# Source paper archive

- Source title: Integrating redox dynamics and gene networks reveals ROS-mediated control of Arabidopsis seed germination under water stress
- Author(s): Bailly, C., Vidal, A., Wagner, S., Meyer, A., Barneche, F., Mu, X.
- Source landing URL: https://www.biorxiv.org/content/10.64898/2026.11.16.766966v1
- PDF URL: https://www.biorxiv.org/content/10.64898/2026.11.16.766966v1.full.pdf
- PDF SHA-256: 428651ee524f06096703ad2cc16729bf518f1f828ad0fb5322d62fe4eded826a
- Converter: pymupdf4llm 0.0.27
- Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version.

---

&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1 **Integrating redox dynamics and gene networks reveals ROS-mediated control of**


2 **Arabidopsis seed germination under water stress**


3


4 **Running title:** ROS control of seed germination under water stress


5


6 **All author names and affiliations:**


7 Xiujie Mu &#91;1,2&#93;, Adrien Vidal &#91;1,2&#93;, Stephan Wagner &#91;3&#93;, Andreas Meyer &#91;3&#93;, Fredy Barneche &#91;1,2+&#93;,


8 Christophe Bailly &#91;1,2&#93;  

9 ¹Sorbonne Université, CNRS, Inserm, Development, Adaptation and Ageing, Dev2A,


10 F-75005 Paris, France


11 ²Sorbonne Université, CNRS, Inserm, Institut de Biologie Paris-Seine, IBPS, F-75005


12 Paris, France


13 3 INRES-Chemical Signalling, University of Bonn, Römerstraße 164, D-53117 Bonn,


14 Germany


15 + co-last author


16 *To whom correspondence should be addressed: Christophe Bailly


17 &#91;(christophe.bailly@sorbonne-universite.fr)&#93;(mailto:christophe.bailly@sorbonne-universite.fr)


18


19 **Author for contact:** Christophe Bailly


20 Sorbonne University


21 &#91;E-mail address: christophe.bailly@sorbonne-universite.fr&#93;(mailto:christophe.bailly@sorbonne-universite.fr)


22 Telephone number: + 33 1 44 27 59 29


23 Mailing address: UMR8263 – Biologie des semences


24 Sorbonne Université


25 Bâtiment C, 2ème étage, Boîte 24


26 4 Place Jussieu


27 75252 Paris Cedex 05


28 France


1


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


29


30 **ABSTRACT**


31 Seed germination is a critical step of plant development that is highly sensitive to water


32 availability. During this delicate transition, water stress responses are tightly regulated


33 by multilevel molecular networks that integrate environmental cues into developmental


34 adaptations. Reactive Oxygen Species (ROS) play a positive signaling role in seed


35 germination, yet their influence on germination under water stress remains poorly


36 understood. Here, we investigated the involvement of ROS in _Arabidopsis thaliana_


37 seed germination upon reproducibly reducing water potential using polyethylene glycol


38 (PEG). The data show that delayed germination under water stress is not simply a


39 consequence of reduced water uptake. Instead, water stress triggers coordinated


40 reprogramming events characterized by altered H 2 O 2 homeostasis, particularly in


41 peroxisomes, which functionally mediates a rebalancing of ABA and GA metabolism


42 toward developmental slowdown. A weighted gene co-expression network analysis of


43 water-stress transcriptomic dynamics identified a gene hub associated with H 2 O 2


44 production during germination, shedding light on multiple oxygen- and redox-related


45 processes that are strongly linked to transcriptional and translational attenuation.


46 Collectively, this study unveils a role for regulating peroxisomal H 2 O 2 homeostasis in


47 _Arabidopsis_ seed water stress responses, offering new perspectives for improving crop


48 resilience to environmental constraints.


49 **Keywords** : _Arabidopsis thaliana,_ germination, reactive oxygen species, seed, water


50 stress


51


52 **INTRODUCTION**


53 Seed germination is a critical stage of plant development that is highly sensitive to


54 water availability. Water deficit significantly reduces germination speed, delays radicle


55 protrusion, and decreases seedling establishment (Vallejo et al., 2010; Liu et al., 2019;


56 Wang et al., 2022; Kolesnikov et al., 2023). The importance of germination for crop


57 productivity is illustrated by numerous studies showing that water deficit during or


58 immediately after sowing reduces emergence rates, delays establishment, and


59 ultimately decreases yield in major crops, including wheat, maize, rice, soybean,


60 sunflower, and rapeseed (Farooq et al., 2009; Liu et al., 2010; Hasanuzzaman et al.,


61 2013; Bailly and Gomez Roldan, 2023). Because seedling establishment represents a


62 major bottleneck in crop production, improving germination performance under water


2


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


63 stress is increasingly recognized as a priority for crop adaptation to climate change


64 (Bailly and Gomez Roldan, 2023). Although the inhibitory effects of water stress on


65 germination have been widely documented, the molecular mechanisms through which


66 germinating seeds perceive water limitation and translate this environmental cue into


67 coordinated changes in hormonal and developmental responses remain poorly


68 understood.


69 Reactive Oxygen Species (ROS) are attractive candidates for constituting the initial


70 sensing layer during germination under water-limited conditions. According to the


71 “oxidative window” concept, successful germination requires ROS accumulation within


72 a permissive range that is sufficient to activate germination-related processes without


73 causing oxidative damage (Bailly et al., 2008). Hence, rather than acting solely as toxic


74 by-products of aerobic metabolism, ROS are broad integrators of environmental


75 signals during germination (Bailly, 2008, 2019, 2022, Jurdak et al., 2020). ROS


76 signaling output depends on their concentration, subcellular localization, and temporal


77 dynamics. In seeds of sunflower, water stress induces significant changes in ROS


78 accumulation and antioxidant enzyme activities during germination (Saux et al., 2020),


79 while similar responses have been reported in seedlings exposed to drought conditions


80 (Vallejo et al., 2010; Liu et al., 2019; Wang et al., 2022; Kolesnikov et al., 2023).


81 Beyond their overall abundance, the cellular origin and subcellular distribution of ROS


82 are increasingly recognized as critical determinants of their signaling functions. During


83 germination, ROS are notably generated in mitochondria, peroxisomes, the apoplast,


84 and plasma membrane-associated NADPH oxidases, each contributing distinct redox


85 signals (Bailly, 2019; Jurdak et al., 2022). Among these compartments, peroxisomes


86 are of particular interest because they are highly active during germination and play


87 central roles in reserve mobilization through fatty acid β-oxidation and the glyoxylate


88 cycle (Graham, 2008). These metabolic pathways generate H 2 O 2 as a by-product,


89 positioning peroxisomes at the interface between metabolism and redox signaling


90 (Graham, 2008; Pracharoenwattana et al., 2010). Peroxisomes may therefore act not


91 only as metabolic organelles but also as important regulators of germination through


92 ROS production (Jurdak et al., 2022). However, despite increasing recognition of the


93 importance of ROS compartmentation, little is known about how environmental


94 stresses influence ROS dynamics within specific organelles during germination. In


95 particular, whether water stress modifies peroxisome-derived ROS signaling and how


96 such changes contribute to the regulation of germination progression remain largely


97 unexplored.


3


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


98 ROS are likely to influence germination by modulating the metabolism and signaling


99 pathways of the phytohormones abscisic acid (ABA) and gibberellic acid (GA) (Bailly,


100 2022), which directly contribute to germination-associated water stress responses.


101 Under water deficit, ABA levels and signaling generally increase, reinforcing


102 dormancy-related processes and inhibiting germination, whereas GA biosynthesis and


103 responsiveness are reduced, limiting embryo growth and radicle protrusion (Llanes et


104 al., 2016; Liu et al., 2019; Hassan et al., 2021). In _A. thaliana_, rice, sunflower, and


105 wheat, low water potential conditions induce genes associated with ABA signaling,


106 osmotic stress responses, antioxidant systems, and reserve mobilization (Liu et al.,


107 2019; Saux et al., 2020; Kolesnikov et al., 2023). Increasing evidence indicates that


108 ROS are closely integrated into this hormonal network. Hydrogen peroxide (H 2 O 2 )


109 addition promotes germination by stimulating GA biosynthesis and ABA catabolism


110 while repressing ABA biosynthesis and GA degradation (Liu et al., 2010; Cheng et al.,


111 2022). Conversely, ABA suppresses ROS accumulation in several species, including


112 _A. thaliana_, notably through the activation of the ABA signaling transcription factor


113 ABA-INSENSITIVE_5 (ABI5), which induces ROS scavenging via CATALASE1 (Ye et


114 al., 2012; Lariguet et al., 2013; Bi et al., 2017). Together, these observations indicate


115 that ROS are important intermediates in linking water stress perception to the


116 transcriptional regulation of ABA-GA signaling and germination progression.


117 In vegetative tissues, drought similarly induces genes involved in ABA signaling,


118 ROS detoxification, osmoprotection, and stress defense, while repressing pathways


119 associated with cell division, protein synthesis, and growth (Shinozaki and Yamaguchi

120 Shinozaki, 2006; Skirycz and Inzé, 2010; Claeys and Inzé, 2013; Nakashima et al.,


121 2014). At the molecular level, these responses involve large-scale activation of ABA

122 responsive transcription factors, including members of the AREB/ABF, NAC, DREB,


123 and WRKY families, together with extensive reorganization of primary metabolism and


124 redox homeostasis (Shinozaki and Yamaguchi-Shinozaki, 2006; Nakashima et al.,


125 2014). Notably, drought-induced growth inhibition is frequently associated with


126 repression of ribosome biogenesis, translation, DNA replication, and chromatin

127 associated functions, suggesting that plants actively redirect resources from growth


128 toward stress adaptation (Skirycz and Inzé, 2010; Claeys and Inzé, 2013). Compared


129 with vegetative tissues, much less is known about transcriptomic responses to water


130 stress during seed germination, and a comprehensive understanding of how water


131 stress triggers a coordinated, temporally regulated reprogramming linking redox


132 signaling, hormonal responses, and developmental competence during germination


4


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


133 remains elusive.


134 We hypothesized that water stress modifies ROS homeostasis and associated


135 signaling pathways, thereby altering the regulatory networks controlling germination


136 progression. Combining germination phenotyping, transcriptomic analyses, ROS


137 biochemical measurements and _in vivo_ monitoring of subcellular redox dynamics, we


138 deciphered how water stress affects ROS homeostasis and associated signaling


139 pathways across successive stages of germination. Our study identifies the temporal


140 sequence of molecular events that link water stress perception to delayed germination


141 and reveals a central role for peroxisome-associated ROS signaling in these regulatory


142 mechanisms.


143


144 **RESULTS**


145 **Seed germination under water stress triggers transcriptome reprogramming**


146 _A. thaliana_ wild-type seeds were germinated on PEG solutions with water potentials


147 ranging from -0.2 MPa to -1.2 MPa. While stratified seeds fully germinated within 3


148 days in water, increasing water potential gradually delayed germination (Figure 1A).


149 For the highest level tested (-1.2 MPa), the final germination rate was severely reduced


150 (Figure 1A). For the remainder of this study, the -0.9 MPa PEG condition was selected


151 because it delayed germination without affecting the final germination percentage


152 (Figure 1A). Based on the germination curves (Figure 1A), three sampling time points


153 corresponding to successive stages of the progression toward germination completion


154 were defined for each condition (Figure 1B). These included an early stage (ET, 3 h


155 after imbibition), a middle stage (MT, corresponding to half of the time required to reach


156 the late stage), and a late stage (LT), defined as one day before the onset of radicle


157 protrusion. Although the time required to reach these stages differed with each water


158 potential tested, the selected sampling points represented equivalent physiological


159 stages of germination progression under both conditions (Figure 1B). The water stress


160 imposed by the -0.9 MPa PEG solution can be considered moderate, as seed water


161 content decreased only slightly, from approximately 0.83 to 0.80 g H 2 O/g fresh weight,


162 during imbibition (Figure 1C), and radicle emergence was not prevented.


5


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


163


164 **Figure 1. Water stress delays seed germination**


165 _(_ _**A**_ _) Germination of stratified seeds in water or a range of PEG solutions at 15°C in the dark (_ _**B**_ _) The 3 time_


166 _points used for analyses, defined according to the germination kinetics on water and on a -0.9 MPa PEG_


167 _solution (_ _**C**_ _) Seed water content at initial time (after stratification, T0) and at the 3 timepoints defined in_


168 _Figure 1B during imbibition at 15°C under the dark. Data presented are means of 3 replicates ± SD._


169 _Student's t-test was used to determine the statistical significance between the two groups at each time_


170 _point. An asterisk below the bar chart indicates a significant difference between the two groups (*, P &lt;_


171 _0.05; **, P &lt; 0.01; ***, P &lt; 0.001)._


172 To investigate transcriptomic responses to water stress during germination,


173 stratified seeds were imbibed in water or in a PEG solution at -0.9 MPa and samples


174 were used for RNA-seq analyses at the 3 time points defined above. Water stress


175 triggered 252, 9844, and 12,653 differentially expressed genes (DEGs) at ET, MT, and


176 LT, respectively (Figure 2A, Table S2-4). At ET, upregulated genes were enriched in


177 secondary metabolism, sulfur compound biosynthesis, glucosinolate metabolism, and


178 stress-related pathways, and included well-characterized stress-responsive genes


179 (Table S1), such as _HSP17.6B_, a member of the heat shock protein (HSP) family


180 (Chen et al., 2024), _LTP3_, encoding a non-specific lipid transfer protein (Gao et al.,


181 2016), and _PUB19_, encoding a U-box E3 ubiquitin ligase involved in ABA and stress


182 signaling (Liu et al., 2011). Concomitantly, several members of the xyloglucan


183 endotransglucosylase/hydrolase family, including _XTH16_, _XTH19_, _XTH22_, and _XTH27_


184 (Rose et al., 2002), were significantly downregulated under PEG treatment (Table S1).


185 In addition, the glycosyltransferase gene _F8H_ (Lee et al., 2009) and the growth

186 promoting gene _EXORDIUM_ ( _EXO_ ; Schröder et al., 2009) were repressed (Table S1).


