Petal abscission is promoted by jasmonic acid-induced autophagy at Arabidopsis petal bases

In angiosperms, the transition from floral-organ maintenance to abscission determines reproductive success and seed dispersion. For petal abscission, cell-fate decisions specifically at the petal-cell base are more important than organ-level senescence or cell death in petals. However, how this transition is regulated remains unclear. Here, we identify a jasmonic acid (JA)-regulated chromatin-state switch at the base of Arabidopsis petals that directs local cell-fate determination via autophagy. During petal maintenance, co-repressors of JA signaling accumulate at the base of petals to block MYC activity, leading to lower levels of ROS. JA acts as an airborne signaling molecule transmitted from stamens to petals, accumulating primarily in petal bases to trigger chromatin remodeling. This allows MYC transcription factors to promote chromatin accessibility for downstream targets, including NAC DOMAIN-CONTAINING PROTEIN102 (ANAC102). ANAC102 accumulates specifically at the petal base prior to abscission and triggers ROS accumulation and cell death via AUTOPHAGY-RELATED GENEs induction. Developmentally induced autophagy at the petal base causes maturation, vacuolar delivery, and breakdown of autophagosomes for terminal cell differentiation. Dynamic changes in vesicles and cytoplasmic components in the vacuole occur in many plants, suggesting JA–NAC-mediated local cell-fate determination by autophagy may be conserved in angiosperms.

1.In figure 1a, the authors defined the stages of petal abscission from e.g.-3 to 3.And in figure 1b, they showed the petal abscission phenotypes in different genotypes, mainly JA related.I suggest that the authors show zoom-in phenotypes at key stages in different genotypes (preferably under high magnitude stereomicroscope), to help the readers better visualize the petal abscission phenotypes.2. Line 173, 'To gain more insight into petal abscission, we sectioned the petal base from position 3 flowers…' .The authors found that 'At position d, WT petal cells were partly filled with large vacuoles, indicating cell differentiation (Fig. 1j; top).Notably, we only noticed partial vacuolation of ag and dad1cells at position d.Furthermore, autophagic body-like structures did not accumulate in the vacuoles of WT petal base cells, while those of ag and dad1 contained several autophagic body-like structures in their vacuoles (Fig. 1k, l).'They concluded that 'JA promote petal abscission through petal cell differentiation at their base'.I wonder whether the cellular position and cell size of the petal base at position 5/6/7/8 in the dad1 mutant are similar compared to WT at position 3? Whether JA-treated dad1, aos mutants showed recovered phenotypes at position d, similar compared to WT?And also, whether the coi1, myc2myc3myc4, anac102 mutants show the similar phenotype?3. Related to the second question, whether the reduced DAB staining and trypan blue staining in different JA-related mutants compared with WT at petal base position 3, is due to delayed ROS accumulation in the JA-related, i.e., normal staining at position 5 or 6 or 7 or 8 in JA mutants, compared with WT position 3; or on the other hand, ROS accumulation is reduced at whatever petal positions in the JA mutants? 4. Line 204, the authors 'detected DAD1 promoter activity (pDAD1::GUS) and DAD1 protein accumulation (pDAD1::DAD1-GFP) at the same stages during stamen development (Fig. 2f, g)… DAD1-GFP did not accumulate in petals (Fig. 2h).'They discussed that '…JA production in the stamens for proper pollen maturation and anther dehiscence and coordination of petal development and flower opening by diffusion or active transport to petals.'I am curious whether other JA biosynthesis genes are also expressed in stamens, but not expressed at the base of petals?In addition, whether the stamen abscission and petal abscission are correlated?And whether JA is also involved in stamen abscission?
In this manuscript, the authors investigated the JA-induced transcriptional changes that lead to petal abscission in Arabidopsis.I am not an expert on hormones or transcriptional responses, so I will focus on the cell biology part in my report.In the current state, the manuscript is not conclusively linking autophagy to petal abscission.Please see below my detailed comments: 1/ Title-Jasmonic acid-induced instead of mediated to avoid having 2 mediated in the title?2/ Fig1k/l-autophagic body-like structures is a very vague term that is not backed up by any data.There are various vacuolar trafficking pathways that could be the source of those vesicles.The authors need to at least back this up with some ATG8 immunogold labeling to be able to say that these are autophagic vacuoles.
3/ My main concern for the autophagy part is the lack of causal evidence directly linking JAinduced transcriptional responses to autophagy.The author's data and previous studies have shown that JA could induce autophagy, but how this autophagy mediates petal abscission is not clear not us.In addition, the authors are well aware of the pleiotropic phenotypes associated with atg mutants.Stress hormones and other stress responses are all induced in these mutants since they are constantly lacking a core homeostatic pathway.To directly link abscission with autophagy, the authors need to use a tissue-specific CRISPR knockout strategy and use petal abscission-specific promoters.Without this evidence, all the other evidence will be circumstantial.4/ Fig8 ideally should be performed in a +/-Concanamycin A setup, so that we can see the autophagic flux.

