Hydrogen sulfide modulates plant hypoxic responses through the persulfidation of Plant Cysteine Oxidases

17 Hydrogen sulfide (H₂S) is a gaseous molecule historically regarded as toxic. Nevertheless, increasing 18 evidence has brought to light important physiological roles in both animals and plants. In plants, H₂S 19 is involved in environmental and developmental responses, such as stomatal closure and seed 20 germination, and in tolerance mechanisms to different stress conditions like salinity, drought and 21 waterlogging. In this study, we report a function of H₂S as a modulator of hypoxic responses in 22 Arabidopsis thaliana . A combination of biochemical and genetic evidence demonstrates that H₂S 23 inhibits the activity of Plant Cysteine Oxidases, the molecular sensors of oxygen, through protein 24 persulfidation to modulate hypoxia-associated responses. Furthermore, we show that H₂S physiology 25 contributes to responses to low oxygen, as disturbing H2S production impaired activation of hypoxia-26 responsive genes and submergence tolerance. Overall, this work introduces H₂S as signalling 27 modulator in plant hypoxic responses and adds a regulatory layer to the plant oxygen -sensing 28 mechanism. 29 30 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Main text 31

32 Plants have evolved intricate mechanisms to perceive and respond to low oxygen conditions 33 (hypoxia), which are common in natural environments such as waterlogged or compacted soils 34 (Habibi et al., 2023). A central component of the hypoxic response in Arabidopsis thaliana is the Arg 35 N-degron pathway, which coordinates the activity of Ethylene Responsive Factor VII (ERFVII) 36 transcription factors according to O2 availability (Gibbs et al., 2011; Licausi et al., 2011). In presence 37 of sufficient O2, Plant Cysteine Oxidases (PCOs) catalyse the oxidation of their ERFVII substrates at 38 the level of the N-terminal Cys residue, allowing subsequent ERFVII proteasomal degradation via 39 the Arg N-degron pathway (Licausi et al., 2020) . Low O2 conditions, instead, impair PCO activity. 40 This makes the ERFVII proteins stable, allowing them to migrate into the nucleus and induce the 41 expression of hypoxia-responsive genes (HRGs) (Gibbs et al., 2011; Licausi et al., 2011) . The 42 Arabidopsis ERFVII family include five members: Hypoxia-Responsive Element (HRE1 and HRE2) 43 and Related to Apetala2 proteins (RAP2.2, RAP2.3, and RAP2.12) (Licausi et al., 2013) . These 44 proteins share a distinctive [MCGGAII(A/S)D] motif, which has been identified as an N -degron 45 (Licausi et al., 2011; Nakano et al., 2006). RAP-type ERFVIIs play a key role in O2 sensing, whereas 46 HRE-type ERFVIIs, encoded by hypoxia -inducible genes, rather function downstream in the 47 signalling pathway (Licausi et al., 2010) . The ERFVII s coordinate HRG expression by binding 48 Hypoxia-Responsive Promoter Elements (HRPEs) (Gasch et al., 2016). A core set of HRGs has been 49 identified, across different plant species, that encode proteins involved in hypoxic metabolism, redox 50 regulation, and other processes that remain to be fully characterized (Mustroph et al., 2010; 51 Renziehausen et al., 2024). Hypoxia signalling, however, is not limited to the perception of variations 52 in subcellular O2 availability. During O2 deprivation, indeed, various signalling molecules accumulate 53 locally, playing a crucial role in coordinating responses to hypoxia. Among them, ethylene, nitric 54 oxide (NO), and reactive oxygen species (ROS) have been extensively studied (Hartman et al., 2019; 55 Sasidharan et al., 2018). 56 More recently, the gasotransmitter molecule, hydrogen sulfide (H₂S), has been gaining increasing 57 attention within the scientific community. Historically regarded a toxic gas that inhibits respiration, 58 H₂S has emerged in recent years as a signalling molecule in plants and animals (Aroca et al., 2015). 59 In plants, H₂S has been recognized as a key regulator of environmental sensing and developmental 60 programs, including ABA-dependent stomatal closure (Scuffi et al., 2014; Zhang et al., 2020) and 61 seed germination (Sharma et al., 2022) . Partially related, a protective role of H 2S in the biotic and 62 abiotic stress defence response was also suggested (Bhadwal et al., 2024; Xiang et al., 2023 ; 63 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Pantaleno et al., 2024) . Notably, pre -treatment with NaHS, a H₂S source, was demonstrated to 64 enhance tolerance to flooding stress and induce the expression of hypoxia-responsive genes in 65 Arabidopsis (Yang et al., 2021) . Evidence indicating H₂S accumulation in plant tissues under 66 submergence (Xiao et al., 2020; Yang et al., 2021) highlights its potential involvement in modulating 67 hypoxia-related signalling pathways, a role that remains to be elucidated. 68 The biosynthesis of H₂S in plants occurs through several pathways; the well-characterized routes are 69 the sulfate (SO₄²⁻) reduction pathway in sulfur assimilation and cysteine metabolism (Saito, 2004). 70 In the former, sulfate is transported into the plastid where its conversion to sulfite (SO₃²⁻) occurs, 71 followed by reduction to sulfide (S²⁻)(Takahashi et al., 2011). Most of the sulfide from the plastid is 72 released as H 2S and fixed into cysteine , the product of sulfur assimilation, by the enzyme O -73 acetylserine(thiol)lyase (OASTL), mostly in the cytosol, although other OASTLs are located in the 74 plastid and mitochondria (Hell & Wirtz, 2008). Cysteine participates in the endogenous synthesis of 75 H₂S through the degradation of Cys by various types of cysteine -degrading enzymes (Gotor et al., 76 2019). Particularly, through the action of L-Cys Desulfhydrases (LCDESs), it leads to the release of 77 H₂S, pyruvate and ammonia , and in Arabidopsis , one cytosolic LCDES, named DES1, has been 78 characterized in detail (Álvarez et al., 2010) . Cys catabolism plays a crucial role when plants are 79 exposed to oxidative stress or O2 deprivation. Under these stresses, Cys catabolism is upregulated, 80 leading to increased H₂S production, which can act as a signalling molecule to mitigate the effects of 81 the stress (Yang et al., 2021). 82 While the precise mechanisms of H₂S signalling in plants are yet to be completely understood, one 83 of the proposed modes of action involves protein persulfidation (Aroca et al., 2015; Mustafa et al., 84 2009). This post-translational modification occurs on the thiol groups ( -SH) of Cys residues, 85 converting them into persulfides (-SSH) (Aroca et al., 2017; Aroca et al., 2018). However, the direct 86 interaction between H₂S and Cys thiols is thermodynamically disfavoured (Cuevasanta et al., 2015; 87 Filipovic et al., 2018) . Therefore, persulfidation typically requires the prior oxidation of thiols to 88 sulfenic acids (-SOH) or S-nitrosylation to nitrothiols (-SNO) via ROS or NO, respectively (Vignane 89 & Filipovic, 2023). This adds a layer of complexity into the biochemistry of persulfidation, suggesting 90 a finely tuned regulatory mechanism involving multiple signalling molecules, with significant 91 consequences for protein function (Aroca et al., 2021; Aroca et al., 2017). Persulfidation is conserved 92 across all domains of life and plays a key role in regulating the localization, conformation and activity 93 of numerous proteins, with profound effects on cellular regulation (Aroca et al., 2018). A proteomic 94 analysis revealed that , out of over 900 proteins in Arabidopsis susceptible to S -nitrosylation, more 95 than 600 also represent persulfidation targets (Aroca et al., 2017) . Among these dual targets, Plant 96 Cysteine Oxidase 3 (PCO3) was identified, which is particularly notable in the context of hypoxia. 97 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint In this study, we investigate the role of H₂S in the response to hypoxia in Arabidopsis. Using 98 biochemical and genetic analyses, we observed that H₂S directly modulates the activity of PCO 99 enzymes, influencing the ability of plants to sense O2. 100 101

102 Sulfate burst stimulates ERFVII-mediated responses in aerobic conditions 103 Sulfate metabolism is the route of H₂S biosynthesis in plant cells ; therefore, we speculated that 104 modulating this process could influence hypoxia responses of Arabidopsis seedlings. We sought to 105 stimulate a transient burst in H2S production through the reductive sulfate assimilation pathway by 106 modifying S availability in the nutrient solution. Col-0 seedlings grown for 5 days under S deficiency 107 (S-, 25 µM MgSO₄) were transferred to optimal S conditions (S+, 750 µM MgSO₄) for 6 hours (Fig. 108 1a) and the effects of S re-supply on marker gene expression was evaluated. Seedlings grown under 109 low S for 5 days showed signs of starvation, as revealed by elevated expression of the S-deficiency 110 markers SULF ATE TRANSPORTERS (SULTR1;1 and SULTR4;2) in comparison with plants grown 111 for the same time on S -replete media (Fig. 1b). Their expression decreased after S re-integration, 112 suggesting the restoration of a normal S status. On the contrary, hypoxia marker genes were not 113 elevated during S depletion, but three out of the four measured markers were significantly induced 114 after S-reintegration (specifically, ADH1, PCO1, and PDC1) (Fig. 1b). The induction of these genes 115 in fully aerated seedlings indicates that low O2 signalling can be modulated by S assimilation. A meta-116 analysis of transcriptomic data from an existing study of S re-integration in A. thaliana (Bielecka et 117 al., 2015) showed, compatibly, the induction of several HRGs already 30 minutes after sulf ate 118 resupply, including HRA1, PDC1, PCO1 and WIP4 measured here (Suppl. Table 1). 119 To understand the role of the ERFVIIs in this phenomenon, we compared the response to 6- or 24-120 hour S re-supply in aerobic seedlings of a pentuple erfVII mutant (Abbas, Berckhan, Rooney, Gibbs, 121 Vicente Conde, et al., 2015) with the wild -type. S-deficiency markers had comparable expression 122 profiles in both genotypes, indicating no involvement of the ERFVIIs in their regulation in response 123 to variations in S provision (Fig. 1c). S-resupply for 6 hours was again associated with the induction 124 of the hypoxia markers in Col-0. The expression of PCO1, PDC1 and LDB41 reverted to basal levels 125 by 24 hours, whereas ADH1 was still induced. In contrast, the erfVII mutant showed no induction of 126 the hypoxia markers, except for a marginal response of PCO1 (Fig. 1c). These results indicate that 127 the transition from S deficiency to optimal S availability, under aerobic conditions, triggers HRG 128 expression in an ERFVII-dependent manner. To further support the role of ERFVII in this response, 129 we used a 35S:RAP2.121-28:Fluc reporter line of Arabidopsis (hereafter, 28RAPFluc), to monitor the 130 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint activity of the Cys N -degron pathway (Weits et al., 2014) . This translational fusion link s the 131 luminescence signal to the N -terminal modifications of RAP2.12 , while excluding any regulation 132 impinging on downstream domains of the protein such as phosphorylation (Kunkowska et al., 2023) 133 or SINAT-dependent proteolysis (Papdi et al., 2015) . S re-supply stimulated reporter activity ( Fig. 134 1d), indicating that, consistent with the observed HRG upregulation, the S burst promoted RAP2.121-135 28 stabilization under aerobic conditions through modulation of the Cys N -degron pathway. Re-136 integration of phosphate or iron, after growth on the corresponding deficient media, had no effects on 137 28RAPFluc output (Suppl. Fig. 1). 138 We next considered the possibility that the anaerobic response induced by S deficiency/re-supply was 139 triggered by regulation of metabolism and respiration (Martin & Maricle, 2015; Pietri et al., 2011) . 