6


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


187 At MT, up-regulated genes were significantly enriched in GO categories related to


188 oxygen deprivation and redox metabolism (Figure 2B), including _GPX8_ (Glutathione


189 peroxidase 8; Gaber, 2013); _GSTF11_ (Glutathione S-transferase F11; Horváth et al.,


190 2023); _GSTU14_ and _GSTU2_ (Glutathione S-transferase U14/2; Horváth et al., 2020),


191 _SRO5_ (Similar to RCD-ONE 5; (Babajani et al., 2009), as well as hypoxia-responsive


192 genes, including _AHB1_ ( _Arabidopsis_ nonsymbiotic hemoglobin 1; Perazzolli et al.,


193 2004), _HRU1_ (Hypoxia responsive universal stress protein 1; Gonzali et al., 2015), and


194 _ADH1_ (Alcohol dehydrogenase 1; Peng et al., 2001) together with members of the


195 small heat shock protein family, were strongly induced (Table S2). Down-regulated


196 genes were mainly associated with processes linked to RNA metabolism and seed


197 developmental programs (Figure 2B). At LT, the transcriptional response shifted


198 toward the regulation of signaling and enzymatic activities with up-regulated genes


199 predominantly enriched in GO terms associated with protein phosphorylation and


200 kinase regulation, including regulation of cyclin-dependent protein serine/threonine


201 kinase activity, and more generally the regulation of catalytic activity (Figure 2D). The


202 “electron transport chain” GO terms remained over-represented at this stage,


203 suggesting a persistent adjustment of respiratory and redox-related processes under


204 water stress. Conversely, down-regulated genes at LT were enriched in categories


205 related to mRNA and non-coding RNA metabolism and developmental regulation


206 (Figure 2D). Lastly, key genes involved in peroxisomal ROS production were


207 downregulated, including _ACX1_ and _ACX2_, which encode acyl-CoA oxidases


208 catalyzing the H 2 O 2 -generating step of β-oxidation (Eastmond et al., 2000; del Río et


209 al., 2006). Expression of the peroxin 5 ( _PEX5_ ) gene, which encodes a major receptor


210 required for peroxisomal protein import (Skowyra and Rapoport, 2022), was also


211 reduced (Table S3), suggesting a broader attenuation of peroxisomal metabolic


212 activity.


7


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


213


214 **Figure 2. Water stress during germination induces specific seed transcriptomic**


215 **responses**


216 _(_ _**A**_ _) Number of differentially expressed genes (DEGs), including up- and down-regulated genes (FDR &lt;_


217 _0.01), at the 3 germination time points at 15°C under water stress conditions (ET, early timepoint; MT,_


218 _mid timepoint; LT, late timepoint). (_ _**B**_ _-_ _**E**_ _) Gene set enrichment analysis (GSEA) of DEGs. The scatter plots_


219 _show significantly enriched Gene Ontology (GO) terms and KEGG pathways under water stress at MT_


220 _(_ _**B**_ _,_ _**C**_ _) and LT (_ _**D**_ _,_ _**E**_ _). Each dot represents a biological process or pathway, and its regulation status (up-_


221 _or down-regulated) is indicated by the normalized enrichment score. Dot size represents the proportion of_


222 _regulated genes relative to the total number of genes in the corresponding pathway, while color indicates_


223 _the adjusted p-value._


224 KEGG pathway enrichment analyses further showed that, at MT, up-regulated


225 genes are strongly enriched for pathways related to the translational machinery and


226 RNA processing, notably ribosome and spliceosome terms, as well as photosynthesis


227 and pentose and glucuronate interconversions (Figure 2C). By contrast, down

228 regulated genes were enriched in pathways associated with intracellular trafficking and


8


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


229 transport. At LT, functions related to ribosome activity, photosynthesis, RNA


230 degradation, and “spliceosome” pathways were still significantly upregulated, whereas


231 down-regulated pathways included nucleocytoplasmic transport, ABC transporters,


232 ribosome biogenesis, and ATP-dependent chromatin remodeling (Figure 2E).


233 Altogether, these analyses indicate that water stress induces a major transcriptional


234 reprogramming during _Arabidopsis_ seed germination, characterized by activation of


235 oxygen- and redox-responsive pathways at MT, followed by the induction of kinase

236 related signaling pathways at LT, while maintaining strong modulation of ribosome

237 associated functions throughout germination.


238 **Mitochondrial electron transport chain activity is altered by water stress during**


239 **seed germination**


240 Considering that water stress specifically induces gene sets associated with


241 mitochondrial electron transport chain (mETC), which plays a critical role in energy


242 production during germination (Nietzel et al., 2020), we assessed seed respiratory


243 activity during germination. As estimated by changes in oxygen uptake within a 10 min


244 interval before and after the selected time point, respiration rate progressively


245 decreased with increasing PEG concentrations, with the strongest inhibition observed


246 24 h after imbibition (Figure 3A). Consistent with these results, respiration rates were


247 significantly reduced at the late stage of germination under -0.9 MPa (Figure 3B).


248


249 **Figure 3. Water stress during germination alters seed respiration**


250 _(_ _**A**_ _) O_ _2_ _consumption rate of stratified seeds imbibed on water or on a range of PEG solutions at 15°C in_


251 _the dark. Data are means of 2 replicates of each concentration. (_ _**B**_ _) O_ _2_ _consumption rate of stratified seeds_


252 _imbibed at 3 timepoints of germination under control or water stress conditions (ET, early timepoint; MT,_


253 _mid timepoint; LT, late timepoint). Data presented are means of 6 replicates ± SD. Student's t-test was_


254 _used to determine the statistical significance between the two groups at each time point. An asterisk below_


255 _the bar chart indicates a significant difference between the two groups (*, P &lt; 0.05)._


256 The mitochondrial redox state was monitored during germination using the


9


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


257 mitochondria-targeted roGFP2-Orp1 sensor with calibrated 405/488 nm excitation


258 ratios as a proxy for redox potentials (Nietzel et al., 2019). To validate sensor


259 functionality and determine its dynamic range in seeds, embryos were treated with 10


260 mM DTT and 100 mM H 2 O 2 to obtain fully reduced and oxidized sensor states,


261 respectively (Figure 4A, S1A). These treatments produced clear shifts in fluorescence


262 ratios, confirming the probe's responsiveness under the imaging conditions used


263 (Figure 4B). In control conditions, the sensor oxidation level showed slight fluctuations


264 during seed germination, suggesting the involvement of mitochondrial redox regulation


265 in the germination process (Fig. 4A, B). A similar oxidation pattern was observed in


266 seeds exposed to water stress (-0.9 MPa) throughout germination, although with a


267 slight reduction in the sensor oxidation level (Figure 4B, C). We concluded that


268 moderate water stress did not substantially alter the mitochondrial redox environment


269 during germination, despite its inhibitory effect on germination kinetics and respiratory


270 activity.


271


272 **Figure 4** . **Mitochondrial roGFP2-Orp1 redox state during seed germination**


273 _(_ _**A**_ _) Confocal microscopy imaging of Mito-roGFP2-Orp1-emitted 405/488 fluorescence ratios in hypocotyl_


274 _area upon seed imbibition in water (control) or a -0.9 MPa PEG solution (water stress) at 15°C in the dark._


275 _The Mito-roGFP2-Orp1 biosensor signal (left panel) was calibrated using 10 mM dithiothreitol (DTT) for_


276 _full reduction or 100 mM H_ _2_ _O_ _2_ _for full oxidation. (ET, early timepoint; MT, mid timepoint; LT, late timepoint)_


277 _Scale bars = 10 μm. (_ _**B**_ _) Redox ratio analysis at the three time points (_ _**C**_ _) Redox potentials at the three_


278 _time points. In B and C, data were obtained from confocal imaging of at least 14 embryos in two_


279 _independent experiments._


10


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


280 **Water stress perturbs ROS homeostasis**


281 To further assess the involvement of ROS during germination under water stress, we


282 used the ROS wheel model (Willems et al., 2016) to search for specific ROS footprints


283 in our transcriptomic data (Table S4). The genes induced by water stress showed a


284 significant overlap with Cluster III (“HL early”) following imbibition, indicating activation


285 of an early redox signaling program typically associated with ROS-mediated stress


286 responses and alterations in cellular oxygen homeostasis.


287 With increasing PEG concentrations, H 2 O 2 production gradually decreased after 24


288 h of imbibition (Figure 5A). Consistently, H 2 O 2 production increased progressively


289 during germination under both control and water stress conditions (Figure 5B).


290 However, water stress significantly decreased H 2 O 2 production at all timepoints,


291 resulting in approximately 1.3- and 1.5-fold lower levels at MT and LT, respectively,


292 compared with the control. Likewise, catalase (CAT) activity also increased during


293 germination but was significantly higher in PEG-treated seeds at the late stage (Figure


294 5C). Despite these changes, malondialdehyde (MDA) content remained unchanged


295 between treatments at all time points (Figure 5D), indicating that water stress did not


296 induce detectable oxidative damage to lipids. To further assess the contribution of ROS


297 to germination under water stress, seeds were germinated in the presence of ROS


298 scavengers. All scavengers had no effect on germination under control conditions, but


299 an inhibitory effect was observed under water stress, except for DMTU, an H 2 O 2


300 scavenger (Figure 5E). These results suggest that ROS production contributes


301 positively to germination and that the reduced germination observed under water


302 stress may be associated with altered ROS signaling.


303 To gain a spatial view of ROS production under water stress, we probed the


304 fluorescent signal of H2DCFDA application. During the early and middle stages of


305 germination, fluorescence signals appeared as discrete punctate structures distributed


306 both within the cell and at the cell periphery (Figure 5F). As germination progressed,


307 fluorescence became predominantly localized to intracellular punctate structures.


308 Interestingly, fluorescence intensity appeared reduced under water stress conditions,


309 particularly at the late stage of germination, consistent with the lower ROS


310 accumulation observed in seed tissues. The punctate pattern of fluorescence suggests


311 that ROS production is associated with specific subcellular compartments. Although


312 H2DCFDA does not allow their precise identification, these structures may correspond


313 to organelles involved in ROS metabolism, including peroxisomes, which are known to


314 play important roles during seed germination and dormancy release


11


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


315 (Pracharoenwattana et al., 2010; Jurdak et al., 2022). These observations indicate that


316 ROS production is subject to a spatiotemporal regulation during germination and may


317 be altered by water stress.


318


319 **Figure 5. ROS homeostasis is involved in the response to water stress during**


320 **germination**


321 _(_ _**A**_ _) H_ _2_ _O_ _2_ _production of stratified seed imbibed with water or a range of PEG solutions for 24 h at 15°C_


322 _under dark conditions. (_ _**B**_ _) H_ _2_ _O_ _2_ _production (_ _**C**_ _) CAT activity and malondialdehyde (MDA) accumulation_


323 _(_ _**D**_ _) at 3 time points during germination under normal or water stress conditions at 15°C in the dark. (_ _**E**_ _)_


324 _Germination of stratified Col-0 seed after 10th days of imbibition at 15°C in the dark under normal or water_


325 _stress conditions in the presence of ROS scavengers. Trion (0.1 mM; superoxide scavenger;_ Wise &amp;


326 Naylor, 1987 _); Sodium benzoate (0.1 mM; free radical scavenger;_ Hung &amp; Kao, 2004 _); Dimethylthiourea_


327 _(0.1 mM; Hydrogen peroxide scavenger;_ Levine &amp; Lamb, 1994 _). (_ _**F**_ _) Localization of ROS production in_


328 _embryonic axes of non-dormant seeds using 2′,7′-dichlorofluorescein diacetate (H2DCFDA) at 15 °C in_


329 _the dark in the presence or absence of water stress. Data presented are means ± SD of three replicates_


330 _(A and E), or 12 replicates (B and F), or six replicates (C and D). Student's t-test was used to determine_


331 _the statistical significance between treatments at each time point; an asterisk above the bar chart indicates_


12


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


332 _a significant difference between the two groups (*, P &lt; 0.05; **, P &lt; 0.01; ***, P &lt; 0.001)._


333 To investigate a possible role of peroxisomes in ROS production, we monitored their


334 redox status using the stably expressed peroxisome-targeted roGFP2-Orp1 transgenic


335 reporter (Arnaud et al., 2023). Treatment of embryos with DTT and H 2 O 2 led to


336 significant differences in fluorescence ratios, thereby validating the sensor’s dynamic


337 response to H 2 O 2 (Figure S1B). Under control conditions, the redox level exhibited a


338 slight increase at the middle time point, followed by a slight decrease at the late time


339 point. However, these changes were not statistically significant. (Figure 6 A-C). Water


340 stress significantly reduced peroxisomal sensor oxidation while preserving its overall


341 temporal dynamics, with the strongest effect at the middle and late stages of


342 germination (Figure 6A-C). Accordingly, the redox potential shifted to more negative


343 values !"#$%&amp;'()%$*%+,%-.%,#%&amp;'//%$*%+,%0.1%23425+,236%,7+,%89"#:2;#$9;%"9$+2394%


344 $#"9%"94&lt;594%&lt;349"%=+,9"%;,"9;;%,7"#&lt;67#&lt;,%69"$23+,2#3&gt;


345 To confirm the role of peroxisomes in water stress, we assessed the expression of


346 genes encoding peroxisomal glycolate oxidases, which produce H 2 O 2 (Esser et al.,


347 2014; Engqvist et al., 2015) and as _CAT2_, which encodes the major peroxisomal


348 catalase responsible for H₂O₂ scavenging (Frugoli et al., 1996). These enzymes are


349 therefore likely to contribute to the changes in peroxisomal redox status observed


350 during germination under water stress. _GOX2_ and _GOX3_ transcript levels were lower


351 under water stress than in control conditions (Figure S2A). Moreover, knock-out mutant


352 lines for both genes are highly sensitive to water stress during germination (Figure


353 S2C). In contrast, although _CAT2_ transcript abundance was reduced under water


354 stress, the _cat2_ mutant did not display altered tolerance to water stress during


355 germination (Figure S2B and S2D). Together, these findings indicate that peroxisomal


356 ROS production contributes to the regulation of germination under water stress, and


357 its lowering upon PEG treatment functionally contributes to seed germination delay.