Reviewer #3 (Remarks to the Author):
This study tells the story of the mechanism of JA / NAC module in Arabidopsis petal abscission by regulating ROS and autophagy in the petal base, and this article reveals the role of proper autophagy regulated by JA in organ abscission.The research evidence is detailed and accurate, and I have some points to communicate with the author.1.The significance annotation in the chart confused me.For example, in Figure 1c, a, b, c are shown, but an asterisk is marked above the bar chart, and the standard should be unified.2.How do the authors understand that treating ag and dad1 mutants with JA does not fully complement their petal shedding phenotype?3.Whether genes downstream of ANAC102 regulates ROS production in the results of RNAseq and Chip-seq.4.Figure 7 needs to complement the abscission phenotypes of atg8a, the staining results of DAB and Trypan blue, and the abscission phenotype of atg7 can be included in the supplementary data.Extended Data Fig. 3 should be annotated with position.

Point-by-point responses to reviewers' comments
We wish to thank the reviewers very much for their thoughtful and helpful comments that have resulted in the manuscript being strengthened.Please find below our responses to each point.Best wishes, Nobutoshi Yamaguchi, on behalf of all the authors.

Responses to Reviewer #1 General comment by Reviewer #1
In this manuscript, Yamamoto et al. reported that petal abscission is regulated in a jasmonic acid (JA)-dependent spatiotemporal manner, through directing the local cell fate determination via autophagy at the base of petals.The authors found that JA biosynthesis mutants showed delayed petal abscission compared with WT.And during petal abscission, JA is accumulated at the base of the petals and triggers cell differentiation, and ROS levels are also regulated by JA.Besides, the authors performed omics and found that ANAC102, one of the targets of the JA signaling pathway, controls petal abscission.The authors examined the spatiotemporal expression of ANAC102 during petal abscission.And they found that ANAC102 triggers local ROS accumulation and cell death via AUTOPHAGY-RELATED GENES induction, and autophagy at the petal base causes maturation.The authors presented substantial data, and revealed some interesting findings.And most of the conclusions are supported by data presented.And I have the following suggestions.

General Response
We appreciate the constructive feedback provided by Reviewer 1.As indicated in the following Responses, we have carefully considered all of your comments and suggestions during the revision of our manuscript.In addition, we have included 16 new supplementary figures.
The major changes can be summarized as: 1. Observation of JA and atg mutants under a high-magnification stereomicroscope.2. Temporal observation of petal-cell dynamics in JA-related mutants via SEM.
3. Temporal observation of ROS accumulation in JA-related mutants using DAB staining.4. Stamen-abscission measurements conducted on selected JA-related mutants. 5. Expression analysis of JA biosynthetic genes.
We believe that the suggested experiments have provided more precise support for our hypothesis compared to the original version.Please refer to our point-bypoint responses below.

Comment 1 by Reviewer #1
In figure 1a, the authors defined the stages of petal abscission from e.g.-3 to 3.And in figure 1b, they showed the petal abscission phenotypes in different genotypes, mainly JA related.I suggest that the authors show zoom-in phenotypes at key stages in different genotypes (preferably under high magnitude stereomicroscope), to help the readers better visualize the petal abscission phenotypes.