140 Specifically, we tested whether S re-supplementation might induce a rapid and intense increase in O2 141 consumption, thereby generating local hypoxia and inducing anaerobic responses. We thus measured 142 the O2 consumption rate in the same conditions as those in Fig. 1a . Following the treatments, 143 seedlings were immersed in water in sealed vials, where the decline in dissolved O 2 due to plant 144 metabolism could be monitored over time thanks to a fluorescent O₂-sensitive sensor (oxyspot). To 145 enable data comparison, we expressed O 2 consumption as OCI (Oxygen Consumption Indicator), 146 which represents the time required for oxygen saturation to decrease from 85% to 70% of the initial 147 value (full saturation under 21% O 2 atmosphere), normalized to fresh weight. OCI was unchanged 148 after 6 hours S re-integration, suggesting that flux through respiratory metabolism is not affected to 149 any extent by the treatment (Fig. 1e). A control treatment with 8 mM pyruvate, which has previously 150 been observed to induce a higher respiratory rate (Zabalza et al., 2009), resulted instead in increased 151 OCI (Suppl. Fig. 2a), however this was not associated with changes in HRG expression (Suppl. Fig. 152 2b-c). Together, the data confirm that the S-induced hypoxic response observed in Fig. 1a-b was not 153 caused by altered respiration rates. 154 155 156 157 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 158 159 Figure 1. Effects of S-resupply on the responses to hypoxia and O2 consumption. a) Schematic 160 representation of the S re-integration experiments: seedlings were sampled after 5 days of growth on 161 S deficiency (S-, 25 µM) or 6 h after S re-integration (S+, 750 µM) in the liquid media. S was provided 162 as MgSO4. b) Gene expression in Col-0 seedlings treated as in (a). Data (mean ± SD, n=5) are relative 163 to a control sample growth for 5 days under S optimal condition ( “Ctrl”, 750 µM S) . c) Gene 164 expression in Col-0 or erfVII mutant treated as in (a), after 6- or 24-hours re-integration. Data (mean 165 ± SD, n=5) are relative to one S- sample. Letters indicate statistically significant differences among 166 conditions and genotypes after two -way ANOV A and Tukey-Kramer post-hoc test (p<0.05). d) O2 167 consumption by Col-0 seedlings treated as in (a) and transferred to closed vials equipped with a 168 fluorescent O2-sensitive spot (see Materials and Methods). OCI, Oxygen Consumption Indicator (see 169 main text). e) Reporter activity in 35S:RAP2.121-28:Fluc (28RAPFluc) seedlings treated as in (a). 170 Firefly luciferase activity was measured before the transfer and after 30 minutes and 1 hour from the 171 recovery and normalised to total soluble proteins (Fluc/µg total proteins). Data are mean ± SD (n=4). 172 Letters in (b) and ( d) indicate statistically significant differences between different conditions after 173 one-way ANOV A and Tukey-Kramer post-hoc test (p<0.05). 174 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 175 Hypoxic responses generated by sulfate re-integration are linked to H2S 176 Aiming to verify that H₂S was involved in the normoxic induction of the hypoxic response, we used 177 NaHS as a sulfide donor. RAP28Fluc seedlings exposed to different concentrations of NaHS for 2 178 hours showed that H₂S-mediated RAP2.121-28 stabilization dose-dependent and non-linear: 50 µM 179 NaHS treatment had no significant effects, whereas concentrations ranging from 100 µM to 1 mM 180 caused reporter stabilization to a similar extent, as compared to untreated samples (Fig. 2a). We then 181 investigated the temporal dynamics of RAP28Fluc response to 100 µM NaHS treatment (the lowest 182 dose associated with an effect) . Luciferase activity was significantly increased after 1 and 2 hours, 183 returning to control levels after 4 hours (Fig. 2b). A similar transient response was observed upon 184 treatment with 500 µM NaHS (Suppl. Fig. 3a). Higher luciferase activity in the reporter lines after 2 185 hours NaHS treatment was compatible with RAP2.3-HA protein stabilization in a 35S:RAP2.33xHA 186 transgenic line (Gibbs et al., 2014) (Suppl. Fig. 3b). 187 To verify whether, alongside ERFVII stabilization, H₂S also triggered the induction of hypoxia 188 marker transcripts, we made use of the transcriptional reporter line HRPE:Nluc of Arabidopsis (Akter 189 et al., 2024), in which NANOLUCIFERASE expression is driven by a synthetic 5xHRPE:ADH15’-UTR 190 promoter (Akter et al., 2024) . Following treatment with 100 µM NaHS, the Nluc signal increased 191 significantly after 1 hour, remained stable up to 2 hours (Fig. 2c). Treating HRPE:Nluc seedlings with 192 500 µM NaHS resulted in a more pronounced and faster response (Suppl. Fig. 3c). 193 Signal reversal in the previous assays might have been partially concealed by the stability of the 194 luciferase proteins (Urquiza-García et al., 2019) leading us to turn to mRNA assessment to 195 complement the above observations . Profiling of endogenous gene expression changes at different 196 NaHS treatment levels showed the transient and dose -responsive signature of the hypoxia -like 197 response. A set of hypoxic genes composed by HB1, HUP7, WIP4, and LBD41 was rapidly induced 198 by NaHS supplementation (as early as 15 minutes after treatment), reversed to baseline levels within 199 30 minutes and underwent inhibition afterwards (Fig. 2d). This behaviour was consistently observed 200 in a treatment range from 150 to 500 µM, whereas any transcriptional response was barely visible at 201 100 µM ( Suppl. Fig. 4 ). PCO1 and PCO2, showed a biphasic response characterized by later re-202 induction, after rapid and more stable induction initial induction followed by repression (Fig. 2d). A 203 slower response was, instead, observed for HRE1 and HRE2 (Fig. 2d). Again, the same profiles were 204 observed across a range of NaHS treatments (Suppl. Fig. 5). Finally, HRA1, PDC1 and ADH1 were 205 unaffected by NaHS, except for slight induction of ADH1 at 500 µM (Suppl. Fig. 6). Altogether, 206 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint these results suggest that H₂S is associated with a rapid and reversible modulation of the activity of 207 proteins involved in the Cys N-degron pathway, causing distinct changes in downstream transcription. 208 209 210 211 Figure 2. Characterization of the hypoxia-like response to H2S supplementation. a) Luciferase 212 activity in 7-days old 35S:RAP2.121-28:Fluc (28RAPFluc) seedlings treated with different 213 concentrations of NaHS for 2 hours (n=5) . Different letters indicate statistically significant 214 differences according to one-way ANOV A followed by Tukey’s post hoc test (p<0.05). b) Luciferase 215 signal in 7-days old 28RAPFluc seedlings in a time-course experiment with 100 µM NaHS (n=6). c) 216 Nanoluciferase activity in of 7-days old HRPE:Nluc seedlings in a time-course experiment with 100 217 µM NaHS (n=5) . d) Hypoxic marker gene expression in 7 -days old Col -0 seedlings at different 218 timepoints after the supplementation of 150 µM NaHS. All data are shown as mean ± SD. Data in (a-219 c) were normalised with total proteins in the extracts. Asterisks indicate statistically significant 220 differences between treated and t0 samples (b-d), or treated and control samples (a) after Student’s t-221 test (* p<0.05; ** p<0.01; *** p<0.001). 222 223 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint PCO enzymatic activity is impaired by H2S 224 ERFVII stabilization and HRG triggering relies on the inactivation of the PCO-initiated Cys N-degron 225 pathway (Gibbs et al., 2011; Lavilla -Puerta et al., 2023; Licausi et al., 2011). Given the swift 226 dynamics of the hypoxic response observed during NaHS treatment, we hypothesized that H₂S may 227 directly affect PCO activity. To test this, we performed an in vitro assay to examine the role of H₂S in 228 modulating the activity of recombinant AtPCO4, as a representative member of the PCO family 229 (White et al., 2018) . The activity of AtPCO4 was monitored by examining Nt-Cys oxidation in a 230 RAP22-15 peptide in the presence of O2 (White et al., 2017). AtPCO4 was at first pre-incubated for 10 231 minutes with or without 1 mM NaHS or with 1 mM H₂O₂, a known inhibitor of PCO activity (Akter 232 et al., 2024) . The data showed that H2S inhibits AtPCO4 activity (Fig. 3a). Pre-incubation periods 233 ranging from 5 to 30 minutes were then tested, to assess the effect of NaHS exposure duration on 234 AtPCO4. A progressive reduction in RAP2 2-15 oxidation was observed with longer pre-incubations, 235 indicating that H2S inhibits AtPCO4 activity in a time-dependent manner (Fig. 3b). When the dose-236 response effect was examined, an “apparent’’ IC50 value of 1092 μM was determined, albeit the data 237 does not completely reflect a competitive inhibition model, potentially due to a persulfidation effect 238 (Fig. 3c). We next examined whether reducing agents could restore the enzymatic activity of H 2S-239 inhibited AtPCO4 through reversal of persulfidation. Recombinant AtPCO4 was initially treated with 240 5 mM NaHS for 30 minutes, followed by incubation either with the biological reducing agent 10 mM 241 glutathione (GSH) or 10 mM tris(2-carboxyethyl)phosphine (TCEP) for varying durations (1, 5, 15, 242 or 30 minutes ), before exposure to the RAP2.12 substrate. Both reducing agents partially restored 243 AtPCO4 activity in a time -dependent manner; after 30 minutes of treatment, TCEP recovered 244 approximately 20% of the enzyme’s maximal activity, whereas GSH restored about 14% (Fig. 3d-e). 245 To assess whether H₂S -mediated inhibition is a general feature of the Arabidopsis PCO family, we 246 extended our analysis to all five isoforms under two conditions: (i) pre-incubation with 10 mM GSH 247 followed by 5 mM NaHS, and (ii) NaHS treatment followed by GSH incubation. In both experimental 248 setups, partial restoration of activity was observed for all PCO isoforms, with the most substantial 249 recovery occurring in PCO1 and PCO5 ( Fig. 3f). Although the reducing potential of GSH can be 250 variable in vitro (due to trace amounts of oxidised GSH), these results support a redox-sensitive and 251 partially reversible mechanism of H₂S-mediated inhibition which likely involves one or more of the 252 Cys residues in PCOs. 253 254 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 255 256 Figure 3. Modulation of PCOs activity in vitro induced by H2S exposure. a) H2S inhibitory effects 257 on AtPCO4 enzymatic activity in vitro measured as oxidation of RAP2 2-15 peptide. The enzyme was 258 pre-incubated for 10 minutes with the indicated amount of NaHS, or with H 2O2, before the assays. 259 Non-treated AtPCO4 and 1 mM H2O2 were included to compare the effect of H2S on AtPCO4 activity 260 (n=3). b) Effect of different exposure times (5, 10, 20, or 30 min) to 5 mM NaHS on AtPCO4 activity 261 (n=3) under the same conditions of a). c) Dose-response curve for NaHS treatments of 2 µM AtPCO4 262 activity; reactions were performed exposing AtPCO4 for 30 minutes to NaHS ranging from 20 μM to 263 20 mM. Data are mean ± SD (n = 3). d) Recovery of AtPCO4 enzymatic activity induced by 264 glutathione (GSH) and e) tris(2-carboxyethyl)phosphine (TCEP). Data are mean ± SD (n=3-4). In (d) 265 and (e) letters indicate significant differences between treatments after one-way ANOV A followed by 266 Tukey HSD test (p<0.05). f) H2S-mediated inhibition of AtPCOs and their recovery of enzymatic 267 activity after 30 minutes pre -incubation with 5 mM NaHS, in presence or absence of 5 mM GSH 268 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint (n=3). Data are mean ± SD (n = 3). Letters indicate statistical differences evaluated using One-way 269 ANOV A followed by Tukey HSD test (p<0.05). 270 271 AtPCO4 is modulated by H2S through persulfidation 272 H₂S has been proposed to regulate protein function through persulfidation, a redox -sensitive post-273 translational modification (PTM) in which Cys thiol groups (-SH) are converted to persulfides (-SSH) 274 (Aroca et al., 2015; Mustafa et al., 2009) . Given the observed sensitivity of PCO enzymatic activity 275 to H₂S treatment (Fig. 3d), we investigated the potential interaction between H₂S and AtPCO4 Cys 276 residues. We examined the effect of H₂S on cysteine modification of AtPCO4 using the BioDiaAlk 277 probe, which selectively reacts with cysteine-derived sulfinic acid moieties (Akter et al., 2018) . 