13


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


358


359 **Figure 6. Peroxisomal redox status is altered by water stress during germination**


360 _(_ _**A**_ _) Confocal microscopy imaging of 405/488 sensor redox ratios depicted as false-colour scales in_


361 _hypocotyls upon seed imbibition with water (control) or -0.9 MPa PEG solution (water stress) at 15°C in_


362 _the dark. RoGFP2-Orp1 (left panel) was calibrated using 10 mM dithiothreitol (DTT) for full reduction (Red)_


363 _or 100 mM H_ _2_ _O_ _2_ _for full oxidation (Ox) of the biosensor in vivo. (ET, early timepoint; MT, mid timepoint;_


364 _LT, late timepoint) Scale bars = 10 μm. (_ _**B**_ _) RoGFP2-Orp1Roxydation levels at the three time points. (_ _**C**_ _)_


365 _Redox potentials calculated from confocal microscopy images. In B and C, data were obtained from at_


366 _least 14 embryos in two independent experiments. Statistical significance was determined using Student’s_


367 _t-test at each time point (*, P &lt; 0.05; **, P &lt; 0.01)._


368


369 **ROS-associated gene networks and hormonal pathways regulate germination**


370 **under water stress**


371 To identify the key regulatory genes associated with H₂O₂ production, we performed a


372 weighted gene co-expression network analysis (WGCNA) integrating our


373 transcriptomic and H₂O₂ production data (Figures 7A and S3). A soft-thresholding


374 power of β = 6 was selected to construct the co-expression network, resulting in the


375 identification of 31 co-expression modules (Figures S3A, S3B and 7A). Among these,


376 the purple module, comprising 335 genes, showed the strongest positive correlation


377 with ROS production (Figure 7A). The eigengene of this module exhibited distinct


378 temporal expression changes during germination (Figure S3C), indicating coordinated


379 regulation of its constituent genes. Gene Ontology enrichment analysis revealed a


14


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


380 significant overrepresentation of biological processes related to cell wall organization,


381 external encapsulating structure organization, transmembrane transport, and negative


382 regulation of abscisic acid (ABA)-activated signaling (Figure 7B), suggesting a close


383 association between this module and H₂O₂ production. We further visualized the genes


384 within the purple module using Cytoscape and identified five highly connected hub


385 genes, including AT1G20440 ( _COLD-REGULATED 47_, _COR47_ ), AT1G73220


386 ( _ORGANIC CATION TRANSPORTER 1_, _OCT1_ ), AT1G28290 ( _ARABINOGALACTAN_


387 _PROTEIN 31_, _AGP31_ ), and two genes with no annotated function, AT4G11100 and


388 AT3G15940 (Figure 7C). The module also contained several transcription factor


389 genes, including members of the MYB, C2H2 zinc-finger, NAC, and GRF families


390 (Table S5), suggesting that transcriptional regulation contributes to the coordination of


391 ROS-associated responses during germination.


392


393 **Figure 7. A central gene hub is associated with germination and ROS production under**


394 **water stress**


395 _(_ _**A**_ _) Correlation matrix between WGCNA modules and ROS production under water stress conditions. The_


396 _x-axis represents ROS production, the y-axis represents the name of each module, the P value is_


397 _presented in each cell, and the color of each cell indicates the correlation between the module and the_


398 _ROS production. (_ _**B**_ _) Simplified gene ontology classification of the purple module . (_ _**C**_ _) Correlation network_


399 _visualization of the interactions between the five genes with highest connectivity in the purple module._


400 _Each node is labelled with its corresponding gene IDs. Node size represents the degree of correlation_


401 _among genes._


402 We examined the transcriptomic data to gain deeper information on the possible


403 role of the ABA pathway in ROS-mediated responses to water stress during


15


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


404 germination (Figure 8A). Consistent with the WGCNA analysis, water stress markedly


405 altered the expression of genes associated with ABA signaling (Figure 8B). In


406 particular, _SnRK2.2_ and _SnRK2.5_, which encode protein kinases required for ABA


407 signaling (Fujii et al., 2007; Wang et al., 2015), were strongly induced under water


408 stress conditions. In addition, expression of the ABA-responsive transcription factor


409 ABF3, whose activity depends on ABA signaling (Yoshida et al., 2010), was


410 significantly upregulated, both indicating an activation of the ABA signaling pathway in


411 response to water stress during germination. Several genes encoding key enzymes of


412 the ABA biosynthetic pathway, including _NCED2_ (9-cis-epoxycarotenoid dioxygenase


413 2; Zhang et al., 2025) and _ABA2_ (short-chain dehydrogenase/reductase; González

414 Guzmán et al., 2002), were significantly upregulated by water stress (Figure 8B).


415 Conversely, _CYP707A2_ and _CYP707A3_, which encode ABA 8′-hydroxylases involved


416 in ABA catabolism (Saito et al., 2004; Kushiro et al., 2004), were significantly


417 downregulated under water-stress conditions (Figure 8). To assess the role of ABA in


418 water-stress-induced germination inhibition, seeds were treated with the ABA


419 biosynthesis inhibitor fluridone (Yoshioka et al., 1998). Surprisingly, inhibition of ABA


420 biosynthesis did not alleviate the inhibitory effect of water stress on germination (Figure


421 8C). In contrast, genetic analyses supported a role for ABA in this response.


422 Germination analysis of the _nced2,5,6,9_ quadruple mutant line affected in ABA


423 biosynthesis or the _abi4_ ABA signaling mutant line showed that ABA is required for


424 water-stress-induced germination inhibition (Figure 8D). Remarkably, seeds of these


425 mutant lines also exhibited significantly higher ROS accumulation than wild-type seeds


426 under water stress conditions (Figure 8E). Collectively, these results demonstrate that


427 ABA signaling contributes to the inhibition of germination under water stress and


428 unravel a functional link between ABA and ROS pathways.


16


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


429


430 **Figure 8. ABA pathways respond to water stress during germination**


431 _(_ _**A**_ _) Schematic representation of the ABA pathway, including synthesis, catabolism, and signaling. (_ _**B**_ _)_


432 _Heat maps showing differentially-expressed genes in the ABA pathway in seeds exposed to water_


433 _(control) or PEG at -0.9 MPa. For each gene, transcript abundance is expressed as normalized read_


434 _counts relative to the early time point (ET) in the control condition, which was arbitrarily set to 1. Asterisks_


435 _indicate significant differences relative to the corresponding control condition at the same time point (FDR_


436 _&lt; 0.01). (_ _**C**_ _) Germination of stratified seeds at 15°C in the dark under normal or water stress conditions in_


437 _the presence of 10 µM fluridone. (_ _**D**_ _) Germination at 15°C in the dark under normal (left) or water stress_


438 _(right) conditions of seeds of ABA pathway-associated mutants. (_ _**E**_ _) H_ _2_ _O_ _2_ _production of ABA-associated_


439 _mutants at 15°C in the dark under water stress conditions after 48h of imbibition (LT). Data presented are_


440 _means ± SD of three replicates (B, C, and D) or six replicates (E). Student's t-test was used to determine_


441 _the statistical significance between two lines; an asterisk above the bar chart indicates a significant_


442 _difference between the two groups (***, P &lt; 0.001; ****, P &lt; 0.0001)._


17


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


443 Considering that the antagonistic interplay between ABA and GA is a central


444 determinant of the balance between dormancy maintenance and germination


445 completion, we also examined the expression of genes involved in GA biosynthesis,


446 catabolism, and signaling (Figure 9A and 9B). The _GA3ox2_ gene, encoding a key


447 enzyme for GA biosynthesis (Mitchum et al., 2006), was strongly downregulated under


448 water stress (Figures 9A and 9B). Conversely, the _GA2ox6_ and _GA2ox8_ genes, which


449 encode GA 2-oxidases involved in GA catabolism (Kubalová et al., 2025), were


450 significantly upregulated. These observations suggest that water stress reduces the


451 pathways leading to bioactive GA accumulation during germination. Consistently,


452 water stress also affected the expression of GA signaling genes, notably _GID1_


453 encoding the GA receptor (Olszewski et al., 2002) (Figure 9A and 9B). To assess the


454 role of GA in the germination response to water stress, seeds were treated with


455 paclobutrazol (Pac), an inhibitor of GA biosynthesis. Pac treatment significantly


456 exacerbated the inhibitory effect of water stress on germination (Figure 9C), supporting


457 a positive role for GA in promoting germination under low water potential. Interestingly,


458 Pac treatment also triggered a significant decrease in ROS accumulation (Figure 9D),


459 suggesting a functional link between GA signaling and ROS homeostasis during


460 germination.


18


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


461


462 **Figure 9. GA pathways respond to water stress during germination**


463 _(_ _**A**_ _) Schematic representation of the GA pathway, including synthesis, catabolism, and signaling. (_ _**B**_ _) Heat_


464 _maps showing differentially-expressed genes associated with GA metabolism and signaling pathways in_


465 _seeds exposed to water (control) or PEG at -0.9 MPa. For each gene, transcript abundance is expressed_


466 _as normalized read counts relative to the early time point (ET in the control condition, which was arbitrarily_


467 _set to 1. Asterisks indicate significant differences relative to the corresponding control condition at the_


468 _same time point (FDR &lt; 0.01). (_ _**C**_ _) Germination assay of seeds treated with GA inhibitor (Paclobutrazol,_


469 _Pac) under control or water-stress conditions. (_ _**D**_ _) H_ _2_ _O_ _2_ _production in non-dormant Col seeds imbibed with_


470 _-0.9 MPa PEG for 48 h (LT) in the presence or absence of Paclobutrazol. Data are presented as means_


471 _± SD of 3 replicates (C) or at least 5 replicates (D). Student's t-test was used to determine the statistical_


472 _significance between two lines; an asterisk above the bar chart indicates a significant difference between_


473 _the two groups (***, P &lt; 0.001)._


474


475 **DISCUSSION**


476 **Water stress induces a developmental slowdown rather than a metabolic arrest**


477 Decreasing water potential to -0.9 MPa delayed germination without substantially


478 affecting its completion, indicating that seeds retained the capacity to germinate


19


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


479 (Figure 1A). Similar effects have been reported in several plant species (Vallejo et al.,


480 2010; Saux et al., 2020; Wang et al., 2022; Ouzakar et al., 2026). Importantly, the


481 moderate reduction in seed water content observed at -0.9 MPa indicates that delayed


482 germination cannot be explained solely by impaired hydration. Instead, the progressive


483 increase in the number of differentially expressed genes throughout germination


484 (Figure 2A) indicates that water stress triggers a continuous transcriptional


485 reprogramming rather than a transient response during imbibition. Thus, moderate


486 water stress temporarily restrains developmental progression, which delays radicle


487 protrusion while preserving germination competence until conditions become more


488 favorable.


489 Consistently, water-stress-induced transcriptome dynamics were found to be highly


490 selective rather than indicative of a generalized metabolic suppression. Consistent with


491 our observations of H 2 O 2 accumulation under optimal germination conditions, genes


492 associated with oxygen responses, redox metabolism, signaling, and stress adaptation


493 were induced in response to water stress. In contrast, genes involved in RNA


494 processing, ribosome biogenesis, DNA replication, and chromatin organization were


495 strongly downregulated (Figure 2). Related transcriptomic reprogramming in response


496 to water limitation during germination has been reported in seeds of lettuce (Hwang et


497 al., 2024), _Allium mogolicum_ (Men et al., 2026), and sunflower (Saux et al., 2020). This


498 pattern suggests that water stress responses actively prioritize stress adaptation over


499 growth, a deceleration strategy also noted in drought-exposed vegetative tissues,


500 where growth is rapidly restricted while protective pathways remain active (Skirycz and


501 Inzé, 2010; Wilkins et al., 2010). Here, as discussed below, we identified a temporal


502 regulatory framework linking water stress perception to delayed germination through


503 coordinated events that unfold during seed imbibition.


504 **Water stress rapidly reallocates cellular resources from germination-associated**


505 **growth to stress-adaptation processes before major physiological inhibition**


506 **occurs**


507 A striking observation of this study is that the transcriptional response to PEG occurred


508 before the appearance of major physiological differences. For instance, at ET,


509 relatively few genes were differentially expressed, yet their functional identities provide


510 important insights into the initial perception of water stress and specific stress signaling


511 pathways, altogether showing that seeds rapidly activate protective mechanisms in


512 response to reduced water availability. Concomitantly, genes associated with cell wall


513 metabolism and xyloglucan remodeling were repressed. Because cell wall loosening


20


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


514 and endosperm weakening are prerequisites for radicle emergence (Finch-Savage


515 and Leubner-Metzger, 2006; Steinbrecher and Leubner-Metzger, 2017), repression of


516 these pathways suggests that water stress affects germination from its earliest stages


517 by limiting the acquisition of growth competence. Interestingly, sulfur metabolism was


518 among the earliest pathways to be activated. Sulfur assimilation supplies cysteine for


519 glutathione biosynthesis and is therefore closely linked to cellular antioxidant capacity


520 (Kopriva and Rennenberg, 2004). Its induction suggests that redox homeostasis is


521 adjusted early during germination, before measurable changes in H 2 O 2 accumulation


522 become apparent. Together, these observations reveal that perception of water deficit


523 by germinating seeds rapidly redirects transcriptional and downstream resources


524 toward stress adaptation while postponing developmental processes required for


525 embryo expansion and radicle protrusion.


526 **Water stress reduces germination-associated H** **2** **O** **2** **accumulation while**


527 **activating redox-sensitive pathways**


528 ROS have emerged as central regulators of seed dormancy release and germination


529 (Bailly, 2004, 2019, 2022). According to the "oxidative window" concept, successful


530 germination requires ROS to accumulate within an optimal range that promotes


531 signaling while avoiding oxidative damage (Bailly et al., 2008). Our data indicate that


532 the transition from the early to the middle stage of germination represents a critical


533 checkpoint at which water stress progressively shifts seeds away from this permissive


534 redox state. At the middle stage, PEG treatment induced extensive transcriptional


535 reprogramming dominated by oxygen- and redox-related processes, including


536 responses to hypoxia, oxygen deprivation, electron transport, oxidative


537 phosphorylation, and redox metabolism. Consistent with this, PEG-responsive genes


538 were significantly enriched within the high-light (HL) ROS cluster defined by Willems


539 _et al_ . (2016), indicating activation of a coordinated redox-responsive transcriptional


540 program. However, physiological analyses revealed that this transcriptional signature


541 is not reflected by enhanced ROS accumulation. Instead, PEG-treated seeds


542 accumulated less H₂O₂, exhibited higher catalase activity, and showed no increase in


543 lipid peroxidation, demonstrating that moderate water stress does not induce oxidative


544 damage but rather restricts the accumulation of signaling ROS. This interpretation is


545 further supported by the inhibitory effect of ROS scavengers on germination,


546 confirming that ROS are required for germination progression. These findings illustrate


547 that transcriptomic signatures associated with oxidative stress should not be


21


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


548 interpreted as direct evidence of increased ROS production but instead reflect


549 activation of adaptive responses to altered redox homeostasis.


550 Subcellular imaging of redox status revealed that this response was highly


551 compartmentalized. Whereas the mitochondrial redox potential remained remarkably


552 stable despite reduced respiration, the oxidative status of peroxisomes decreased


553 markedly under water stress. Given the established roles of peroxisomes in reserve


554 mobilization, β-oxidation, dormancy release, and H 2 O 2 production during germination


555 (Oracz et al., 2007; Pracharoenwattana et al., 2010; Jurdak et al., 2022), these results


556 identify peroxisomes as a major site of redox regulation under water deficit. We


557 propose that moderate water stress imposes a reversible growth restraint by limiting


558 the accumulation of peroxisome-derived signaling H 2 O 2 to levels below the threshold


559 required to sustain the developmental transitions leading to radicle protrusion, rather


560 than by triggering a damaging oxidative burst.


561 This mechanism contrasts with the increased ROS production typically observed in


562 drought-stressed vegetative tissues (Liu et al., 2019; Wang et al., 2022), highlighting


563 the importance of considering developmental context when interpreting plant redox


564 responses. More broadly, our findings reinforce the concept that ROS function


565 primarily as developmental signals during germination and demonstrate that


566 successful adaptation to water stress relies on maintaining ROS within an appropriate


567 signaling window rather than maximizing antioxidant protection. They also emphasize


568 the importance of integrating physiological measurements with transcriptomic


569 analyses to accurately resolve the mechanisms underlying stress responses during


570 seed germination.


571 **Peroxisomes emerge as a central redox signaling hub controlling germination**


572 **under water stress**


573 The compartment-specific redox analyses identify peroxisomes as a major redox


574 signaling hub linking water stress to delayed germination. While peroxisomes have


575 long been associated with reserve mobilization, dormancy release, and germination


576 (Oracz et al., 2007; Leymarie et al., 2012; Jurdak et al., 2022), our transcriptomic


577 analyses indicate that water stress selectively reprograms peroxisomal functions


578 rather than globally suppressing it. Because acyl-CoA oxidases constitute a major


579 source of peroxisomal H 2 O 2 during germination (Eastmond et al., 2000; del Río et al.,


580 2006), the downregulation of key genes involved in peroxisomal H 2 O 2 production


581 provides a molecular explanation for both the reduced H 2 O 2 accumulation and the


22


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


582 more reduced peroxisomal redox state observed under water stress. Interestingly, this


583 response was not associated with a complete shutdown of peroxisomal metabolism.