Response
Based on this suggestion, we have added magnified images of inflorescences to the revised version of our supplementary figures.To show the progression of petal abscission in both WT and JA-related mutants, as well as in anac and atg mutants, we examined petal abscission in floral-bud clusters and in petals at positions +3, +5, +7, and +9 (Extended Data Figures 1,13,22,32,33).WT petals were typically detached by position +5, followed by subsequent fruit elongation.Unlike the WT, most of the mutants examined in this study still retained their petals at position +5.We believe that these five new figures will help the readers better visualize the progression of petal abscission.

Comment 2 by Reviewer #1
Line 173, 'To gain more insight into petal abscission, we sectioned the petal base from position 3 flowers...' .The authors found that 'At position d, WT petal cells were partly filled with large vacuoles, indicating cell differentiation (Fig. 1j;top).Notably, we only noticed partial vacuolation of ag and dad1cells at position d.Furthermore, autophagic body-like structures did not accumulate in the vacuoles of WT petal base cells, while those of ag and dad1 contained several autophagic body-like structures in their vacuoles (Fig. 1k,l).'They concluded that 'JA promote petal abscission through petal cell differentiation at their base'.I wonder whether the cellular position and cell size of the petal base at position 5/6/7/8 in the dad1 mutant are similar compared to WT at position 3? Whether JA-treated dad1, aos mutants showed recovered phenotypes at position d, similar compared to WT?And also, whether the coi1, myc2myc3myc4, anac102 mutants show the similar phenotype?Response Indeed, we should have examined petal cell differentiation not only at earlier stages, but also at later stages.We have now conducted all three suggested experiments to better explore temporal changes in cell vacuolation, the effects of JA on cell vacuolation in the respective mutants, and the phenotypes of other JArelated mutants using SEM. 1.We compared petal cells at position +3 in WT with those at position +7 in dad1.We observed that petal cells in dad1 become vacuolated at position +7 (Extended Data Figure 6).2. We performed SEM using JA-treated WT and dad1 petals to observe the rescue of the mutant phenotype.In accordance with the timing of petal abscission, the partial-vacuolation phenotype observed in dad1 petal cells was rescued after JA treatment (Extended Data Figure 7).Thus, petal cells in JA-treated dad1 at position +3 were filled with large vacuoles, like the WT. 3. Finally, we also observed cellular changes in coi1 and myc2 myc3 myc4 petals (Extended Data Figure 6).Similar to what was observed in ag and dad1 cells, we observed partial vacuolation in the coi1 and myc2 myc3 myc4 mutants.These suggested experiments support the original hypothesis that JA promotes petal abscission by affecting petal-cell differentiation at the petal base.We thank you for these suggestions.
Comment 3 by Reviewer #1 3. Related to the second question, whether the reduced DAB staining and trypan blue staining in different JA-related mutants compared with WT at petal base position 3, is due to delayed ROS accumulation in the JA-related, i.e., normal staining at position 5 or 6 or 7 or 8 in JA mutants, compared with WT position 3; or on the other hand, ROS accumulation is reduced at whatever petal positions in the JA mutants?Response We fully agree with this point.In the revised manuscript, we have included images of DAB staining at two different floral positions: position +3 and +7 (Extended Data Figures 5,14,23).At position +3, just prior to petal abscission, the WT exhibited significantly higher ROS accumulation at its petal base compared to the JA-related mutants.However, at position +7, ROS accumulation in the JA-related mutants was similar to that at position +3 of the WT.These findings suggest that ROS accumulation in JA-related mutants is closely linked to the timing of petal abscission rather than to a general deficiency in ROS production.The timecourse analysis of ROS accumulation in both WT and dad1 further supports this interpretation (Extended Data Figure 4).
Comment 4 by Reviewer #1 4. Line 204, the authors 'detected DAD1 promoter activity (pDAD1::GUS) and DAD1 protein accumulation (pDAD1::DAD1-GFP) at the same stages during stamen development (Fig. 2f, g)... DAD1-GFP did not accumulate in petals (Fig. 2h).'They discussed that '...JA production in the stamens for proper pollen maturation and anther dehiscence and coordination of petal development and flower opening by diffusion or active transport to petals.'I am curious whether other JA biosynthesis genes are also expressed in stamens, but not expressed at the base of petals?In addition, whether the stamen abscission and petal abscission are correlated?And whether JA is also involved in stamen abscission?Response Based on this comment, we examined the spatiotemporal expression of the major JA biosynthesis genes AOS (Laudert and Weiler, 1998) and OPR3 (Sander et al., 2000) (Extended Data Figure 9).AOS and OPR3 encode a cytochrome P450 and a 12-oxo-phytodienoic acid (OPDA) reductase, respectively, both of which are required for proper JA biosynthesis.As observed for DAD1 expression, AOS and OPR3 were specifically expressed in stamen filaments, but not in petals.Therefore, not only is the JA intermediate generated in stamens (via DAD1), but the resulting metabolites might also be used to produce JA in stamens via a process catalyzed by AOS and OPR3.
We also observed the timing of petal and stamen abscission in WT and dad1 plants and their correlation (Extended Data Figure 2).In the WT, the timing of petal and stamen abscission were well correlated.In the JA-related mutants, the correlation coefficient between the timing of petal and stamen abscission was close to 1.0, suggesting that JA is involved in stamen abscission as well.
Although GUS expression and stamen-abscission data included in this revision provide additional support for our original hypothesis, the mechanisms of JA diffusion or transport between floral organs is not fully addressed yet.In the revised version, we have discussed future prospects for JA in organ abscission (Page 14 line 604-610).