278 AtPCO4 was treated with 10 mM NaHS, either alone or in combination with 1 mM H₂O₂, followed 279 by BioDiaAlk labeling. In the presence of H₂O₂ alone, a marked increase in sulfinylation of AtPCO4 280 was observed, indicating the formation of sulfinic acid. However, when H₂S was co -applied with 281 H₂O₂, sulfinylation signals were substantially reduced or completely absent (Fig. 4a and Suppl. Fig. 282 7). These results imply that H₂S directly interacts with oxidized Cys residues on AtPCO4, potentially 283 preventing their further oxidation to sulfinic acid. 284 As all five AtPCOs were inhibited by H2S treatment, we looked for conserved Cys residues that may 285 act as potential PTM targets. Cys12, Cys79, Cys88, Cys172, and Cys190 (AtPCO4-1 numbering) (Dirr et 286 al., 2025) are conserved across all isoforms, and an additional cysteine residue (Cys 165) is present in 287 all isoforms except AtPCO1 ( Suppl. Fig. 8a ). To investigate which of those residues were 288 persulfidated by H 2S treatment, we performed LC -MS/MS–based proteomic analysis on NaHS -289 treated and untreated AtPCO4 samples, adapting a differential alkylation -based method for 290 persulfidation detection, following the principles described by Aroca et al. (2015) and Zivanovic et 291 al. (2019). In this approach, both free thiols and persulfidated cysteines were initially alkylated with 292 iodoacetamide (IAM). After tryptic digestion, the peptide mixture was split: one aliquot was subjected 293 to DTT reduction, selectively releasing persulfide -linked IAM adducts, while the other was left 294 untreated (Fig. 4b). While more than one Cys residue was found to be oxidised or modified by IAM, 295 under the conditions tested, Cys172 of AtPCO4 was the only residue identified by MS/MS to undergo 296 persulfidation in response to H₂S (Fig. 4c) . In the non -reduced ( –DTT) sample, persulfidated 297 cysteines retained an additional +89 Da mass shift (comprising +32 Da from S and +57 Da from 298 IAM). In contrast, the DTT -treated sample showed conversion of these sites into reduced thiols or 299 further oxidation products (e.g., sulfinic or sulfonic acids), lacking the +89 Da signature. The 300 identification of the modified cysteine was further supported by LC-MS/MS fragmentation analysis, 301 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint in which the b- and y-ion series confirmed the site of modification (Fig. 4d and Suppl. Fig. 8b). An 302 error map showed minimal deviation in fragment ion masses, confirming the confidence of site 303 assignment (Suppl. Fig. 8c). 304 305 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 306 307 308 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Figure 4. Detection of persulfidation in AtPCO4 by differential alkylation -mass spectrometry 309 strategy. a) Detection of AtPCO4 sulfinylation in vitro. Purified AtPCO4 was treated in combination 310 with 10 mM H 2S and/or 1 mM H 2O2 and the amount of AtPCO4 sulfinylated detected using the 311 BioDiaAlk probe specific for sulfinic acid (Akter et al., 2018). The full-size images of the membranes 312 can be found in Suppl. Fig. 8. b) Schematic representation of the differential IAM labeling workflow 313 used for persulfidation detection. Proteins were first treated with NaHS (H₂S) to induce 314 persulfidation, either alone or in combination with H₂O₂. Free thiols and persulfidated cysteines were 315 then alkylated with iodoacetamide (IAM), blocking accessible –SH and –SSH groups. Following 316 tryptic digestion, peptides were incubated with or without dithiothreitol (DTT) to reduce persulfides 317 or other reversible oxidative modifications (e.g., disulfides). By comparing the thiol content in DTT-318 treated and untreated samples, putatively persulfidated cysteine residues were identified. c) Summary 319 of potential cysteine modifications and corresponding mass shifts in H₂O₂ + NaHS -treated AtPCO4 320 (top), and AtPCO4 peptide sequence coverage after MS/MS analysis. Modifications observed at 321 specific cysteine residues following DTT or no -DTT treatment are annotated. d) LC-MS/MS 322 spectrum showing the fragmentation pattern of the modified peptide, including annotated b - and y-323 ions of interest. 324 325 These results led us to investigate whether AtPCO4 is also subject to persulfidation in vivo . In 326 particular, we asked whether under hypoxia dynamic changes in H₂S could modulate the hypoxic 327 response by inhibiting PCO activity through persulfidation. We introduced a pPCO4:PCO4:GFP 328 construct into the 4pco mutant background, to enable AtPCO4 immunodetection. Persulfidation levels 329 were quantified as the ratio between persulfidated AtPCO4 and total AtPCO4 input using the 330 dimedone switch assay (Aroca et al., 2022). Total protein extracts were split into two fractions, one 331 of which was labelled for persulfidated residues, using the fluorescent probe NBF-Cl/DCP-Bio1, and 332 subsequently purified with streptavidin-bound beads. Both fractions were subsequently probed with 333 an anti-GFP antibody to detect persulfidated AtPCO4 from the total batch of persulfidated proteins in 334 the extract or total AtPCO4 input. S even-day-old seedlings were exposed to hypoxia. Increased 335 AtPCO4 persulfidation was observed both after 1 and 4 hours of H₂S treatmen t (Suppl. Fig. 9). 336 Interestingly, hypoxia also induced a strong response, particularly at the 4 hour s timepoint (Suppl. 337 Fig. 9, Fig. 5a and Suppl. Fig. 10a). In parallel, AtPCO4 sulfenylation levels in hypoxic seedlings 338 were assessed using the DCP-Bio1 probe, which selectively reacts with sulfenic groups (-SOH), and 339 were quantified relative to the corresponding AtPCO4 input. Sulfenylation progressively decreased 340 with prolonged hypoxia exposure ( Fig. 5b and Suppl. Fig. 1 0b), suggesting a redox -based shift in 341 AtPCO4 from a sulfenylated state under normoxia to a persulfidated state under hypoxia. 342 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 343 Figure 5. H 2S-mediated post-translational modification o f AtPCO4 Cys residues. Immunoblot 344 and relative quantification of a) persulfidated and b) sulfenylated PCO4 in 7 -day-old 345 pPCO4:PCO4:GFP/4pco seedlings exposed to 1% O₂ for 1 or 4 hours. For a) and b) protein extracts 346 were split into two fractions: one was subjected to specific labeling: (persulfidation: NBF-Cl blocking 347 followed by DCP-Bio1; sulfenylation: DCP-Bio1 labelling of –SOH groups), and the other was used 348 directly as input control. Both fractions were analyzed by immunoblotting with anti-GFP antibodies. 349 Ponceau staining was used as loading control (for further details, see Material and Methods) . Data 350 are presented as the ratio between modified and input PCO4 (mean ± SD; n = 6). Different letters 351 indicate statistically significant differences (one -way ANOV A followed by Tukey’s post hoc test). 352 Uncropped blots are shown in Suppl. Fig. 10. 353 354 H₂S modulates H2O2 levels in a dose-dependent and compartment-specific manner 355 These new insights into S-modification dynamics of PCO4 highlight a potential redox-based shift in 356 PTMs that may influence PCO function under varying oxygen and H₂S conditions. This prompted us 357 to investigate whether H₂S affects H 2O2 levels under normoxic and hypoxic conditions. NaHS 358 supplementation had very mild effects on the induction of transcripts associated with ROS 359 homeostasis, such as MSD1 and CAT2 (Suppl. Fig. 11), while it stimulated a major response in the 360 transcription factor ZAT12, known to be regulated by both oxidative stress and low oxygen 361 (Pucciariello et al., 2012). We used the roGFP2-based biosensors roGFP2-Orp1 and Grx1-roGFP2 to 362 monitor intracellular redox dynamics in vivo; these sensors, targeted to the cytosol and mitochondria, 363 allow real-time monitoring of H₂O₂ and glutathione redox potential, respectively (Nietzel et al., 2019). 364 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Experiments were conducted in Col -0 and erfVII Arabidopsis seedlings expressing the biosensors 365 under normoxia, hypoxia and reoxygenation , in combination with different concentrations of 366 exogenous H₂S (NaHS). 367 Under normoxic conditions (21% O₂), H₂S treatment caused a transient increase in cytosolic roGFP2-368 Orp1 oxidation in a dose -dependent manner, indicating a fast increase in H 2O2 levels (Fig. 6a). At 369 lower H 2S concentrations, sensor oxidation quickly returned to baseline within 30 minutes, 370 suggesting that endogenous antioxidant systems effectively counteract the H₂S -induced oxidative 371 challenge. However, at higher H₂S concentrations (5 mM), a sustained increase in biosensor oxidation 372 was observed. This effect may result from H₂S-induced stimulation of H2O2 production, or disruption 373 of antioxidant pathways (or both). Also, a direct interaction of H 2S (and derived polysulfides ) 374 (Greiner et al., 2013) with the Cys moieties of the sensor itself cannot be ruled out (Li & Lancaster, 375 2013). 376 Under severe hypoxia (0.1% O₂ for 6 h), a similar early H 2O2 increase was detected following H₂S 377 application. However, H₂S did not significantly affect the gradual sensor oxidation associated with 378 hypoxic stress, nor did it modulate the prominent oxidative burst observed during reoxygenation (6 h 379 at 21% O₂) ( Fig. 6b -c) (Jethva et al., 2023) . These effects were consistently observed with both 380 sensors, roGFP2-Orp1 and Grx1-roGFP2-Grx. 381 Interestingly, H₂S treatment led to less pronounced oxidation of sensors expressed in the 382 mitochondrial matrix under hypoxia, suggesting that H₂S may attenuate mitochondrial H₂O₂ 383 production or enhance specific antioxidant responses, or both ( Fig. 6d, e). This is consistent with 384 prior studies describing H₂S as a mitochondrial electron donor and as an inhibitor of cytochrome c 385 oxidase, both of which could influence mitochondrial ROS generation (Cooper & Brown, 2008; Fu 386 et al., 2012.; Pedroletti et al., 2023; Szabo, 2010) 387 Despite this, the reoxygenation (6 h at 21% O₂) -induced burst of oxidation in the matrix remained 388 unaffected by H₂S. To determine whether the ERFVII transcription factors influence this H₂S -389 mediated redox response in mitochondria, we compared the redox sensor dynamics in Col -0 and 390 erfVII mutant backgrounds using mitochondrial probes. Even though the progressive oxidation during 391 hypoxia was strongly increased in the erfVII mutant as compared to the wild-type, the responses to 392 H₂S treatment were similar between genotypes (Fig. 6f), suggesting that this apparent ability of H₂S 393 to counteract sensor oxidation in the matrix is independent of ERFVII signalling pathways. 394 395 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 396 Figure 6. H₂S alters cytosolic and mitochondrial redox states in Arabidopsis under hypoxic 397 conditions. a) Time-course measurements of cytosolic redox changes in Arabidopsis seedlings 398 expressing the cytosolic roGFP2 -Orp1 sensor treated with varying concentrations of H₂S (NaHS 399 0.05–5 mM), along with DTT (20 mM, reducing control), H₂O₂ (50 mM, oxidizing control), or mock 400 solution. Data are shown as the log₁₀ of the 400/482 nm excitation ratio, indicative of sensor oxidation. 401 b) Redox dynamics in cytosolic roGFP2 -Orp1-expressing seedlings under mild H₂S stress (0.125 –402 0.5 mM) in hypoxic conditions (0.1% O₂). Right y -axis shows oxygen concentration (% O₂) over 403 time, indicating the hypoxia treatment and subsequent reoxygenation phase. c) Cytosolic redox 404 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint changes in seedlings expressing Grx1-roGFP2 exposed to 0.125–0.5 mM H₂S under hypoxia. Sensor 405 oxidation is represented as log ₁₀ (400/482 nm). d) Mitochondrial redox changes in seedlings 406 expressing mitochondrial-targeted roGFP2-Orp1 (mt -roGFP2-Orp1) under hypoxia and 0.125 –0.5 407 mM H₂S treatment. Oxygen levels (% O₂) are shown in black right y-axis. e) Quantitative analysis of 408 mitochondrial roGFP2 -Orp1 oxidation at distinct time points during the hypoxia time course. 