584 In particular, genes encoding acyl-CoA oxidases, which constitute the major source of


585 peroxisomal H 2 O 2 during germination (Eastmond et al., 2000; del Río et al., 2006),


586 together with _PEX5_, were repressed under PEG treatment, providing a molecular


587 explanation for the reduced peroxisomal oxidation observed in vivo. In contrast, genes


588 involved in downstream β-oxidation and the glyoxylate cycle, including 3-ketoacyl-CoA


589 thiolase-2 ( _KAT2,_ Germain et al., 2001) and Isocitrate Lyase ( _ICL_, Graham, 2008),


590 remained expressed, suggesting that carbon mobilization is maintained while ROS

591 production is selectively attenuated.


592 The functional significance of this regulation is supported by the enhanced


593 sensitivity of _gox1_ **,** _gox2_ **,** and _gox3_ mutant seeds to water stress, demonstrating that


594 impairment of peroxisomal H 2 O 2 production compromises germination under low water


595 potential. By contrast, the absence of a germination phenotype in the _cat2_ mutant


596 indicates that reducing peroxisomal H 2 O 2 scavenging alone is insufficient to modify the


597 response to water stress. Together with the transcriptomic and redox imaging data,


598 these genetic analyses indicate that the production of peroxisome-derived H 2 O 2, rather


599 than its detoxification, is a critical determinant of germination progression under water


600 deficit. We therefore propose that water stress selectively suppresses a permissive


601 peroxisomal H 2 O 2 signal while preserving the metabolic functions required for reserve


602 mobilization and seedling establishment.


603 **WGCNA identifies developmental and hormonal pathways associated with**


604 **peroxisomal H** **2** **O** **2** **signaling**


605 The WGCNA analysis provided a systems-level perspective on the biological


606 significance of the altered peroxisomal redox state observed under water stress.


607 Notably, one of the modules showing the strongest correlation with H 2 O 2 accumulation


608 was enriched in genes involved in cell wall organization and hormonal regulation. The


609 strong coregulation between H 2 O 2 -related modules and genes involved in cell wall


610 organization suggests that reduced peroxisomal H 2 O 2 under water stress delays


611 germination by impairing the cell wall remodeling processes required for radicle


612 protrusion. The identification of ABA-related genes within a H 2 O 2 -associated gene hub,


613 together with the strong correlation between H 2 O 2 accumulation and developmental


614 processes required for germination, suggests that altered redox signaling may


615 represent an upstream event contributing to the hormonal reprogramming induced by


616 PEG. This is consistent with the extensive crosstalk between H 2 O 2, ABA, and GA


23


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


617 signaling pathways during seed germination (Liu et al., 2010; Ishibashi et al., 2012; Bi


618 et al., 2017; Cheng et al., 2022) and raises the possibility that suppression of


619 peroxisome-derived H 2 O 2 constitutes one of the primary signals through which water


620 stress modifies hormonal balance and developmental timing.


621 Increasing evidence indicates that H 2 O 2 functions both upstream and downstream


622 of ABA and GA signaling, forming an interconnected regulatory network that controls


623 seed germination (Bailly, 2008, 2022; Liu et al., 2010). Our results support the


624 existence of such a regulatory network under water stress. Transcriptomic analyses


625 revealed activation of ABA biosynthesis and signaling together with repression of ABA


626 catabolic genes. Consistent with these observations, both ABA-deficient and ABA

627 insensitive mutants exhibited a reduced germination delay and accumulated higher


628 levels of H 2 O 2 under water stress (Figure 8D), indicating that ABA contributes to


629 limiting H 2 O 2 accumulation during germination. This concurs with previous studies


630 showing that ABA modulates cellular redox homeostasis through the regulation of both


631 antioxidant systems and ROS-producing pathways. Conversely, water stress


632 repressed the expression of genes associated with GA biosynthesis and signaling.


633 Moreover, pharmacological inhibition of GA biosynthesis further reduced H 2 O 2


634 accumulation, supporting previous reports that GA promotes germination, at least in


635 part, by stimulating ROS production (Müller et al., 2009; Bahin et al., 2011). Together,


636 these findings indicate that ABA and GA exert opposite effects on ROS homeostasis


637 during germination: ABA restricts, whereas GA promotes, the accumulation of


638 signaling H 2 O 2 .


639 Based on our physiological, genetic, and transcriptomic data, we propose that


640 peroxisome-derived H 2 O 2 constitutes a central regulatory hub coordinating ABA and


641 GA interplay under water stress. Moderate water deficit initially reduces peroxisomal


642 H 2 O 2 production, thereby shifting the hormonal balance toward enhanced ABA


643 signaling and reduced GA responsiveness. These hormonal changes further reinforce


644 the decline in H 2 O 2 accumulation, establishing a positive feedback loop that


645 progressively stabilizes a low-redox state. As a consequence, downstream processes


646 required for germination, including cell wall remodeling, embryo growth, and ultimately


647 radicle protrusion, are delayed. Rather than acting as independent signaling pathways,


648 ABA, GA, and H 2 O 2 therefore appear to constitute an integrated regulatory network


649 organized around peroxisomal redox signaling.


650 **Water stress maintains a transcriptionally and translationally restrained**


651 **developmental state**


24


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


652 Our transcriptomic analyses indicated that water stress progressively triggers a


653 transcriptionally and translationally restrained developmental state during germination.


654 This response was pronounced at mid- and late-germination time points, where the


655 most significantly enriched downregulated GO categories were associated with


656 ribosome biogenesis, rRNA processing, RNA metabolism, ribonucleoprotein complex


657 assembly, DNA replication, chromosome organization, and ATP-dependent chromatin


658 remodeling. Given the downregulation of multiple genes encoding ribosomal and


659 nucleolar proteins, water stress may, therefore, limit the capacity of germinating seeds


660 to sustain the increase in translational activity required for embryo growth, thereby


661 contributing to delayed germination under water stress.


662 Importantly, these regulatory mechanisms were accompanied by a broad


663 downregulation of genes involved in chromatin assembly, transcriptional regulation,


664 RNA processing, and genome organization, suggesting that water stress limits the


665 chromatin reconfiguration normally associated with developmental progression.


666 Chromatin modification and remodeling are increasingly recognized as critical


667 components of dormancy release and germination (Nonogaki, 2019; Sano et al.,


668 2022), and the widespread repression of chromatin-associated functions observed


669 during germination suggests that PEG-treated seeds fail to fully activate the


670 transcriptional programs required for rapid growth.


671 Finally, these changes did not reflect developmental arrest. Genes associated with


672 chloroplast biogenesis and photosynthesis were efficiently induced under water stress,


673 indicating that developmental progression continued despite its delay. Instead, the


674 downregulation of translational and chromatin-remodeling pathways suggests that


675 water stress maintains seeds in a state of reduced developmental competence,


676 thereby delaying the transition toward rapid embryo growth and radicle protrusion.


677 Altogether, we provided evidence that delayed germination under water stress


678 results from a coordinated developmental reprogramming rather than a simple


679 consequence of reduced water availability, and we identified a temporal sequence of


680 events linking water-stress perception to delayed radicle protrusion. It involves an


681 initial activation of stress-acclimation responses, and subsequent reduction in


682 peroxisome-derived ROS accumulation, the promotion of ABA dominance over GA


683 responsiveness, and plausibly, the establishment of a transcriptionally and


684 translationally restrained developmental state that contributes to germination delay.


685 These data support a model in which water stress primarily uncouples the metabolic


686 and signaling functions of peroxisomes, maintaining reserve mobilization while


25


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


687 attenuating ROS-dependent developmental signaling. This altered redox state


688 propagates through hormonal and developmental networks controlling cell wall


689 remodeling, embryo growth, and germination timing. Together, these findings provide


690 a mechanistic framework explaining how seeds dynamically adjust germination to


691 water availability and identify peroxisome-derived ROS as a key regulator of


692 environmental control of germination.


693


694 **MATERIALS AND METHODS**


695 **Plant material**


696 _Arabidopsis thaliana_ seeds used in this study were from the Columbia (Col-0) genetic


697 background. All mutant lines and reduction–oxidation sensitive Green Fluorescent


698 Protein 2 fused to Oxidant Receptor Peroxidase 1 (roGFP2-Orp1) lines (Nietzel et al.,


699 2019) were derived from the same background. We used the ABA related mutants,


700 nced2.5.6.9 quadruple mutant (Chauffour et al., 2019) and abi4 single mutant (Dong


701 et al., 2016), glycolate oxidase mutants (gox1-2, gox2-2, and gox3-2; Yang et al.,


702 2025 ) and catalase mutant (cat2-2; Yang et al., 2025). Prior to sowing, seeds were


703 stratified at 4°C for 4 days, then planted in soil and transferred to growth chambers


704 maintained under controlled conditions (21°C, 16-hour daily photoperiod, and light


705 intensity of 150 µmol m &#91;-2&#93; s &#91;-1&#93; ). Once matured, seeds were collected and stored at -20°C


706 until required for further experiments.


707 **Seed germination assays**


708 Seeds were stratified prior to germination assays by placing them on a layer of cotton


709 wool moistened with deionized water in darkness at 4 °C for 4 days. For germination


710 assays, stratified seeds were transferred to petri dishes containing two layers of cotton


711 overlaid with filter paper, moistened with either distilled water or polyethylene glycol


712 8000 (PEG 8000) solutions. A range of PEG solutions was prepared to achieve water


713 potential values from -0.2 to -1.2 MPa, as described by Michel and Kaufmann (1973),


714 depending on experimental requirements. The concentrations of the chemical


715 reagents used for germination are provided in the main text. All assays were conducted


716 in the dark and included at least 50 seeds per genotype or treatment with a minimum


717 of 3 biological replicates. Germination was monitored daily for 10 days. A seed was


718 considered germinated when the radicle had emerged through the seed coat.


719 **Seed moisture content**


26


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


720 Approximately 10 mg of _Arabidopsis_ seeds were imbibed with water (control) or -0.9


721 MPa PEG (water stress) at 15°C in darkness. At the indicated time points, seeds were


722 collected and immediately weighed to determine fresh weight (FW). The collected


723 samples were then dried at 100°C for 24 h to determine dry weight (DW). Water


724 content was calculated and expressed as g H 2 O g &#91;-1&#93; FW. Data represent the mean ±


725 SD of three biological replicates.


726 **Measurement of hydrogen peroxide production**


727 Hydrogen peroxide production was assessed by a decrease in scopoletin


728 fluorescence, as described by Leymarie et al. (2012). Twenty-four mg of seeds were


729 incubated in 200 µl of potassium phosphate buffer (20 mM, pH 6.0) containing 5 µM


730 scopoletin and 1 U/mL of horseradish peroxidase (HRP) in darkness at 25°C on a


731 shaker for 1 h, and known H 2 O 2 concentrations were used for calibration.


732 Measurements were carried out at excitation/emission wavelengths of 346 nm/455 nm.


733 Results correspond to the mean ± SD of 3 replicates and are expressed per mg of the


734 fresh weight.


735 **Malondialdehyde (MDA) content**


736 Malondialdehyde (MDA) was assessed as described by Heath and Packer (1968).


737 Approximately twelve mg of _Arabidopsis_ seeds were homogenized in 0.1%


738 trichloroacetic acid (TCA) on ice. Following centrifugation at 12,000 × g for 10 min, the


739 supernatant was collected and mixed with an equal volume of 0.5% thiobarbituric acid


740 (TBA). The reaction mixture was incubated at 95°C for 30 min, then cooled on ice for


741 10 min to stop the reaction. Absorbance was measured at 532 and 600 nm using a


742 spectrophotometer. MDA content was determined using an extinction coefficient of 155


743 mM &#91;-1&#93; cm &#91;-1&#93; . Results correspond to the mean ± SD of 3 replicates and are expressed


744 per g of the fresh weight.


745 **Measurement of catalase activity**


746 Approximately fifty mg of _Arabidopsis_ seed were ground with liquid nitrogen, and


747 proteins were extracted in potassium phosphate buffer (0.1 M, pH 7.8) containing 2


748 mM dithiotreitol, 0.1 mM EDTA, and 1.25 mM polyethylene glycol 4000 as described


749 by Bailly et al (1996). After centrifugation at 13000 g for 15 min, the supernatant of


750 each sample was desalted on Sephadex PD10 columns and used for measuring


751 catalase (CAT) activity. CAT activity is expressed as nanomole H 2 O 2 per mg per


752 protein per min, and data presented are means ± SD.


753 _**In situ**_ **localization of ROS**


27


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


754 Intracellular localization of ROS was visualized using 2',7'-dichlorodihydrofluorescein


755 diacetate (H2DCFDA) with a confocal microscope as described by Leymarie et al


756 (2012). Embryos were dissected from imbibed seeds and incubated in 50 mM


757 potassium phosphate buffer (pH 7.4) containing 25 µM H2DCFDA for 30 min at 20 °C


758 in the dark. Following incubation, embryos were rinsed in potassium phosphate buffer


759 for 10 min. Images were acquired using a confocal laser scanning microscope (Leica


760 TCS SP5 AOBS) with excitation at 488 nm and emission collected between 498 and


761 550 nm, using a 63 X objective. Z-stack images were collected with a step size of 0.5


762 µm. To compare H2DCFDA fluorescence intensity, embryonic axes from different


763 experimental conditions were fixed simultaneously and imaged using identical confocal


764 settings. The acquired images were further processed using ImageJ software.


765 **Confocal imaging of roGFP2-Orp1 lines**


766 Confocal imaging of imbibed _Arabidopsis thaliana_ seed expressing roGFP2-Orp1 was


767 acquired using a confocal laser scanning microscope (Zeiss LSM 980 inverted)


768 equipped with a Plan-Apochromat 63X/1.40 water-immersion objective. For


769 calibration, de-coated seeds were fully reduced or oxidized by treatment with 10 mM


770 dithiothreitol (DTT) or 10 mM H 2 O 2, respectively. roGFP2-Orp1 was sequentially


771 excited at 405 nm and 488 nm, and fluorescence emission was collected between 508


772 and 530 nm. The acquired images were visualized, and fluorescence quantified using


773 Matlab-based software (Fricker, 2016), and the redox potential was calculated


774 according to the method described by (Schwarzländer et al., 2008).


775 **Respiration measurement**


776 Approximately 5 mg of seeds were stratified at 4°C for 4 days and then transferred to


777 2 mL glass vials containing cotton pads saturated with either distilled water or PEG


778 solution (-0.9 MPa). Oxygen content in the vials was monitored using a fixed-needle


779 oxygen microsensor (PyroScience; OXF500PT-OI) calibrated with air and nitrogen


780 before measurement. Oxygen content was recorded every 10 min for 2 days with


781 FireStingGO2 Manager software connected to the sensor. Respiration rate was


782 defined as the change in oxygen uptake within 10 min intervals before and after the


783 selected time point. Data correspond to the mean ± SD of six replicates and are


784 expressed as nmol L-1 of oxygen consumed per min per mg of fresh weight.