Responses to Reviewer #2 General comment by Reviewer #2
In this manuscript, the authors investigated the JA-induced transcriptional changes that lead to petal abscission in Arabidopsis.I am not an expert on hormones or transcriptional responses, so I will focus on the cell biology part in my report.In the current state, the manuscript is not conclusively linking autophagy to petal abscission.

General response
We are grateful to Reviewer #2 for the critical comments, which have helped us improve our paper.As indicated below, we have taken all these comments and suggestions into account in the revised manuscript.
Importantly, we obtained causal evidence to directly link JA-induced transcriptional responses to autophagy, based on tissue-specific knockdown and complementation lines, as suggested.We believe that phenotyping of these two types of transgenic lines has strengthened our results.Furthermore, we performed concanamycin A treatment and observed changes in response to inhibited autophagic flux, similar to those of defects in jasmonic acid.Finally, we carefully rephrased our descriptions, as suggested.
We believe that the suggested experiments provide stronger support for our hypothesis.For details, please refer to our point-by-point responses below.
Comment 1 by Reviewer #2 1. Title-Jasmonic acid-induced instead of mediated to avoid having 2 mediated in the title?

Response
Thank you for pointing this out.Based on this suggestion, we have changed the title of our manuscript to the following: 'Petal abscission is promoted by jasmonic acid-induced autophagy at Arabidopsis petal bases'.
Comment 2 by Reviewer #2 2. Fig1k/l-autophagic body-like structures is a very vague term that is not backed up by any data.There are various vacuolar trafficking pathways that could be the source of those vesicles.The authors need to at least back this up with some ATG8 immunogold labeling to be able to say that these are autophagic vacuoles.Response Thank you for raising this important point.One of our authors, Kiminori Toyooka, successfully performed immunogold labeling using anti-ATG8 antibody in a previous study (Yoshimoto et al., 2014, Journal of Cell Science).However, despite his expertise, sectioning of the petal base for immunogold labeling would be technically challenging because the fixation solution is transparent, unlike the osmium solution used for TEM, rendering the flowers invisible during fixation.
To avoid using such an ambiguous term, we rephrased and toned down the descriptions of Fig. 1k and 1l in the revised manuscript.What we had intended to say is that round membrane structures containing cytoplasmic components were observed within vacuoles in the ag or JA-related mutant backgrounds.In the revised version, we now refer to them as small vesicles within vacuoles (Results, Page 5, lines 189-190) (Discussion, Page 14, lines 589-591).In addition, as indicated by Reviewer 2, we now stress that these vesicles could be generated not only due to defects in autophagy, but also through various vacuolar trafficking pathways (Results,Page 5,.(Discussion, Page 14, lines 589-591).
Comment 3 by Reviewer #2 3. My main concern for the autophagy part is the lack of causal evidence directly linking JA-induced transcriptional responses to autophagy.The author's data and previous studies have shown that JA could induce autophagy, but how this autophagy mediates petal abscission is not clear not us.In addition, the authors are well aware of the pleiotropic phenotypes associated with atg mutants.Stress hormones and other stress responses are all induced in these mutants since they are constantly lacking a core homeostatic pathway.To directly link abscission with autophagy, the authors need to use a tissue-specific CRISPR knockout strategy and use petal abscission-specific promoters.Without this evidence, all the other evidence will be circumstantial.

Response
We fully agree with this critical point.To provide causal evidence, we have performed two parallel experiments: 1) Tissue-specific knockdown and 2) tissuespecific genetic complementation.In the revised manuscript, these data are presented in the new Figure 8.
1. Tissue-specific knockdown.As suggested by Reviewer 2, CRISPR-TSKO would be one of the best techniques for efficient mutagenesis in specific cell types, tissues, or organs (Decaestecker et al., 2019, Plant Cell).Although one of our authors, Tatsuaki Goh, has a CRISPR-TSKO gene-knockout tool set based on the Golden Gate system, we have not yet established this gene-knockout technique due to some technical issues.Instead, we generated transgenic Arabidopsis lines expressing an artificial microRNA targeting ATG5 (amiR-ATG5) under the control of the NAC102 promoter, which drives transcription at the base of position +2 flowers onward in the atg5 gATG5-GFP background (Schwab et al., 2006, Plant Cell;Goh et al., 2022, Development).The amiR construct leads to lower ATG5-GFP fluorescence specifically at the petal base prior to abscission.The NAC102pro:amiR-ATG5 and NAC102pro:amiR-ATG7 plants have delayed petal abscission, fewer dead cells, and lower DAB accumulation in petal bases.These results suggest that local autophagy activity at the petal base is required for petal abscission.

Tissue-specific rescue.
To further clarify whether activation of autophagy at the petal base is sufficient for petal abscission, we generated plants expressing mVENUS-ATG7 in the atg7 mutant background under the control of the NAC102 promoter, which drives transcription at the base of position +2 flowers onward.We detected dot-like mVENUS-ATG7 signals at the petal base.Phenotyping, as well as trypan blue and DAB staining, revealed that the delay in petal abscission seen in the atg7 mutant was restored to WT in these transgenic plants.These results suggest that the activation of autophagy at the petal base is sufficient to trigger petal abscission.
Decaestecker W, Buono RA, Pfeiffer ML, Vangheluwe N, Jourquin J, Karimi M, Van Isterdael G, Beeckman T, Nowack MK, Jacobs TB.CRISPR-TSKO: A Technique for Efficient Mutagenesis in Specific Cell Types, Tissues, or Organs in Arabidopsis ( 2019 Comment 4 by Reviewer #2 4. Fig8 ideally should be performed in a +/-concanamycin A setup, so that we can see the autophagic flux.

Response
In Fig. 8, we have examined the effect of JA signaling on GFP-ATG8a dynamics during petal abscission.However, how inhibition of autophagy affects GFP-ATG8a dynamics during petal abscission was unclear.To determine whether we would observe similar changes in GFP-ATG8a dynamics in plants with inhibited autophagic flow and in plants with jasmonic acid defects, we have now conducted concanamycin A treatment during petal development (Extended Data Figure 35).As with the JA-related mutants, the degradation of ATG8a was delayed at the petal base upon concanamycin A treatment.This significant delay was confirmed using immunoblotting-based autophagic-flux assays with concanamycin Atreated petals.These results suggest that the inhibition of autophagy and defects in JA signaling have similar effects on GFP-ATG8a dynamics during petal abscission.
Interestingly, we observed differences in autophagosome formation in petals treated with concanamycin A and in petals from JA-related mutants.Concanamycin A is a specific inhibitor of vacuolar type H + -ATPase activity that causes autophagic bodies to accumulate in vacuoles without affecting autophagosome shape.Indeed, we often observed typical ring-shaped autophagosomes in both mock-and concanamycin A-treated petal bases.By contrast, we rarely saw ring-shaped autophagosomes seen in the JA-related mutants, suggesting that JA signaling also affects autophagosome formation either directly or indirectly due to the lack of a core homeostatic pathway.In the revised manuscript, we have included a discussion about both the similarities and differences in the JA-and concanamycin A-mediated pathways.

