis in maize (Zea mays, L.). Genetics 162, 1979–1993 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Harper L., Golubovskaya I., Cande W. Z., A bouquet of chromosomes. J. Cell Sci. 117, 4025–4032 (2004). [DOI] [PubMed] [Google Scholar]
  • 29.Ku J. C., et al. , Dynamic localization of SPO11-1 and conformational changes of meiotic axial elements during recombination initiation of maize meiosis. PLoS Genet. 16, e1007881 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Cowan C. R., Carlton P. M., Cande W. Z., Reorganization and polarization of the meiotic bouquet-stage cell can be uncoupled from telomere clustering. J. Cell Sci. 115, 3757–3766 (2002). [DOI] [PubMed] [Google Scholar]
  • 31.Sheehan M. J., Pawlowski W. P., Live imaging of rapid chromosome movements in meiotic prophase I in maize. Proc. Natl. Acad. Sci. U.S.A. 106, 20989–20994 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Koszul R., Kleckner N., Dynamic chromosome movements during meiosis: A way to eliminate unwanted connections?. Trends Cell. Biol. 19, 716–724 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kelliher T., Walbot V., Emergence and patterning of the five cell types of the Zea mays anther locule. Dev. Biol. 350, 32–49 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Sanchez-Moran E., Santos J. L., Jones G. H., Franklin F. C., ASY1 mediates AtDMC1-dependent interhomolog recombination during meiosis in Arabidopsis. Genes Dev. 21, 2220–2233 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Li J., Mahajan A., Tsai M. D., Ankyrin repeat: A unique motif mediating protein-protein interactions. Biochemistry 45, 15168–15178 (2006). [DOI] [PubMed] [Google Scholar]
  • 36.Welch M. D., Mullins R. D., Cellular control of actin nucleation. Annu. Rev. Cell Dev. Biol. 18, 247–288 (2002). [DOI] [PubMed] [Google Scholar]
  • 37.Collins K. M., et al. , An RRM-ZnF RNA recognition module targets RBM10 to exonic sequences to promote exon exclusion. Nucleic Acids Res. 45, 6761–6774 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zhan X., et al. , An Arabidopsis PWI and RRM motif-containing protein is critical for pre-mRNA splicing and ABA responses. Nat. Commun. 6, 8139 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Sun L., et al. , SRSF1 is crucial for male meiosis through alternative splicing during homologous pairing and synapsis in mice. Sci. Bull. 68, 1100–1104 (2023). [DOI] [PubMed] [Google Scholar]
  • 40.Hoopes G. M., et al. , An updated gene atlas for maize reveals organ-specific and stress-induced genes. Plant J. 97, 1154–1167 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wang Y., et al. , rMATS-turbo: An efficient and flexible computational tool for alternative splicing analysis of large-scale RNA-seq data. Nat. Protoc. 19, 1083–1104 (2024). [DOI] [PubMed] [Google Scholar]
  • 42.Ding D., et al. , LOTUS domain is a novel class of G-rich and G-quadruplex RNA binding domain. Nucleic Acids Res. 48, 9262–9272 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Collart M. A., The Ccr4-Not complex is a key regulator of eukaryotic gene expression. Wiley Interdiscip Rev RNA 7, 438–454 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Chapat C., et al. , Alternative splicing of CNOT7 diversifies CCR4-NOT functions. Nucleic Acids Res. 45, 8508–8523 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ur S. N., Corbett K. D., Architecture and dynamics of meiotic chromosomes. Annu. Rev. Genet. 55, 497–526 (2021). [DOI] [PubMed] [Google Scholar]
  • 46.Schwenk P., et al. , Uncovering a novel function of the CCR4-NOT complex in phytochrome A-mediated light signalling in plants. Elife 10, e63697 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Ozgur S., et al. , Structure of a human 4E-T/DDX6/CNOT1 complex reveals the different interplay of DDX6-binding proteins with the CCR4-NOT complex. Cell Rep. 13, 703–711 (2015). [DOI] [PubMed] [Google Scholar]
  • 48.Haugen R. J., Arvola R. M., Connacher R. P., Roden R. T., Goldstrohm A. C., A conserved domain of Drosophila RNA-binding protein Pumilio interacts with multiple CCR4-NOT deadenylase complex subunits to repress target mRNAs. J. Biol. Chem. 298, 102270 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Kruk J. A., Dutta A., Fu J., Gilmour D. S., Reese J. C., The multifunctional Ccr4-Not complex directly promotes transcription elongation. Genes Dev. 25, 581–593 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Marquardt S., Petrillo E., Manavella P. A., Cotranscriptional RNA processing and modification in plants. Plant. Cell. 35, 1654–1670 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Ding D. Q., et al. , Meiosis-specific noncoding RNA mediates robust pairing of homologous chromosomes in meiosis. Science 336, 732–736 (2012). [DOI] [PubMed] [Google Scholar]
  • 52.Mogessie B., Schuh M., Actin protects mammalian eggs against chromosome segregation errors. Science 357, eaal1647 (2017). [DOI] [PubMed] [Google Scholar]