785 **RNA isolation and transcriptomic analysis**


786 Total RNA was extracted from approximately 50 mg of seeds using the


787 cetyltrimethylammonium bromide (CTAB) method as described by Chang et al (1993).


28


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


788 Genomic DNA of the sample was removed using the TURBO DNA-free™ Kit and


789 purified using the NucleoSpin™ RNA Clean-up XS kit. RNA samples with RNA


790 concentration &gt; 100 ng/µL and RNA integrity number (RIN) values &gt; 4.0 were deemed


791 suitable for further sequencing.


792 Library construction and subsequent sequencing were performed on the


793 Novogene Illumina platform. Sequencing was performed using paired-end sequencing,


794 with each end containing 150 bp reads (PE150); approximately 6 Gb of data per


795 sample was acquired. After sequencing, raw reads were quality-filtered and aligned to


796 the _Arabidopsis_ reference genome (Araport11 annotation of the TAIR10 genome)


797 using Hisat2 v2.0.5. Differential expression analysis was conducted using the DESeq2


798 package (v1.20.0) in R software. Genes with a false discovery rate (FDR) &lt; 0.01 were


799 considered significantly differentially expressed under water stress conditions. Gene


800 set enrichment analysis (GSEA) was used to assess the enrichment of differentially


801 expressed genes in the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and


802 Genomes (KEGG) databases. Significantly enriched pathways were identified based


803 on normalized enrichment scores (NES) and corrected p-values.


804 **Weighted gene co-expression network analysis**


805 To identify hub genes associated with ROS generation, weighted gene co-expression


806 network analysis (WGCNA) was performed using the WGCNA R software package.


807 Genes with expression levels below 1 in more than 50% of the samples were excluded.


808 The filtered genes were used to determine the soft-thresholding power (β) using the


809 pickSoftThreshold function, with the optimal value selected when the scale-free


810 topology fit index first exceeded 0.9. Using the selected soft-thresholding power (β =


811 6, Figure S3A), the gene expression data were clustered into 31 modules based on


812 their co-expression patterns. Module eigengenes were then calculated, and their


813 correlations with ROS generation (including control and water stress conditions) were


814 assessed using Pearson correlation analysis. A module with an FDR &lt; 0.01 and an


815 absolute correlation coefficient (|r|) &gt; 0.75 was considered significantly associated with


816 ROS production. Gene co-expression networks within the selected module were


817 visualized using Cytoscape software (v3.10.2). Genes with the top five degree values


818 were defined as hub genes for further analysis.


819 **Statistical analyses**


820 Statistical analyses were performed using Student’s t-test for comparisons between


821 two groups and analysis of variance (ANOVA) for comparisons among three or more


29


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


822 groups with GraphPad Prism. Differences were considered statistically significant at _P_


823 &lt; 0.05.


824 **DATA AVAILIBILITY STATEMENT**


825 &#91;The RNA-seq data generated in this study have been have been deposited in the NCBI&#93;(https://www.ncbi.nlm.nih.gov/geo/)


826 &#91;Gene Expression Omnibus (GEO)&#93;(https://www.ncbi.nlm.nih.gov/geo/) under the GSE338026 accession.


827


828 **ACKNOWLEDGEMENTS**


829 This work benefited from a China Scholarship Council (CSC) doctoral scholarship to


830 XM, the research grant SEED (ANR-25-CE20-6227-02) from Agence Nationale pour


831 la Recherche (ANR) to CB and FB. F.B. was also supported by Velux Stiftung (project


832 No. 1747). The authors thank Pr Yi Han (China Agricultural University) for providing


833 the Arabidopsis _gox1-2_, _gox2-2_, _gox3-2_, and _cat2-2_ mutant seeds, and Dr Annie


834 Marion-Poll (INRAe) for providing the Arabidopsis _nced2 nced5 nced6 nced9_


835 quadruple mutant and _abi4_ mutant seeds.


836


837 **CONFLICTS OF INTEREST**


838 The authors declare no conflicts of interest.


839


840 **AUTHOR CONTRIBUTIONS**


841 CB and FB conceived and designed the research. XM performed the experiments and


842 analyzed the data. AV contributed to the transcriptome data analysis. SW provided


843 technical assistance to XM. XM wrote the original draft of the manuscript. CB, FB, and


844 AM revised the manuscript. All authors read and approved the final manuscript.


845


846 **REFERENCES**


847 **Arnaud, D., Deeks, M.J., Smirnoff, N.** (2023). Organelle-targeted biosensors reveal


848 distinct oxidative events during pattern-triggered immune responses. Plant Physiol.


849 **191:** 2551–2569.


850 **Babajani, G., Effendy, J., Plant, A.L.** (2009). _Sl-SROl1_ increases salt tolerance and


851 is a member of the _radical-induced cell death 1_ — _similar to RCD1_ gene family of


852 tomato. Plant Sci. **176:** 214–222.


30


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


853 **Bahin, E., Bailly, C., Sotta, B., Kranner, I., Corbineau, F., Leymarie, J.** (2011).


854 Crosstalk between reactive oxygen species and hormonal signalling pathways


855 regulates grain dormancy in barley. Plant Cell Environ. **34:** 980–993.


856 **Bailly, C.** (2019) The signalling role of ROS in the regulation of seed germination and


857 dormancy. Biochem. J. **476:** 3019–3032.


858 **Bailly, C.** (2022). ROS in seed germination. In: Mittler R, Van Breusegem F, Jacquot


859 JP, eds. Advances in Botanical Research, Vol. 105: Oxidative Stress Response in


860 Plants. Academic Press, pp. 177–204.


861 **Bailly, C.** (2004). Active oxygen species and antioxidants in seed biology. Seed Sci.


862 Res. **14:** 93–107.


863 **Bailly, C., Benamar, A., Corbineau, F., Come, D.** (1996). Changes in


864 malondialdehyde content and in superoxide dismutase, catalase and glutathione


865 reductase activities in sunflower seeds as related to deterioration during accelerated


866 aging. Physiol. Plant. **97:** 104–110.


867 **Bailly, C., El-Maarouf-Bouteau, H., Corbineau, F.** (2008). From intracellular


868 signaling networks to cell death: the dual role of reactive oxygen species in seed


869 physiology. C. R. Biol. **331:** 806–814.


870 **Bailly, C., Gomez Roldan, M.V.** (2023). Impact of climate perturbations on seeds and


871 seed quality for global agriculture. Biochem. J. **480:** 177–196.


872 **Bi, C., Ma, Y., Wu, Z., Yu, Y.T., Liang, S., Lu, K., Wang, X.F.** (2017). _Arabidopsis_


873 ABI5 plays a role in regulating ROS homeostasis by activating _CATALASE 1_


874 transcription in seed germination. Plant Mol. Biol. **94:** 197–213.


875 **Chang, S., Puryear, J., Cairney, J.** (1993). A simple and efficient method for isolating


876 RNA from pine trees. Plant Mol. Biol. Rep. **11:** 113–116.


877 **Chauffour, F., Bailly, M., Perreau, F., Cueff, G., Suzuki, H., Collet, B., Frey, A.,**


878 **Clément, G., Soubigou-Taconnat, L., Balliau, T., et al.** (2019). Multi-omics


879 analysis reveals sequential roles for ABA during seed maturation. Plant Physiol.


880 **180:** 1198–1218.


881 **Chen, W., Zhao, J., Tao, Z., Zhang, S., Bei, X., Lu, W., Qi, X.** (2024). Short heat


882 shock factor A2 regulates heat resistance and growth balance in _Arabidopsis_ . eLife


883 **13:** RP99937.


884 **Cheng, M., Guo, Y., Liu, Q., Nan, S., Xue, Y., Wei, C., Zhang, Y., Luan, F., Zhang,**


31


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


885 **X., Li, H.** (2022). H 2 O 2 and Ca &#91;2+&#93; signaling crosstalk counteracts aba to induce seed


886 germination. Antioxidants (Basel) **11:** 1594.


887 **Claeys, H., Inzé, D.** (2013). The agony of choice: how plants balance growth and


888 survival under water-limiting conditions. Plant Physiol. **162:** 1768–1779.


889 **Dong, Z., Yu, Y., Li, S., Wang, J., Tang, S., Huang, R.** (2016). Abscisic acid


890 antagonizes ethylene production through the ABI4-mediated transcriptional


891 repression of _ACS4_ and _ACS8_ in _Arabidopsis_ . Molecular Plant **9:** 126–135.


892 **Eastmond, P.J., Hooks, M.A., Williams, D., Lange, P., Bechtold, N., Sarrobert, C.,**


893 **Nussaume, L., Graham, I.A.** (2000). Promoter trapping of a novel medium-chain


894 acyl-coa oxidase, which is induced transcriptionally during _Arabidopsis_ seed


895 germination. J. Biol. Chem. **275:** 34375–34381.


896 **Engqvist, MKM., Schmitz, J., Gertzmann, A., Florian, A., Jaspert, N., Arif, M.,**


897 **Balazadeh, S., Mueller-Roeber, B., Fernie, A.R., Maurino, V.G.** (2015).


898 GLYCOLATE OXIDASE3, a glycolate oxidase homolog of yeast l-lactate


899 cytochrome c oxidoreductase, supports l-lactate oxidation in roots of _Arabidopsis_ .


900 Plant Physiol. **169:** 1042–1061.


901 **Esser, C., Kuhn, A., Groth, G., Lercher, M.J., Maurino, V.G.** (2014). Plant and


902 animal glycolate oxidases have a common eukaryotic ancestor and convergently


903 duplicated to evolve long-chain 2-hydroxy acid oxidases. Mol. Biol. Evol. **31:** 1089–


904 1101.


905 **Farooq, M., Wahid, A., Lee, D.J., Ito, O., Siddique, KHM.** (2009). Advances in


906 drought resistance of rice. Crit. Rev. Plant Sci. **28:** 199–217.


907 **Finch-Savage, W.E., Leubner-Metzger, G.** (2006). Seed dormancy and the control


908 of germination. New Phytol. **171:** 501–523.


909 **Fricker, M.D.** (2016). Quantitative redox imaging software. Antioxid. Redox Signal. **24:**


910 752–762.


911 **Frugoli, J.A., Zhong, H.H., Nuccio, M.L., McCourt, P., McPeek, MA., Thomas, TL.,**


912 **McClung, CR.** (1996). Catalase is encoded by a multigene family in _Arabidopsis_


913 _thaliana_ (L.) Heynh. Plant Physiol. **112:** 327–336.


914 **Fujii, H., Verslues, PE., Zhu, J.K.** (2007). Identification of two protein kinases required


915 for abscisic acid regulation of seed germination, root growth, and gene expression


916 in _Arabidopsis_ . Plant Cell **19:** 485–494.


32


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


917 **Gaber, A** . (2013). The importance of _Arabidopsis_ glutathione peroxidase 8 for


918 protecting _Arabidopsis_ plant and E. coli cells against oxidative stress. GM crops &amp;


919 food. **5:** 20–26.


920 **Gao, S., Guo, W., Feng, W., Liu, L., Song, X., Chen, J., Hou, W., Zhu, H., Tang, S.,**


921 **Hu, J.** (2016). _LTP3_ contributes to disease susceptibility in _Arabidopsis_ by


922 enhancing abscisic acid (ABA) biosynthesis. Mol. Plant Pathol. **17:** 412–426.


923 **Germain, V., Rylott, E.L., Larson, T.R., Sherson, S.M., Bechtold, N., Carde, J.P.,**


924 **Bryce, J.H., Graham, I.A., Smith, S.M.** (2001). Requirement for 3-ketoacyl-CoA


925 thiolase-2 in peroxisome development, fatty acid β-oxidation and breakdown of


926 triacylglycerol in lipid bodies of _Arabidopsis_ seedlings. Plant J. **28:** 1–12.


927 **González-Guzmán, M., Apostolova, N., Bellés, J.M., Barrero, J.M., Piqueras, P.,**


928 **Ponce, M.R., Micol, J.L., Serrano, R., Rodríguez, P.L.** (2002). The short-chain


929 alcohol dehydrogenase ABA2 catalyzes the conversion of xanthoxin to abscisic


930 aldehyde. Plant Cell **14:** 1833–1846.


931 **Gonzali, S., Loreti, E., Cardarelli, F., Novi, G., Parlanti, S., Pucciariello, C.,**


932 **Bassolino, L., Banti, V., Licausi, F., Perata, P.** (2015). Universal stress protein


933 HRU1 mediates ROS homeostasis under anoxia. Nature Plants **1:** 15151.


934 **Graham, I.A.** (2008). Seed storage oil mobilization. Annu. Rev. Plant Biol. **59:** 115–


935 142.


936 **Hasanuzzaman, M., Nahar, K., Gill, S.S., Fujita, M.** (2013). Drought stress responses


937 in plants, oxidative stress, and antioxidant defense. In: Tuteja N, Gill SS, eds.


938 Climate Change and Plant Abiotic Stress Tolerance. Wiley-VCH Verlag GmbH &amp; Co.


939 KGaA, Weinheim, pp 209–250.


940 **Hassan, M.J., Geng, W., Zeng, W., Raza, M.A., Khan, I., Iqbal, M.Z., Peng, Y., Zhu,**


941 **Y., Li, Z.** (2021). Diethyl aminoethyl hexanoate priming ameliorates seed


942 germination _via_ involvement in hormonal changes, osmotic adjustment, and


943 dehydrins accumulation in white clover under drought stress. Front Plant Sci. **12:**


944 709187.


945 **Heath, RL., Packer, L.** (1968). Photoperoxidation in isolated chloroplasts: I. Kinetics


946 and stoichiometry of fatty acid peroxidation. Arch. Biochem. Biophys. **125:** 189–198.


947 **Horváth, E., Bela, K., Gallé, Á., Riyazuddin, R., Csomor, G., Csenki, D., Csiszár,**


948 **J.** (2020). Compensation of mutation in _Arabidopsis_ glutathione _transferase_


949 ( _AtGSTU_ ) genes under control or salt stress conditions. Int. J. Mol. Sci. **21:** 2349.


33


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


950 **Horváth, E., Bela, K., Kulman, K., Faragó, N., Riyazuddin, R., Gallé, Á., Puskás,**


951 **L.G., Csiszár, J.** (2023). Glutathione transferases are involved in salicylic acid

952 induced transcriptional reprogramming. J. Plant Growth Regul. **42:** 4497–4510.


953 **Hung, K.T., Kao, C.H.** (2004). Nitric oxide acts as an antioxidant and delays methyl


954 jasmonate-induced senescence of rice leaves. J. Plant Physiol. **161:** 43–52.


955 **Hwang, S., Simko, I., Mou, B.** (2024). QTL mapping and transcriptome analysis of


956 seed germination under PEG-induced water stress in Lactuca spp. Sci Rep. **14:**


957 27157.