409 Seedlings were pre -incubated in normoxia (21% O₂), exposed to H₂S and hypoxia (0.1% O₂), and 410 then reoxygenated. Bars represent means ± SD. Different letters indicate statistically significant 411 differences (ANOVA with Tukey’s post hoc test, p < 0.05). f) Comparison of mitochondrial matrix 412 redox responses in wild-type (Col-0) (data from panel d) and erfVII mutant seedlings expressing mt-413 roGFP2-Orp1 under hypoxia with or without 0.25 mM H₂S treatment. Oxygen levels (% O₂) are 414 plotted in black on right y-axis. 415 416 Collectively, these results indicate that H₂S plays a context - and concentration -dependent role in 417 cellular redox regulation. While H₂S can transiently increase H 2O2 under normoxic and hypoxic 418 conditions, it may also serve a protective role in mitochondria by dampening H₂O₂ production under 419 hypoxia, a phenomenon that should be further investigated . These findings provide mechanistic 420 support for a model in which H₂S modulates redox status through both pro -oxidant and antioxidant 421 mechanisms, with implications for PCO4 function and hypoxia signalling. 422 423 Decreased H₂S levels are associated with a weaker hypoxia responses and lower tolerance to low-424 oxygen conditions 425 The increase in PCO4 persulfidation in seedlings exposed to hypoxia ( Fig. 5a) hints at a possible 426 impact of H₂S on the establishment of the hypoxic response to low oxygen deprivation. Free H₂S 427 quantification in Col-0 seedlings exposed to hypoxia showed a decrease after both 1 hour and 4 hours 428 treatment (Fig. 7a). This may be due to its utilization for the formation of persulfide groups on Cys 429 residues of proteins, which may help protect them from irreversible oxidation caused by ROS 430 accumulation under hypoxic condition (Pucciariello et al., 2012) . We adopted a genetic strategy t o 431 verify whether the physiological H₂S levels present in seedlings under hypoxia may contribute to 432 determine the dynamics of hypoxic responses. We exposed the des1 mutant, characterized by a 30% 433 reduction in free H2S (Álvarez et al., 2012), to short-term hypoxia (1% O2), followed by analysis of 434 HRG expression. In the mutant, the early induction of most transcripts tested was significantly lower 435 than in Col -0 (Fig. 7b). All markers but PCO1 exhibited lower expression in des1 after 1 hour of 436 hypoxia exposure, whereas HRA1 decrease was already visible after 30 minutes (Fig. 7b). We also 437 evaluated impact of des1 under submergence, a condition that favours the entrapment of gaseous 438 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint signals like H₂S. Seedlings were completely covered with water to simulate a dark flooding condition 439 in the well. Also in this case, a less pronounced hypoxic response was observed in the des1 mutant 440 compared to Col-0 (Fig. 7c). No differences were seen after 30 minutes, but all genes except ADH1 441 showed lower expression in des1 1 hour into the treatment (Fig. 7c). 442 443 Figure 7. H2S involvement under low oxygen. a) Relative quantification of free H2S on fresh weight 444 (FW) in 7-days old Col-0 seedlings exposed to dark hypoxia (1% O 2 v/v) for 1 or 4 hours. Data are 445 mean ± SD (n=4). Different letters denote significant differences (one -way ANOV A followed by 446 Tukey’s post hoc test). b) HRG expression in 7-days old Col-0 and des1 seedlings exposed to dark 447 hypoxia (1% O2 v/v) for 30 minutes or 1 hour. In control condition seedlings were treated in the dark 448 at atmospheric oxygen concentration (21% O 2 v/v). Data are mean ± SD (n=5). c) HRG expression 449 after 30 minutes or 1 hour submergence. Controls samples (t0) were treated in the dark under 21% 450 O2. d) Luciferase activity in three independent HRPE-based reporter lines in des1 background. 451 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint HRPE:Nluc seedlings in Col -0 or des1 mutant background were grown for seven days on vertical 452 plates and treated with dark hypoxia (1% O2 v/v) for 30 minutes or 2 hours. e) Extended response to 453 hypoxia in 7-days old HRPE:Nluc #4 seedlings in des1 or Col-0 background. Nanoluciferase signals 454 were normalised to the total soluble protein (Nluc/µg total protein); data are mean ± SD (n=4). 455 Asterisk indicates statistically significant differences between genotypes after Student’s t -test 456 (*p<0.05; **p<0.01; *** p<0.001). 457 458 We further characterized the dynamics of the hypoxic responses in the mutant by introducing the 459 HRPE:Nluc reporter in the des1 background. Three independent lines showed a comparable signal 460 when exposed to low oxygen, with a lower signal after 2 hours of hypoxia compared to the Col -0 461

( Fig. 7d ). We selected the HRPE:Nluc #4 to monitor the response under prolonged 462 hypoxia. The signal showed a lower intensity after 2, 4, and 8 hours of exposure to 1% of O2 than in 463 the wild-type background (Fig. 7e). 464 Given the accumulating evidence suggesting a role for H₂S in mediating the hypoxic response during 465 oxygen deprivation, we test ed the consequences of H 2S biosynthesis impairment on submergence 466 tolerance. The des1 mutant showed lower stress tolerance as compared to the wild-type after 60 hours 467 dark submergence ( Fig. 8a, Suppl. Fig. 12) . No significant differences were instead observed 468 following 72 hours of submergence, which represented a lethal treatment for both genotypes. The 469 survival rate after 1 week of recovery from 60 hours of submergence was 93% for Col -0, whereas it 470 was only 26% for the des1 mutant (Fig. 8b). Additionally, under these conditions, the mutant showed 471 a significant decrease in Plant Leaf Area (PLA) compared to the wild-type, a difference that was not 472 observed either in the controls or after 72 hours of submergence ( Fig. 8c). Together, these findings 473 suggest a role for H 2S in modulating the hypoxic response, emphasizing how appropriate levels of 474 H2S contribute to the initiation and maintenance of an adequate response during oxygen deprivation. 475 476 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 477 478 Figure 8. Effect of submergence on Col-0 and des1 plants. a) Images of representative Col-0 and 479 des1 plants after 1 week of recovery from dark submergence. The control plants were treated under 480 dark conditions for 60h. b) Survival rate of Col -0 and des1 after 1 week of recovery from 481 submergence. Categories: Alive = all or most old leaves alive and new leaves produced; stunted = 482 new leaves produced but most or all old leaves dead; dead = all or most old leaves dead, no or very 483 few new leaves (n=4 for control, n=15 for submergence). c) Plants Leaf Area (PLA) in the Col-0 and 484 des1 after 1 week from the submergence. Data are mean ± SD (n=4 for the control, n = 15 for the 485 treated plants). The asterisks indicate a statistical difference after the student’s t -test (*0.01≤p<0.05; 486 **0.001≤p<0.01; ***p<0.001). 487 488 489 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint

490 Once considered as a toxic gas, H₂S is now recognized for its regulatory role in several physiological 491 processes and responses to both biotic and abiotic stresses, such as salinity, drought and waterlogging 492 (Chen et al., 2015; Jurado -Flores et al., 2023; Y . Li et al., 2021) . Although several studies have 493 reported that exogenous H₂S can mimic aspects of the hypoxic response and improve survival rate of 494 A. thaliana plants exposed to submergence (Xiao et al., 2020; Li et al., 2021; Zhou et al., 2021), its 495 role as an endogenous modulator of hypoxic signalling remains poorly understood. Here, we describe 496 PCO enzyme persulfidation as a molecular mechanism connecting H₂S with the modulation of the 497 hypoxic responses in A. thaliana. 498 We first investigated whether alterations in sulfate assimilation, a primary route of endogenous H₂S 499 production (Aroca et al., 2018) , could influence the hypoxic response. Bielecka et al. (2015) have 500 previously reported HRG up-regulation in A. thaliana seedlings grown under S-deficient conditions 501 early after S re -supplementation (Suppl. Table 1 ). Notably, no significant changes in metabolite 502 levels were detected within the first hour of nutritional shift, suggesting that signalling events, rather 503 than metabolic regulation, should have caused HRG induction. Our results indicate that restoration 504 of S following S starvation caused induction of the hypoxic genes, under aerobic conditions, in an 505 ERFVII-dependent manner ( Fig. 1b -d). Having ruled out the occurrence of local hypoxia due to 506 increased O2 consumption (Fig. 1e, Suppl. Fig. 1 ), we put forward that the observed hypoxia-like 507 responses might be caused by a short -term direct effect of S re -integration on the O2-sensing 508 machinery. 509 The involvement of H₂S in this phenomenon was confirmed through the application of sodium 510 hydrosulfide (NaHS), as H₂S source. A minimum concentration of 100 µM NaHS , a non-toxic dose 511 commonly applied for short term treatments in Arabidopsis (Zhou et al., 2025) was sufficient to 512 stabilize RAP2.12 and promote HRPE-driven reporter expression under normoxic conditions (Fig. 513 2a-c). The induction of hypoxic genes was weak at 100 µM , but became progressively more 514 pronounced at higher concentrations , highlighting the transient and dose-dependent effect of the 515 exogenous treatment (Fig. 2d ). High concentrations of NaHS (e.g. 500 µM) also triggered the 516 expression of oxidative stress marker genes (Suppl. Fig. 11), suggesting a ROS burst. Although H₂S 517 is well established as a ROS scavenger (García-Calderón et al., 2023; Jaiswal et al., 2024; Wang et 518 al., 2022), some studies indicate that it can also trigger indirect ROS production when supplied at 519 high doses (Shen et al., 2020; Zhang et al., 2017) . Consistently, NaHS supplementation resulted in 520 detectable cellular redox changes, as evidenced by the roGFP2-derived biosensors (Fig. 6), under 521 both hypoxic and normoxic conditions. Recently, the inhibitory effect of H₂O₂ on PCO activity has 522 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint been demonstrated (Akter et al., 2024) . Therefore, we cannot rule out that the indirect mediation of 523 ROS might concur to the promotion of hypoxia-like responses by H₂S. However, HRG expression 524 patterns were maintained all over the range of H₂S concentration (Suppl. Fig. 4-6), both at low doses, 525 that are likely to be associated with ROS scavenging, and at higher doses, that might trigger oxidative 526 stress. This suggests that the direct effect of H₂S might be predominant over any indirect effects 527 mediated by ROS. 528 Persulfidation of susceptible Cys residues are considered the predominant mechanism mediating H2S-529 dependent signalling, through changes in target protein biochemical activity, interactions or structure 530 (Shen et al., 2020; Wang et al., 2023) . Rapid induction of HRG expression can follow from the 531 inhibition of PCO enzymes (Lavilla-Puerta et al., 2023) , within minutes of hypoxia (Lavilla-Puerta 532 et al., 2025) . We thus hypothesized that f ast stimulation of hypoxic responses by H₂S (Fig. 2c-d) 533 might be linked to the persulfidation of PCOs. In a previous proteomic study, Aroca et al., (2017) 534 have identified PCO3 as a target of both S-nitrosylation and persulfidation, highlighting the potential 535 for redox regulation within this enzyme family. Compatibly, we observed persulfidation on a 536 recombinant AtPCO4 in vitro after H₂S incubation ( Fig. 4 ), leading to the inhibition of its Cys 537 oxidation activity. Although the concentrations of AtPCO4 and H₂S are likely to be much lower in 538 vivo than in our experiments with recombinant enzyme, we nevertheless observed an inhibitory effect 539 that was proportional to H₂S dose and exposure time (Fig. 3a-c). The inhibitory effect was present on 540 all Arabidopsis PCO isoforms and partially reversible by reducing agents, supporting the model of a 541 redox-dependent inhibition of PCO enzymes through persulfidation (and likely oxidised derivatives 542 thereof) (Fig. 3d-f). Through LC -MS/MS, we identified Cys172 as a AtPCO4 persulfidation site. 543 While we cannot rule out that other Cys residues may also be persulfidated, this conserved residue 544 has been shown to play a role in enzyme interaction with O₂ and catalysis (Dirr et al., 2025) . The 545 AtPCO4-2 C173A variant (corresponding to AtPCO4-1 Cys172) exhibited lower catalytic efficiency 546 of dioxygenation in vitro than the wild-type version, without affecting substrate binding, and, when 547 expressed in Arabidopsis, conferred higher tolerance to submergence (Dirr et al., 2025). Our analysis 548 revealed that Cys172 can exist in several oxidized forms ( Fig. 4c), in line with its proximity to the 549 active site and confirming its interaction with O 2. The fact that Cys172 can undergo irreversible 550 oxidation ( Fig. 4 ), but is also susceptible to persulfidation, suggests a protective role of H₂S in 551 maintaining AtPCOs functionality under oxidative conditions (Vignane & Filipovic, 2023). 552 We also observed in vivo that persulfidation levels of AtPCO4 increased in seedlings exposed to NaHS 553 and are further elevated under hypoxic conditions (Suppl. Fig. 9), suggesting a direct involvement of 554 H₂S as physiological messenger in the modulation of hypoxic responses. Enhanced persulfidation 555 was accompanied by a decrease in sulfenylation levels ( Fig. 5a-b), supporting the idea of a finely 556 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint tuned redox regulation of AtPCOs according to oxygen availability. Although counterintuitive, ROS 557 accumulation is among the common consequences of hypoxia and has been detected as early as 30 558 minutes into oxygen deprivation (Gonzali et al., 2015). The ROS-scavenging properties of H₂S, which 559 under hypoxia are supposed to help mitigate oxidative stress caused by ROS excessive production, 560 seem to be reflected by the observed decrease of H₂S levels under hypoxia ( Fig. 7a). Rapidly 561 synthesized H₂S might thus have been consumed to increase the persulfidation levels in hypoxia (Fig. 562 5a), resulting in lower steady-state levels of H₂S after 1- or 4- hours hypoxia. Previous studies have 563 shown H₂S accumulation under low -oxygen conditions, particularly during submergence or anoxia 564 (Yang et al., 2021; Zhou et al., 2021). However, these conditions may induce effects not taking place 565 in our experimental setting of gaseous hypoxia ( Fig. 5a), such as the local entrapment of gaseous 566 molecules due to water coverage. A comparative analysis of H₂S dynamics under different oxygen -567 deficient conditions could help clarify the discrepancy observed in current data. 568 Use of the des1 mutant (Álvarez et al., 2010) impaired in Cys desulfhydration capacity in the cytosol, 569 underscored the importance of H₂S as a physiological messenger during the establishment of the 570 anaerobic response. Lowering of the cytosolic H₂S levels impact ed on HRG induction under low 571 oxygen (Fig. 7b-e), compatible with the (partial) preservation of PCO activity in comparison with 572 wild-type seedlings. Furthermore, des1 impaired tolerance to submergence (Fig. 8) suggests a broader 573 role for H₂S in low -oxygen stress, beyond the inhibition of PCOs to enable full activation of 574 transcription following hypoxia . H2S impact under low oxygen might extend to safeguard key 575 proteins from irreversible oxidation occurring in the post -submergence phase of reoxygenation , 576 thereby helping plant recovery after stress . Quantitative analysis of the persulfidated proteome 577 (persulfidome) under submergence would provide valuable insights into this hypothesis. 578 Here, we provided the first direct evidence for an endogenous role for H₂S to set up the hypoxic 579 response in plants and highlights its contribution to the complex biochemical and molecular networks 580 for survival under submergence stress. 581 582

583 Plant material 584 The Columbia-0 ecotype (Col-0) was used as the wild-type background in all experiments. The des1 585 mutant (SALK_103855) was obtained from NASC (National Arabidopsis Stock Center). The 586 quintuple erf-VII mutant was previously described in Abbas et al. (2015). The transgenic lines 587 35S:RAP2.121-28:Fluc (Licausi et al., 2011), 35S:RAP2.33xHA (Gibbs et al., 2014) and the HRPE:Nluc 588 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint (Akter et al., 2024) were previously described. The HRPE:Nluc/des1 line was obtained by crossing 589 des1 with the HRPE:Nluc line. The homozygous seeds were selected on ½ MS supplemented with 590 10 g l-1 of sucrose, 9 g l-1 of agar and glufosinate ammonium 1mM. F3 seeds were collected and used 591 for the experiments. 592 Wild-type Arabidopsis thaliana lines stably expressing the cytosolic and nuclear -localized roGFP2-593 Orp1 biosensor were previously described (Nietzel et al., 2019). To generate sensor lines in the erfVII 594 background, Agrobacterium tumefaciens-mediated floral dip transformation was performed using a 595 pH2GW7:mt-roGFP2-Orp1 construct. Transgenic lines were selected on hygromycin B -containing 596 media and screened for sensor fluorescence. 597 For the generation of the PCO4:GFP construct, a 1519 -nt fragment was synthesized by Integrated 598 DNA Technologies (IDT), including the attB1 and attB2 sequences upstream and downstream of the 599 PCO4:GFP sequence, respectively. The coding sequence corresponding to the PCO4 -1 version was 600 adopted. The fragment was first cloned into pDONR207 and subsequently into pH7pPCO4GW 601 (White et al., 2020) via BP and LR reactions, respectively, following the manufacturer's instruction. 602 For sequence details see Supplemental Table 3. Agrobacterium-mediated transformation using the 603 floral dip method (Clough & Bent, 1998) was employed to generate a stable pPCO4:PCO4:GFP 604 transgenic line in the 4pco background (Masson et al., 2019). T0 seeds were selected for hygromycin 605 resistance on solid MS medium and subsequently transplanted into soil. Integration of the construct 606 was confirmed by PCR using GoTaq® DNA polymerase (Promega). Experiments were conducted 607 using seeds from the T3 generation. 608 Growth conditions 609 For axenic experiments, seeds were sterilized with ethanol 70% (v/v) for 1 minute, incubated in 10% 610 (v/v) sodium hypochlorite (NaClO) for 10 minutes, followed by 5 times washes with sterile water. 611 Seeds were incubated in the dark for 48 h at 4°C before the growing. For the nutrient reintegration 612 experiments, seedlings were grown aerobically in 6 -well plates for 5 days in 2.5 mL of deficiency 613 liquid medium and then transferred into 2.5 mL of complete liquid media. The media compositions 614 are described in Supplemental Table 2. Experiments in Figure 2 and Figure 4b, 4c were conducted 615 on 7-days old seedlings grown for 6 days on ½ MS (Duchefa) supplemented with 10 g l -1 di sucrose 616 and 9 g l-1 agar (Duchefa) and transferred for one day in 6-well plates with 2.5 mL of ½ MS (Duchefa) 617 with the addition of 10 g L -1 of sucrose. Soil experiments were carried out on soil:vermiculite 3:1 618 ratio mixture. Plants were grown under atmospheric conditions with 23°C day/19°C night 619 temperature and a 12 h light period with a light intensity of 120 μmol photons m-2s-1. 620 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Low oxygen treatments 621 For hypoxia treatments , seedlings were grown on vertical plates for 7 days and exposed with 1% 622 O2/N2 (v/v) atmosphere recreated in a sealed glovebox (Coy) , or normoxic atmosphere, for the 623 indicated duration, in the dark. In partial submergence experiments, the seedlings were grown for 6 624 days on vertical plates and transferred for one day into 6 -well plates. Subsequently, the media was 625 removed, and the seedlings were covered with 5 mL MilliQ water and exposed to 1% O 2/N2 (v/v) in 626 darkness for the specified duration. Seedlings in control conditions were exposed to dark and 627 atmospheric oxygen without the removal of the liquid media. All the treatments were performed using 628 the Gloveless Anaerobic chamber (COY Laboratory Products). The submergence was performed in 629 21-days old plants grown in soil. After 60h or 72h of dark submergence the photoperiodic period was 630 restored (23°C day/19°C night temperature and a 12h light period with a light intensity of 120 μmol 631 photons m-2 s-1). Control plants were exposed to dark conditions for 60h. Survival rate was determined 632 using the following parameters: “Alive”: all or most old leaves alive, new leaves produced; “Stunted”: 633 new leaves produced, but most or all old leaves dead; “Dead”: all or most old leaves dead, absence 634 of new leaves. Before the submergence and after 1 week of recovery the Plant Leaf Surfaces (PLA) 635 was calculated using a LabScanalyzer digital phenotyping machine (LemnaTec GmbH, Aachen, 636 Germany) equipped with a Manta G-1236 camera and a Kowa LM12XC lens. 637 Chemical treatments 638 H2S treatments were induced by supplementation of NaHS (Sigma -Aldrich) dissolved in MilliQ 639 water, freshly prepared. Pyruvate treatments were performed by supplementation of 8 mM pyruvate 640 (Sigma-Aldrich) dissolved in MilliQ water. 641 O2 consumption measurements 642 O2 consumption over time was assessed in seedlings placed inside cuvettes (Pyroscience) containing 643 2 mL of sterile water, kept stirring with a magnet. The variation of O2 concentration was measured 644 with a phosphorescent sensor (Fiber-Optic Oxygen Meter 2 channels FSO2-C2, Pyroscience), which 645 detected the data through an O2 sensitive spot attached on the internal surface of the cuvette. The 646 collected data were recorded and analyzed using the software PyroDataInspector. To compare O2 647 consumption between different samples, the parameter Oxygen Consumption Indicator (OCI) was 648 introduced, defined as the time required for dissolved O2 to decrease from 85% to 70% saturation (in 649 water initially equilibrated with a 21% O2 atmosphere), normalized with respect to the fresh weight. 650 Biosensing of intracellular H2O2 dynamics in living Arabidopsis tissue 651 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Mt-roGFP-Orp1 sensor lines in the erfVII background were used in both T 2 (segregating for 652 biosensor) and T 3 (homozygous) generations, depending on the experiment: T 2 lines for leaf disc 653 assays and T 3 for seedling experiments. Wild -type sensor lines were homozygous for the insertion. 654 To account for positional effects, two independent sensor lines per genotype were analyzed, and 655 replicates were pooled for final data analysis. Leaf discs (from 5 -week-old plants) and 7 -day-old 656 seedlings were submerged in 200 µL standard assay buffer (10 mM MES, pH 5.8 adjusted with KOH, 657 10 mM MgCl₂, 10 mM CaCl₂, 5 mM KCl) in a 96 -well plate. Each well received a single leaf disc 658 (abaxial side up) or 5 –6 seedlings. Ratiometric fluorescence measurements were performed using a 659 ClarioStar Plus microplate reader (BMG Labtech) in top-optic mode with 30 excitation flashes and 3 660 mm averaging diameter (orbital for discs, spiral for seedlings). Excitation/emission settings were: 661 Ex1: 400 ± 10 nm, Ex2: 482 ± 16 nm; Dichroic mirror: LP504; Em: 520 ± 10 nm. Data were collected 662 every 243 seconds. Oxygen concentration was controlled using an Atmospheric Control Unit (BMG 663 Labtech) by targeted N₂ influx. Separate experimental repetitions were used for each O₂ treatment. 664

correction was performed using wildtype Col-0 samples and background was subtracted 665 before calculating 400/482 nm ratios. Ratio data were log₁₀-transformed for statistical normalization. 666 RNA extraction and gene expression analysis 667 RNA extraction was performed as previously described (Kosmacz et al., 2015). The RNA quality and 668 quantity were checked by gel electrophoresis on 1% (w/v) agarose and by spectrophotometric 669 analysis. Maxima First -Strand complementary DNA (cDNA) Synthesis Kit (Thermo Fischer 670 Scientific) was used for the reverse transcription, according to the manufacturer’s recommendation. 671 ABI Prism 7300 sequence detection system (Applied Biosystems) was used for the RT -qPCR using 672 12.5 ng cDNA with PowerUp SYBR® Green Master Mix (Thermo Fisher Scientific). The relative 673 gene expression was calculated using UBQ10 (AT4G05320) as housekeeping gene through the ΔΔCt 674

(Livak & Schmittgen, 2001). Primer sequences are specified in Supplemental Table 3. 675 Luciferase assays 676 Seedlings were ground in liquid nitrogen and the total protein extracted in passive lysis buffer 677 (Promega). ONE -Glo Luciferase Assay kit (Promega) was used to quantify the Firefly luciferase 678 activity and the NanoLuc luciferase enzyme was measured with the Nano -Glo® Luciferase Assay 679 System (Promega). Luciferase signals were normalized on the total protein concentration quantified 680 using Bradford assay (Bradford, 1976). 681 Persulfidation and sulfenylation detection 682 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Protein extracts were obtained from plant material in 1x phosphate-buffered saline (PBS) pH 7.4, 683 supplemented with 1 mM EDTA, 2% w/v SDS, and 1x protease inhibitor cocktail (Thermo). The 684 total amount of protein was determined using the DC Protein Assay (Bio -Rad). The protein extract 685 was divided into two batches, one of which was used for immunoblot analysis (input extract) and the 686 second was subjected to the dimedone switch method followed by immunoblot analysis (persulfidated 687 and sulfenylated extract). The persulfidation and sulfenylation of PCO was detected after treatment 688 following the dimedone switch method previously described by Aroca et al., (2022) . Briefly, for 689 persulfidation analysis, NBF-Cl (Merck) was employed to derivatize all cysteine residues and amino 690 groups, and then a labelling step with DCP-Bio1 (Merck) was performed for tagging the persulfidated 691 adducts. The labelled proteins were purified by incubation with Sera -Mag Magnetic Streptavidin 692 beads (Cytiva) overnight at 4 °C and subsequent elution from the beads with 2.25 M NH4OH and 2% 693 SDS at 95ºC for 10 minutes. Eluted proteins were then precipitated with acetone/TCA and 694 resuspended in 1x PBS and 2% SDS. Proteins of the input and the persulfidated extract were separated 695 in an SDS–PAGE 10% (w/v) polyacrylamide gel and transferred to a nitrocellulose membrane. 696 PCO:GFP was detected with the antibodies anti-GFP (Bioscience, dilution 1:1000) in PBS containing 697 0.1% Tween 20 (Sigma-Aldrich) and 5% milk powder. The ECL Select Western blotting Detection 698 Reaction (GE Healthcare) was used to detect proteins with horseradish peroxidase -conjugated anti-699 rabbit secondary antibodies. For a protein loading control, the membrane before immunodetection 700 was stained with Ponceau S (Sigma -Aldrich) to detect all protein bands. Sulfenylated proteins were 701 detected via DCP-Bio1 labelling and visualized using Alexa Fluor™ 488-conjugated streptavidin, as 702 previously described (Zivanovic et al., 2019) . Sulfenylation levels were quantified using β -tubulin, 703 detected by western blotting (As10680; Agrisera, Vännäs, Sweden; dilution 1:5000), as an internal 704 loading control. 705 H2S quantification 706 100 mg of leaf tissue were homogenised in liquid nitrogen and metabolites were extracted in 707 homogenous mixture of Tris-HCl buffer (100 mM, pH 8.5; EDTA 1 mM), with shaking for 30 min 708 at 4 ◦C. Samples were then sonicated for 5 min in ice -bath and centrifuged for 15 min at 12,500× g 709 at 4 ◦C; 100 µL of supernatants were derivatised with 25 µL of 15 mM monobromobimane (MBB, 710 Merck) solution for 30 min at room temperature, and stopped by adding 25 µL of 5% formic acid. 711 The mixture was subjected to centrifugation at 800× g for 10 min, and 1 µL of the supernatant was 712 injected into the LC–MS/MS system for analysis. A calibration curve for NaHS concentrations was 713 established ranging from 2.5 µM to 100 µM, and the H2S concentration in the samples was determined 714 using this standard curve. The results are presented mean values ± SD of three different biological 715 replicates. 716 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Recombinant protein production 717 Recombinant PCO proteins from A. thaliana were expressed and purified following previously 718 established protocols (White et al., 2018). His6-tag affinity purification was performed, after which 719 the His6-tag was cleaved using TEV protease and subsequently removed with a HisTrap HP column 720 (GE Healthcare). The proteins underwent further purification using a HiLoad 26/600 Superdex 75 721 prep-grade size exclusion column (GE Healthcare) equilibrated with 50 mM Tris (pH 7.5) and 0.4 M 722 NaCl. The purity of the isolated proteins was confirmed by SDS-PAGE. 723 In vitro H2S oxidation assay of PCOs 724 For the in vitro H₂S oxidation assay, 10 μM recombinant PCO ( AtPCO1-5) was incubated with 725 defined concentrations of H₂S, H₂O₂, or an equivalent volume of H₂O at 4 °C for specified time 726 duration. Subsequently, 200 μM RAP2 2-15 peptide (CGGAIISDFIPPPR, purchased from GL 727 Biochem, China) was reacted with 1 μM PCO (treated or untreated with H₂S/H₂O₂) at 25°C for the 728 desired duration in 50 mM HEPES buffer (pH 7.4), hereafter referred to as reaction buffer. The 729 reaction was halted by quenching 5 μL of the sample in 45 μL of 5% formic acid, and peptide masses 730 were analyzed using an Agilent RapidFire RF360 sampling robot coupled to an Agilent 6530 731 Accurate-Mass Q-ToF mass spectrometer operating in positive electrospray mode. The distribution 732 of reaction products was determined based on the relative integrated areas of corresponding peaks, 733 and spectra were visualized using Qualitative Analysis (Version B.07.00). Agilent RapidFire 734 Integrator (Version 4.3.0.17235) was used to compute the integrated peak areas. 735 A related experiment examined the protective effects of GSH or TCEP against H₂S -mediated 736 inhibition of PCO activity. In this experiment, 2 μM PCO was treated with or without 5 mM H₂S for 737 30 minutes on ice, followed by 10 mM GSH/TCEP treatment (30 minutes on ice). Control samples, 738 which were not exposed to H₂S, were simultaneously treated with 10 mM GSH. Following treatment, 739 1 μM AtPCO was reacted with 100 μM AtRAP22-15 at 25°C for defined time periods in the reaction 740 buffer. Reactions were quenched with 5% formic acid, and two replicates were performed for each 741 condition. 742 Detection of Cys modification by H2S in vitro 743 To detect Cys oxidative modifications in AtPCO4, BioDiaAlk, a biotinylated form of DiaAlk which 744 is a chemical probe specific to sulfinic acid formation, was utilized. 100 μM of recombinant AtPCO4 745 enzyme was incubated with H₂S, H₂O₂, or an equivalent volume of H₂O in reaction buffer for 1 hour 746 at 25°C. Excess H₂S/H₂O₂ was removed using a Micro Bio -Spin P-6 chromatography column (Bio-747 Rad) equilibrated with reaction buffer. AtPCO4 samples were then incubated with 1 mM BioDiaAlk 748 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint in the dark for 1 hour at 25°C, followed by reduction with 10 mM DTT for 1 hour at 25°C. Proteins 749 were separated using SDS -PAGE, transferred onto a polyvinylidene difluoride (PVDF) membrane, 750 and subjected to streptavidin -HRP blotting (1:1000 dilution) or anti -His-HRP blotting (1:10,000 751 dilution). The protein signals were visualized using chemiluminescence (ECL Plus, Pierce). 752 Trypsin digest and LC-MS/MS to detect AtPCO4 modification sites 753 To identify modification sites on AtPCO4, 15 μg recombinant enzyme was incubated with H 2S or 754 H2O2 or equal volume of H2O in reaction buffer for 1 h at 25°C, followed by 100 mM iodoacetamide 755 (IAM) treatment to block free thiols for 1h in dark at 25°C. After removing excess H 2S/H2O2/IAM 756 by spin column, in-solution trypsin digestion was performed by adding trypsin in a 1:50 (w/w) ratio 757 overnight at 37 °C, followed by 85 mM DTT reduction for 1 h at RT to break the persulfide bridge. 758 Reduced peptides were purified by C18 ZipTip column and resulting tryptic peptides were 759 resuspended in 40 μL of Milli-Q water with 2 % acetonitrile and 0.1 % formic acid. 2 µL resuspended 760 peptides were analysed on a NanoAcquity -UPLC system (Waters) connected to an Orbitrap Elite 761 mass spectrometer (Thermo Fischer Scientific) possessing an EASY -Spray nano -electrospray ion 762 source (Thermo Fischer Scientific). The peptides were trapped on an in-house packed guard column 763 (75 μm i.d. x 20 mm, Acclaim PepMap C18, 3μm, 100 Å) using solvent A (0.1 % Formic Acid in 764 water) at a pressure of 140 bar. The peptides were separated on an EASY -spray Acclaim 765 PepMap® analytical column (75 μm i.d. × 50 mm, RSLC C18, 3 μm, 100 Å) using a linear gradient 766 (length: 100 minutes, 3 % to 60 % solvent B (0.1 % formic acid in acetonitrile), flow rate: 300 767 nL/min). The separated peptides were electrosprayed directly into the mass spectrometer operating in 768 a data-dependent mode using a CID-based method. Full scan MS spectra (scan range 350-1500 m/z, 769 resolution 120000, AGC target 1e6, maximum injection time 250 ms) and subsequent CID MS/MS 770 spectra (AGC target 5e4, maximum injection time 100 ms) of 10 most intense peaks were acquired 771 in the Ion Trap. CID fragmentation was performed at 35 % of normalised collision energy, and the 772 signal intensity threshold was kept at 500 counts. The CID method used performs beam -type CID 773 fragmentation of the peptides. Data analysis was performed with Peaks 8.5. The raw MS file was 774 searched against the TAIR database. Trypsin with a maximum of 3 missed cleavages and one 775 unspecific end was selected as the protease. Carbamidomethylation (Cys) was set as a fixed 776 modification, and oxidation (Methionine) and deamination (Asparagine, Glutamine) were set as 777 variable modifications. Precursor mass tolerance was set as 15 ppm. Fragment mass tolerances for 778 CID were set to 0.8 Da, respectively. All spectra were manually validated. For all peptides present at 779 -10*Log(P) > 20, considered as high -confidence, spectra were manually checked, validated, or 780 disqualified. 781 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint 782

783 The BioDiaAlk probe was kindly provided by Prof. Kate Carroll (Florida Atlantic University, USA). 784 Author contributions 785 Y .T., S.A., P.P., E.F. and B.G. conceptualized and designed the study. Y .T., S.A., A.A., L.P., Z.D., 786 G.N., M.L-P., D.M.G., N.L.M. and S.L. performed the experiments. Y .T., S.A., A.A., M.L-P., E.F. and 787 B.G contributed to data analysis and figure preparation. P.P., C.G., M.S. , E.F. and B.G. provided 788 funding acquisition for the study. Y .T., S.A., E.F., and B.G. wrote the manuscript with critical input 789 from all authors. 790 Funding 791 This research was supported by the University of Pisa (Italy) and Sant’Anna School of Advanced 792 Studies (Pisa, Italy). Y .T. was supported by funding from the Italian National Recovery and Resilience 793 Plan 2021-2027 (PNRR). S.A., D.M.G. and E.F. were supported by the European Research Council 794 (European Union’s Horizon 2020 Research and Innovation Programme Grant 864888). S.A. was 795 financially supported by a Universität Münster Fellowship from the University of Münster 796 Internationalization Fund to conduct research on the H₂O₂ biosensor . This work was supported by 797 ERDF A way of making Europe and MCIN/AEI/10.13039/501100011033 (grant No. PID2022 -798 141885NB-I00), and Junta de Andalucía (grant No PROYEXCEL_00177). 799 Competing interest 800 The authors declare no competing interests. 801

802 Abbas, M., Berckhan, S., Rooney, D. J., Gibbs, D. J., Vicente Conde, J., Sousa Correia, C., Bassel, 803 G. W., Marín-De La Rosa, N., León, J., Alabadí, D., Blázquez, M. A., & Holdsworth, M. J. 804 (2015). Oxygen Sensing Coordinates Photomorphogenesis to Facilitate Seedling Survival. 805 Current Biology, 25(11), 1483. https://doi.org/10.1016/J.CUB.2015.03.060 806 Akter, S., Perri, M., Lavilla-Puerta, M., Ferretti, B., Dalle Carbonare, L., Shukla, V ., Telara, Y ., Zhang, 807 D., Gunawardana, D. M., Myers, W. K., Giuntoli, B., Flashman, E., & Licausi, F. (2024). H 2 O 808 2 repurposes the plant oxygen -sensing machinery to control the transcriptional response to 809 oxidative stress. https://doi.org/10.1101/2024.10.21.619351 810 Álvarez, C., Calo, L., Romero, L. C., García, I., & Gotor, C. (2010). An O-Acetylserine(thiol)lyase 811 Homolog with l -Cysteine Desulfhydrase Activity Regulates Cysteine Homeostasis in 812 Arabidopsis. Plant Physiology, 152(2), 656–669. https://doi.org/10.1104/PP.109.147975 813 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Álvarez, C., García, I., Moreno, I., Pérez -Pérez, M. E., Crespo, J. L., Romero, L. C., & Gotor, C. 814 (2012). Cysteine-generated sulfide in the cytosol negatively regulates autophagy and modulates 815 the transcriptional profile in arabidopsisw. Plant Cell , 24(11), 4621 –4634. 816 https://doi.org/10.1105/tpc.112.105403 817 Aroca, A., Benito, J. M., Gotor, C., & Romero, L. C. (2017). Persulfidation proteome reveals the 818 regulation of protein function by hydrogen sulfide in diverse biological processes in Arabidopsis. 819 Journal of Experimental Botany, 68(17), 4915–4927. https://doi.org/10.1093/jxb/erx294 820 Aroca, A., Gotor, C., & Romero, L. C. (2018). Hydrogen sulfide signaling in plants: emerging roles 821 of protein persulfidation. Frontiers in Plant Science , 9, 1369. 822 https://doi.org/10.3389/FPLS.2018.01369/BIBTEX 823 Aroca, A., Jurado -Flores, A., Filipovic, M. R., Gotor, C., & Romero, L. C. (2022). Detection of 824 protein persulfidation in plants by the dimedone switch method. In Methods in Enzymology (1st 825 ed., V ol. 676). Elsevier Inc. https://doi.org/10.1016/bs.mie.2022.07.024 826 Aroca, A., Schneider, M., Scheibe, R., Gotor, C., & Romero, L. C. (2017). Hydrogen Sulfide 827 Regulates the Cytosolic/Nuclear Partitioning of Glyceraldehyde-3-Phosphate Dehydrogenase by 828 Enhancing its Nuclear Localization. Plant and Cell Physiology , 58(6), 983 –992. 829 https://doi.org/10.1093/PCP/PCX056 830 Aroca, Á., Serna, A., Gotor, C., & Romero, L. C. (2015). S-sulfhydration: A cysteine posttranslational 831 modification in plant systems. Plant Physiology , 168(1), 334 –342. 832 https://doi.org/10.1104/pp.15.00009 833 Aroca, A., Zhang, J., Xie, Y ., Romero, L. C., & Gotor, C. (2021). Hydrogen sulfide signaling in plant 834 adaptations to adverse conditions: Molecular mechanisms. Journal of Experimental Botany , 835 72(16), 5893–5904. https://doi.org/10.1093/jxb/erab239 836 Bhadwal, S. S., Verma, S., Hassan, S., & Kaur, S. (2024). Unraveling the potential of hydrogen sulfide 837 as a signaling molecule for plant development and environmental stress responses: A state -of-838 the-art review. Plant Physiology and Biochemistry , 212, 108730. 839 https://doi.org/10.1016/J.PLAPHY .2024.108730 840 Bielecka, M., Watanabe, M., Morcuende, R., Scheible, W. R., Hawkesford, M. J., Hesse, H., & 841 Hoefgen, R. (2015). Transcriptome and Metabolome analysis of plant sulfate starvation and 842 resupply provides novel information on transcriptional regulation of metabolism associated with 843 sulfur, nitrogen and phosphorus nutritional responses in Arabidopsis. Frontiers in Plant Science, 844 5(JAN), 1–18. https://doi.org/10.3389/fpls.2014.00805 845 Chen, J., Wang, W. H., Wu, F. H., He, E. M., Liu, X., Shangguan, Z. P., & Zheng, H. L. (2015). 846 Hydrogen sulfide enhances salt tolerance through nitric oxide -mediated maintenance of ion 847 homeostasis in barley seedling roots. Scientific Reports, 5. https://doi.org/10.1038/srep12516 848 Clough, S. J., & Bent, A. F. (1998). Floral dip: a simplified method for Agrobacterium -mediated 849 transformation of Arabidopsis thaliana. The Plant Journal : For Cell and Molecular Biology , 850 16(6), 735–743. https://doi.org/10.1046/J.1365-313X.1998.00343.X 851 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Cooper, C. E., & Brown, G. C. (2008). The inhibition of mitochondrial cytochrome oxidase by the 852 gases carbon monoxide, nitric oxide, hydrogen cyanide and hydrogen sulfide: Chemical 853 mechanism and physiological significance. Journal of Bioenergetics and Biomembranes, 40(5), 854 533–539. https://doi.org/10.1007/S10863-008-9166-6/METRICS 855 Cuevasanta, E., Lange, M., Bonanata, J., Coitiño, E. L., Ferrer-Sueta, G., Filipovic, M. R., & Alvarez, 856 B. (2015). Reaction of Hydrogen Sulfide with Disulfide and Sulfenic Acid to Form the Strongly 857 Nucleophilic Persulfide. The Journal of Biological Chemistry , 290(45), 26866 –26880. 858 https://doi.org/10.1074/JBC.M115.672816 859 Dirr, A., Greco, A. Del, Howe, S., Walker, L. S., Perri, M., Shukla, V ., Licausi, F., & Flashman, E. 860 (2025). Engineering reduced activity in the oxygen-sensing Arabidopsis thaliana plant cysteine 861 oxidase 4 enzyme results in improved flood resilience. BioRxiv, 2025.07.24.666444. 862 https://doi.org/10.1101/2025.07.24.666444 863 Filipovic, M. R., Zivanovic , J., Alvarez, B., & Banerjee, R. (2018). Chemical Biology of H2S 864 Signaling through Persulfidation. Chemical Reviews , 118(3), 1253 –1337. 865 https://doi.org/10.1021/ACS.CHEMREV .7B00205 866 Fu, M., Zhang, W., Wu, L., Yang, G., Li, H., & Wang, R. (n.d.). Hydrogen sulfide (H 2 S) metabolism 867 in mitochondria and its regulatory role in energy production . 868 https://doi.org/10.1073/pnas.1115634109 869 García-Calderón, M., Vignane, T., Filipovic, M. R., Ruiz, M. T., Romero, L. C., Márquez, A. J., Gotor, 870 C., & Aroca, A. (2023). Persulfidation protects from oxidative stress under nonphotorespiratory 871 conditions in Arabidopsis. New Phytologist , 238(4), 1431 –1445. 872 https://doi.org/10.1111/nph.18838 873 Gasch, P., Fundinger, M., Müller, J. T., Lee, T., Bailey-Serres, J., & Mustropha, A. (2016). Redundant 874 ERF-VII Transcription Factors Bind to an Evolutionarily Conserved cis -Motif to Regulate 875 Hypoxia-Responsive Gene Expression in Arabidopsis. The Plant Cell , 28(1), 160 –180. 876 https://doi.org/10.1105/TPC.15.00866 877 Gibbs, D. J., Lee, S. C., Md Isa, N., Gramuglia, S., Fukao, T., Bassel, G. W., Correia, C. S., Corbineau, 878 F., Theodoulou, F. L., Bailey-Serres, J., & Holdsworth, M. J. (2011). Homeostatic response to 879 hypoxia is regulated by the N -end rule pathway in plants. Nature 2011 479:7373, 479(7373), 880 415–418. https://doi.org/10.1038/nature10534 881 Gibbs, D. J., MdIsa, N., Movahedi, M., Lozano -Juste, J., Mendiondo, G. M., Berckhan, S., Marín -882 delaRosa, N., VicenteConde, J., SousaCorreia , C., Pearce, S. P., Bassel, G. W., Hamali, B., 883 Talloji, P., Tomé, D. F. A., Coego, A., Beynon, J., Alabadí, D., Bachmair, A., León, J., … 884 Holdsworth, M. J. (2014). Nitric Oxide Sensing in Plants Is Mediated by Proteolytic Control of 885 Group VII ERF Transcription Factors. Molecular Cell , 53(3), 369 –379. 886 https://doi.org/10.1016/j.molcel.2013.12.020 887 Gonzali, S., Loreti, E., Cardarelli, F., Novi, G., Parlanti, S., Pucciariello, C., Bassolino, L., Banti, V ., 888 Licausi, F., & Perata, P. (2015). Universal stress protein HRU1 mediates ROS homeostasis under 889 anoxia. Nature Plants, 1. https://doi.org/10.1038/nplants.2015.151 890 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Gotor, C., García, I., Aroca, Á., Laureano -Marín, A. M., Arenas -Alfonseca, L., Jurado -Flores, A., 891 Moreno, I., & Romero, L. C. (2019). Signaling by hydrogen sulfide and cyanide through post -892 translational modification. Journal of Experimental Botany , 70(16), 4251 –4265. 893 https://doi.org/10.1093/JXB/ERZ225 894 Greiner, R., Pálinkás, Z., Bäsell, K., Becher, D., Antelmann, H., Nagy, P., & Dick, T. P. (2013). 895 Polysulfides link H2S to protein thiol oxidation. Antioxidants and Redox Signaling , 19(15), 896 1749–1765. https://doi.org/10.1089/ARS.2012.5041;CTYPE:STRING:JOURNAL 897 Habibi, F., Liu, T., Shahid, M. A., Schaffer, B., & Sarkhosh, A. (2023). Physiological, biochemical, 898 and molecular responses of fruit trees to root zone hypoxia. Environmental and Experimental 899 Botany, 206, 105179. https://doi.org/10.1016/J.ENVEXPBOT.2022.105179 900 Hartman, S., Liu, Z., van Veen, H., Vicente, J., Reinen, E., Martopawiro, S., Zhang, H., van Dongen, 901 N., Bosman, F., Bassel, G. W., Visser, E. J. W., Bailey-Serres, J., Theodoulou, F. L., Hebelstrup, 902 K. H., Gibbs, D. J., Holdsworth, M. J., Sasidharan, R., & V oesenek, L. A. C. J. (2019). Ethylene-903 mediated nitric oxide depletion pre -adapts plants to hypoxia stress. Nature Communications, 904 10(1). https://doi.org/10.1038/S41467-019-12045-4 905 Hell, R., & Wirtz, M. (2008). Metabolism of Cysteine in Plants and Phototrophic Bacteria . 59–91. 906 https://doi.org/10.1007/978-1-4020-6863-8_4 907 Jaiswal, S., Singh, S. P., Singh, S., Gupta, R., Tripathi, D. K., Corpas, F. J., & Singh, V . P. (2024). 908 Hydrogen Sulphide: A Key Player in Plant Development and Stress Resilience. Plant, Cell & 909 Environment. https://doi.org/10.1111/PCE.15309 910 Jethva, J., Lichtenauer, S., Schmidt -Schippers, R., Steffen -Heins, A., Poschet, G., Wirtz, M., van 911 Dongen, J. T., Eirich, J., Finkemeier, I., Bilger, W., Schwarzländer, M., & Sauter, M. (2023). 912 Mitochondrial alternative NADH dehydrogenases NDA1 and NDA2 promote survival of 913 reoxygenation stress in Arabidopsis by safeguarding photosynthesis and limiting ROS 914 generation. New Phytologist , 238(1), 96 –112. 915 https://doi.org/10.1111/NPH.18657;WGROUP:STRING:PUBLICATION 916 Jurado-Flores, A., Aroca, A., Romero, L. C., & Gotor, C. (2023). Sulfide promotes tolerance to 917 drought through protein persulfidation in Arabidopsis. Journal of Experimental Botany, 74(15), 918 4654–4669. https://doi.org/10.1093/jxb/erad165 919 Kosmacz, M., Parlanti, S., Schwarzländer, M., Kragler, F., Licausi, F., & Van Dongen, J. T. (2015). 920 The stability and nuclear localization of the transcription factor RAP2.12 are dynamically 921 regulated by oxygen concentration. Plant Cell and Environment , 38(6), 1094 –1103. 922 https://doi.org/10.1111/pce.12493 923 Kunkowska, A. B., Fontana, F., Betti, F., Soeur, R., Beckers, G. J. M., Meyer, C., De Jaeger, G., Weits, 924 D. A., Loreti, E., & Perata, P. (2023). Target of rapamycin signaling couples energy to oxygen 925 sensing to modulate hypoxic gene expression in Arabidopsis. Proceedings of the National 926 Academy of Sciences of the United States of America , 120(3). 927 https://doi.org/10.1073/PNAS.2212474120 928 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Lavilla-Puerta, M., He, Y ., Piccinini, L., Paco, L. Di, Papachristodoulou, A., Licausi, F., & Giuntoli, 929 B. (2025). A synthetic ERFVII-dependent circuit in yeast sheds light on the regulation of early 930 hypoxic responses of plants. BioRxiv, 2025.09.01.673502. 931 https://doi.org/10.1101/2025.09.01.673502 932 Lavilla-Puerta, M., Latter, R., Bellè, F., Cervelli, T., Galli, A., Perata, P., Chini, A., Flashman, E., & 933 Giuntoli, B. (2023). Identification of novel plant cysteine oxidase inhibitors from a yeast 934 chemical genetic screen. The Journal of Biological Chemistry , 299(12). 935 https://doi.org/10.1016/J.JBC.2023.105366 936 Li, Q., & Lancaster, J. R. (2013). Chemical Foundations of Hydrogen Sulfide Biology. Nitric Oxide : 937 Biology and Chemistry / Official Journal of the Nitric Oxide Society , 0, 938 10.1016/j.niox.2013.07.001. https://doi.org/10.1016/J.NIOX.2013.07.001 939 Li, Y ., Sun, D., Xu, K., Jin, L., & Peng, R. (2021). Hydrogen Sulfide Enhances Plant Tolerance to 940 Waterlogging Stress. Plants 2021, V ol. 10, Page 1928 , 10(9), 1928. 941 https://doi.org/10.3390/PLANTS10091928 942 Licausi, F., Giuntoli, B., & Perata, P. (2020). Similar and Yet Different: Oxygen Sensing in Animals 943 and Plants. Trends in Plant Science, 25(1), 6–9. https://doi.org/10.1016/j.tplants.2019.10.013 944 Licausi, F., Kosmacz, M., Weits, D. A., Giuntoli, B., Giorgi, F. M., V oesenek, L. A. C. J., Perata, P., 945 & Dongen, J. T. van. (2011). Oxygen sensing in plants is mediated by an N-end rule pathway for 946 protein destabilization. Nature 2011 479:7373 , 479(7373), 419 –422. 947 https://doi.org/10.1038/nature10536 948 Licausi, F., Ohme -Takagi, M., & Perata, P. (2013). APETALA2/Ethylene Responsive Factor 949 (AP2/ERF) transcription factors: mediators of stress responses and developmental programs. 950 The New Phytologist, 199(3), 639–649. https://doi.org/10.1111/NPH.12291 951 Livak, K. J., & Schmittgen, T. D. (2001). Analysis of relative gene expression data using real -time 952 quantitative PCR and the 2 -ΔΔCT method. Methods, 25(4), 402 –408. 953 https://doi.org/10.1006/meth.2001.1262 954 Martin, N. M., & Maricle, B. R. (2015). Species -specific enzymatic tolerance of sulfide toxicity in 955 plant roots. Plant Physiology and Biochemistry , 88, 36 –41. 956 https://doi.org/10.1016/j.plaphy.2015.01.007 957 Masson, N., Keeley, T. P., Giuntoli, B., White, M. D., Puerta, M. L., Perata, P., Hopkinson, R. J., 958 Flashman, E., Licausi, F., & Ratcliffe, P. J. (2019). Conserved N-terminal cysteine dioxygenases 959 transduce responses to hypoxia in animals and plants. Science, 365(6448), 65 –69. 960 https://doi.org/10.1126/SCIENCE.AAW0112 961 Mustafa, A. K., Gadalla, M. M., Sen, N., Kim, S., Mu, W., Gazi, S. K., Barrow, R. K., Yang, G., Wang, 962 R., & Snyder, S. H. (2009). H2S signals through protein S -sulfhydration. Science Signaling, 963 2(96). https://doi.org/10.1126/SCISIGNAL.2000464 964 Mustroph, A., Lee, S. C., Oosumi, T., Zanetti, M. E., Yang, H., Ma, K., Yaghoubi-Masihi, A., Fukao, 965 T., & Bailey-Serres, J. (2010). Cross -Kingdom Comparison of Transcriptomic Adjustments to 966 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Low-Oxygen Stress Highlights Conserved and Plant -Specific Responses. Plant Physiology , 967 152(3), 1484–1500. https://doi.org/10.1104/PP.109.151845 968 Nietzel, T., Elsässer, M., Ruberti, C., Steinbeck, J., Ugalde, J. M., Fuchs, P., Wagner, S., Ostermann, 969 L., Moseler, A., Lemke, P., Fricker, M. D., Müller-Schüssele, S. J., Moerschbacher, B. M., Costa, 970 A., Meyer, A. J., & Schwarzländer, M. (2019). The fluorescent protein sensor roGFP2 -Orp1 971 monitors in vivo H 2 O 2 and thiol redox integration and elucidates intracellular H 2 O 2 972 dynamics during elicitor -induced oxidative burst in Arabidopsis. New Phytologist , 221(3), 973 1649–1664. https://doi.org/10.1111/NPH.15550 974 Pantaleno, R., Scuffi, D., Schiel, P., Schwarzländer, M., Alex Costa, |, Carlos García -Mata, |, & 975 García-Mata, C. (2024). Mitochondrial ß-Cyanoalanine Synthase Participates in flg22-Induced 976 Stomatal Immunity. https://doi.org/10.1111/pce.15155 977 Papdi, C., Pérez-Salamõ, I., Joseph, M. P., Giuntoli, B., Bögre, L., Koncz, C., & Szabados, L. (2015). 978 The low oxygen, oxidative and osmotic stress responses synergistically act through the ethylene 979 response factor VII genes RAP2.12, RAP2.2 and RAP2.3. The Plant Journal, 82(5), 772–784. 980 https://doi.org/10.1111/TPJ.12848 981 Pedroletti, L., Moseler, A., Timm, S., Poschet, G., Homagk, M., The, J. X. L., Wagner, S., Wirtz, M., 982 Hell, R., & Meyer, A. J. (2023). Altered iron-sulfur cluster transfer in Arabidopsis mitochondria 983 reveals lipoyl synthase as a Janus -faced enzyme that generates toxic sulfide . 984 https://doi.org/10.1101/2023.08.30.555573 985 Pietri, R., Román -Morales, E., & López -Garriga, J. (2011). Hydrogen sulfide and hemeproteins: 986 Knowledge and mysteries. Antioxidants and Redox Signaling , 15(2), 393 –404. 987 https://doi.org/10.1089/ARS.2010.3698;WGROUP:STRING:PUBLICATION 988 Pucciariello, C., Parlanti, S., Banti, V ., Novi, G., & Perata, P. (2012). Reactive oxygen species-driven 989 transcription in Arabidopsis under oxygen deprivation. Plant Physiology , 159(1), 184 –196. 990 https://doi.org/10.1104/pp.111.191122 991 Renziehausen, T., Chaudhury, R., Hartman, S., Mustroph, A., & Schmidt-Schippers, R. R. (2024). A 992 mechanistic integration of hypoxia signaling with energy, redox, and hormonal cues. Plant 993 Physiology, 00, 596. https://doi.org/10.1093/plphys/kiae596 994 Saito, K. (2004). Sulfur Assimilatory Metabolism. The Long and Smelling Road. Plant Physiology, 995 136(1), 2443–2450. https://doi.org/10.1104/PP.104.046755 996 Sasidharan, R., Hartman, S., Liu, Z., Martopawiro, S., Sajeev, N., Van Veen, H., Yeung, E., & 997 V oesenek, L. A. C. J. (2018). Signal Dynamics and Interactions during Flooding Stress. Plant 998 Physiology, 176(2), 1106–1117. https://doi.org/10.1104/PP.17.01232 999 Sharma, P., Meyyazhagan, A., Easwaran, M., Sharma, M. M. M., Mehta, S., Pandey, V ., Liu, W. C., 1000 Kamyab, H., Balasubramanian, B., Baskaran, R., Klemeš, J. J., Mesbah, M., & Chelliapan, S. 1001 (2022). Hydrogen Sulfide: A new warrior in assisting seed germination during adverse 1002 environmental conditions. Plant Growth Regulation 2022 98:3 , 98(3), 401 –420. 1003 https://doi.org/10.1007/S10725-022-00887-W 1004 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint Shen, J., Zhang, J., Zhou, M., Zhou, H., Cui, B., Gotor, C., Romero, L. C., Fu, L., Yang, J., Foyer, C. 1005 H., Pan, Q., Shen, W., & Xie, Y . (2020). Persulfidation -based modification of cysteine 1006 desulfhydrase and the NADPH oxidase RBOHD controls guard cell abscisic acid signaling. 1007 Plant Cell, 32(4), 1000–1017. https://doi.org/10.1105/tpc.19.00826 1008 Szabo, C. (2010). Gaseotransmitters: New frontiers for translational science. Science Translational 1009 Medicine, 2(59). https://doi.org/10.1126/SCITRANSLMED.3000721/ASSET/060FFD11 -1010 5A95-408B-B9C1-54337A1818F0/ASSETS/GRAPHIC/259PS54-F2.JPEG 1011 Takahashi, H., Kopriva, S., Giordano, M., Saito, K., & Hell, R. (2011). Sulfur assimilation in 1012 photosynthetic organisms: molecular functions and regulations of transporters and assimilatory 1013 enzymes. Annual Review of Plant Biology , 62, 157–184. https://doi.org/10.1146/ANNUREV-1014 ARPLANT-042110-103921 1015 Urquiza-García, U., & Millar, A. J. (2019). Expanding the bioluminescent reporter toolkit for plant 1016 science with NanoLUC. Plant Methods, 15(1). https://doi.org/10.1186/s13007-019-0454-4 1017 Vignane, T., & Filipovic, M. R. (2023). Emerging Chemical Biology of Protein Persulfidation. In 1018 Antioxidants and Redox Signaling (V ol. 39, Issues 1–3, pp. 19 –39). Mary Ann Liebert Inc. 1019 https://doi.org/10.1089/ars.2023.0352 1020 Wang, L., Mu, X., Chen, X., & Han, Y . (2022). Hydrogen sulfide attenuates intracellular oxidative 1021 stress via repressing glycolate oxidase activities in Arabidopsis thaliana. BMC Plant Biology , 1022 22(1). https://doi.org/10.1186/s12870-022-03490-3 1023 Wang, X., Shi, C., Hu, Y ., Ma, Y ., Yi, Y ., Jia, H., Li, F., Sun, H., Li, T., Wang, X., Li, T., & Li, J. 1024 (2023). Persulfidation maintains cytosolic G6PDs activity through changing tetrameric structure 1025 and competing cysteine sulfur oxidation under salt stress in Arabidopsis and tomato. New 1026 Phytologist, 240(2), 626–643. https://doi.org/10.1111/NPH.19188 1027 Weits, D. A., Giuntoli, B., Kosmacz, M., Parlanti, S., Hubberten, H. M., Riegler, H., Hoefgen, R., 1028 Perata, P., Van Dongen, J. T., & Licausi, F. (2014). Plant cysteine oxidases control the oxygen -1029 dependent branch of the N -end-rule pathway. Nature Communications , 5. 1030 https://doi.org/10.1038/ncomms4425 1031 White, M. D., Carbonare, L. D., Puerta, M. L., Iacopino, S., Edwards, M., Dunne, K., Pires, E., Levy, 1032 C., McDonough, M. A., Licausi, F., & Flashman, E. (2020). Structures of Arabidopsis thaliana 1033 oxygen-sensing plant cysteine oxidases 4 and 5 enable targeted manipulation of their activity. 1034 Proceedings of the National Academy of Sciences , 117(37), 23140 –23147. 1035 https://doi.org/10.1073/PNAS.2000206117 1036 White, M. D., Kamps, J. J. A. G., East, S., Taylor Kearney, L. J., & Flashman, E. (2018). The plant 1037 cysteine oxidases from Arabidopsis thaliana are kinetically tailored to act as oxygen sensors. 1038 Journal of Biological Chemistry , 293(30), 11786 –11795. 1039 https://doi.org/10.1074/jbc.RA118.003496 1040 White, M. D., Klecker, M., Hopkinson, R. J., Weits, D. A., Mueller, C., Naumann, C., O’Neill, R., 1041 Wickens, J., Yang, J., Brooks-Bartlett, J. C., Garman, E. F., Grossmann, T. N., Dissmeyer, N., & 1042 Flashman, E. (2017). Plant cysteine oxidases are dioxygenases that directly enable arginyl 1043 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint transferase-catalysed arginylation of N -end rule targets. Nature Communications , 8. 1044 https://doi.org/10.1038/ncomms14690 1045 Xiang, Z. X., Li, W., Lu, Y . T., & Yuan, T. T. (2023). Hydrogen sulfide alleviates osmotic stress -1046 induced root growth inhibition by promoting auxin homeostasis. Plant Journal, 114(6), 1369–1047 1384. https://doi.org/10.1111/TPJ.16198 1048 Xiao, Y ., Wu, X., Sun, M., & Peng, F. (2020). Hydrogen Sulfide Alleviates Waterlogging -Induced 1049 Damage in Peach Seedlings via Enhancing Antioxidative System and Inhibiting Ethylene 1050 Synthesis. Frontiers in Plant Science , 11, 533438. 1051 https://doi.org/10.3389/FPLS.2020.00696/BIBTEX 1052 Yang, T., Yuan, G., Zhang, Q., Xuan, L., Li, J., Zhou, L., Shi, H., Wang, X., & Wang, C. (2021). 1053 Transcriptome and metabolome analyses reveal the pivotal role of hydrogen sulfide in promoting 1054 submergence tolerance in Arabidopsis. Environmental and Experimental Botany, 183, 104365. 1055 https://doi.org/10.1016/J.ENVEXPBOT.2020.104365 1056 Zabalza, A., Van Dongen, J. T., Froehlich, A., Oliver, S. N., Faix, B., Gupta, K. J., Schmälzlin, E., 1057 Igal, M., Orcaray , L., Royuela, M., & Geigenberger, P. (2009). Regulation of respiration and 1058 fermentation to control the plant internal oxygen concentration. Plant Physiology, 149(2), 1087–1059 1098. https://doi.org/10.1104/pp.108.129288 1060 Zhang, J., Zhou, M., Ge, Z., Shen, J., Zhou, C., Gotor, C., Romero, L. C., Duan, X., Liu, X., Wu, D., 1061 Yin, X., & Xie, Y . (2020). Abscisic acid-triggered guard cell l -cysteine desulfhydrase function 1062 and in situ hydrogen sulfide production contributes to heme oxygenase -modulated stomatal 1063 closure. Plant, Cell & Environment, 43(3), 624–636. https://doi.org/10.1111/PCE.13685 1064 Zhang, P., Luo, Q., Wang, R., & Xu, J. (2017). Hydrogen sulfide toxicity inhibits primary root growth 1065 through the ROS -NO pathway. Scientific Reports, 7(1), 1–11. https://doi.org/10.1038/s41598-1066 017-01046-2 1067 Zhou, L., Gao, X., Weits, D. A., Zeng, P., Wang, X., & Cai, J. (2021). H2S regulates low oxygen 1068 signaling via integration with the unfolded protein response in Arabidopsis thaliana. Plant and 1069 Soil, 467(1–2), 531–547. https://doi.org/10.1007/S11104-021-05091-9/FIGURES/8 1070 Zhou, M., Xie, Y ., Van Breusegem, F., & Huang, J. (2025). Hydrogen sulfide and protein 1071 persulfidation in plant stress signaling. Journal of Experimental Botany , 76(13), 3738–3757. 1072 https://doi.org/10.1093/JXB/ERAF100 1073 Zivanovic, J., Kouroussis, E., Kohl, J. B., Adhikari, B., Bursac, B., Schott -Roux, S., Petrovic, D., 1074 Miljkovic, J. L., Thomas-Lopez, D., Jung, Y ., Miler, M., Mitchell, S., Milosevic, V ., Gomes, J. 1075 E., Benhar, M., Gonzales-Zorn, B., Ivanovic-Burmazovic, I., Torregrossa, R., Mitchell, J. R., … 1076 Filipovic, M. R. (2019). Selective Persulfide Detection Reveals Evolutionarily Conserved 1077 Antiaging Effects of S -Sulfhydration. Cell Metabolism , 30(6), 1152 -1170.e13. 1078 https://doi.org/10.1016/j.cmet.2019.10.007 1079 1080 1081 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 6, 2025. ; https://doi.org/10.1101/2025.11.05.686772doi: bioRxiv preprint