Responses to Reviewer #3 General comment by Reviewer #3
This study tells the story of the mechanism of JA / NAC module in Arabidopsis petal abscission by regulating ROS and autophagy in the petal base, and this article reveals the role of proper autophagy regulated by JA in organ abscission.The research evidence is detailed and accurate, and I have some points to communicate with the author.

General Response
We appreciate the favorable comments and useful suggestions provided by Reviewer 3. As indicated in the following responses, we have carefully considered all of these comments and suggestions in the revised manuscript.
The major revisions are as follows: 1.We observed the phenotypes of the atg8a mutant.2. We added descriptions of the direct targets of ANAC102 involved in ROS accumulation.3. We added a detailed explanation of the rescue of defects in ag and dad1 by JA treatment.
We believe that the suggested experiments provided more precise support for our hypothesis compared to the original version.For details, please refer to our point-by-point responses below.
Comment 1 by Reviewer #3 1.The significance annotation in the chart confused me.For example, in Figure 1c, a, b, c are shown, but an asterisk is marked above the bar chart, and the standard should be unified.

Response
We made the requested changes to our figures.We have also expanded and updated the description of the figure labels throughout the manuscript.
Comment 2 by Reviewer #3 2.How do the authors understand that treating ag and dad1 mutants with JA does not fully complement their petal shedding phenotype?Response We conducted JA treatment via a single spray application with methyl jasmonate.This regimen rescued petal abscission without causing visible secondary effects (e.g.cellular damage), but petal abscission in only 2 or 3 flowers per plant were fully rescued 2 days after application.The remaining flowers were either rescued only partially or not at all.Previously, one of our authors, Sumie Ishiguro, also demonstrated the rescue of a limited number of flowers in JA biosynthesis mutants via a single spray treatment (Ishiguro et al., 2001, Plant Cell: Please see Fig. 3E in the Plant Cell paper).Therefore, obtaining the proper floral stage before treatment is critical for successful rescue.We believe that the partial rescue was largely due to subtle differences in the stages of flowers before JA treatment.
We believe that the petal-abscission phenotype could be rescued by performing multiple JA treatments.Nevertheless, we are inclined not to pursue this approach because it could potentially trigger a JA-mediated defense response and/or lead to secondary effects and confound the experiment.In the revised manuscript, we have provided a comprehensive description of our methods and the outcomes of the rescue process in the Results section.
Comment 3 by Reviewer #3 3.Whether genes downstream of ANAC102 regulates ROS production in the results of RNA-seq and Chip-seq.Response Indeed, ANAC102 directly controls the expression of genes involved in ROS production based on RNA-seq and ChIP-seq analysis.For example, the PEROXIDASE 33 (PRX33) gene encodes a Class III peroxidase that plays a role in generating H2O2 (Arnaud et al., 2023, Plant Physiology).Another example is the PEROXIN11d (PEX11d) gene, which encodes a peroxisomal membrane protein that regulates peroxisome proliferation (Orth et al., 2007, Plant Cell).In the revised manuscript, we have added descriptions of a few key genes involved in ROS production that are directly regulated by ANAC102, along with expression analysis, binding peaks, and important citations (Extended Data Figure 28, and 29).Orth T, Reumann S, Zhang X, Fan J, Wenzel D, Quan S, Hu J (2007).The PEROXIN11 protein family controls peroxisome proliferation in Arabidopsis.Plant .
Comment 4 by Reviewer #3 4.Figure 7 needs to complement the abscission phenotypes of atg8a, the staining results of DAB and Trypan blue, and the abscission phenotype of atg7 can be included in the supplementary data.

Response
We have added an analysis of the atg8a mutant.For this purpose, we obtained the atg8a mutant from TAIR and conducted phenotypic analysis.atg8a (SALK_012133) harbors a T-DNA within the gene body, resulting in undetectable levels of ATG8a transcripts.The atg8a mutant has delayed petal abscission, accompanied by lower ROS accumulation, and fewer dead cells at the petal base than the WT.
In the revised manuscript, we present the atg8a characterization in Fig. 7.We describe genotyping of atg8a and the atg8a and atg7 mutant phenotypes in the supplementary figures .
Comment 5 by Reviewer #3 Extended Data Fig. 3 should be annotated with position.

Response
We have made the requested change.