958 **Ishibashi, Y., Tawaratsumida, T., Kondo, K., Kasa, S., Sakamoto, M., Aoki, N.,**


959 **Zheng, S.H., Yuasa, T., Iwaya-Inoue, M.** (2012). Reactive oxygen species are


960 involved in gibberellin/abscisic acid signaling in barley aleurone cells. Plant Physiol.


961 **158:** 1705–1714.


962 **Jurdak, R., Launay-Avon, A., Roux, CP.L., Bailly, C.** (2020). Retrograde signalling


963 from the mitochondria to the nucleus translates the positive effect of ethylene on


964 dormancy breaking of _Arabidopsis thaliana_ seeds. New Phytol. **229:** 2192–2205.


965 **Jurdak, R., Rodrigues, G. de. AG., Chaumont, N., Schivre, G., Bourbousse, C.,**


966 **Barneche, F., Bou, Dagher., Kharrat, M., Bailly, C.** (2022). Intracellular reactive


967 oxygen species trafficking participates in seed dormancy alleviation in _Arabidopsis_


968 seeds. New Phytol. **234:** 850–866.


969 **Kolesnikov, M., Gerasko, T., Paschenko, Yu., Pokoptseva, L., Onyschenko, O.,**


970 **Kolesnikova, A.** (2023). Effect of water deficit on maize seeds ( _Zea mays_ L.) during


971 germination. Agron. Res. **21:** e028.


972 **Kopriva, S., Rennenberg, H** . (2004). Control of sulphate assimilation and glutathione


973 synthesis: interaction with N and C metabolism. J. Exp. Bot. **55:** 1831–1842.


974 **Kubalová, M., Griffiths, J., Müller, K., Levenets, L., Tylová, E., Tarkowská, D.,**


975 **Jones, A.M., Fendrych, M.** (2025). Gibberellin-deactivating GA2OX enzymes act


976 as a hub for auxin–gibberellin cross talk in _Arabidopsis thaliana_ root growth


977 regulation. Proc. Natl. Acad. Sci. U S A. **122:** e2425574122.


978 **Kushiro, T., Okamoto, M., Nakabayashi, K., Yamagishi, K., Kitamura, S., Asami,**


979 **T., Hirai, N., Koshiba, T., Kamiya, Y., Nambara, E.** (2004). The _Arabidopsis_


980 cytochrome P450 CYP707A encodes ABA 8′-hydroxylases: key enzymes in ABA


981 catabolism. EMBO J. **23:** 1647–1656.


982 **Lariguet, P., Ranocha, P., De Meyer, M., Barbier, O., Penel, C., Dunand, C.** (2013).


34


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


983 Identification of a hydrogen peroxide signalling pathway in the control of light

984 dependent germination in _Arabidopsis_ . Planta **238:** 381–395.


985 **Lee, C., Teng, Q., Huang, W., Zhong, R., Ye, Z.H.** (2009). The F8H


986 glycosyltransferase is a functional paralog of fra8 involved in glucuronoxylan


987 biosynthesis in _Arabidopsis_ . Plant Cell Physiol. **50:** 812–827.


988 **Levine, A., Lamb, C.** (1994). H 2 O 2 from the oxidative burst orchestrates the plant


989 hypersensitive disease resistance response. Cell **79:** 583–593.


990 **Leymarie, J., Vitkauskaité, G., Hoang, H.H., Gendreau, E., Chazoule, V.,**


991 **Meimoun, P., Corbineau, F., El-Maarouf-Bouteau, H., Bailly, C.** (2012). Role of


992 reactive oxygen species in the regulation of _Arabidopsis_ seed dormancy. Plant Cell


993 Physiol. **53:** 96–106.


994 **Liu, J., Hasanuzzaman, M., Wen, H., Zhang, J., Peng, T., Sun, H., Zhao, Q.** (2019).


995 High temperature and drought stress cause abscisic acid and reactive oxygen


996 species accumulation and suppress seed germination growth in rice. Protoplasma


997 **256:** 1217–1227.


998 **Liu, Y., Ye, N., Liu, R., Chen, M., Zhang, J.** (2010). H 2 O 2 mediates the regulation of


999 ABA catabolism and GA biosynthesis in _Arabidopsis_ seed dormancy and


1000 germination. J. Exp. Bot, **61:** 2979–2990.


1001 **Liu, Y.C., Wu, Y.R., Huang, X.H., Sun, J., Xie, Q.** (2011). AtPUB19, a U-box e3


1002 ubiquitin ligase, negatively regulates abscisic acid and drought responses in


1003 _Arabidopsis thaliana_ . Mol. Plant. **4:** 938–946.


1004 **Llanes, A., Andrade, A., Masciarelli, O., Alemano, S., Luna, V.** (2016). Drought and


1005 salinity alter endogenous hormonal profiles at the seed germination phase. Seed


1006 Sci. Res. **26:** 1–13.


1007 **Men, Y., Wei, P., Cao, Z., Yang, L., Wei, J., Zhang, D., Zhang, X., Zhang, F., Yang,**


1008 **Z.** (2026). A transcriptome-based study using PEG-induced drought simulation in


1009 _Allium mongolicum_ seeds. Grassl. Sci. **72:** 16–30.


1010 **Michel, B.E., Kaufmann, M.R.** (1973), The osmotic potential of polyethylene glycol


1011 6000. Plant Physiol. **51:** 914–916.


1012 **Mitchum, M.G., Yamaguchi, S., Hanada, A., Kuwahara, A., Yoshioka, Y., Kato, T.,**


1013 **Tabata, S., Kamiya, Y., Sun, T.P.** (2006). Distinct and overlapping roles of two


1014 gibberellin 3-oxidases in _Arabidopsis_ development. Plant J. **45:** 804–818.


35


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1015 **Müller, K., Linkies, A., Vreeburg, RAM., Fry, S.C., Krieger-Liszkay, A., Leubner-**


1016 **Metzger, G.** (2009). In vivo cell wall loosening by hydroxyl radicals during cress seed


1017 germination and elongation growth. Plant Physiol. **150:** 1855–1865.


1018 **Nakashima, K., Yamaguchi-Shinozaki, K., Shinozaki, K.** (2014). The transcriptional


1019 regulatory network in the drought response and its crosstalk in abiotic stress


1020 responses including drought, cold, and heat. Front Plant Sci. **5:** 170.


1021 **Nietzel, T., Elsässer, M., Ruberti, C., Steinbeck, J., Ugalde, J.M., Fuchs, P.,**


1022 **Wagner, S., Ostermann, L., Moseler, A., Lemke, P., et al** (2019) The fluorescent


1023 protein sensor roGFP2-Orp1 monitors in _vivo_ H 2 O 2 and thiol redox integration and


1024 elucidates intracellular H 2 O 2 dynamics during elicitor-induced oxidative burst in


1025 _Arabidopsis_ . New Phytologist **221:** 1649–1664.


1026 **Nietzel, T., Mostertz, J., Ruberti, C., Née, G., Fuchs, P., Wagner, S., Moseler, A.,**


1027 **Müller-Schüssele, S.J., Benamar, A., Poschet, G., et al.** (2020). Redox-mediated


1028 kick-start of mitochondrial energy metabolism drives resource-efficient seed


1029 germination. Proc. Natl. Acad. Sci. U S A. **117:** 741–751.


1030 **Nonogaki, H.** (2019). The long-standing paradox of seed dormancy unfolded? Trends


1031 Plant Sci. **24:** 989–998.


1032 **Olszewski, N., Sun, T., Gubler, F.** (2002) Gibberellin signaling. Plant Cell **14:** s61–


1033 s80.


1034 **Oracz, K., El-Maarouf Bouteau, H., Farrant, JM., Cooper, K., Belghazi, M., Job, C.,**


1035 **Job, D., Corbineau, F., Bailly, C.** (2007). ROS production and protein oxidation as


1036 a novel mechanism for seed dormancy alleviation. Plant J. **50:** 452–465.


1037 **Ouzakar, S., Senhaji, NS., El Harsa,l A., Abrini, J.** (2026). Effects of water stress on


1038 the germination of _Origanum compactum_ Benth, _Origanum elongatum_ (Bonnet)


1039 Emb. &amp; Maire, _Origanum grosii_ Pau &amp; Font Quer, and _Thymus capitatus_ (L.)


1040 Hoffmanns. &amp; Link. J. Appl. Res. Med. Aromat. Plants **51:** 100695.


1041 **Peng, H.P., Chan, C.S., Shih, M.C., Yang, S.F.** (2001). Signaling events in the


1042 hypoxic induction of alcohol dehydrogenase gene in _Arabidopsis_ . Plant Physiol. **126:**


1043 742–749.


1044 **Perazzolli, M., Dominici, P., Romero-Puertas, M.C., Zago, E., Zeier, J., Sonoda,**


1045 **M., Lamb, C., Delledonne, M.** (2004), _Arabidopsis_ nonsymbiotic hemoglobin AHb1


1046 modulates nitric oxide bioactivity. Plant Cell **16:** 2785–2794.


1047 **Pracharoenwattana, I., Zhou, W., Smith, S.M,** (2010). Fatty acid beta-oxidation in


36


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1048 germinating _Arabidopsis_ seeds is supported by peroxisomal hydroxypyruvate


1049 reductase when malate dehydrogenase is absent. Plant Mol. Biol. **72:** 101–109.


1050 **del Río, LA., Sandalio, L.M., Corpas, F.J., Palma, J.M., Barroso, J.B.** (2006).


1051 Reactive oxygen species and reactive nitrogen species in peroxisomes. production,


1052 scavenging, and role in cell signaling. Plant Physiol. **141:** 330–335.


1053 **Rose, JKC., Braam, J., Fry, S.C., Nishitani, K.** (2002). The XTH family of enzymes


1054 involved in xyloglucan endotransglucosylation and endohydrolysis: current


1055 perspectives and a new unifying nomenclature. Plant Cell Physiol. **43:** 1421–1435.


1056 **Saito, S., Hirai, N., Matsumoto, C., Ohigashi, H., Ohta, D., Sakata, K., Mizutani, M.**


1057 (2004). _Arabidopsis_ _CYP707As_ encode (+)-abscisic acid 8′-hydroxylase, a key


1058 enzyme in the oxidative catabolism of abscisic acid. Plant Physiol. **134:** 1439–1449.


1059 **Sano, N., Malabarba, J., Chen, Z., Gaillard, S., Windels, D., Verdier, J.** (2022).


1060 Chromatin dynamics associated with seed desiccation tolerance/sensitivity at early


1061 germination in _Medicago truncatula_ . Front Plant Sci. **3:** 1059493.


1062 **Saux, M., Ponnaiah, M., Langlade, N., Zanchetta, C., Balliau, T., El-Maarouf-**


1063 **Bouteau, H., Bailly, C.** (2020). A multiscale approach reveals regulatory players of


1064 water stress responses in seeds during germination. Plant Cell Environ. **43:** 1300–


1065 1313.


1066 **Schröder, F., Lisso, J., Lange, P., Müssig, C.** (2009). The extracellular EXO protein


1067 mediates cell expansion in _Arabidopsis_ leaves. BMC Plant Biol. **9:** 20.


1068 **Schwarzländer, M., Fricker, M.d., Müller, C., Marty, L., Brach, T., Novak, J.,**


1069 **Sweetlove, L. j., Hell, R., Meyer, A. j.** (2008). Confocal imaging of glutathione redox


1070 potential in living plant cells. J. Microsc. **231:** 299–316.


1071 **Shinozaki, K., Yamaguchi-Shinozaki, K.** (2006). Gene networks involved in drought


1072 stress response and tolerance. J. Exp. Bot. **58:** 221–227.


1073 **Skirycz, A., Inzé, D.** (2010). More from less: plant growth under limited water. Curr.


1074 Opin. Biotechnol. **21:** 197–203.


1075 **Skowyra, M.L., Rapoport, T.A.** (2022). PEX5 translocation into and out of


1076 peroxisomes drives matrix protein import. Mol. Cell **82:** 3209-3225.e7.


1077 **Steinbrecher, T., Leubner-Metzger, G.** (2017). The biomechanics of seed


1078 germination. J. Exp. Bot. **68:** 765–783.


1079 **Vallejo, A.J., Yanovsky, M.J., Botto, J.F.** (2010) Germination variation in _Arabidopsis_


37


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1080 _thaliana_ accessions under moderate osmotic and salt stresses. Ann. Bot. **106:** 833–


1081 842.


1082 **Wang, P., Zhu, J.K., Lang, Z.** (2015). Nitric oxide suppresses the inhibitory effect of


1083 abscisic acid on seed germination by S-nitrosylation of SnRK2 proteins. Plant Signal


1084 Behav. **10:** e1031939.


1085 **Wang, W., Zhang, C., Zheng, W., Lv, H., Li, J., Liang, B., Zhou, W.** (2022). Seed


1086 priming with protein hydrolysate promotes seed germination via reserve


1087 mobilization, osmolyte accumulation and antioxidant systems under PEG-induced


1088 drought stress. Plant Cell Rep. **41:** 2173–2186.


1089 **Wilkins, O., Bräutigam, K., Campbell, M.M.** (2010). Time of day shapes _Arabidopsis_


1090 drought transcriptomes. Plant J. **63:** 715–727.


1091 **Willems, P., Mhamdi, A., Stael, S., Storme, V., Kerchev, P., Noctor, G., Gevaert,**


1092 **K., Van Breusegem, F.** (2016). The ROS wheel: refining ROS transcriptional


1093 footprints. Plant Physiol. **171:** 1720–1733.


1094 **Wise, R.R., Naylor, A.W.** (1987). Chilling-enhanced photooxidation: evidence for the


1095 role of singlet oxygen and endogenous antioxidants. Plant Physiol. **83:** 278–282.


1096 **Yang, T., Mu, X., Yu, M., Ergashev, U., Zhu, Y., Shi, N., Li, N., Luo, L., Zhang, K.,**


1097 **Han, Y.** (2025). Consecutive oxidative stress in CATALASE2-deficient _Arabidopsis_


1098 negatively regulates Glycolate Oxidase1 activity through _S_ -nitrosylation. Physiol.


1099 Plant **177:** e70040.


1100 **Ye, N., Zhu, G., Liu, Y., Zhang, A., Li, Y., Liu, R., Shi, L., Jia, L., Zhang, J.** (2012).


1101 Ascorbic acid and reactive oxygen species are involved in the inhibition of seed


1102 germination by abscisic acid in rice seeds. J. Exp. Bot. **63:** 1809–1822.


1103 **Yoshida, T., Fujita, Y., Sayama, H., Kidokoro, S., Maruyama, K., Mizoi, J.,**


1104 **Shinozaki, K., Yamaguchi-Shinozaki, K.** (2010). AREB1, AREB2, and ABF3 are


1105 master transcription factors that cooperatively regulate ABRE-dependent ABA


1106 signaling involved in drought stress tolerance and require ABA for full activation.


1107 Plant J. **61:** 672–685.


1108 **Yoshioka, T., Endo, T., Satoh, S.** (1998). Restoration of seed germination at


1109 supraoptimal temperatures by fluridone, an inhibitor of abscisic acid biosynthesis.


1110 Plant Cell Physiol. **39:** 307–312.


1111 **Zhang, R.R., Wang, Y.H., Peng, X.F., Sun, Y.J., Zhang, N., Xiong, A.S.** (2025).


1112 DcNCED2 promotes ABA synthesis via carotenoid degradation and enhances


38


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1113 drought resistance in carrot. Plant Cell Rep. **44:** 75.


1114


1115 **Arnaud, D., Deeks, M.J., Smirnoff, N.** (2023). Organelle-targeted biosensors reveal


1116 distinct oxidative events during pattern-triggered immune responses. Plant Physiol.


1117 **191:** 2551–2569.


1118 **Babajani, G., Effendy, J., Plant, A.L.** (2009). _Sl-SROl1_ increases salt tolerance and


1119 is a member of the _radical-induced cell death 1_ — _similar to RCD1_ gene family of


1120 tomato. Plant Sci. **176:** 214–222.


1121 **Bahin, E., Bailly, C., Sotta, B., Kranner, I., Corbineau, F., Leymarie, J.** (2011).


1122 Crosstalk between reactive oxygen species and hormonal signalling pathways


1123 regulates grain dormancy in barley. Plant Cell Environ. **34:** 980–993.


1124 **Bailly, C.** (2019) The signalling role of ROS in the regulation of seed germination and


1125 dormancy. Biochem. J. **476:** 3019–3032.


1126 **Bailly, C.** (2022). ROS in seed germination. In: Mittler R, Van Breusegem F, Jacquot


1127 JP, eds. Advances in Botanical Research, Vol. 105: Oxidative Stress Response in


1128 Plants. Academic Press, pp. 177–204.


1129 **Bailly, C.** (2004). Active oxygen species and antioxidants in seed biology. Seed Sci.


1130 Res. **14:** 93–107.


1131 **Bailly, C., Benamar, A., Corbineau, F., Come, D.** (1996). Changes in


1132 malondialdehyde content and in superoxide dismutase, catalase and glutathione


1133 reductase activities in sunflower seeds as related to deterioration during accelerated


1134 aging. Physiol. Plant. **97:** 104–110.


1135 **Bailly, C., El-Maarouf-Bouteau, H., Corbineau, F.** (2008). From intracellular


1136 signaling networks to cell death: the dual role of reactive oxygen species in seed


1137 physiology. C. R. Biol. **331:** 806–814.


1138 **Bailly, C., Gomez Roldan, M.V.** (2023). Impact of climate perturbations on seeds and


1139 seed quality for global agriculture. Biochem. J. **480:** 177–196.


1140 **Bi, C., Ma, Y., Wu, Z., Yu, Y.T., Liang, S., Lu, K., Wang, X.F.** (2017). _Arabidopsis_


1141 ABI5 plays a role in regulating ROS homeostasis by activating _CATALASE 1_


1142 transcription in seed germination. Plant Mol. Biol. **94:** 197–213.


1143 **Chang, S., Puryear, J., Cairney, J.** (1993). A simple and efficient method for isolating


1144 RNA from pine trees. Plant Mol. Biol. Rep. **11:** 113–116.


39


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1145 **Chauffour, F., Bailly, M., Perreau, F., Cueff, G., Suzuki, H., Collet, B., Frey, A.,**


1146 **Clément, G., Soubigou-Taconnat, L., Balliau, T., et al.** (2019). Multi-omics


1147 analysis reveals sequential roles for ABA during seed maturation. Plant Physiol.


1148 **180:** 1198–1218.


1149 **Chen, W., Zhao, J., Tao, Z., Zhang, S., Bei, X., Lu, W., Qi, X.** (2024). Short heat


1150 shock factor A2 regulates heat resistance and growth balance in _Arabidopsis_ . eLife


1151 **13:** RP99937.


1152 **Cheng, M., Guo, Y., Liu, Q., Nan, S., Xue, Y., Wei, C., Zhang, Y., Luan, F., Zhang,**


1153 **X., Li, H.** (2022). H 2 O 2 and Ca &#91;2+&#93; signaling crosstalk counteracts aba to induce seed


1154 germination. Antioxidants (Basel) **11:** 1594.


1155 **Claeys, H., Inzé, D.** (2013). The agony of choice: how plants balance growth and


1156 survival under water-limiting conditions. Plant Physiol. **162:** 1768–1779.


1157 **Dong, Z., Yu, Y., Li, S., Wang, J., Tang, S., Huang, R.** (2016). Abscisic acid


1158 antagonizes ethylene production through the ABI4-mediated transcriptional


1159 repression of _ACS4_ and _ACS8_ in _Arabidopsis_ . Molecular Plant **9:** 126–135.


1160 **Eastmond, P.J., Hooks, M.A., Williams, D., Lange, P., Bechtold, N., Sarrobert, C.,**


1161 **Nussaume, L., Graham, I.A.** (2000). Promoter trapping of a novel medium-chain


1162 acyl-coa oxidase, which is induced transcriptionally during _Arabidopsis_ seed


1163 germination. J. Biol. Chem. **275:** 34375–34381.


1164 **Engqvist, MKM., Schmitz, J., Gertzmann, A., Florian, A., Jaspert, N., Arif, M.,**


1165 **Balazadeh, S., Mueller-Roeber, B., Fernie, A.R., Maurino, V.G.** (2015).


1166 GLYCOLATE OXIDASE3, a glycolate oxidase homolog of yeast l-lactate


1167 cytochrome c oxidoreductase, supports l-lactate oxidation in roots of _Arabidopsis_ .


1168 Plant Physiol. **169:** 1042–1061.


1169 **Esser, C., Kuhn, A., Groth, G., Lercher, M.J., Maurino, V.G.** (2014). Plant and


1170 animal glycolate oxidases have a common eukaryotic ancestor and convergently


1171 duplicated to evolve long-chain 2-hydroxy acid oxidases. Mol. Biol. Evol. **31:** 1089–


1172 1101.


1173 **Farooq, M., Wahid, A., Lee, D.J., Ito, O., Siddique, KHM.** (2009). Advances in


1174 drought resistance of rice. Crit. Rev. Plant Sci. **28:** 199–217.


1175 **Finch-Savage, W.E., Leubner-Metzger, G.** (2006). Seed dormancy and the control


1176 of germination. New Phytol. **171:** 501–523.


1177 **Fricker, M.D.** (2016). Quantitative redox imaging software. Antioxid. Redox Signal. **24:**


40


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1178 752–762.


1179 **Frugoli, J.A., Zhong, H.H., Nuccio, M.L., McCourt, P., McPeek, MA., Thomas, TL.,**


1180 **McClung, CR.** (1996). Catalase is encoded by a multigene family in _Arabidopsis_


1181 _thaliana_ (L.) Heynh. Plant Physiol. **112:** 327–336.


1182 **Fujii, H., Verslues, PE., Zhu, J.K.** (2007). Identification of two protein kinases required


1183 for abscisic acid regulation of seed germination, root growth, and gene expression


1184 in _Arabidopsis_ . Plant Cell **19:** 485–494.


1185 **Gaber, A** . (2013). The importance of _Arabidopsis_ glutathione peroxidase 8 for


1186 protecting _Arabidopsis_ plant and E. coli cells against oxidative stress. GM crops &amp;


1187 food. **5:** 20–26.


1188 **Gao, S., Guo, W., Feng, W., Liu, L., Song, X., Chen, J., Hou, W., Zhu, H., Tang, S.,**


1189 **Hu, J.** (2016). _LTP3_ contributes to disease susceptibility in _Arabidopsis_ by


1190 enhancing abscisic acid (ABA) biosynthesis. Mol. Plant Pathol. **17:** 412–426.


1191 **Germain, V., Rylott, E.L., Larson, T.R., Sherson, S.M., Bechtold, N., Carde, J.P.,**


1192 **Bryce, J.H., Graham, I.A., Smith, S.M.** (2001). Requirement for 3-ketoacyl-CoA


1193 thiolase-2 in peroxisome development, fatty acid β-oxidation and breakdown of


1194 triacylglycerol in lipid bodies of _Arabidopsis_ seedlings. Plant J. **28:** 1–12.


1195 **González-Guzmán, M., Apostolova, N., Bellés, J.M., Barrero, J.M., Piqueras, P.,**


1196 **Ponce, M.R., Micol, J.L., Serrano, R., Rodríguez, P.L.** (2002). The short-chain


1197 alcohol dehydrogenase ABA2 catalyzes the conversion of xanthoxin to abscisic


1198 aldehyde. Plant Cell **14:** 1833–1846.


1199 **Gonzali, S., Loreti, E., Cardarelli, F., Novi, G., Parlanti, S., Pucciariello, C.,**


1200 **Bassolino, L., Banti, V., Licausi, F., Perata, P.** (2015). Universal stress protein


1201 HRU1 mediates ROS homeostasis under anoxia. Nature Plants **1:** 15151.


1202 **Graham, I.A.** (2008). Seed storage oil mobilization. Annu. Rev. Plant Biol. **59:** 115–


1203 142.


1204 **Hasanuzzaman, M., Nahar, K., Gill, S.S., Fujita, M.** (2013). Drought stress responses


1205 in plants, oxidative stress, and antioxidant defense. In: Tuteja N, Gill SS, eds.


1206 Climate Change and Plant Abiotic Stress Tolerance. Wiley-VCH Verlag GmbH &amp; Co.


1207 KGaA, Weinheim, pp 209–250.


1208 **Hassan, M.J., Geng, W., Zeng, W., Raza, M.A., Khan, I., Iqbal, M.Z., Peng, Y., Zhu,**


1209 **Y., Li, Z.** (2021). Diethyl aminoethyl hexanoate priming ameliorates seed


1210 germination _via_ involvement in hormonal changes, osmotic adjustment, and


41


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1211 dehydrins accumulation in white clover under drought stress. Front Plant Sci. **12:**


1212 709187.


1213 **Heath, RL., Packer, L.** (1968). Photoperoxidation in isolated chloroplasts: I. Kinetics


1214 and stoichiometry of fatty acid peroxidation. Arch. Biochem. Biophys. **125:** 189–198.


1215 **Horváth, E., Bela, K., Gallé, Á., Riyazuddin, R., Csomor, G., Csenki, D., Csiszár,**


1216 **J.** (2020). Compensation of mutation in _Arabidopsis_ glutathione _transferase_


1217 ( _AtGSTU_ ) genes under control or salt stress conditions. Int. J. Mol. Sci. **21:** 2349.


1218 **Horváth, E., Bela, K., Kulman, K., Faragó, N., Riyazuddin, R., Gallé, Á., Puskás,**


1219 **L.G., Csiszár, J.** (2023). Glutathione transferases are involved in salicylic acid

1220 induced transcriptional reprogramming. J. Plant Growth Regul. **42:** 4497–4510.


1221 **Hung, K.T., Kao, C.H.** (2004). Nitric oxide acts as an antioxidant and delays methyl


1222 jasmonate-induced senescence of rice leaves. J. Plant Physiol. **161:** 43–52.


1223 **Hwang, S., Simko, I., Mou, B.** (2024). QTL mapping and transcriptome analysis of


1224 seed germination under PEG-induced water stress in Lactuca spp. Sci Rep. **14:**


1225 27157.


1226 **Ishibashi, Y., Tawaratsumida, T., Kondo, K., Kasa, S., Sakamoto, M., Aoki, N.,**


1227 **Zheng, S.H., Yuasa, T., Iwaya-Inoue, M.** (2012). Reactive oxygen species are


1228 involved in gibberellin/abscisic acid signaling in barley aleurone cells. Plant Physiol.


1229 **158:** 1705–1714.


1230 **Jurdak, R., Launay-Avon, A., Roux, CP.L., Bailly, C.** (2020). Retrograde signalling


1231 from the mitochondria to the nucleus translates the positive effect of ethylene on


1232 dormancy breaking of _Arabidopsis thaliana_ seeds. New Phytol. **229:** 2192–2205.


1233 **Jurdak, R., Rodrigues, G. de. AG., Chaumont, N., Schivre, G., Bourbousse, C.,**


1234 **Barneche, F., Bou, Dagher., Kharrat, M., Bailly, C.** (2022). Intracellular reactive


1235 oxygen species trafficking participates in seed dormancy alleviation in _Arabidopsis_


1236 seeds. New Phytol. **234:** 850–866.


1237 **Kolesnikov, M., Gerasko, T., Paschenko, Yu., Pokoptseva, L., Onyschenko, O.,**


1238 **Kolesnikova, A.** (2023). Effect of water deficit on maize seeds ( _Zea mays_ L.) during


1239 germination. Agron. Res. **21:** e028.


1240 **Kopriva, S., Rennenberg, H** . (2004). Control of sulphate assimilation and glutathione


1241 synthesis: interaction with N and C metabolism. J. Exp. Bot. **55:** 1831–1842.


1242 **Kubalová, M., Griffiths, J., Müller, K., Levenets, L., Tylová, E., Tarkowská, D.,**


42


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1243 **Jones, A.M., Fendrych, M.** (2025). Gibberellin-deactivating GA2OX enzymes act


1244 as a hub for auxin–gibberellin cross talk in _Arabidopsis thaliana_ root growth


1245 regulation. Proc. Natl. Acad. Sci. U S A. **122:** e2425574122.


1246 **Kushiro, T., Okamoto, M., Nakabayashi, K., Yamagishi, K., Kitamura, S., Asami,**


1247 **T., Hirai, N., Koshiba, T., Kamiya, Y., Nambara, E.** (2004). The _Arabidopsis_


1248 cytochrome P450 CYP707A encodes ABA 8′-hydroxylases: key enzymes in ABA


1249 catabolism. EMBO J. **23:** 1647–1656.


1250 **Lariguet, P., Ranocha, P., De Meyer, M., Barbier, O., Penel, C., Dunand, C.** (2013).


1251 Identification of a hydrogen peroxide signalling pathway in the control of light

1252 dependent germination in _Arabidopsis_ . Planta **238:** 381–395.


1253 **Lee, C., Teng, Q., Huang, W., Zhong, R., Ye, Z.H.** (2009). The F8H


1254 glycosyltransferase is a functional paralog of fra8 involved in glucuronoxylan


1255 biosynthesis in _Arabidopsis_ . Plant Cell Physiol. **50:** 812–827.


1256 **Levine, A., Lamb, C.** (1994). H 2 O 2 from the oxidative burst orchestrates the plant


1257 hypersensitive disease resistance response. Cell **79:** 583–593.


1258 **Leymarie, J., Vitkauskaité, G., Hoang, H.H., Gendreau, E., Chazoule, V.,**


1259 **Meimoun, P., Corbineau, F., El-Maarouf-Bouteau, H., Bailly, C.** (2012). Role of


1260 reactive oxygen species in the regulation of _Arabidopsis_ seed dormancy. Plant Cell


1261 Physiol. **53:** 96–106.


1262 **Liu, J., Hasanuzzaman, M., Wen, H., Zhang, J., Peng, T., Sun, H., Zhao, Q.** (2019).


1263 High temperature and drought stress cause abscisic acid and reactive oxygen


1264 species accumulation and suppress seed germination growth in rice. Protoplasma


1265 **256:** 1217–1227.


1266 **Liu, Y., Ye, N., Liu, R., Chen, M., Zhang, J.** (2010). H 2 O 2 mediates the regulation of


1267 ABA catabolism and GA biosynthesis in _Arabidopsis_ seed dormancy and


1268 germination. J. Exp. Bot, **61:** 2979–2990.


1269 **Liu, Y.C., Wu, Y.R., Huang, X.H., Sun, J., Xie, Q.** (2011). AtPUB19, a U-box e3


1270 ubiquitin ligase, negatively regulates abscisic acid and drought responses in


1271 _Arabidopsis thaliana_ . Mol. Plant. **4:** 938–946.


1272 **Llanes, A., Andrade, A., Masciarelli, O., Alemano, S., Luna, V.** (2016). Drought and


1273 salinity alter endogenous hormonal profiles at the seed germination phase. Seed


1274 Sci. Res. **26:** 1–13.


1275 **Men, Y., Wei, P., Cao, Z., Yang, L., Wei, J., Zhang, D., Zhang, X., Zhang, F., Yang,**


43


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1276 **Z.** (2026). A transcriptome-based study using PEG-induced drought simulation in


1277 _Allium mongolicum_ seeds. Grassl. Sci. **72:** 16–30.


1278 **Michel, B.E., Kaufmann, M.R.** (1973), The osmotic potential of polyethylene glycol


1279 6000. Plant Physiol. **51:** 914–916.


1280 **Mitchum, M.G., Yamaguchi, S., Hanada, A., Kuwahara, A., Yoshioka, Y., Kato, T.,**


1281 **Tabata, S., Kamiya, Y., Sun, T.P.** (2006). Distinct and overlapping roles of two


1282 gibberellin 3-oxidases in _Arabidopsis_ development. Plant J. **45:** 804–818.


1283 **Müller, K., Linkies, A., Vreeburg, RAM., Fry, S.C., Krieger-Liszkay, A., Leubner-**


1284 **Metzger, G.** (2009). In vivo cell wall loosening by hydroxyl radicals during cress seed


1285 germination and elongation growth. Plant Physiol. **150:** 1855–1865.


1286 **Nakashima, K., Yamaguchi-Shinozaki, K., Shinozaki, K.** (2014). The transcriptional


1287 regulatory network in the drought response and its crosstalk in abiotic stress


1288 responses including drought, cold, and heat. Front Plant Sci. **5:** 170.


1289 **Nietzel, T., Elsässer, M., Ruberti, C., Steinbeck, J., Ugalde, J.M., Fuchs, P.,**


1290 **Wagner, S., Ostermann, L., Moseler, A., Lemke, P., et al** (2019) The fluorescent


1291 protein sensor roGFP2-Orp1 monitors in _vivo_ H 2 O 2 and thiol redox integration and


1292 elucidates intracellular H 2 O 2 dynamics during elicitor-induced oxidative burst in


1293 _Arabidopsis_ . New Phytologist **221:** 1649–1664.


1294 **Nietzel, T., Mostertz, J., Ruberti, C., Née, G., Fuchs, P., Wagner, S., Moseler, A.,**


1295 **Müller-Schüssele, S.J., Benamar, A., Poschet, G., et al.** (2020). Redox-mediated


1296 kick-start of mitochondrial energy metabolism drives resource-efficient seed


1297 germination. Proc. Natl. Acad. Sci. U S A. **117:** 741–751.


1298 **Nonogaki, H.** (2019). The long-standing paradox of seed dormancy unfolded? Trends


1299 Plant Sci. **24:** 989–998.


1300 **Olszewski, N., Sun, T., Gubler, F.** (2002) Gibberellin signaling. Plant Cell **14:** s61–


1301 s80.


1302 **Oracz, K., El-Maarouf Bouteau, H., Farrant, JM., Cooper, K., Belghazi, M., Job, C.,**


1303 **Job, D., Corbineau, F., Bailly, C.** (2007). ROS production and protein oxidation as


1304 a novel mechanism for seed dormancy alleviation. Plant J. **50:** 452–465.


1305 **Ouzakar, S., Senhaji, NS., El Harsa,l A., Abrini, J.** (2026). Effects of water stress on


1306 the germination of _Origanum compactum_ Benth, _Origanum elongatum_ (Bonnet)


1307 Emb. &amp; Maire, _Origanum grosii_ Pau &amp; Font Quer, and _Thymus capitatus_ (L.)


1308 Hoffmanns. &amp; Link. J. Appl. Res. Med. Aromat. Plants **51:** 100695.


44


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1309 **Peng, H.P., Chan, C.S., Shih, M.C., Yang, S.F.** (2001). Signaling events in the


1310 hypoxic induction of alcohol dehydrogenase gene in _Arabidopsis_ . Plant Physiol. **126:**


1311 742–749.


1312 **Perazzolli, M., Dominici, P., Romero-Puertas, M.C., Zago, E., Zeier, J., Sonoda,**


1313 **M., Lamb, C., Delledonne, M.** (2004), _Arabidopsis_ nonsymbiotic hemoglobin AHb1


1314 modulates nitric oxide bioactivity. Plant Cell **16:** 2785–2794.


1315 **Pracharoenwattana, I., Zhou, W., Smith, S.M,** (2010). Fatty acid beta-oxidation in


1316 germinating _Arabidopsis_ seeds is supported by peroxisomal hydroxypyruvate


1317 reductase when malate dehydrogenase is absent. Plant Mol. Biol. **72:** 101–109.


1318 **del Río, LA., Sandalio, L.M., Corpas, F.J., Palma, J.M., Barroso, J.B.** (2006).


1319 Reactive oxygen species and reactive nitrogen species in peroxisomes. production,


1320 scavenging, and role in cell signaling. Plant Physiol. **141:** 330–335.


1321 **Rose, JKC., Braam, J., Fry, S.C., Nishitani, K.** (2002). The XTH family of enzymes


1322 involved in xyloglucan endotransglucosylation and endohydrolysis: current


1323 perspectives and a new unifying nomenclature. Plant Cell Physiol. **43:** 1421–1435.


1324 **Saito, S., Hirai, N., Matsumoto, C., Ohigashi, H., Ohta, D., Sakata, K., Mizutani, M.**


1325 (2004). _Arabidopsis_ _CYP707As_ encode (+)-abscisic acid 8′-hydroxylase, a key


1326 enzyme in the oxidative catabolism of abscisic acid. Plant Physiol. **134:** 1439–1449.


1327 **Sano, N., Malabarba, J., Chen, Z., Gaillard, S., Windels, D., Verdier, J.** (2022).


1328 Chromatin dynamics associated with seed desiccation tolerance/sensitivity at early


1329 germination in _Medicago truncatula_ . Front Plant Sci. **3:** 1059493.


1330 **Saux, M., Ponnaiah, M., Langlade, N., Zanchetta, C., Balliau, T., El-Maarouf-**


1331 **Bouteau, H., Bailly, C.** (2020). A multiscale approach reveals regulatory players of


1332 water stress responses in seeds during germination. Plant Cell Environ. **43:** 1300–


1333 1313.


1334 **Schröder, F., Lisso, J., Lange, P., Müssig, C.** (2009). The extracellular EXO protein


1335 mediates cell expansion in _Arabidopsis_ leaves. BMC Plant Biol. **9:** 20.


1336 **Schwarzländer, M., Fricker, M.d., Müller, C., Marty, L., Brach, T., Novak, J.,**


1337 **Sweetlove, L. j., Hell, R., Meyer, A. j.** (2008). Confocal imaging of glutathione redox


1338 potential in living plant cells. J. Microsc. **231:** 299–316.


1339 **Shinozaki, K., Yamaguchi-Shinozaki, K.** (2006). Gene networks involved in drought


1340 stress response and tolerance. J. Exp. Bot. **58:** 221–227.


45


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1341 **Skirycz, A., Inzé, D.** (2010). More from less: plant growth under limited water. Curr.


1342 Opin. Biotechnol. **21:** 197–203.


1343 **Skowyra, M.L., Rapoport, T.A.** (2022). PEX5 translocation into and out of


1344 peroxisomes drives matrix protein import. Mol. Cell **82:** 3209-3225.e7.


1345 **Steinbrecher, T., Leubner-Metzger, G.** (2017). The biomechanics of seed


1346 germination. J. Exp. Bot. **68:** 765–783.


1347 **Vallejo, A.J., Yanovsky, M.J., Botto, J.F.** (2010) Germination variation in _Arabidopsis_


1348 _thaliana_ accessions under moderate osmotic and salt stresses. Ann. Bot. **106:** 833–


1349 842.


1350 **Wang, P., Zhu, J.K., Lang, Z.** (2015). Nitric oxide suppresses the inhibitory effect of


1351 abscisic acid on seed germination by S-nitrosylation of SnRK2 proteins. Plant Signal


1352 Behav. **10:** e1031939.


1353 **Wang, W., Zhang, C., Zheng, W., Lv, H., Li, J., Liang, B., Zhou, W.** (2022). Seed


1354 priming with protein hydrolysate promotes seed germination via reserve


1355 mobilization, osmolyte accumulation and antioxidant systems under PEG-induced


1356 drought stress. Plant Cell Rep. **41:** 2173–2186.


1357 **Wilkins, O., Bräutigam, K., Campbell, M.M.** (2010). Time of day shapes _Arabidopsis_


1358 drought transcriptomes. Plant J. **63:** 715–727.


1359 **Willems, P., Mhamdi, A., Stael, S., Storme, V., Kerchev, P., Noctor, G., Gevaert,**


1360 **K., Van Breusegem, F.** (2016). The ROS wheel: refining ROS transcriptional


1361 footprints. Plant Physiol. **171:** 1720–1733.


1362 **Wise, R.R., Naylor, A.W.** (1987). Chilling-enhanced photooxidation: evidence for the


1363 role of singlet oxygen and endogenous antioxidants. Plant Physiol. **83:** 278–282.


1364 **Yang, T., Mu, X., Yu, M., Ergashev, U., Zhu, Y., Shi, N., Li, N., Luo, L., Zhang, K.,**


1365 **Han, Y.** (2025). Consecutive oxidative stress in CATALASE2-deficient _Arabidopsis_


1366 negatively regulates Glycolate Oxidase1 activity through _S_ -nitrosylation. Physiol.


1367 Plant **177:** e70040.


1368 **Ye, N., Zhu, G., Liu, Y., Zhang, A., Li, Y., Liu, R., Shi, L., Jia, L., Zhang, J.** (2012).


1369 Ascorbic acid and reactive oxygen species are involved in the inhibition of seed


1370 germination by abscisic acid in rice seeds. J. Exp. Bot. **63:** 1809–1822.


1371 **Yoshida, T., Fujita, Y., Sayama, H., Kidokoro, S., Maruyama, K., Mizoi, J.,**


1372 **Shinozaki, K., Yamaguchi-Shinozaki, K.** (2010). AREB1, AREB2, and ABF3 are


46


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.16.766966; this version posted November 18, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.16.766966)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1373 master transcription factors that cooperatively regulate ABRE-dependent ABA


1374 signaling involved in drought stress tolerance and require ABA for full activation.


1375 Plant J. **61:** 672–685.


1376 **Yoshioka, T., Endo, T., Satoh, S.** (1998). Restoration of seed germination at


1377 supraoptimal temperatures by fluridone, an inhibitor of abscisic acid biosynthesis.


1378 Plant Cell Physiol. **39:** 307–312.


1379 **Zhang, R.R., Wang, Y.H., Peng, X.F., Sun, Y.J., Zhang, N., Xiong, A.S.** (2025).


1380 DcNCED2 promotes ABA synthesis via carotenoid degradation and enhances


1381 drought resistance in carrot. Plant Cell Rep. **44:** 75.


1382


1383 **SUPPORTING INFORMATION**


1384 The following Additional Supporting Information is available for this article:


1385 Figure S1: Calibration of the Mito-roGFP2-Orp1 and Pero-roGFP2-Orp1 redox sensors


1386 in Arabidopsis seeds.


1387 Figure S2: Peroxisome-derived ROS contribute to the germination response to water


1388 stress.


1389 Figure S3: Identification of hub genes by weighted gene co-expression network


1390 analysis (WGCNA).


1391 Table S1: Differentially expressed genes under water stress condition vs control


1392 condition at ET


1393 Table S2: Differentially expressed genes under water stress condition vs control


1394 condition at MT


1395 Table S3: Differentially expressed genes under water stress condition vs control


1396 condition at LT


1397 Table S4: ROS-related transcriptional signatures associated with the response to


1398 water stress during seed germination.


1399 Table S5: Gene list of purple module acquired by gene set enrichment analysis


1400


47


**Figure S1. Calibration of the Mito-roGFP2-Orp1 and Pero-roGFP2-Orp1 redox**
**sensors in Arabidopsis seeds.**


_Calibration of Arabidopsis seeds expressing Mito-roGFP2-Orp1 (_ _**A**_ _) or Pero-roGFP2-Orp1 (_ _**B**_ _)._
_Biosensors were fully oxidized by treatment with 10 mM H_ _2_ _O_ _2_ _(Ox) or fully reduced with 10 mM DTT_
_(Red). Representative fluorescence images acquired following excitation at 405 nm and 488 nm,_
_merged images, and the corresponding false-color ratio images (405/488) are shown. Scale bars = 10_

μ _m._


**Figure S2. Peroxisome-derived ROS contribute to the germination response to**
**water stress.**


_Transcript abundance of genes involved in peroxisomal ROS metabolism, including ROS-producing_
_enzymes (GOXs,_ _**A**_ _) and ROS-scavenging enzymes (CAT2,_ _**B**_ _), during germination under control and_
_water stress conditions. Data are presented as the mean ± SD of three biological replicates. An asterisk_
_indicates genes that were significantly differentially expressed between control and water stress_
_conditions (FDR &lt; 0.01). Germination assays of loss-of-function mutants affected in peroxisomal ROS_
_production (_ _**C**_ _) or ROS scavenging (_ _**D**_ _) under control and water stress conditions. Data are presented as_
_the mean ± SD of five biological replicates. Statistical significance between treatments at each time_
_point was determined using Student's t-test. Asterisks indicate significant differences between_
_treatments_ _(P_ _&lt;_ _0.05;_ _*P_ _&lt;_ _0.01;_ _***P_ _&lt;_ _0.0001)._


**Figure S3. Identification of hub genes by weighted gene co-expression network**
**analysis (WGCNA).**


_**(A)**_ _Analysis of the scale-free topology fit index and mean connectivity as a function of the soft-_
_thresholding power (_ β _)._ _**(B)**_ _Hierarchical clustering dendrogram of 24,640 transcripts and the_
_corresponding module assignments. Different colors represent distinct co-expression modules (soft-_
_thresholding power_ β _= 6; scale-free topology fit index,_ R _² &gt; 0.9)._ _**(C)**_ _Expression profile of the purple_
_module eigengene under control (blue) and water stress (red) conditions during germination. Data are_
_presented as the mean ± SD of normalized transcript counts from three biological replicates. ET, early_
_time point; MT, middle time point; LT, late time point._