  • 53.Roeles J., Tsiavaliaris G., Actin-microtubule interplay coordinates spindle assembly in human oocytes. Nat. Commun. 10, 4651 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Nelms B., Walbot V., Defining the developmental program leading to meiosis in maize. Science 364, 52–56 (2019). [DOI] [PubMed] [Google Scholar]
  • 55.Dukowic-Schulze S., et al. , The transcriptome landscape of early maize meiosis. BMC Plant Biol. 14, 118 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Nan G. L., et al. , Global transcriptome analysis of two ameiotic1 alleles in maize anthers: Defining steps in meiotic entry and progression through Prophase I. BMC Plant Biol. 11, 120 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Bensen R. J., et al. , Cloning and characterization of the maize An1 gene. Plant. Cell. 7, 75–84 (1995). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Meeley R., Briggs S. P., Reverse genetics for maize. Maize Newsl. 69, 67–82 (1995). [Google Scholar]
  • 59.Du Q., Pawlowski W. P., Preparing immunolocalization slides of maize meiotic chromosomes for three-dimensional microscopy. Bio-protocol 10, e3691 (2020). [Google Scholar]
  • 60.Armstrong S. J., Franklin F. C., Jones G. H., Nucleolus-associated telomere clustering and pairing precede meiotic chromosome synapsis in Arabidopsis thaliana. J. Cell Sci. 114, 4207–4217 (2001). [DOI] [PubMed] [Google Scholar]
  • 61.Golubovskaya I. N., et al. , Alleles of afd1 dissect REC8 functions during meiotic prophase I. J. Cell Sci. 119, 3306–3315 (2006). [DOI] [PubMed] [Google Scholar]
  • 62.Lee D. H., et al. , The axial element protein DESYNAPTIC2 mediates meiotic double-strand break formation and synaptonemal complex assembly in maize. Plant Cell 27, 2516–2529 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Kato A., Lamb J. C., Birchler J. A., Chromosome painting using repetitive DNA sequences as probes for somatic chromosome identification in maize. Proc. Natl. Acad. Sci. U.S.A. 101, 13554–13559 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Armstrong S. J., Sanchez-Moran E., Franklin F. C., Cytological analysis of Arabidopsis thaliana meiotic chromosomes. Methods Mol. Biol. 558, 131–145 (2009). [DOI] [PubMed] [Google Scholar]
  • 65.Alexander M. P., Differential staining of aborted and nonaborted pollen. Stain Technol. 44, 117–122 (1969). [DOI] [PubMed] [Google Scholar]
  • 66.Gou M., et al. , Opposing effects on two phases of defense responses from concerted actions of HEAT SHOCK COGNATE70 and BONZAI1 in Arabidopsis. Plant Physiol. 169, 2304–2323 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Schutze K., Harter K., Chaban C., Bimolecular fluorescence complementation (BiFC) to study protein-protein interactions in living plant cells. Methods Mol. Biol. 479, 189–202 (2009). [DOI] [PubMed] [Google Scholar]
  • 68.Schmid L. M., et al. , PUMPKIN, the sole plastid UMP kinase, associates with group II introns and alters their metabolism. Plant Physiol. 179, 248–264 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Jiao Y., et al. , Improved maize reference genome with single-molecule technologies. Nature 546, 524–527 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Dobin A., et al. , STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Love M. I., Huber W., Anders S., Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Langmead B., Salzberg S. L., Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Zhang Y., et al. , Model-based analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.Ramirez F., et al. , deepTools2: A next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44, W160–W165 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Bailey T. L., Johnson J., Grant C. E., Noble W. S., The MEME suite. Nucleic Acids Res. 43, W39–W49 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Tian T., et al. , agriGO v2.0: A GO analysis toolkit for the agricultural community, 2017 update. Nucleic Acids Res. 45, W122–W129 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Du Q., Wang M., Pawlowski W., RRM protein PAM1 regulates meiosis by interacting with chromosome axis and the CCR4-NOT complex. Gene Expression Omnibus. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE216990. Accessed 28 April 2026.

Associated Data

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

Supplementary Materials

Appendix 01 (PDF)

pnas.2535316123.sapp.pdf (26.5MB, pdf)

Dataset S01 (XLSX)

pnas.2535316123.sd01.xlsx (163.8KB, xlsx)

Dataset S02 (XLSX)

pnas.2535316123.sd02.xlsx (25.4KB, xlsx)

Dataset S03 (XLSX)

pnas.2535316123.sd03.xlsx (46.2KB, xlsx)

Dataset S04 (XLSX)

pnas.2535316123.sd04.xlsx (14.2KB, xlsx)

Dataset S05 (XLSX)

Dataset S06 (XLSX)

pnas.2535316123.sd06.xlsx (12.3KB, xlsx)

Data Availability Statement

RNA-seq and RNAIP sequence data were deposited at NCBI under accession number GSE216990 (77).


Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES