# Source paper archive - Source title: The GSNOR1-VDAC3 complex acts as a cargo co-receptor for mitophagy in response to mitochondrial stress in plants - Author(s): Huifang Lin, Hongyan Guo, Rong Wu, Wan Wang, Peng-Fei Jia, Shina Sun, Ru Bie, Hong-Ju Li, Wei-Cai Yang, Jianru Zuo - Source landing URL: https://www.biorxiv.org/content/10.64898/2026.10.28.758855v1 - PDF URL: https://www.biorxiv.org/content/10.64898/2026.10.28.758855v1.full.pdf - PDF SHA-256: 74bdc6788be497408c5fe8781385da325bea289eb16571023a6810181b8f7521 - Converter: pymupdf4llm 0.0.27 - Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version. --- [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1 _**S**_ **-nitrosoglutathione reductase acts as a mitophagy receptor to control mitochondrial** 2 **quality in plants** 3 4 Huifang Lin [1,2,3], Hongyan Guo [1,3], Rong Wu [1,2], Wan Wang [1], Peng-Fei Jia [1], Shina Sun [1,2], Ru 5 Bie [1,2], Hong-Ju Li [1,2], Wei-Cai Yang [1,2] & Jianru Zuo [1,2]  6 7 1 State Key Laboratory of Seed Innovation, Institute of Genetics and Developmental Biology, 8 Chinese Academy of Sciences, Beijing, China 9 2 College of Advanced Agricultural Sciences, University of Chinese Academy of Sciences, 10 Beijing, China 11 3 These authors contributed equally 12  [e-mail: jrzuo@genetics.ac.cn](mailto:jrzuo@genetics.ac.cn) 13 1 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 14 **Abstract** 15 16 Mitochondria are central for cellular energy metabolism, and their quality is warranted by 17 multiple mechanisms, including mitophagy, a process of selectively removing damaged, 18 aged, or excess mitochondria. However, the regulatory mechanism of mitophagy remains 19 elusive in plants. Here, we report that Arabidopsis _S_ -nitrosoglutathione reductase 1 20 (GSNOR1), an enzyme that has long been known to degrade _S_ -nitrosoglutathione, interacts 21 with the outer mitochondrial membrane protein Voltage-Dependent Anion Channel 3 22 (VDAC3) to mediate mitophagy. GSNOR1 translocates to depolarized mitochondria, binds 23 to VDAC3, and interacts with Autophagy-related 8 (ATG8) via its ATG8-interacting motifs, 24 thereby engulfing damaged mitochondria in autophagosomes for mitophagic degradation. 25 Mutations in _GSNOR1_ or _VDAC3_ cause the accumulation of misshapen mitochondria and 26 hypersensitivity to mitochondrial stress. We propose that GSNOR1 acts as a cargo receptor 27 interacting with VDAC3 and ATG8 to target depolarized mitochondria to mitophagy, thereby 28 regulating cellular homeostasis. 29 30 **Keywords** : Mitophagy; mitochondrial stress; cargo receptor; GSNOR1; VDAC3; ATG8; 31 Arabidopsis 32 2 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 33 **Main** 34 35 Autophagy is a cellular process that targets macromolecules or organelles to the double 36 membraned autophagosome, followed by fusing to the lysosome or vacuole where the 37 targets are degraded into small molecules for recycling. This highly conserved process is 38 vital for both cellular homeostasis and survival under stress and starvation conditions. 39 Autophagy is generally classified into three types based on the different pathways of 40 substrate transport to lysosomes or vacuoles, namely macroautophagy, microautophagy, and 41 molecular chaperone-mediated autophagy (Gross et al., 2025; Kuchitsu and Taguchi, 2024; 42 Liu and Bassham, 2012; Marshall and Vierstra, 2018). Depending on the availability of 43 nutrients, autophagy can also be classified into starvation-induced non-selective autophagy 44 and selective autophagy. While non-selective autophagy is a key strategy for survival under 45 starvation conditions, selective autophagy occurs under nutrient sufficient conditions and is 46 generally initiated in response to cellular and environmental stresses to maintain cellular 47 homeostasis (Gross et al., 2024; Marshall and Vierstra, 2018; Park et al., 2023; Wang et al., 48 2026). In plants, autophagy is involved in regulating various physiological processes, 49 including growth, development, and both biotic and abiotic stress responses (Gross _et al._, 50 2025; Liu and Bassham, 2012; Marshall and Vierstra, 2018; Wang _et al._, 2026). 51 In selective autophagy, a target or cargo destined for degradation is specifically 52 recognized by a receptor protein that in turn interacts with Autophagy-related 8 (ATG8), 53 resulting in the engulfment of the target in autophagosome and subsequent autophagic 54 degradation. The interaction of a cargo or a receptor with ATG8 is mediated by ATG8 55 Interacting Motif (AIM; also known as LC3-interacting region or LIR motif) located in a  56 sheet (Noda et al., 2010; North et al., 2025). A typical AIM/LIR consists of a core motif of 57 four amino acid residues [W/F/Y]XX[L/I/V], where X stands for any amino acid (Johansen 58 and Lamark, 2020; Noda _et al._, 2010; North _et al._, 2025). In addition to AIM, the receptor 59 ATG8 interaction has also been shown to be mediated by the Ubiquitin-Interacting Motif 60 (UIM), characterized as Ψ-ζ-X-A-Ψ-X-X-S (Ψ stands for hydrophobic amino acids; ζ stands 3 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 61 for hydrophilic amino acids; and X stands for any amino acid) (Marshall et al., 2019; 62 Stephani and Dagdas, 2020). 63 Mitochondria are central for cellular metabolism and energy supplies. When challenged 64 by nutrient starvation or stresses, excessive, aged, or damaged mitochondria are subjected to 65 autophagic degradation, a process known as mitophagy, which acts as a key strategy to 66 maintain mitochondrial homeostasis (Broda et al., 2018; Pickles et al., 2018; Yang et al., 67 2024). Defects in mitophagy have been associated with abnormal growth and development 68 as well as a variety of diseases in humans (Broda _et al._, 2018; D'Arcy, 2024; Pickles _et al._, 69 2018; Yang _et al._, 2024). In plants, mitophagy is largely induced by nutrient starvation, 70 senescence, and mitochondrial stresses, and is involved in regulating growth, development, 71 and stress responses (Broda _et al._, 2018; Duckney et al., 2024; Kacprzak and Van Aken, 72 2022). Mitophagy is initiated by recognizing and binding of damaged or depolarized 73 mitochondria by autophagic cargo receptors, followed by the formation of autophagosome or 74 mitophagosome. In principle, a mitophagy cargo receptor should contain an AIM motif, a 75 UIM motif, or other types of motifs capable of binding to ATG8, and should localize at or 76 associate with mitochondria. These features allow a cargo receptor recognizing damaged 77 mitochondria. A number of mitophagy receptors have been identified in yeast and animals 78 (Cen et al., 2020; Kanki et al., 2009; Liu et al., 2012; Okamoto et al., 2009; Sandoval et al., 79 2008; Wei et al., 2017; Yamashita et al., 2024). Recent studies have identified several 80 putative receptors or receptor-like proteins for mitophagy in Arabidopsis (Li et al., 2022; 81 Schmitt and Van Aken, 2023; Yang et al., 2023). Among those, TraB proteins, with dual 82 mitochondrial and endoplasmic reticulum (ER) localization, interact with ATG8 to control 83 mitophagy. The _trb1 trb2_ double mutant accumulates an increased number of damaged 84 mitochondria and shows altered stress responses (Li _et al._, 2022). Moreover, a mitochondrial 85 E3 ligase SPL2 is found to degrade two mitochondrial outer membrane proteins that interact 86 with an ER protein to mediate ER-mitochondrial interaction during seedling emergence 87 (Tian et al., 2026). When challenged by carbon starvation, the mitochondria-localized FCS 88 like zinc finger (FLZ) proteins negatively regulate starvation-induced autophagy by 4 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 89 inhibiting the activation of SnRK1, a key metabolic sensor (Yang _et al._, 2023). As FLZ 90 proteins interact with ATG8 via the canonical AIM motifs, they may act as receptors for 91 carbon starvation-induced mitophagy (Schmitt and Van Aken, 2023; Yang _et al._, 2023). Also 92 in Arabidopsis, Friendly Mitochondria (Friendly or FMT), a homolog of Clustered 93 Mitochondria Protein in animals, interacts with ATG8 and is possibly localized to damaged 94 mitochondria during mitophagy (El Zawily et al., 2014; Kacprzak and Van Aken, 2023; Ma 95 et al., 2021). FMT is proposed to act as a potential cargo receptor or to mediate mitophagy of 96 depolarized mitochondria by preventing mitochondrial clustering to render autophagosomes 97 more effectively capturing damaged mitochondria (Kacprzak and Van Aken, 2023; Ma _et al._, 98 2021; Schmitt and Van Aken, 2023). During later stage of mitophagy, mitochondrial fission 99 and subsequent formation of mitophagosome require coordinated action of several key 100 players, including ELM1, DRP3, and SH3P2 in response to heat stress (Ma et al., 2025a). 101 Notably, core regulatory components of autophagy, including those for initiation, elongation 102 and translocation, are highly conserved in eukaryotes. However, cargo receptors specifically 103 targeting damaged mitochondria appear diverse in yeast, animals, and plants (Broda _et al._, 104 2018; Duckney _et al._, 2024; Schmitt and Van Aken, 2023). 105 Among numerous AIM-containing proteins, _S-_ nitrosoglutathione reductase (GSNOR) is 106 of particular interest. GSNOR proteins are highly conserved from bacteria to humans that 107 irreversibly degrade _S-_ nitrosoglutathione (GSNO), a major bioactive and reservoir form of 108 nitric oxide (NO) (Liu et al., 2001; Sakamoto et al., 2002). Thus, GSNOR plays a vital role 109 in regulating intracellular GSNO levels, thereby controlling downstream signaling processes 110 mediated by NO and protein _S-_ nitrosylation as a denitrosylase (Feng et al., 2019; Li et al., 111 2021). Loss _-_ of-function mutations in Arabidopsis _GSNOR1_, known as _gsnor1_ / _sensitive to_ 112 _hot temperatures5_ ( _hot5_ )/ _paraquat resistant2_ ( _par2_ ), cause severe growth and developmental 113 defects (Chen et al., 2009; Feechan et al., 2005; Kwon et al., 2012; Lee et al., 2008). As a 114 key regulator of intracellular GSNO levels, the biochemical activity and stability of 115 Arabidopsis GSNOR1 are in turn controlled by NO-mediated _S-_ nitrosylation (Guerra et al., 116 2016; Zhan et al., 2018). In Arabidopsis, _S-_ nitrosylation of GSNOR1 at a highly conserved 5 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 117 cysteine promotes its selective autophagic degradation mediated by a highly conserved AIM 118 motif recognized by ATG8 (Zhan _et al._, 2018). In mouse, GSNOR is suggested to regulate 119 mitophagy as a denitrosylase by modulating the _S-_ nitrosylation levels of key components of 120 mitophagy (Cirotti et al., 2021; Rizza et al., 2018; Tang et al., 2023). Because of the presence 121 of a functional AIM motif (Zhan _et al._, 2018), GSNOR1 may also act as a cargo receptor for 122 selective autophagy of other yet unidentified clients. 123 In this study, we present evidence showing that Arabidopsis GSNOR1 acts as a cargo 124 receptor binding to the outer mitochondrial membrane protein Voltage-Dependent Anion 125 Channel 3 (VDAC3) and ATG8 to target depolarized mitochondria to mitophagy in response 126 to mitochondrial stress. 127 6 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 128 **Results** 129 130 _**GSNOR1**_ **controls mitochondrial number and quality** 131 Previous studies show that Arabidopsis GSNOR1 protein is degraded via selective 132 autophagy (Zhan _et al._, 2018) and that mouse GSNOR protein is involved in regulation of 133 mitophagy by modulating de _-_ nitrosylation of key components associated with the autophagic 134 degradation of mitochondria (Rizza _et al._, 2018). Because GSNOR1 contains a highly 135 conserved AIM motif, which is critical for its own selective autophagy and also is a key 136 feature of cargo receptors for selective autophagy, we reasoned that GSNOR1 might regulate 137 mitophagy. To test this possibility, we first examined the mitochondrial phenotype of _gsnor1_ . 138 An analysis of the number of mitochondria, judged by the copy numbers of the 139 mitochondrial genome-specific genes, revealed that the copy numbers of the tested 140 mitochondrial genome-specific genes ( _COX1_, _ATP1_, _RPS4_, and _NAD6_ ) in _gsnor1_ were 141 significantly higher than those in wild type (Fig. 1a). A marginally increased copy number of 142 these genes was observed in _atg2_ and _atg5_, two autophagy deficient mutants (Fig. 1a), 143 consistent with the observation that those two mutants had more mitochondria than wild type 144 in dark-grown cotyledons by a different assay (Ma _et al._, 2021). Moreover, the accumulation 145 of VDAC (an outer mitochondrial membrane marker), COXII (an inner mitochondrial 146 membrane marker), and mtHSP70 (a matrix mitochondrial marker) proteins was at a 147 substantially higher level in _gsnor1_ than that in wild type (Fig. 1b,c). 148 To understand structural impacts of the _gsnor1_ mutation, we examined the 149 mitochondrial morphology by transmission electron microscopy of ultra-thin sections 150 prepared from the root elongation zone. In wild type, spherical or rod-shaped mitochondria 151 with well-organized cristae were observed. By contrast, irregular mitochondria with bubble 152 like cristae and the reduced number of cristae were found in _gsnor1_ (Fig. 1d,e). Analysis of 153 the surface areas of mitochondria showed that their sizes largely ranged smaller than 0.35 154 µm [2] (Groups I-III), and a small population was larger than 0.35 µm [2] (Groups IV and V). 155 While the surface areas of mitochondria in wild type ranged between 0.15–0.25 µm [2] (47.0%; 7 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 156 Group II) and smaller than 0.15 µm [2] (29.9%; Group I), the pattern was reversed in _gsnor1_ 157 with reduced population in Group II (37.5%) and increased population in Group I (34.2%) 158 (Fig. 1f). The surface areas were marginally altered in Groups III and IV between wild type 159 and _gsnor1_ (Fig. 1f). Notably, Group V (larger than 0.45 µm [2] ), constituted the smallest 160 population, was nearly undetectable in wild type, but showed a significantly higher ratio in 161 _gsnor1_ (1.7%) (Fig. 1f), further suggesting that the _gsnor1_ mutation causes irregular 162 morphology. These results suggest that GSNOR1 is crucial for the maintenance of structure 163 and membrane integrity of mitochondria. 164 Because GSNOR, a highly conserved enzyme catalyzing the degradation of GSNO, is 165 an important regulator of NO homeostasis (Liu _et al._, 2001; Sakamoto _et al._, 2002), and 166 Arabidopsis _gsnor1/hot5/par2_ allelic mutants show a remarkably increased NO level (Chen 167 _et al._, 2009; Feechan _et al._, 2005; Lee _et al._, 2008), we then asked if the observed 168 abnormalities of mitochondria in _gsnor1_ were associated with the increased NO level or 169 alterations in protein _S_ -nitrosylation levels. To address this question, we examined the 170 mitochondrial phenotype of _nox1_ and _gsnor1 rog1_, two mutants that both accumulate 171 excessive amount of NO (Chen et al., 2020; He et al., 2004). _NO overproducer 1_ ( _nox1_ ) is 172 characterized as a mutant containing an elevated NO level to repress floral transition (He _et_ 173 _al._, 2004). ROG1 (REPRESSOR OF GSNOR1) is proposed as a transnitrosylase catalyzing 174 protein _S_ -nitrosylation, and the _rog1_ mutation restores defective growth and development of 175 the _gsnor1_ mutant to nearly normal, accompanying with proteome-wide reduction of protein 176 _S_ -nitrosylation in the _gsnor1_ mutant background (Chen _et al._, 2020). Analysis of 177 mitochondrial cristae morphology revealed that the _nox1_ mutant largely showed a wild type 178 like phenotype with a higher ratio of the misshapen mitochondria (7/122; 5.7%) compared to 179 wild type (2/136; 1.5%; Fig. 1d,e), suggesting that _NOX1_ may play a minor role in 180 mitochondrial quality control. By contrast, the _gsnor1_ _rog1_ (23/100; 23.0%) double mutant 181 displayed a phenotype similar to that of _gsnor1_ (36/163; 22.1%; Fig. 1d,e). Consistent with 182 these observations, _gsnor1_ _rog1_ showed a phenotype of the mitochondrial area size similar 183 to that of _gsnor1_, whereas _nox1_ showed a pattern similar to that of wild type but opposite to 8 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 184 that of _gsnor1_ (Fig. 1f). These results suggest that NO and NO-mediated protein _S_ 185 nitrosylation are not directly related to the dysfunctional mitochondria observed in _gsnor1_ . 186 Taken together, we conclude that _GSNOR1_ plays an important role in regulating 187 mitochondrial number and mitochondrial quality control. 188 189 _**GSNOR1**_ **positively controls mitophagy** 190 Data presented above show that the _gsnor1_ mutant accumulates excessive number of 191 damaged mitochondria, suggestive of impaired mitochondrial quality control. Under 192 nitrogen-deficient conditions, the _gsnor1_ mutant showed a phenotype similar to that of wild 193 type, including significantly inhibited growth and non-specific degradation of inner and 194 outer mitochondrial membrane marker proteins, whereas the macroautophagy-related _atg5_ 195 and _atg7_ mutants were more sensitive to nitrogen deficiency (Extended Data Fig. 1a-e). 196 Dark-induced carbon starvation has been reported to trigger mitophagy (Kacprzak and Van 197 Aken, 2022). No apparent phenotypic difference was observed between wild type and the 198 _gsnor1_ mutant under the assay conditions (Extended Data Fig. 1a-d). These results suggest 199 that _GSNOR1_ does not have detectable effects in nitrogen- or dark-induced specific 200 macroautophagy under the assay conditions. 201 To explore possible regulatory role of GSNOR1 in mitophagy, we tested the response of 202 the _gsnor1_ mutant to 2,4-dinitrophenol (DNP), a specific uncoupler of oxidative 203 phosphorylation, which disrupts the electrochemical potential of the inner mitochondrial 204 membrane, leading to the depolarized mitochondria targeting to degradation via _ATG5_ 205 dependent mitophagy (Kacprzak and Van Aken, 2023; Li _et al._, 2022; Ma _et al._, 2021). 206 Consistent with previous observations (Kacprzak and Van Aken, 2023; Li _et al._, 2022; Ma _et_ 207 _al._, 2021), when treated with DNP, the accumulation of the inner and outer mitochondrial 208 membrane marker proteins (VDAC and COXII) was reduced in wild type, characteristics of 209 the uncoupler-induced mitophagy. In the _atg5_ and _gsnor1_ mutants, while the accumulation of 210 VDAC and COXII was comparable or slightly increased compared to that in wild type, the 211 DNP-induced reduction of these proteins was compromised or slightly increased (Fig. 2a,b), 9 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 212 suggesting that DNP-induced mitophagy is genetically dependent on _GSNOR1_ . 213 To monitor dynamic changes of mitophagy in planta, we generated transgenic plants 214 carrying a mitochondrial marker _Mito-GFP_ (Kohler et al., 1997) in the wild type and _gsnor1_ 215 backgrounds, referred to as _Mito-GFP_ /WT and _Mito-GFP_ / _gsnor1_, respectively. A key 216 hallmark of mitophagy is the depolarization of mitochondrial membrane, which is 217 structurally distinguishable from normal mitochondria. Normal mitochondria, but not 218 depolarized mitochondria, can be specifically labelled by tetramethylrhodamine ethyl ester 219 (TMRE), a mitochondrial membrane potential-sensitive fluorescence dye (Qiao et al., 2025; 220 Roussos et al., 2025; Zhou et al., 2016). Analysis of the root elongation zone showed that 221 Mito-GFP and TMRE fluorescence signals were overlapped in nearly all foci in wild type 222 (Fig. 2c), suggesting that the majority of mitochondria is normal. In the _gsnor1_ mutant, a 223 large portion of mitochondria was depolarized, showing largely as green foci when Mito 224 GFP and TMRE fluorescence signals were merged (Fig. 2d), consistent with the data 225 presented above that the _gsnor1_ mutation causes the accumulation of damaged mitochondria. 226 When treated with DNP, depolarized mitochondria were dramatically increased in wild type; 227 however, the _gsnor1_ mutant, which contained a large portion of depolarized mitochondria, 228 did not show apparent responses to DNP (Fig. 2c-e). These results suggest that the _gsnor1_ 229 mutation causes the accumulation of depolarized mitochondria in a manner similar to that 230 induced by DNP. We noticed that _Mito-GFP_ /WT plants had substantially less mitochondria 231 than that observed in _Mito-GFP_ / _gsnor1_ plants (Fig. 2c-e), providing an additional line of 232 evidence supporting a role of _GSNOR1_ in controlling mitochondrial number. 233 To reinforce the above observations, the _Mito-GFP_ transgenic plants were treated with 234 concanamycin A (ConA), a vacuolar H [+] -ATPase inhibitor that blocks the autophagic flux, 235 thus allowing the visualization of fluorescence in the vacuole (Tamura et al., 2003; 236 Thompson et al., 2005). Treatment with ConA caused the accumulation of depolarized 237 mitochondria in both wild type and _gsnor1_ plants (Fig. 2c–e), suggesting that the vacuole is 238 the final destination for the damaged mitochondria. Co-treatment of wild type with DNP and 239 ConA showed similar phenotypes as treatments with individual reagents. However, the _Mito-_ 10 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 240 _GFP_ / _gsnor1_ plants had a high level of depolarized mitochondria, nearly comparable to that 241 of DNP-treated wild type plants, and showed little or no response to DNP or ConA (Fig. 2c 242 e). In a mitophagic flux assay, the cleavage of Mito-GFP was enhanced upon DNP treatment 243 in wild type, indicated by an increased GFP/Mito-GFP ratio; however, the cleavage was 244 slightly suppressed in the _gsnor1_ mutant, indicated by a deceased GFP/Mito-GFP ratio upon 245 DNP treatment (Fig. 2f,g). Moreover, partial cleavage of GFP-TOM20-3 was detected in 246 wild type, and the cleavage was enhanced by DNP treatment in wild type but not _gsnor1_ 247 (Extended Data Fig. 2a). These results suggest that the _gsnor1_ mutation causes impaired 248 mitophagy. 249 Transmission electron microscopy revealed that damaged mitochondria were engulfed in 250 single-layer autophagosomes in wild type treated with DNP but not in the control samples 251 (Fig. 2h). In the _gsnor1_ mutant, while damaged mitochondria were found in both the control 252 and DNP-treated samples, a significant portion of damaged mitochondria were not engulfed 253 in autophagosomes under both control and DNP-treated conditions (Fig. 2h), suggesting that 254 GSNOR1 plays a critical role in targeting damaged mitochondria to autophagosomes. 255 Collectively, these results suggest that mitophagy is genetically dependent on _GSNOR1_ . 256 257 **GSNOR1 interacts with ATG8 to engulf depolarized mitochondria to autophagosomes** 258 Given the genetic dependency of mitophagy on _GSNOR1_, we next explored the function of 259 GSNOR1 protein in the process. We generated transgenic plants carrying 260 _pGSNOR1_ :: _GSNOR1-GFP_ (referred to as _GSNOR1-GFP_ hereafter) and _p35S_ :: _IDH1_ 261 _mCherry_ (a mitochondrial marker gene) transgenes in the _gsnor1_ mutant background. 262 Similar to the observation previously reported (Zhan _et al._, 2018), the _GSNOR1-GFP_ 263 transgene fully rescued the _gsnor1_ mutant phenotype. Under normal growth conditions, 264 GSNOR1-GFP distributed uniformly in the cytosol, barely co-localized with IDH1-mCherry 265 (Fig. 3a). ConA treatment caused substantial co-localization of GSNOR1-GFP and IDH1 266 mCherry (Extended Data Fig. 2b). Similarly, when treated with ConA, co-localization of 267 GSNOR1-GFP with MitoTracker Red-labeled mitochondria (Ma _et al._, 2021; Narendra et 11 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 268 al., 2008) was readily detected (Extended Data Fig. 2c). When treated with DNP, GSNOR1 269 GFP was also observed in punctuated foci co-localized with IDH1-mCherry in a time 270 dependent manner (Fig. 3a,b), suggesting that GSNOR1 is translocated to depolarized 271 mitochondria during the onset of DNP-induced mitophagy. When treated with both DNP and 272 ConA to block the autophagic flux, the co-localization of GSNOR1-GFP and IDH1-mCherry 273 formed enlarged focal structures that were readily detected (66.6% ± 14.8% co-localization; 274 Fig. 3a). In a live imaging analysis, we found that GSNOR1-GFP was dynamically localized 275 with the IDH1-mCherry-labeled mitochondria (Supplementary Video 1). Collectively, these 276 results suggest that GSNOR1 is preferentially localized to depolarized mitochondria. 277 During autophagy, a key step is the link of ATG8 proteins to selective targets subjected 278 to autophagic degradation. Previous studies showed that ATG8e interacted with GSNOR1 279 via an AIM (residues 152-155, YTVV) located in a  -sheet of GSNOR1 (Zhan _et al._, 2018). 280 A closer inspection identified an additional highly conserved AIM-like motif in GSNOR1 281 (residues 269-272; FECI) located in a  -sheet (Extended Data Fig. 3a). A pull-down 282 experiment showed that substitution mutations in both AIM motifs significantly reduced the 283 interaction between GSNOR1 and ATG8e (Extended Data Fig. 3b,c). When transiently 284 expressed in Arabidopsis protoplasts, co-localization of GSNOR1-mCherry and GFP 285 ATG8a, another member of the ATG8 family proteins, was observed (Fig. 3c and Extended 286 Data Fig. 3c). In the _GFP-ATG8a_ transgenic plants, DNP substantially enhanced the 287 interaction between GSNOR1 and GFP-ATG8a in a dose-dependent manner as revealed by 288 co-immunoprecipitation assay (Fig. 3d). These results suggest that GSNOR1 interacts with 289 ATG8 proteins, targeting depolarized mitochondria to autophagosomes during mitochondrial 290 stress. 291 To examine the physiological significance of these AIMs, we generated transgenic 292 plants carrying double ( _pGSNOR1_ :: _GSNOR1_ _[Y152A V155A F269A I272A]_ _-mCherry_ ; referred to as 293 _GSNOR1_ _[m4]_ _-mCherry_ hereafter) or single ( _pGSNOR1_ :: _GSNOR1_ _[V155A]_ _-GFP_ ) mutations in the 294 _gsnor1_ background (we failed to obtained single mutants in the second AIM). While 295 _GSNOR1_ _[V155A]_ _- GFP_ fully rescued the developmental defects of _gsnor1_ under the assay 12 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 296 condition, _GSNOR1_ _[m4]_ _-mCherry_ transgenic plants showed a _gsnor1_ -like phenotype (Fig. 3e 297 and Extended Data Fig. 3d), indicating that these two AIMs are essential for the GSNOR1 298 function. The nonfunctional GSNOR1 [m4] -mCherry protein could be attributed to its 299 instability (Fig. 3f). We noticed that the accumulation of GSNOR1 [V155A] -GFP (line #20; 300 relative level: 0.99) was comparable to that of endogenous GSNOR1 level (relative level: 301 1.0), approximately one-fold over that of GSNOR1 [m4] -mCherry (line #40; relative level: 302 0.40) (Fig. 3f). However, these two transgenic proteins functioned completely different in 303 rescuing the _gsnor1_ mutant phenotype (Fig. 3e and Extended Data Fig. 3d), implying that the 304 _GSNOR1_ _[m4]_ mutations in both AIM motifs may not only cause protein instability, but also 305 cause the loss of the biochemical activity of the protein. The regulatory mechanism of 306 GSNOR1 [m4] protein remains explored. 307 308 **GSNOR1 interacts with outer mitochondrial membrane-localized VDAC proteins** 309 Proteins containing AIMs are considered as targets of selective autophagy and, in many 310 cases, also act as receptors for cargos subjected to selective autophagy (Gross _et al._, 2025; 311 Liu et al., 2024). Because GSNOR1 contains two functional AIMs and plays a critical role in 312 mitophagy, we speculated that GSNOR1 might also act as a cargo receptor to mediate 313 selective autophagy of mitochondria. If this is true, GSNOR1 should interact with a 314 mitochondrion-localized protein, preferably localized at the outer membrane accessible by 315 the cytosolic GSNOR1. In a yeast two-hybrid screen, we identified 25 putative GSNOR1 316 interacting proteins, including VDAC2 (1 clone) and VDAC3 (11 clones) (Supplementary 317 Table 1). VDAC proteins, also known as porins, are a small family of highly conserved and 318 abundant proteins localized on the outer mitochondrial membrane, which are also used as 319 markers for the outer mitochondrial membrane (Doan et al., 2019; Kanwar et al., 2021; 320 Najbauer et al., 2021; Oh et al., 2022; Shoshan-Barmatz et al., 2010) (Extended Data Fig. 4). 321 In the Arabidopsis genome, 5 members have been identified, namely _VDAC1_ through 322 _VDAC5_, of which _VDAC5_ was considered as a pseudogene (Lee et al., 2009; Tateda et al., 323 2011). The remaining four members of Arabidopsis VDACs showed 50%–68% homology. 13 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 324 Notably, the positions of α-helices and β-sheets of VDACs are highly conserved across the 325 plant kingdom and animal kingdom (Extended Data Fig. 4), implying that these structures 326 are functionally important. Mutations in the four Arabidopsis _VDAC_ genes ( _VDAC1_ through 327 _VDAC4_ ) caused a variety of developmental abnormalities (Robert et al., 2012; Tateda _et al._, 328 2011). 329 When recapitulated in yeast cells, the interaction between GSNOR1/BD and 330 VDAC3/AD was detected but not in other three VDAC proteins, and all VDAC/AD 331 displayed varying degrees of auto-activated transcription activity (Extended Data Fig. 5a). In 332 a pull-down experiment, the interaction of His-GSNOR1 recombinant protein with GST 333 VDAC recombinant proteins was detected at varying degrees, of which GST-VDAC3 334 showed the strongest interaction (Fig. 4a), consistent with the result that more VDAC3 335 clones were identified in the yeast two-hybrid screen (Supplementary Table 1). A co 336 immunoprecipitation experiment revealed that GSNOR1-FLAG was complexed with VDAC 337 proteins as detected with polyclonal anti-VDAC antibodies that recognized VDAC1 through 338 VDAC4 under normal growth conditions, and the interaction was enhanced by DNP (Fig. 339 4b). Moreover, the interaction of GSNOR1 with all four members of VDACs was also 340 detected by the luciferase complementation experiment and bimolecular fluorescence 341 complementation when transiently expressed in tobacco leaves (Fig. 4c,d). Notably, the co 342 localized GSNOR1 and VDAC proteins were observed in punctuated foci when transiently 343 expressed in tobacco leaves and Arabidopsis protoplasts (Fig. 4d,e). These results suggest 344 that GSNOR1 interacts with the outer mitochondrial membrane-localized VDAC family 345 proteins, which may act to monitor the quality of mitochondria under normal growth 346 conditions and promote mitophagy in response to stresses. Because VDAC3 showed the 347 strongest interaction with GSNOR1, we analyzed VDAC3 in most experiments hereafter. 348 349 **VDAC3 interacts with GSNOR1 and ATG8 via its N-terminal region** 350 If VDAC proteins indeed play a role in mediating mitophagy, they should directly or 351 indirectly interact with ATG8. In a pull-down experiment, we found that His-ATG8a 14 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 352 recombinant protein directly interacted with GST-VDAC proteins (Fig. 5a). The interaction 353 between VDAC and GFP-ATG8a proteins was also detected in a co-immunoprecipitation 354 assay in the _GFP-ATG8a_ transgenic plants (Fig. 5b). In a bimolecular fluorescence 355 complementation assay, ATG8a was found to interact with VDAC proteins by forming 356 punctuated co-localization signals (Fig. 5c and Extended Data Fig. 6a). When transiently 357 expressed in Arabidopsis protoplasts, mCherry-VDAC3 and GFP-ATG8a were found to co 358 localize in punctuated foci (Fig. 5d). In a proteome-wide analysis, VDAC2 and VDAC3 359 were identified to interact with ATG8 via their UDS (UIM-docking site) and LDS 360 (LIR/AIM-docking site) motifs, respectively (Marshall _et al._, 2019). Mutations in an LDS of 361 ATG8a (Y65-L66) caused reduced interaction with both VDAC2 and VDAC3 (Extended 362 Data Fig. 6b-d). However, mutations in a UDS of ATG8a (I92-F93-V94) showed enhanced 363 interactions with VDAC2 and VDAC3 for unknown reasons (Extended Data Fig. 6b-d). 364 These results suggest that VDAC3 proteins interact and co-localize with ATG8. 365 To map the regions interacting with ATG8a and GSNOR1, we generated truncated 366 mutations of VDAC3 (Fig. 5e). When transiently expressed in Arabidopsis protoplasts, we 367 noticed that VDAC-eGFP fusion proteins showed diffused subcellular localization patterns 368 but not specifically localized in mitochondria for unknown reasons. We therefore used 369 eGFP-VDAC fusion proteins in this study. We found that eGFP-VDAC3, eGFP-D1, and 370 eGFP-D5 fusion proteins were localized to mitochondria in a punctate pattern and partially 371 overlapped with MitoTracker Red, whereas eGFP-D2, eGFP-D3, and eGFP-D4 showed a 372 diffuse cytosolic distribution (Fig. 5f and Extended Data Fig. 6e), suggesting that the D1 and 373 D5 deletions do not have detectable effects on the subcellular localization of the fusion 374 proteins. In a GST pull-down assay, the deletion of residues 1-34 (D1) dramatically reduced 375 the interaction between His-ATG8a and GST-VDAC3 (Fig. 5g), suggesting that this region is 376 important for the interaction. Consistent with this result, a recent study identified functional 377 AIM motifs in the N-terminal region of VDAC3 (Ma et al., 2025b). Deletion of additional 378 sequences (D2 and D3) showed a phenotype similar to that of D1. The D4 mutant, 379 containing only the carboxyl terminal 41 residues (residues 203-274), showed the enhanced 15 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 380 interaction with His-ATG8a, while the deletion of this region (D5) did not have detectable 381 effect on the interaction (Fig. 5e,g). This result might be attributed to a cryptic signal or 382 structure in D4. The interactions of His-GSNOR1 with these mutant proteins showed a 383 pattern similar to that of His-ATG8a (Fig. 5h and Extended Data Fig. 5b). However, in 384 contrast to that of His-ATG8a, D4 did not show an increased interaction with His-GSNOR1, 385 implying that these two proteins may bind to VDAC3 with possibly different structural 386 requirements. Based on the resolved 3-D structures of human and mouse VDAC proteins 387 (Hiller et al., 2008; Ujwal et al., 2008), a putative structure model of Arabidopsis VDAC3 388 was constructed, in which the region spanning residues 1-34 contains two α-helices and the 389 first β-sheet of a barrel structure that is formed with 19 anti-parallel β strands (Fig. 5i). To 390 narrow down the interacting motifs of VDAC3, we made three additional mutants with 391 smaller deletions, designated as D6 through D8, which covered the putative α1, α2, or β1 392 regions, respectively (Fig. 5e). All these three mutants showed substantially reduced 393 interaction with His-ATG8a (Fig. 5j). Notably, these three mutants also nearly abolished 394 their interactions with His-GSNOR1 (Fig. 5k). These results suggest that the amino terminal 395 region (residues 1-34) of VDAC3 is critical for its interaction with both ATG8a and 396 GSNOR1. Collectively, these results suggest that VDAC3 interacts with both GSNOR1 and 397 ATG8a through the first 34 amino acid residues. 398 399 **The GSNOR1-VDAC3-ATG8 complex modulates mitochondrial stress responses** 400 Data presented above showed that GSNOR1, ATG8a, and VDAC3 interact with one another. 401 We next asked if these three proteins are in a same complex. To test this possibility, we co 402 transformed _FLAG-mCherry-VDAC3_ and _HA-BFP-ATG8a_ transgenes driven by a 35S 403 promoter into protoplasts prepared from _GSNOR1-GFP_ transgenic plants. Co-localization of 404 these three proteins were detected in punctuated foci (Fig. 6a), indicating that these proteins 405 form a complex in autophagosomes. 406 The _gsnor1_ mutant shows significantly altered responses to a variety of stress (Chen _et_ 407 _al._, 2009; Feechan _et al._, 2005; Lee _et al._, 2008). Given that GSNOR1 may act as a cargo 16 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 408 receptor to mediate mitophagy, we examined the response of the _gsnor1_ mutant to DNP. 409 Seven-day-old seedlings grown under normal conditions were treated with 10 µM DNP for 410 an additional 9 days, and the relative fresh weight was analyzed. Upon treatment with DNP, 411 the relative fresh weight of the _gsnor1_ and _atg5_ mutants was significantly lower than wild 412 type (Fig. 6b,c). In a growth recovery experiment, seedlings were treated with DNP for 4 hr 413 and then transferred to normal growth conditions culturing for 48 hr. Analysis of root 414 elongation post DNP treatment showed that _gsnor1_ and _vdac3_ mutants had lower recovery 415 growth rate than that of wild type, while _gsnor1 rog1_ and _atg5_ mutants had marginally 416 reduced recovery growth rates (Fig. 6d-f). 417 Consistent with the above results, transmission electron microscopy revealed severely 418 damaged mitochondrial cristae in the _vdac3_ mutant (Fig. 6g), similar to that observed in the 419 _gsnor1_ mutant. However, these damaged mitochondria were not frequently associated with 420 double-membrane autophagic structures under the assay conditions. Taken together, these 421 results suggest that the GSNOR1-VDAC3 complex regulates stress responses via mitophagy. 422 423 **Discussion** 424 425 In this study, we present multiple lines of evidence, obtained from genetic, cell biological, 426 and biochemical studies, showing that GSNOR1 may act as a cargo receptor docking 427 damaged mitochondria for mitophagic degradation via interacting with VDAC3 and ATG8. 428 First, mutations in _GSNOR1_ and _VDAC3_ cause a similar phenotype characterized as the 429 accumulation of dysfunctional mitochondria and hypersensitive to the oxidative 430 phosphorylation uncoupler DNP, thus demonstrating the genetic dependency of mitophagy 431 on these two genes. Second, GSNOR1 dynamically and preferentially localizes to 432 depolarized mitochondria, which is a prerequisite of a regulator for mitophagy, 433 accompanying with the co-localization of GSNOR1, VDAC3, and ATG8a in 434 autophagosomes. These observations directly link GSNOR1 and VDAC3 to ATG8-mediated 435 mitophagy. Third, GSNOR1 interacts and co-localizes with both ATG8 and VDAC3 in 17 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 436 autophagosomes, characteristics of a cargo receptor of selective autophagy. Based on these 437 observations, we propose that GSNOR1 functions as a cargo receptor, interacting with the 438 outer mitochondrial membrane protein VDAC3 and targeting the depolarized mitochondria 439 to ATG8, eventually docking to autophagosomes for degradation (Fig. 7). These findings not 440 only identify a previously unrecognized mechanism regulating mitophagy, a highly 441 conserved process in eukaryotes, and also uncover an unexpected role of GSNOR1 that has 442 long been known as an enzyme degrading GSNO. 443 VDACs are highly conserved proteins localized on the outer mitochondrial membrane. 444 In animals, the leafhopper VDAC1 interacts with ATG8 to mediate mitophagy upon insect 445 borne virus infection (Chen et al., 2023) and the localization of the E3 ligase Parkin to 446 depolarized mitochondria causes poly-ubiquitylation of VDAC1 at the onset of mitophagy 447 (Geisler et al., 2010). Observations made in this study show that the Arabidopsis _vdac3_ 448 mutant is hypersensitive to DNP and has malformed mitochondrial structures, accompanying 449 with the physical interaction of VADC3 with ATG8, implying a possible role of VADC3 in 450 regulating mitophagy. Consistent with this notion, the Arabidopsis VDACs have been 451 identified as ATG8-interacting proteins in a proteomic study (Marshall _et al._, 2019). 452 Moreover, a recent study shows that VDAC1, VDAC2, and VDAC3 acts as mitophagy 453 receptors recruiting ATG8 to depolarized mitochondria under DNP-induced mitochondrial 454 stress (Ma _et al._, 2025b). Biochemically, VDAC3 directly interacts with both ATG8 and 455 GSNOR1 via its amino-terminal region. The human VDAC1 (hVDAC) structure is 456 characterized as a tunnel-like structure comprising 19 anti-parallel β-sheets, connecting to a 457 short and flexible amino-terminal region (PDB code: 2K4T) (Hiller _et al._, 2008). Analog to 458 the amino-terminal region of hVDAC1, a short stretch of Arabidopsis VDAC3 (residues 1 459 34) is critical for its interaction with both GSNOR1 and ATG8a, suggesting that GSNOR1 460 and ATG8a may bind to the outer mitochondrial membrane-localized VDAC3 in a 461 cooperative fashion. 462 While it is not unexpected for the regulatory role of VDAC proteins (Broda _et al._, 463 2018; Geisler _et al._, 2010; Ma _et al._, 2025b), it is somewhat surprising to find that GSNOR1 18 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 464 functions as a receptor to mediate mitophagy, considering its canonical biochemical activity 465 as an _S_ -nitrosoglutathione reductase, also known as a glutathione-dependent formaldehyde 466 dehydrogenase (Liu _et al._, 2001; Sakamoto _et al._, 2002). An interesting clue comes from the 467 observation that while _HOT5/GSNOR1_ is a key regulator of thermotolerance (Lee _et al._, 468 2008), heat stress induces piecemeal mitophagy (Ma _et al._, 2025a). However, it remains 469 elusive if _HOT5/GSNOR1_ is involved in piecemeal mitophagy. GSNOR1 contains two 470 AIMs, one of which has been shown important for its own selective autophagy (Lubega et 471 al., 2021; Zhan _et al._, 2018). Proteins containing functional AIMs are also considered as 472 cargo receptors to target specific proteins or organelles to selective autophagy (Faruk et al., 473 2021; Stephani and Dagdas, 2020), and GSNOR1 is therefore a strong candidate as an 474 autophagy receptor. While evidence presented in this study suggests that GSNOR1 acts as a 475 cargo receptor to mediate mitophagy, it seems difficult to understand why a GSNO 476 degrading enzyme moonlights as a mitophagy receptor. Stress-induced burst of NO and 477 reactive oxygen species (ROS) cause the damage of mitochondria (Dai et al., 2020; Huang et 478 al., 2016; Zhang et al., 2022) and, on the other hand, _S_ -nitrosylation (Guerra _et al._, 2016; 479 Zhan _et al._, 2018) and oxidation (Sun et al., 2025) also negatively regulate the GSNOR1 480 activity. As a cytosol dominantly localized protein, GSNOR1 is presumably capable of 481 rapidly sensing and subsequently responding to cellular redox alterations by adjusting its 482 own enzymatic activity and evoking a mechanism clearing damaged cellular components in 483 response to stresses. Echoed to this scenario, an analogous mechanism has been well 484 appreciated in glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a key enzyme 485 catalyzing the production of 2,3-bisphosphoglyceric acid from glyceraldehyde-3-phosphate 486 in glycolysis. In addition to its role in glycolysis, GAPDH is also involved in the regulation 487 of a wide range of cellular activities, including autophagy, cell death, cell differentiation, and 488 responses to carbon metabolic flux and other stimuli (Cornett et al., 2022; Dodson et al., 489 2013; Feng et al., 2024; Henry et al., 2015). Analogously, GSNOR1 may play dual and 490 distinctive roles in NO signaling and mitophagy in a fashion similar to GAPDH. Moreover, 491 considering the pleiotropic phenotype of the _gsnor1_ mutant and the nature of a cargo 19 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 492 receptor, it is predictable that GSNOR1 may function as a receptor of other yet unidentified 493 clients during selective autophagy. 494 Given that GSNOR1 catalyzes the degradation of GSNO and that the _gsnor1_ mutant 495 accumulates excessive amount of NO and causes proteome-wide over-nitrosylation (Chen _et_ 496 _al._, 2020; Chen _et al._, 2009; Feechan _et al._, 2005; Hu et al., 2015; Lee _et al._, 2008), one 497 would expect that NO or NO-mediated _S_ -nitrosylation may also play an important role in 498 regulating mitophagy. In this regard, a large portion of mitochondrial proteins have been 499 identified as _S_ -nitrosylated proteins by proteomics studies (Chang et al., 2014; Foster and 500 Stamler, 2004; Hu _et al._, 2015), implying possible involvement of these potential targets in 501 the NO-regulated mitophagy. A line of more direct evidence comes from the observation that 502 the knockout of the mouse _GSNOR_ gene, similar to that in Arabidopsis, also causes impaired 503 mitophagy, which is attributed to excessive _S_ -nitrosylation of Drp1 and Parkin (Rizza _et al._, 504 2018). Because both Drp1 and Parkin homologs are not found in plants, it is not likely a 505 similar mechanism operating mitophagy in plant cells. Therefore, while the lack of the 506 GSNOR activity evokes mitophagy in both mammals and plants, different mechanisms are 507 found in these two systems, in which GSNOR acts as a denitrosylase and a cargo receptor, 508 respectively. In fact, the observation that minimal mitochondrial phenotype is observed in 509 the _nox1_ mutant rules out the possibility of a major role of NO in regulating mitophagy 510 under the assay conditions. Moreover, the _rog1_ mutation genetically suppresses the _gsnor1_ 511 phenotype in multiple aspects, including the proteome-wide over-nitrosylation of _gsnor1_ . 512 However, the restoration or reduction of the _S_ -nitrosylation level of _gsnor1_ to near the wild 513 type level by _rog1_ does not suppress the defective mitophagy phenotype, implying that 514 neither NO nor NO-mediated protein _S_ -nitrosylation plays a decisive role in regulating 515 mitophagy under the assay condition. Despite of this notion, we do not exclude the 516 possibility that GSNOR1 also functions as a denitrosylase to regulate mitophagy in plants. 517 Instead, it will be of great interests to identify _S_ -nitrosylated proteins regulating mitophagy 518 in plants. Lastly, the interactions among ATG8, GSNOR1, and VADC3 are detected under 519 normal growth conditions, supporting a notion that the machinery may act to monitor the 20 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 520 quality of mitochondria under both permissive and nonpermissive growth conditions. 521 In conclusion, discoveries made in this study identify a unique and unexpected 522 mechanism regulating mitophagy by GSNOR1 in plants. Because all major components in 523 the pathway, including GSNOR, VDAC, and ATG8, are highly conserved, a similar 524 regulatory scheme may also be functional to execute mitophagy in animals as well. 525 21 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 526 **Methods** 527 528 **Plant materials and growth conditions** 529 The Columbia-0 (Col-0) accession of _Arabidopsis_ was used in this study. The _atg2_, _atg5_, 530 _atg7_, _nox1_, and _vdac3_ (SALK_127899) mutant seeds were provided by Dingzhong Tang, 531 Yikun He, and ABRC, respectively (He _et al._, 2004; Wang et al., 2011). The _gsnor1_ and 532 _gsnor1 rog1_ mutants have been previously described (Chen _et al._, 2020; Chen _et al._, 2009). 533 Transgenic _Arabidopsis_ plants were generated using the _Agrobacterium_ -mediated floral dip 534 transformation method (Bechtold and Pelletier, 1998). 535 Seeds were surface-sterilized and sown on 1/2 MS medium agar plates with 1% sucrose 536 and imbibed at 4°C for 2 days. The plates were then placed at 21°C under a 16-h light/8-h 537 dark photoperiod. 538 539 **Chemical treatment** 540 Chemical treatments were performed by transferring seedlings (typically 7-day-old unless 541 otherwise indicated) into liquid 1/2 MS medium containing 50  M DNP (Sigma-Aldrich, 542 Cat # D198501) or 1  M ConA (MedChemExpress, Cat # HY-N1724). Equivalent 543 concentrations of the solvent dimethyl sulfoxide (DMSO) were used as controls in all 544 experiments where applicable. 545 In 2,4-dinitrophenol (DNP) treatment and recovery assays, Arabidopsis seedlings were 546 grown on 1/2 MS medium agar plates for 7 days and then transferred into liquid 1/2 MS 547 medium supplemented with either DMSO (control) or 50 μM DNP for 4 h. Following 548 treatment, the seedlings were transferred onto fresh 1/2 MS medium agar plates and allowed 549 to recover for 2 days. Images were captured using a digital camera, and root length was 550 measured using ImageJ software. 551 552 **Chlorophyll extraction and measurement** 553 Arabidopsis seedlings (fresh weight ~ 10 mg) were collected and incubated in 1 mL of 95% 22 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 554 ethanol for 48 h in the dark at 4°C to extract chlorophyll. The extracts were centrifuged at 555 12,000 rpm for 10 min at 4°C, and the absorbance of the supernatant was measured at 665 556 nm and 649 nm. Total chlorophyll concentration was calculated using the following 557 equation: total chlorophyll (mg/L) = 6.63 _A_ 665 + 18.08 _A_ 649 . 558 559 **Plasmid construction** 560 The mitochondria-targeting sequence (Mito) derived from yeast _COXIV_ transit-peptide (5’ 561 ATGTTGTCACTACGTCAATCTATAAGATTTTTCAAGCCAGCCACAAGAACTTTGTG 562 TAGCTCTAGATATCTGCTTGAGCAAAAACCC-3’) (Kohler _et al._, 1997) was synthesized 563 and cloned into the _Xba_ I and _Sal_ I sites of pCAMBIA1300-GFP to generate pCAMBIA1300 564 35S::Mito-GFP vector. The coding sequences (CDS) of _ATG8a_, _ATG8e_, _GSNOR1_, _IDH1_, 565 _VDAC1_, _VDAC2_, _VDAC3_, and _VDAC4_ were PCR-amplified from wild-type Col-0 cDNA. 566 The resulting _IDH1_ and _ATG8a_ CDS were cloned into pCAMBIA3301-35S-mCherry and 567 pCAMBIA1300-35S-GFP to generate pCAMBIA3301-35S::IDH1-mCherry and 568 pCAMBIA1300-35S::GFP-ATG8a vectors, respectively. 569 For recombinant protein expression, the CDS of _GSNOR1_ was cloned into the _Sal_ I and 570 _Eco_ RI sites of pET28a or pGEX-4T-1 to generate pET28a-GSNOR1 and pGEX-4T-1 571 GSNOR1 vectors, respectively. Site-directed mutagenesis was performed using appropriate 572 PCR primer pairs to generate the following constructs: pGEX-4T-1-GSNOR1 [Y152A], pGEX 573 4T-1-GSNOR1 [V155A], pGEX-4T-1-GSNOR1 [Y152A V155A], pGEX-4T-1-GSNOR1 [F269A], pGEX 574 4T-1-GSNOR1 [I272A], and pGEX-4T-1-GSNOR1 [F269A I272A] . The CDS of _VDAC1_, _VDAC2_, 575 _VDAC3_, and _VDAC4_ were cloned into the _Bam_ HI and _Not_ I sites of pGEX-4T-1 to generate 576 pGEX-4T-1-VDAC1, pGEX-4T-1-VDAC2, pGEX-4T-1-VDAC3, and pGEX-4T-1-VDAC4 577 vectors. Truncated mutants of VDAC3 were generated using appropriate PCR primer pairs to 578 produce pGEX-4T-1-VDAC3 [D1] through pGEX-4T-1-VDAC3 [D8] constructs. The _VDAC2_ and 579 _VDAC3_ CDS were cloned into the _Bam_ HI and _Sac_ I sites of pMal-C2X to generate pMal 580 C2X-VDAC2 and pMal-C2X-VDAC3 vectors. The CDS of _ATG8a_ and _ATG8e_ were cloned 581 into the _Bam_ HI and _Sal_ I sites of pET28a or pGEX-4T-1 to generate pET28a-ATG8a, 23 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 582 pET28a-ATG8e, and pGEX-4T-1-ATG8a vectors. Site-directed mutagenesis was performed 583 to generate pGEX-4T-1-ATG8a [YL65AA] and pGEX-4T-1-ATG8a [IFV92AAA] constructs. 584 For luciferase complementation assays, CDS of _GSNOR1_, _VDAC1_, _VDAC2_, _VDAC3_, 585 and _VDAC4_ were cloned into _Kpn_ I and _Sal_ I sites of pCAMBIA1300-NLUC or 586 pCAMBIA1300-CLUC to generate pCAMBIA1300-GSNOR1-NLUC, pCAMBIA1300 587 CLUC-VDAC1, pCAMBIA1300-CLUC-VDAC2, pCAMBIA1300-CLUC-VDAC3, and 588 pCAMBIA1300-CLUC-VDAC4 vectors. 589 For bimolecular fluorescence complementation (BiFC) assays, CDS of _ATG8a_, _VDAC1_, 590 _VDAC2_, _VDAC3_, and _VDAC4_ were cloned into pCAMBIA1300-35S-YCE or 591 pCAMBIA1300-35S-YNE to generate pCAMBIA1300-35S-YCE-ATG8a, pCAMBIA1300 592 35S-YNE-VDAC1, pCAMBIA1300-35S-YNE-VDAC2, pCAMBIA1300-35S-YNE 593 VDAC3, and pCAMBIA1300-35S-YNE-VDAC4 vectors. 594 For subcellular localization assays in protoplast, the _TagBFP_ sequence was PCR 595 amplified from the pLL3.7-TagBFP plasmid and cloned into pUC to generate the pUC-35S 596 TagBFP vector. CDS of _ATG8a_, _ATG8e_, _GSNOR1_, _IDH1_, and _VDAC3_ were cloned into 597 pUC-35S-eGFP, pUC-35S-mCherry, or pUC-35S-TagBFP to generate pUC-35S::eGFP 598 ATG8a, pUC-35S::eGFP-ATG8e, pUC-35S::IDH1-eGFP, pUC-35S::GSNOR1-mCherry, 599 pUC-35S::FLAG-mCherry-VDAC3, and pUC-35S::HA-TagBFP-ATG8a vectors. Site 600 directed mutagenesis was performed to generate pUC-35S::GSNOR1 [Y152A V155A] -mCherry 601 and pUC-35S::GSNOR1 [I272A] -mCherry constructs. 602 All primers used in this study are listed in Supplementary Table 2. 603 604 **Mitochondrial DNA copy number analysis** 605 Mitochondrial DNA (mtDNA) copy number was determined by quantitative real-time PCR 606 (qPCR) as previously described with minor modifications (Cirotti _et al._, 2021; Preuten et al., 607 2010). Total plant DNA was extracted using CTAB extraction buffer. Approximately 5 ng of 608 total DNA was used as templates for each qPCR reaction. qPCR was performed using 609 primers specific for mitochondrial genes and nuclear reference genes with UltraSYBR 24 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 610 Mixture (CoWin Biotech, Cat # 0957) on a CFX96 Real-Time PCR Detection System (Bio 611 Rad). The relative mtDNA copy number was calculated by normalizing mitochondrial DNA 612 copy number to nuclear DNA copy number (the nuclear-encoded single-copy gene 613 _AtRpoTp_ ), using the 2 [−ΔΔCt] method. All primers used for qPCR are listed in Supplementary 614 Table 2. 615 616 **Recombinant protein expression and purification** 617 The recombinant protein expression vectors were transformed into _Escherichia coli_ strain 618 BL21 (DE3). Bacterial cultures were grown in LB medium at 37°C until OD₆₀₀ reached 0.6– 619 0.8, and protein expression was induced with 0.1 mM isopropyl-  -D-thiogalactoside (IPTG) 620 at 16°C overnight. Cells were harvested by centrifugation and lysed. His-tagged, GST 621 tagged, and MBP-tagged recombinant proteins were purified using Ni-NTA agarose 622 (QIAGEN, Cat # 30230), Glutathione Sepharose 4B (GE Healthcare, Cat # 17-0756-01), and 623 Amylose Resin (New England Biolabs, Cat # E8021), respectively, according to the 624 manufacturers’ instructions. 625 626 **Antibodies and immunoblotting** 627 The anti-GSNOR1 antibody (also known as anti-PAR2 antibody) was described previously 628 (Chen _et al._, 2009; Zhan _et al._, 2018). The following antibodies were obtained from 629 commercial sources: anti-ATG8 (Agrisera, Cat # AS142811), anti-COXII (Agrisera, Cat # 630 AS04053A), anti-GFP (Invitrogen, Cat # MA5-15256), anti-GST (CoWin Biotech, Cat # 631 CW0084), anti-His (Engibody, Cat # AT0025), anti-HSP70 (Agrisera, Cat # AS08347), anti 632 HSP82 (Beijing Protein Innovation, Cat # AbM51099-31-PU), anti-MBP (EMAR, Cat # 633 EM33019), and anti-VDAC (Agrisera, Cat # AS07212). Immunoblotting was performed as 634 previously described (Chen _et al._, 2009; Zhan _et al._, 2018). Quantification of the 635 immunoblot signals was performed using ImageJ software. 636 637 **GST pull-down assay** 25 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 638 GST pull-down assays were performed as previously described (Ren et al., 2013) with minor 639 modifications. Briefly, purified GST or GST-fusion proteins were incubated with purified 640 His-tagged or MBP-tagged proteins and Glutathione Sepharose 4B beads for 2 h at 4°C in 641 pull-down buffer [137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4, 1.8 mM KH 2 PO 4, pH 7.4, 642 0.05% octylphenyl-polyethylene glycol (brand name: IGEPAL CA-630)]. Beads were 643 washed six times with pre-chilled pull-down buffer. Bound proteins were eluted by boiling in 644 2× SDS loading buffer and analyzed by immunoblotting. 645 646 **Co-immunoprecipitation (Co-IP) assay** 647 Co-IP experiments were performed as previously described (Ren _et al._, 2013) with 648 modifications. Briefly, 2-week-old seedlings were ground in liquid nitrogen and 649 homogenized in IP buffer [50 mM Tris-HCl (pH 7.4), 1.5 mM MgCl 2, 100 mM NaCl, 10% 650 glycerol, 0.1% octylphenyl-polyethylene glycol (brand name: IGEPAL CA-630)] 651 supplemented with Protease Inhibitor Cocktail (Sigma, Cat # P9599). After centrifugation at 652 16,000 _g_ for 20 min at 4°C, supernatants were collected and incubated with anti-GFP-Nano 653 Agarose beads (LABLEAD, Cat # GNA-20-400) or anti-FLAG M2 Affinity Gel (Sigma 654 Aldrich, Cat # A2220) for 2 h at 4°C. Beads were washed five times with IP buffer, boiled in 655 2× SDS loading buffer, and the eluted proteins were analyzed by immunoblotting. 656 657 **Luciferase complementation imaging (LCI) and bimolecular fluorescence** 658 **complementation (BiFC) assay** 659 _Agrobacterium_ _tumefaciens_ strain GV3101 carrying the indicated expression vectors was 660 cultured overnight in liquid Luria-Bertani medium at 28°C with shaking. _Agrobacterial_ cells 661 were collected by centrifugation at 4000 rpm for 10 min and resuspended in infiltration 662 buffer [10 mM 2-( _N_ -morpholino)ethanesulfonic acid (MES, pH 5.7), 10 mM MgCl 2, 200 μM 663 acetosyringone] to an OD 600 of 0.8. The indicated _Agrobacterium_ suspensions were 664 incubated at room temperature for 3–4 h, mixed at a 1:1 ratio, and co-infiltrated into fully 665 expanded leaves of _Nicotiana benthamiana_ using a needleless syringe. For the luciferase 26 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 666 complementation imaging (LCI) assay, _Agrobacterium_ -mediated transient expression was 667 performed as previously described (Chen _et al._, 2020). Infiltrated plants were grown under a 668 16-h light/8-h dark photoperiod at 21°C for 2 days. The detached leaves were sprayed with 1 669 mM beetle luciferin (Promega, Cat # E1605) and the luminescence signal was captured using 670 a NightSHADE LB 985 in vivo Plant Imaging System. For the BiFC assay, the infiltrated 671 leaves were excised 2–4 days post-infiltration and fluorescence signals were observed under 672 a confocal laser scanning microscope (ZEISS LSM980) with excitation at 514 nm and 673 emission at 517–595 nm for YFP. 674 675 **Yeast two-hybrid assay** 676 The yeast two-hybrid assay was performed using the Matchmaker Gold Yeast Two-Hybrid 677 System (Clontech Laboratories, Cat # 630466) following the manufacturer’s instructions. 678 Briefly, the CDS of GSNOR1 were cloned into the pGBKT7 vector as the bait to screen a 679 Mate & Plate Library-Universal Arabidopsis library (TaKaRa, Cat # 630487). The bait 680 construct was transformed into the yeast strain Y2HGold and mated with the library strain. 681 The mating culture was incubated at 30°C for 20–24 h with gentle shaking at 40 rpm, 682 followed by plating on selective medium SD/-Leu/-Trp/-His/-Ade supplemented with X-  683 Gal (40 mg/L) and Aureobasidin A (200 μg/L). For the yeast two-hybrid pairwise assay, the 684 CDSs of VDAC1, VDAC2, VDAC3, and VDAC4, as well as the truncated variants of 685 VDAC3, were cloned into the pGADT7 vector and co-transformed with pGBKT7-GSNOR1 686 into the Y2HGold strain using the Yeastmaker Yeast Transformation System 2 (Clontech 687 Laboratories, Cat # 630439) according to the manufacturer’s instructions. To assess protein 688 protein interactions, yeast transformants were grown and serially diluted to 10 [-1], 10 [-2], and 10 [-] 689 3, and 5 μL of each dilution was spotted onto SD/-Leu/-Trp medium or SD/-Leu/-Trp/-His/ 690 Ade medium supplemented with X-α-Gal and Aureobasidin A. Images were captured after 691 incubation at 30°C for 3 days. 692 693 **Arabidopsis protoplast preparation and transformation** 27 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 694 Arabidopsis protoplast preparation and transformation were performed as previously 695 described with minor modifications (Yoo et al., 2007). Briefly, protoplasts were prepared 696 from Arabidopsis plants grown in soil for 3–4 weeks at 21°C under a 12-h light/12-h dark 697 photoperiod. Leaves were cut into 0.5–1-mm strips and incubated in enzyme solution [0.4 M 698 mannitol, 20 mM MES (pH 5.7), 1.5% (w/v) cellulase R10, 0.4% (w/v) macerozyme R10, 699 20 mM KCl, 10 mM CaCl 2, 0.1% (w/v) BSA] at room temperature for 3–4 h in the dark with 700 gentle shaking at 50 rpm. The digestion mixture was diluted with an equal volume of W5 701 solution [154 mM NaCl, 125 mM CaCl 2, 5 mM KCl, 2 mM MES (pH 5.7)], filtered through 702 a 35-μm nylon mesh, and centrifuged at 100 × _g_ for 2 min to collect the protoplasts. The 703 protoplasts were washed twice with W5 solution, kept on ice for 30 min, and then 704 resuspended in MMG solution [0.4 M mannitol, 15 mM MgCl 2, 4 mM MES (pH 5.7)] at a 705 concentration of 1 × 10 [6] cells/mL. For polyethylene glycol (PEG)-mediated transformation, 706 10 μg of the indicated plasmid DNA was mixed with 100 μL of protoplast suspension, and 707 110 μL of PEG solution [0.2 M mannitol, 100 mM CaCl 2, 40% (w/v) PEG 4000] was then 708 added. The mixture was incubated at room temperature for 10 min and then washed twice 709 with W5 solution. The protoplasts were resuspended in 1.5 mL of W5 solution and incubated 710 overnight at 23°C in the dark. 711 712 **Confocal microscopy and image processing** 713 Confocal fluorescence images were acquired using an LSM980 laser scanning confocal 714 microscope equipped with a 63× oil-immersion objective lens (Carl Zeiss). Seven-day-old 715 Arabidopsis seedlings were treated with DMSO (control), 50  M DNP (Sigma-Aldrich, Cat 716 # D198501), or 1  M ConA (MedChemExpress, Cat # HY-N1724) for the indicated times in 717 1/2 MS liquid medium before imaging. Mitochondria in root cells were stained with 500 nM 718 tetramethylrhodamine ethyl ester (TMRE; Invitrogen, Cat # T669) or MitoTracker Red 719 (MTR; Thermo Fisher, Cat # M7512) for 10 min. For visualizing depolarized mitochondria 720 in cotyledon cells, samples were stained with TMRE for 15 min before confocal microscopy 721 imaging. Sequential acquisition mode was applied when imaging multiple fluorescent 28 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 722 proteins. The excitation and emission wavelengths were as follows: GFP, 488 nm excitation 723 and 490–550 nm emission; TMRE and MitoTracker Red, 561 nm excitation and 580–630 724 nm emission; mCherry, 561 nm excitation and 580–630 nm emission; TagBFP, 405 nm 725 excitation and 410–585 nm emission; YFP, 514 nm excitation and 517–595 nm emission. 726 Images were processed and analyzed using ZEN software. 727 728 **Transmission electron microscopy (TEM)** 729 Samples were immediately fixed in 2.5% (v/v) glutaraldehyde in 0.1 M phosphate buffer 730 (pH 7.2) for 2 h at room temperature, and then rinsed three times with 0.1 M phosphate 731 buffer (pH 7.2) for 20 min each. Samples were post-fixed in 1% (v/v) osmium tetroxide 732 (OsO 4 ) for 2 h, followed by three rinses with phosphate buffer (15 min each). The samples 733 were then dehydrated through a graded acetone series [30%, 50%, 70%, 90%, and 100% 734 (three times); v/v in ddH 2 O] at room temperature, with 20 min incubation at each 735 concentration. Dehydrated samples were infiltrated with acetone/EMBed-812 resin (Electron 736 Microscopy Sciences, Cat #14120) mixture at ratios of 3:1, 1:1, 1:3 (v/v) for 12 h each, 737 followed by 100% EMBed-812 resin for 12 h. Samples were embedded in fresh EMBed-812 738 resin and polymerized at 37°C for 12 h, 45°C for 12 h, and 60°C for 48 h. Ultrathin sections 739 (70 nm) were prepared using an ultramicrotome (Leica, EM UC7). Sections were observed 740 and photographed using a transmission electron microscope (Hitachi, HT-7700) at an 741 accelerating voltage of 80 kV. 742 743 **Quantification and statistical analysis** 744 For quantification analysis, the mean SD or SEM were calculated and compared to the 745 control. The significance ( _P_ value) was determined using two-tailed Student’s _t_ -test or One 746 way ANOVA test (specified in Figure legends). * _P_ < 0.05, ** _P_ < 0.01, *** _P_ < 0.001. All 747 experiments were repeated at least 3 times, and representative results are shown. 748 749 **Data availability** 29 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 750 All other data are available in the main text or supplementary materials. 751 30 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 752 **References** 753 754 Bechtold, N., and Pelletier, G. (1998). In planta Agrobacterium-mediated transformation of 755 adult Arabidopsis thaliana plants by vacuum infiltration. Methods Mol Biol _82_, 259-266. 756 10.1385/0-89603-391-0:259. 757 Broda, M., Millar, A.H., and Van Aken, O. (2018). Mitophagy: a mechanism for plant growth 758 and survival. Trends Plant Sci _23_, 434-450. 10.1016/j.tplants.2018.02.010. 759 Cen, X., Chen, Y., Xu, X., Wu, R., He, F., Zhao, Q., Sun, Q., Yi, C., Wu, J., Najafov, A., and 760 Xia, H. (2020). Pharmacological targeting of MCL-1 promotes mitophagy and improves 761 disease pathologies in an Alzheimer's disease mouse model. Nat Commun _11_, 5731. 762 10.1038/s41467-020-19547-6. 763 Chang, A.H., Sancheti, H., Garcia, J., Kaplowitz, N., Cadenas, E., and Han, D. (2014). 764 Respiratory substrates regulate S-nitrosylation of mitochondrial proteins through a thiol765 dependent pathway. Chem Res Toxicol _27_, 794-804. 10.1021/tx400462r. 766 Chen, L., Wu, R., Feng, J., Feng, T., Wang, C., Hu, J., Zhan, N., Li, Y., Ma, X., Ren, B., et al. 767 (2020). Transnitrosylation mediated by the non-canonical catalase ROG1 regulates nitric 768 oxide signaling in plants. Dev Cell _53_, 444-457. 10.1016/j.devcel.2020.03.020. 769 Chen, Q., Jia, D., Ren, J., Cheng, Y., Wu, H., Guo, S., and Wei, T. (2023). VDAC1 balances 770 mitophagy and apoptosis in leafhopper upon arbovirus infection. Autophagy _19_, 1678-1692. 771 10.1080/15548627.2022.2150001. 772 Chen, R., Sun, S., Wang, C., Li, Y., Liang, Y., An, F., Li, C., Dong, H., Yang, X., Zhang, J., 773 and Zuo, J. (2009). The _Arabidopsis PARAQUAT RESISTANT2_ gene encodes an _S_ 774 nitrosoglutathione reductase that is a key regulator of cell death. Cell Res _19_, 1377-1387. 775 10.1038/cr.2009.117. 776 Cirotti, C., Rizza, S., Giglio, P., Poerio, N., Allega, M.F., Claps, G., Pecorari, C., Lee, J.H., 777 Benassi, B., Barila, D., et al. (2021). Redox activation of ATM enhances GSNOR translation 778 to sustain mitophagy and tolerance to oxidative stress. EMBO Rep _22_, e50500. 779 10.15252/embr.202050500. 780 Cornett, K., Puderbaugh, A., Back, O., and Craven, R. (2022). GAPDH in neuroblastoma: 781 Functions in metabolism and survival. Front Oncol _12_, 979683. 10.3389/fonc.2022.979683. 782 D'Arcy, M.S. (2024). Mitophagy in health and disease. Molecular mechanisms, regulatory 783 pathways, and therapeutic implications. Apoptosis _29_, 1415-1428. 10.1007/s10495-024 784 01977-y. 785 Dai, C.Q., Guo, Y., and Chu, X.Y. (2020). Neuropathic Pain: the Dysfunction of Drp1, 786 Mitochondria, and ROS Homeostasis. Neurotox Res _38_, 553-563. 10.1007/s12640-020 787 00257-2. 788 Doan, K.N., Ellenrieder, L., and Becker, T. (2019). Mitochondrial porin links protein 789 biogenesis to metabolism. Curr Genet _65_, 899-903. 10.1007/s00294-019-00965-z. 790 Dodson, M., Darley-Usmar, V., and Zhang, J. (2013). Cellular metabolic and autophagic 791 pathways: traffic control by redox signaling. Free Radic Biol Med _63_, 207-221. 792 10.1016/j.freeradbiomed.2013.05.014. 31 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 793 Duckney, P.J., Wang, P.W., and Hussey, P.J. (2024). Mitophagy in plants: Emerging 794 regulators of mitochondrial targeting for selective autophagy. J Microsc-Oxford. 795 10.1111/jmi.13267. 796 El Zawily, A.M., Schwarzlander, M., Finkemeier, I., Johnston, I.G., Benamar, A., Cao, Y., 797 Gissot, C., Meyer, A.J., Wilson, K., Datla, R., et al. (2014). FRIENDLY regulates 798 mitochondrial distribution, fusion, and quality control in _Arabidopsis_ . Plant Physiol _166_, 799 808-828. 10.1104/pp.114.243824. 800 Faruk, M.O., Ichimura, Y., and Komatsu, M. (2021). Selective autophagy. Cancer Sci _112_, 801 3972-3978. 10.1111/cas.15112. 802 Feechan, A., Kwon, E., Yun, B.W., Wang, Y., Pallas, J.A., and Loake, G.J. (2005). A central 803 role for _S_ -nitrosothiols in plant disease resistance. Proc Natl Acad Sci U S A _102_, 8054-8059. 804 10.1073/pnas.0501456102. 805 Feng, J., Chen, L., and Zuo, J. (2019). Protein _S_ -nitrosylation in plants: current progresses 806 and challenges. J Integr Plant Biol _61_, 1206-1223. 10.1111/jipb.12780. 807 Feng, Y., Chen, Y., Wu, X., Chen, J., Zhou, Q., Liu, B., Zhang, L., and Yi, C. (2024). 808 Interplay of energy metabolism and autophagy. Autophagy _20_, 4-14. 809 10.1080/15548627.2023.2247300. 810 Foster, M.W., and Stamler, J.S. (2004). New insights into protein S-nitrosylation. 811 Mitochondria as a model system. J Biol Chem _279_, 25891-25897. 10.1074/jbc.M313853200. 812 Geisler, S., Holmstrom, K.M., Skujat, D., Fiesel, F.C., Rothfuss, O.C., Kahle, P.J., and 813 Springer, W. (2010). PINK1/Parkin-mediated mitophagy is dependent on VDAC1 and 814 p62/SQSTM1. Nat Cell Biol _12_, 119-131. 10.1038/ncb2012. 815 Gross, A.S., Ghillebert, R., Schuetter, M., Reinartz, E., Rowland, A., Bishop, B.C., Stumpe, 816 M., Dengjel, J., and Graef, M. (2024). A metabolite sensor subunit of the Atg1/ULK 817 complex regulates selective autophagy. Nat Cell Biol _26_, 366-377. 10.1038/s41556-024 818 01348-4. 819 Gross, A.S., Raffeiner, M., Zeng, Y., Ustun, S., and Dagdas, Y. (2025). Autophagy in Plant 820 Health and Disease. Annu Rev Plant Biol _76_, 197-227. 10.1146/annurev-arplant-060324 821 094912. 822 Guerra, D., Ballard, K., Truebridge, I., and Vierling, E. (2016). _S_ -nitrosation of conserved 823 cysteines modulates activity and stability of _S_ -nitrosoglutathione reductase (GSNOR). 824 Biochemistry _55_, 2452-2464. 10.1021/acs.biochem.5b01373. 825 He, Y., Tang, R.-H., Hao, Y., Stevens, R.D., Cook, C.W., Ahn, S.M., Jing, L., Yang, Z., Chen, 826 L., and Guo, F. (2004). Nitric oxide represses the Arabidopsis floral transition. Science _305_, 827 1968-1971. 828 Henry, E., Fung, N., Liu, J., Drakakaki, G., and Coaker, G. (2015). Beyond glycolysis: 829 GAPDHs are multi-functional enzymes involved in regulation of ROS, autophagy, and plant 830 immune responses. PLoS Genet _11_, e1005199. 10.1371/journal.pgen.1005199. 831 Hiller, S., Garces, R.G., Malia, T.J., Orekhov, V.Y., Colombini, M., and Wagner, G. (2008). 832 Solution structure of the integral human membrane protein VDAC-1 in detergent micelles. 833 Science _321_, 1206-1210. 10.1126/science.1161302. 834 Hu, J., Huang, X., Chen, L., Sun, X., Lu, C., Zhang, L., Wang, Y., and Zuo, J. (2015). Site 32 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 835 specific nitrosoproteomic identification of endogenously S-nitrosylated proteins in 836 Arabidopsis. Plant Physiol _167_, 1731-1746. 10.1104/pp.15.00026. 837 Huang, S., Van Aken, O., Schwarzlander, M., Belt, K., and Millar, A.H. (2016). The roles of 838 mitochondrial reactive oxygen species in cellular signaling and stress response in plants. 839 Plant Physiol _171_, 1551-1559. 10.1104/pp.16.00166. 840 Johansen, T., and Lamark, T. (2020). Selective Autophagy: ATG8 Family Proteins, LIR 841 Motifs and Cargo Receptors. J Mol Biol _432_, 80-103. 10.1016/j.jmb.2019.07.016. 842 Kacprzak, S.M., and Van Aken, O. (2022). Carbon starvation, senescence and specific 843 mitochondrial stresses, but not nitrogen starvation and general stresses, are major triggers for 844 mitophagy in _Arabidopsis_ . Autophagy _18_, 2894-2912. 10.1080/15548627.2022.2054039. 845 Kacprzak, S.M., and Van Aken, O. (2023). FRIENDLY is required for efficient dark-induced 846 mitophagy and controlled senescence in _Arabidopsis_ . Free Radic Biol Med _204_, 1-7. 847 10.1016/j.freeradbiomed.2023.04.007. 848 Kanki, T., Wang, K., Cao, Y., Baba, M., and Klionsky, D.J. (2009). Atg32 is a mitochondrial 849 protein that confers selectivity during mitophagy. Dev Cell _17_, 98-109. 850 10.1016/j.devcel.2009.06.014. 851 Kanwar, P., Sanyal, S.K., Mahiwal, S., Ravi, B., Kaur, K., Fernandes, J.L., Yadav, A.K., 852 Tokas, I., Srivastava, A.K., Suprasanna, P., and Pandey, G.K. (2021). CIPK9 targets VDAC3 853 and modulates oxidative stress responses in Arabidopsis. Plant J. 10.1111/tpj.15572. 854 Kohler, R.H., Zipfel, W.R., Webb, W.W., and Hanson, M.R. (1997). The green fluorescent 855 protein as a marker to visualize plant mitochondria _in vivo_ . Plant J _11_, 613-621. 856 10.1046/j.1365-313x.1997.11030613.x. 857 Kuchitsu, Y., and Taguchi, T. (2024). Lysosomal microautophagy: an emerging dimension in 858 mammalian autophagy. Trends Cell Biol _34_, 606-616. 10.1016/j.tcb.2023.11.005. 859 Kwon, E., Feechan, A., Yun, B.W., Hwang, B.H., Pallas, J.A., Kang, J.G., and Loake, G.J. 860 (2012). AtGSNOR1 function is required for multiple developmental programs in 861 _Arabidopsis_ . Planta _236_, 887-900. 10.1007/s00425-012-1697-8. 862 Lee, S.M., Hoang, M.H., Han, H.J., Kim, H.S., Lee, K., Kim, K.E., Kim, D.H., Lee, S.Y., 863 and Chung, W.S. (2009). Pathogen inducible voltage-dependent anion channel (AtVDAC) 864 isoforms are localized to mitochondria membrane in _Arabidopsis_ . Mol Cells _27_, 321-327. 865 10.1007/s10059-009-0041-z. 866 Lee, U., Wie, C., Fernandez, B.O., Feelisch, M., and Vierling, E. (2008). Modulation of 867 nitrosative stress by _S_ -nitrosoglutathione reductase is critical for thermotolerance and plant 868 growth in _Arabidopsis_ . Plant Cell _20_, 786-802. 10.1105/tpc.107.052647. 869 Li, B., Sun, C., Lin, X., and Busch, W. (2021). The Emerging Role of GSNOR in Oxidative 870 Stress Regulation. Trends Plant Sci _26_, 156-168. 10.1016/j.tplants.2020.09.004. 871 Li, C., Duckney, P., Zhang, T., Fu, Y., Li, X., Kroon, J., De Jaeger, G., Cheng, Y., Hussey, 872 P.J., and Wang, P. (2022). TraB family proteins are components of ER-mitochondrial contact 873 sites and regulate ER-mitochondrial interactions and mitophagy. Nat Commun _13_, 5658. 874 10.1038/s41467-022-33402-w. 875 Liu, J., Wu, Y., Meng, S., Xu, P., Li, S., Li, Y., Hu, X., Ouyang, L., and Wang, G. (2024). 876 Selective autophagy in cancer: mechanisms, therapeutic implications, and future 33 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 877 perspectives. Mol Cancer _23_, 22. 10.1186/s12943-024-01934-y. 878 Liu, L., Feng, D., Chen, G., Chen, M., Zheng, Q., Song, P., Ma, Q., Zhu, C., Wang, R., Qi, 879 W., et al. (2012). Mitochondrial outer-membrane protein FUNDC1 mediates hypoxia880 induced mitophagy in mammalian cells. Nat Cell Biol _14_, 177-185. 10.1038/ncb2422. 881 Liu, L., Hausladen, A., Zeng, M., Que, L., Heitman, J., and Stamler, J.S. (2001). A metabolic 882 enzyme for S-nitrosothiol conserved from bacteria to humans. Nature _410_, 490. 883 Liu, Y., and Bassham, D.C. (2012). Autophagy: pathways for self-eating in plant cells. Annu 884 Rev Plant Biol _63_, 215-237. 10.1146/annurev-arplant-042811-105441. 885 Lubega, J., Umbreen, S., and Loake, G.J. (2021). Recent advances in the regulation of plant 886 immunity by S-nitrosylation. J Exp Bot _72_, 864-872. 10.1093/jxb/eraa454. 887 Ma, J., Li, C., Ren, K., Zhang, K., Feng, L., Law, K.C., Chung, K.K., Bing, Y., Gao, C., 888 Kang, B.H., and Zhuang, X. (2025a). Plant-specific mitochondrial fission machinery 889 connects mitochondrial dynamics and mitophagosome formation for piecemeal mitophagy. 890 Proc Natl Acad Sci U S A _122_, e2504921122. 10.1073/pnas.2504921122. 891 Ma, J., Liang, Z., Zhao, J., Wang, P., Ma, W., Mai, K.K., Fernandez Andrade, J.A., Zeng, Y., 892 Grujic, N., Jiang, L., et al. (2021). Friendly mediates membrane depolarization-induced 893 mitophagy in _Arabidopsis_ . Curr Biol _31_, 1931-1944. 10.1016/j.cub.2021.02.034. 894 Ma, W., Ma, J., Zhang, K., Zheng, X., Wang, P., Feng, L., Ming, S., Zhuang, X., Zhou, J., 895 Gao, C., and Kang, B.-H. (2025b). Voltage-dependent anion channels are mitophagy 896 receptors mediating the recycling of depolarized mitochondria in _Arabidopsis_ . bioRxiv, 897 2025.2005.2029.656919. 10.1101/2025.05.29.656919. 898 Marshall, R.S., Hua, Z., Mali, S., McLoughlin, F., and Vierstra, R.D. (2019). Atg8-binding 899 UIM proteins define a new class of autophagy adaptors and receptors. Cell _177_, 766-781. 900 10.1016/j.cell.2019.02.009. 901 Marshall, R.S., and Vierstra, R.D. (2018). Autophagy: the master of bulk and selective 902 recycling. Annu Rev Plant Biol _69_, 173-208. 10.1146/annurev-arplant-042817-040606. 903 Najbauer, E.E., Becker, S., Giller, K., Zweckstetter, M., Lange, A., Steinem, C., de Groot, 904 B.L., Griesinger, C., and Andreas, L.B. (2021). Structure, gating and interactions of the 905 voltage-dependent anion channel. Eur Biophys J _50_, 159-172. 10.1007/s00249-021-01515-7. 906 Narendra, D., Tanaka, A., Suen, D.F., and Youle, R.J. (2008). Parkin is recruited selectively 907 to impaired mitochondria and promotes their autophagy. J Cell Biol _183_, 795-803. 908 10.1083/jcb.200809125. 909 Noda, N.N., Ohsumi, Y., and Inagaki, F. (2010). Atg8-family interacting motif crucial for 910 selective autophagy. FEBS Lett _584_, 1379-1385. 911 North, B.J., Fracchiolla, D., Ragusa, M.J., Martens, S., and Shoemaker, C.J. (2025). The 912 rapidly expanding role of LC3-interacting regions in autophagy. J Cell Biol _224_ . 913 10.1083/jcb.202504076. 914 Oh, S.J., Ikeda, M., Ide, T., Hur, K.Y., and Lee, M.S. (2022). Mitochondrial event as an 915 ultimate step in ferroptosis. Cell Death Discov _8_, 414. 10.1038/s41420-022-01199-8. 916 Okamoto, K., Kondo-Okamoto, N., and Ohsumi, Y. (2009). Mitochondria-anchored receptor 917 Atg32 mediates degradation of mitochondria _via_ selective autophagy. Dev Cell _17_, 87-97. 918 10.1016/j.devcel.2009.06.013. 34 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 919 Park, J.M., Lee, D.H., and Kim, D.H. (2023). Redefining the role of AMPK in autophagy 920 and the energy stress response. Nat Commun _14_, 2994. 10.1038/s41467-023-38401-z. 921 Pickles, S., Vigie, P., and Youle, R.J. (2018). Mitophagy and quality control mechanisms in 922 mitochondrial maintenance. Curr Biol _28_, R170-R185. 10.1016/j.cub.2018.01.004. 923 Preuten, T., Cincu, E., Fuchs, J., Zoschke, R., Liere, K., and Borner, T. (2010). Fewer genes 924 than organelles: extremely low and variable gene copy numbers in mitochondria of somatic 925 plant cells. Plant J _64_, 948-959. 10.1111/j.1365-313X.2010.04389.x. 926 Qiao, Y., Wang, J., Wang, B., Zhou, H., Ni, Q., Fu, W., Hu, Z., and Zhong, Q. (2025). 927 Sodium disrupts mitochondrial energy metabolism to execute NECSO. Nat Commun _17_, 928 493. 10.1038/s41467-025-67181-x. 929 Ren, B., Chen, Q., Hong, S., Zhao, W., Feng, J., Feng, H., and Zuo, J. (2013). The 930 _Arabidopsis_ eukaryotic translation initiation factor eIF5A-2 regulates root protoxylem 931 development by modulating cytokinin signaling. Plant Cell _25_, 3841-3857. 932 10.1105/tpc.113.116236. 933 Rizza, S., Cardaci, S., Montagna, C., Di Giacomo, G., De Zio, D., Bordi, M., Maiani, E., 934 Campello, S., Borreca, A., Puca, A.A., et al. (2018). _S_ -nitrosylation drives cell senescence 935 and aging in mammals by controlling mitochondrial dynamics and mitophagy. Proc Natl 936 Acad Sci U S A _115_, E3388-E3397. 10.1073/pnas.1722452115. 937 Robert, N., d'Erfurth, I., Marmagne, A., Erhardt, M., Allot, M., Boivin, K., Gissot, L., 938 Monachello, D., Michaud, M., Duchene, A.M., et al. (2012). Voltage-dependent-anion939 channels (VDACs) in Arabidopsis have a dual localization in the cell but show a distinct role 940 in mitochondria. Plant Mol Biol _78_, 431-446. 10.1007/s11103-012-9874-5. 941 Roussos, A., Kitopoulou, K., Borbolis, F., Ploumi, C., Gianniou, D.D., Li, Z., He, H., Tsakiri, 942 E., Borland, H., Kostakis, I.K., et al. (2025). Urolithin Α modulates inter-organellar 943 communication via calcium-dependent mitophagy to promote healthy ageing. Autophagy _21_, 944 3097-3122. 10.1080/15548627.2025.2561073. 945 Sakamoto, A., Ueda, M., and Morikawa, H. (2002). Arabidopsis glutathione-dependent 946 formaldehyde dehydrogenase is an S-nitrosoglutathione reductase. FEBS Lett _515_, 20-24. 947 10.1016/s0014-5793(02)02414-6. 948 Sandoval, H., Thiagarajan, P., Dasgupta, S.K., Schumacher, A., Prchal, J.T., Chen, M., and 949 Wang, J. (2008). Essential role for Nix in autophagic maturation of erythroid cells. Nature 950 _454_, 232-235. 10.1038/nature07006. 951 Schmitt, V., and Van Aken, O. (2023). Mitophagy: from the dark into the spotlight. Mol Plant 952 _16_, 1487-1489. 10.1016/j.molp.2023.08.015. 953 Shoshan-Barmatz, V., De Pinto, V., Zweckstetter, M., Raviv, Z., Keinan, N., and Arbel, N. 954 (2010). VDAC, a multi-functional mitochondrial protein regulating cell life and death. Mol 955 Aspects Med _31_, 227-285. 10.1016/j.mam.2010.03.002. 956 Stephani, M., and Dagdas, Y. (2020). Plant selective autophagy-still an uncharted territory 957 with a lot of hidden gems. J Mol Biol _432_, 63-79. 10.1016/j.jmb.2019.06.028. 958 Sun, S., Jia, P.F., Wang, W., Chen, L., Gong, X., Lin, H., Wu, R., Yang, W.C., Li, H.J., Zuo, 959 J., and Guo, H. (2025). S-sulfenylation-mediated inhibition of the GSNOR1 activity 960 regulates ovule development in Arabidopsis. J Genet Genomics _52_, 1034-1045. 35 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 961 10.1016/j.jgg.2025.01.007. 962 Tamura, K., Shimada, T., Ono, E., Tanaka, Y., Nagatani, A., Higashi, S.I., Watanabe, M., 963 Nishimura, M., and Hara-Nishimura, I. (2003). Why green fluorescent fusion proteins have 964 not been observed in the vacuoles of higher plants. Plant J _35_, 545-555. 10.1046/j.1365 965 313x.2003.01822.x. 966 Tang, X., Zhao, S., Liu, J., Liu, X., Sha, X., Huang, C., Hu, L., Sun, S., Gao, Y., Chen, H., et 967 al. (2023). Mitochondrial GSNOR alleviates cardiac dysfunction _via_ ANT1 denitrosylation. 968 Circ Res _133_, 220-236. 10.1161/CIRCRESAHA.123.322654. 969 Tateda, C., Watanabe, K., Kusano, T., and Takahashi, Y. (2011). Molecular and genetic 970 characterization of the gene family encoding the voltage-dependent anion channel in 971 _Arabidopsis_ . J Exp Bot _62_, 4773-4785. 10.1093/jxb/err113. 972 Thompson, A.R., Doelling, J.H., Suttangkakul, A., and Vierstra, R.D. (2005). Autophagic 973 nutrient recycling in Arabidopsis directed by the ATG8 and ATG12 conjugation pathways. 974 Plant Physiol _138_, 2097-2110. 10.1104/pp.105.060673. 975 Tian, Z., Huo, Y., Li, C., Zheng, Q., Hu, F., Li, J., Ma, J., Qu, X., Cheng, Y., Kang, B.H., et 976 al. (2026). Ubiquitin-dependent mitochondrial protein degradation ensures seedling 977 emergence by regulating ER-mitochondrial interaction and mitophagy. Nat Plants _12_, 1220 978 1231. 10.1038/s41477-026-02306-8. 979 Ujwal, R., Cascio, D., Colletier, J.P., Faham, S., Zhang, J., Toro, L., Ping, P., and Abramson, 980 J. (2008). The crystal structure of mouse VDAC1 at 2.3 A resolution reveals mechanistic 981 insights into metabolite gating. Proc Natl Acad Sci U S A _105_, 17742-17747. 982 10.1073/pnas.0809634105. 983 Wang, Y., Liu, C.-Y., Chen, L., Guo, S.-H., Xiao, S., and Qi, H. (2026). Transcriptional, 984 epigenetic, and post-translational regulation of plant autophagy. J Genet Genomics. 985 [https://doi.org/10.1016/j.jgg.2026.01.007.](https://doi.org/10.1016/j.jgg.2026.01.007) 986 Wang, Y., Nishimura, M.T., Zhao, T., and Tang, D. (2011). ATG2, an autophagy-related 987 protein, negatively affects powdery mildew resistance and mildew-induced cell death in 988 _Arabidopsis_ . Plant J _68_, 74-87. 10.1111/j.1365-313X.2011.04669.x. 989 Wei, Y., Chiang, W.C., Sumpter, R., Jr., Mishra, P., and Levine, B. (2017). Prohibitin 2 is an 990 inner mitochondrial membrane mitophagy receptor. Cell _168_, 224-238. 991 10.1016/j.cell.2016.11.042. 992 Xie, Q., Tzfadia, O., Levy, M., Weithorn, E., Peled-Zehavi, H., Van Parys, T., Van de Peer, 993 Y., and Galili, G. (2016). hfAIM: A reliable bioinformatics approach for in silico genome994 wide identification of autophagy-associated Atg8-interacting motifs in various organisms. 995 Autophagy _12_, 876-887. 10.1080/15548627.2016.1147668. 996 Yamashita, S.I., Sugiura, Y., Matsuoka, Y., Maeda, R., Inoue, K., Furukawa, K., Fukuda, T., 997 Chan, D.C., and Kanki, T. (2024). Mitophagy mediated by BNIP3 and NIX protects against 998 ferroptosis by downregulating mitochondrial reactive oxygen species. Cell Death Differ _31_, 999 651-661. 10.1038/s41418-024-01280-y. 1000 Yang, C., Li, X., Yang, L., Chen, S., Liao, J., Li, K., Zhou, J., Shen, W., Zhuang, X., Bai, M., 1001 et al. (2023). A positive feedback regulation of SnRK1 signaling by autophagy in plants. Mol 1002 Plant _16_, 1192-1211. 10.1016/j.molp.2023.07.001. 36 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1003 Yang, K., Yan, Y., Yu, A., Zhang, R., Zhang, Y., Qiu, Z., Li, Z., Zhang, Q., Wu, S., and Li, F. 1004 (2024). Mitophagy in neurodegenerative disease pathogenesis. Neural Regen Res _19_, 998 1005 1005. 10.4103/1673-5374.385281. 1006 Yoo, S.D., Cho, Y.H., and Sheen, J. (2007). Arabidopsis mesophyll protoplasts: a versatile 1007 cell system for transient gene expression analysis. Nat Protoc _2_, 1565-1572. 1008 10.1038/nprot.2007.199. 1009 Zhan, N., Wang, C., Chen, L., Yang, H., Feng, J., Gong, X., Ren, B., Wu, R., Mu, J., Li, Y., et 1010 al. (2018). _S_ -nitrosylation targets GSNO reductase for selective autophagy during hypoxia 1011 responses in plants. Mol Cell _71_, 142-154. 10.1016/j.molcel.2018.05.024. 1012 Zhang, T., Liu, Q., Gao, W., Sehgal, S.A., and Wu, H. (2022). The multifaceted regulation of 1013 mitophagy by endogenous metabolites. Autophagy _18_, 1216-1239. 1014 10.1080/15548627.2021.1975914. 1015 Zhou, Q., Li, H., Li, H., Nakagawa, A., Lin, J.L., Lee, E.S., Harry, B.L., Skeen-Gaar, R.R., 1016 Suehiro, Y., William, D., et al. (2016). Mitochondrial endonuclease G mediates breakdown 1017 of paternal mitochondria upon fertilization. Science _353_, 394-399. 10.1126/science.aaf4777. 1018 1019 37 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1020 **Acknowledgements** 1021 We thank Dingzhong Tang, Yikun He, and Ye Tian for kindly providing seeds and plasmids; 1022 Yanbao Tian for help on confocal microscopy and transmission electron microscopy; Byung 1023 Ho Kang for sharing unpublished data; members of Zuo Lab for discussion and suggestions. 1024 This work was supported by grants from the National Natural Science Foundation of China 1025 (NSFC; 31830017 and 32200256) and State Key Laboratory of Plant Genomics 1026 (SKLPG2023-22). 1027 1028 **Author Contributions** 1029 H.L. and H.G. performed most of the experiments, assisted by R.W., W.W., P.-F.J., S.S., and 1030 R.B. J.Z., W.-C.Y., and H.-J.L. supervised the research. J.Z., H.L., H.G., W.-C.Y., and H.-J.L. 1031 analyzed the data. J.Z. and H.L. wrote the manuscript, assisted by H.G. All authors read and 1032 approved the manuscript. 1033 1034 **Ethics declarations** 1035 **Competing interests** 1036 The authors declare no competing interests. 1037 1038 **Supplementary information** 1039 **Supplementary Table 1.** Putative GSNOR1-interacting proteins identified by yeast two 1040 hybrid screen. 1041 **Supplementary Table 2.** Primers used in this study. 1042 **Supplementary Video 1.** Co-localization of GSNOR1-GFP and IDH1-mCherry. 1043 38 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1044 1045 1046 **Fig. 1:** _**GSNOR1**_ **controls mitochondrial quality.** 1047 **a** _,_ Analysis of mitochondrial gene copy numbers of wild type (WT) and the indicated 1048 mutants by quantitative PCR (qPCR). DNA samples were prepared from 20-day-old 1049 seedlings. Bars represent SEM of more than five biological replicates. The nuclear-encoded 1050 single-copy gene _AtRpoTp_ was used as an internal standard. Statistical significance was 1051 determined by one-way ANOVA followed by Tukey’s multiple comparisons test. ns, not 1052 significant; *** _P_ < 0.001. 1053 **b,** Immunoblotting analysis of mitochondrial marker proteins in 20-day-old seedlings with 1054 the indicated genotypes _._ HSP82 served as a loading control. Quantification is shown below 1055 the blot. The relative protein level in wild type is set as 1.0. VDAC: an outer mitochondrial 1056 membrane marker; COXII: an inner mitochondrial membrane marker; mtHSP70: a matrix 1057 mitochondrial marker. 1058 **c,** Quantitative analysis of the relative protein levels of VDAC, COXII, and mtHSP70 shown 39 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1059 in ( **b** ). Data are presented as mean ± SD from three biological replicates. Statistical 1060 significance was determined by two-tailed Student _t_ -test. ns, not significant; * _P_ < 0.05; ** _P_ 1061 < 0.01. 1062 **d,** Transmission electron microscopy (TEM) of ultra-thin sections prepared from root tips of 1063 the indicated genotypes. White arrows denote damaged mitochondria not engulfed by 1064 autophagosomes. M, mitochondrion. Scale bars, 200 nm. 1065 **e,** Quantitative analysis of mitochondrial cristae morphology damage shown in ( **d** ) by 1066 counting the mitochondrial numbers of the defined phenotypes from multiple TEM 1067 micrographs. At least 100 mitochondria were counted for each sample. 1068 **f,** Relative numbers of mitochondrial surface area with the indicated ranges shown in ( **d** ). 1069 The analysis was performed by counting the mitochondrial numbers with the defined surface 1070 area ranges from multiple TEM micrographs. At least 100 mitochondria were counted for 1071 each sample. 1072 40 1073 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 41 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1074 **Fig. 2: GSNOR1 positively regulates mitophagy.** 1075 **a,** Immunoblotting analysis of mitochondrial outer membrane proteins (VDAC) and inner 1076 membrane protein (COXII) in 7-day-old seedlings of wild type (WT) and mutants with the 1077 indicated genotypes treated with DMSO (control) or 50 µM DNP for 4 h. HSP82 served as a 1078 loading control. 1079 **b,** Quantitative analysis of the relative protein levels of VDAC and COXII shown in ( **a** ). 1080 Data are presented as mean ± SD from three biological replicates. Statistical significance 1081 was determined by two-tailed Student _t_ -test. ns, not significant; ** _P_ < 0.01; *** _P_ < 0.001. 1082 **c,d,** Confocal microscopy of mitochondria in roots derived from 7-day-old transgenic 1083 seedlings carrying a mitochondrial marker gene _Mito-GFP_ in the wild type ( **c** ) and _gsnor1_ 1084 ( **d** ) genetic backgrounds. The samples were treated with DMSO (control; 4 h), ConA (1 µM, 1085 4 h), DNP (50 µM, 1.5 h), or ConA + DNP (4 h), followed by staining with TMRE (0.5 µM, 1086 10 min). White arrows denote representative depolarized mitochondria. Scale bars, 5 µm. 1087 **e,** Quantitative analysis of depolarized mitochondria shown in ( **c** ) and ( **d** ). At least 1,000 1088 mitochondria from 8 cells (three root samples) were analyzed. Data are presented as mean ± 1089 SEM. Statistical significance was determined by one-way ANOVA followed by Tukey’s 1090 multiple comparisons test. ns, not significant; ** _P_ < 0.01. 1091 **f,** Analysis of mitophagic flux in _Mito-GFP_ transgenic seedlings in the wild-type (WT) and 1092 _gsnor1_ genetic backgrounds. Cleavage of GFP was detected using an anti-GFP antibody. The 1093 upper and lower bands show full-length Mito-GFP and free GFP, respectively. Seven-day-old 1094 transgenic seedlings were treated with DMSO (control, 2 h) or with 50 µM DNP for the 1095 indicated times. Quantification was performed by calculating the ratio of free GFP to full 1096 length Mito-GFP in each lane. HSP82 served as a loading control. 1097 **g,** Quantitative analysis of the ratio of free GFP to full-length Mito-GFP shown in ( **f** ). Data 1098 are presented as mean ± SD from three biological replicates. Statistical significance was 1099 determined by one-way ANOVA followed by Tukey’s multiple comparisons test. ns, not 1100 significant. 1101 **h,** Transmission electron microscopy of ultra-thin sections prepared from root tips derived 42 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1102 from 7-day-old wild-type (WT) and _gsnor1_ seedlings treated with 50 µM DNP for 4 h. 1103 White and red arrows denote damaged mitochondria engulfed and not engulfed by 1104 autophagosomes, respectively. M, mitochondrion. Scale bars, 200 nm. 1105 43 1106 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 44 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1107 **Fig. 3: GSNOR1 localizes to depolarized mitochondria and interacts with ATG8a.** 1108 **a,** Confocal microscopy of the root elongation zones derived from seedlings carrying both 1109 _GSNOR1-GFP_ and _IDH1-mCherry_ transgenes. Samples were treated with DMSO (control), 1110 DNP (50 µM), or DNP + ConA (1 µM) for 1.5 h prior to imaging. Line scans on the right 1111 side of the images show co-localization of GSNOR1-GFP (green) and IDH1-mCherry (red). 1112 White arrows indicate punctate co-localization signals. Scale bar, 5 µm. 1113 **b,** Confocal microscopy of the root elongation zones derived from 7-day-old seedlings 1114 carrying both _GSNOR1-GFP_ and _IDH1-mCherry_ transgenes. The samples were treated with 1115 ConA (1 µM) for 12 h prior to imaging at the indicated time points (second; s). White arrows 1116 indicate punctate co-localization signals. Scale bar, 5 µm. 1117 **c,** Confocal microscopy of Arabidopsis protoplast cells transformed with _GSNOR1-mCherry_ 1118 and _GFP-ATG8a_ genes. The boxed region shows an enlarged view at the bottom right shows 1119 co-localization of GSNOR1-mCherry and GFP-ATG8a. Scale bar, 5 µm. 1120 **d,** Analysis of the GSNOR1-ATG8a interaction by co-immunoprecipitation (Co-IP). Eight 1121 day-old transgenic seedlings carrying _GFP_ (control) or _GFP-ATG8a_ transgenes were treated 1122 with the indicated concentrations of DNP for 2 h and then subjected to Co-IP assay. 1123 **e,** Six-week-old plants with the indicated genotypes grown under long-day condition (16-h 1124 light/8-h dark) in a growth chamber. Scale bar, 5 cm. 1125 **f,** Analysis of GSNOR1 protein in 6-week-old plants with the indicated genotypes by 1126 immunoblotting using an anti-GSNOR1 antibody. Equal loading was shown by Ponceau S 1127 staining. Quantification is shown below the blot. The relative protein level in wild type is set 1128 as 1.0. See also Extended Fig. 3d for the growth phenotype of other transgenic lines. 1129 45 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1130 1131 1132 **Fig. 4: GSNOR1 interacts with VDAC proteins.** 1133 **a,** Analysis of the interaction of His-GSNOR1 and GST-VDAC recombinant proteins by a 1134 GST pull-down assay. GST protein was used as a negative control. 1135 **b,** Analysis of the interaction between GSNOR1 and VDAC proteins by co 1136 immunoprecipitation (Co-IP) in 8-day-old wild-type (WT) and _GSNOR1-FLAG_ transgenic 1137 seedlings (left), and in _GSNOR1-FLAG_ transgenic seedlings treated with or without 50 µM 46 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1138 DNP for 2 h (right). Quantification is shown below the blot. The relative protein level in 1139 samples without DNP treatment was set as 1.0. The polyclonal anti-VDAC antibody 1140 recognizes VDAC1 through VDAC4 proteins. 1141 **c,** Analysis of the interaction of GSNOR1 and VDAC proteins by a luciferase 1142 complementation imaging assay in tobacco leaves infiltrated with _Agrobacterium_ 1143 transformed with various combinations of transgenes as indicated. The assay was performed 1144 2 days post-infiltration. 1145 **d,** Analysis of the interaction of GSNOR1 and VDAC proteins by a bimolecular fluorescence 1146 complementation assay in tobacco leaves infiltrated with _Agrobacterium_ transformed with 1147 various combinations of transgenes as indicated. Confocal microscopy was performed 2 days 1148 post-infiltration. Co-localization signals are shown in boxed regions. Scale bar, 20 µm. 1149 **e,** Confocal microscopy of Arabidopsis protoplast cells transformed with _GSNOR1-GFP_ and 1150 _FLAG-mCherry-VDAC3_ transgenes. Co-localization signals are shown in enlarged views of 1151 the boxed regions. Scale bar, 10 µm. 1152 47 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1153 1154 **Fig. 5: VDAC proteins interact with ATG8a and GSNOR1.** 1155 **a,** Analysis of the interaction of GST-VDAC and His-ATG8a recombinant proteins by a GST 1156 pull-down assay. GST protein was used as a negative control. 1157 **b,** Analysis of the interaction of VDAC and ATG8a proteins by a co-immunoprecipitation 1158 (Co-IP) assay in 8-day-old seedlings carrying _GFP_ or _GFP-ATG8a_ transgenes. The 48 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1159 polyclonal anti-VDAC antibody recognizes VDAC1 through VDAC4 proteins. 1160 **c,** Analysis of the interaction of VDAC3 and ATG8a proteins by a bimolecular fluorescence 1161 complementation assay in tobacco leaves infiltrated with _Agrobacterium_ transformed with 1162 various combinations of transgenes as indicated. Confocal microscopy was performed 2 days 1163 post-infiltration. Co-localization signals are shown in the boxed regions. Scale bar, 20 µm. 1164 **d,** Confocal microscopy of Arabidopsis protoplasts transformed with _FLAG-mCherry-_ 1165 _VDAC3_ and _eGFP-ATG8a_ transgenes. White arrowheads indicate punctate co-localization 1166 signals. Scale bar, 10 µm. 1167 **e,** Schematic representations of wild-type (WT) VDAC3 and deletion mutants fused to GST 1168 (GST not shown in the diagrams). 1169 **f,** Confocal microscopy of Arabidopsis protoplasts transiently expressing eGFP-VDAC3, 1170 eGFP-D1, or eGFP-D2. Protoplasts were stained with 200 nM MitoTracker Red for 15 min 1171 prior to imaging. Scale bars, 10 µm. 1172 **g,h,** Analysis of the interaction of wild-type (WT) GST-VDAC3 with D1 through D5 mutant 1173 recombinant proteins shown in ( **e** ) with His-ATG8a ( **g** ) or His-GSNOR1 ( **h** ) recombinant 1174 proteins by a GST pull-down assay. 1175 **i,** A projected 3-D structure of Arabidopsis VDAC3 protein trained with human hVDAC1 1176 (PDB code: 2K4T) as a model. N1, N2, and N3 denote the putative α-helixes (N1 and N2) 1177 and a putative β-sheet (N3), of which are covered by deletional mutations D6-D8 shown in 1178 ( **e** ), respectively. 1179 **j,k,** Analysis of the interaction of wild-type (WT) GST-VDAC3 and D6-D8 mutant 1180 recombinant proteins shown in ( **e** ) with His-ATG8a ( **j** ) or His-GSNOR1 ( **k** ) recombinant 1181 proteins by a GST pull-down assay. 1182 49 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1183 1184 1185 **Fig. 6: The GSNOR1-VDAC3-ATG8 complex modulates mitochondrial stress** 1186 **responses.** 1187 **a,** Confocal imaging of Arabidopsis protoplasts prepared from _GSNOR1-GFP_ transgenic 1188 plants transformed with _FLAG-mCherry-VDAC3_ and _HA-BFP-ATG8a_ . The boxed region 1189 shows an enlarged view of the co-localized signal. Scale bar, 10 µm. 1190 **b,** Seven-day-old wild type (WT), _atg5_, and _gsnor1_ seedlings germinated and grown on 1/2 1191 MS agar medium were transferred to 1/2 MS agar medium containing 10 µM DNP and 1192 cultured for additional 9 days. Scale bar, 1 cm. 50 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1193 **c,** Quantification of fresh weight of seedlings shown in ( **b** ). The relative fresh weight of 1194 wild-type seedlings treated with DMSO was set as 1.0. Data are presented as mean ± SD ( _n_ 1195 ≥ 60). Statistical significance was determined by one-way ANOVA followed by Tukey’s 1196 multiple comparisons test. * _P_ < 0.05; *** _P_ < 0.001. 1197 **d,** Seven-day-old wild-type and mutant seedlings with the indicated genotypes germinated 1198 and grown on 1/2 MS medium were transferred to liquid 1/2 MS medium containing 50 µM 1199 DNP and treated for 4 h. After the treatment, seedlings were transferred to 1/2 MS agar 1200 medium (the positions of root tips marked in blue) and cultured for additional 48 hr for 1201 recovery growth. Scale bar, 1 cm. 1202 **e,** Quantification of the root length of seedlings shown in ( **d** ). Data are presented as mean ± 1203 SEM ( _n_ ≥ 20). Statistical significance was determined by two-tailed Student _t_ -test. *** _P_ < 1204 0.001. 1205 **f,** Analysis of root recovery growth rate of seedlings shown in ( **d** ). The relative root recovery 1206 growth rate of seedlings cultured on medium containing DMSO (control) for each genotype 1207 is set as 100%. Root elongation from the marked root tip position was measured after 48 h of 1208 recovery growth. Data are presented as mean ± SEM ( _n_ ≥ 20). Statistical significance was 1209 determined by one-way ANOVA followed by Tukey’s multiple comparisons test. ns, not 1210 significant; * _P_ < 0.05; ** _P_ < 0.01. 1211 **g,** Transmission electron microscopy of ultra-thin sections prepared from root tips derived 1212 from 9-day-old wild-type (WT), _gsnor1_, and _vdac3_ seedlings. White arrowheads denote 1213 damaged mitochondrial with abnormal cristae. Scale bars, 500 nm. 1214 51 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1215 1216 1217 **Fig.7: A model of GSNOR1-regulated plant mitophagy.** 1218 Under normal growth conditions, GSNOR1 dynamically interacts with the outer 1219 mitochondrial membrane-localized VDAC3 protein, which may serve to monitor the quality 1220 of mitochondria. When challenged by stresses, mitochondria become depolarized, initiating 1221 the formation of a complex consisting of the mitophagy receptor GSNOR1, the phagophore 1222 localized ATG8, and VDAC3. The GSNOR1-ATG8-VDAC3 complex recruits phagophore to 1223 encapsulate depolarized mitochondria, which subsequently fuses to vacuoles for autophagic 1224 degradation, thereby safeguarding cellular homeostasis and regulating stress responses. 1225 52 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1226 **Extended data** 1227 1228 1229 1230 **Extended Data** **Fig. 1: GSNOR1 is not involved in starvation-induced mitochondrial** 1231 **degradation.** 1232 **a,** Seven-day-old wild-type (WT; Col-0), _atg5_, _atg7_, and _gsnor1_ seedlings germinated and 1233 grown on 1/2 MS medium were transferred to fresh 1/2 MS medium (control), nitrogen-free 1234 medium (N starvation), or sucrose-free agar medium cultured in the dark (C starvation). 53 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1235 Seedlings were cultured for additional 8 days prior to photographing. Scale bar, 1 cm. 1236 **b,** Analysis of chlorophyll contents of seedlings shown in ( **a** ). The relative chlorophyll 1237 content of each genotype under control conditions was set as 1.0. Data are presented as mean 1238 ± SD of eight biological replicates. Statistical significance was determined by one-way 1239 ANOVA followed by Tukey’s multiple comparisons test. *** _P_ < 0.001. 1240 **c,** Seven-day-old wild type (WT; Col-0), _atg5_, _atg7_, and _gsnor1_ seedlings germinated and 1241 grown on 1/2 MS medium were transferred to fresh liquid 1/2 MS medium (control), 1242 nitrogen-free medium (N starvation), or sucrose-free medium cultured in the dark (C 1243 starvation). Seedlings were cultured for additional 8 days prior to photographing. Scale bar, 1244 1 cm. 1245 **d,** Analysis of chlorophyll contents of seedlings shown in ( **c** ). The relative chlorophyll 1246 content of each genotype under control conditions was set as 1.0. Data are presented as mean 1247 ± SD of eight biological replicates. Statistical significance was determined by one-way 1248 ANOVA followed by Tukey’s multiple comparisons test. ns, not significant; * _P_ < 0.05; *** _P_ 1249 < 0.001. 1250 **e,** Analysis of mitochondrial outer membrane proteins VDACs and the mitochondrial inner 1251 membrane protein COXII in wild-type (WT), _atg5_, and _gsnor1_ seedlings by 1252 immunoblotting. Seven-day-old seedlings germinated and grown on 1/2 MS agar medium 1253 were transferred to liquid 1/2 MS medium or nitrogen-free medium for 24 h. Equal loading 1254 was shown by Ponceau S staining. Quantification of band intensities is shown below the 1255 blot. The relative protein level in WT grown under control conditions was set as 1.0. 1256 54 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1257 1258 **Extended Data** **Fig. 2: GSNOR1 localizes to depolarized mitochondria.** 1259 **a,** Analysis of mitophagic flux in _GFP-TOM20-3_ transgenic seedlings in the wild-type (WT) 1260 and _gsnor1_ genetic backgrounds. Cleavage of GFP from GFP-TOM20-3 was detected using 1261 an anti-GFP antibody. Arrows indicate the bands of full-length GFP-TOM20-3 and free GFP, 1262 respectively. Seven-day-old transgenic seedlings were treated with DMSO (control, 2 h) or 1263 with 50 µM DNP for the indicated times. HSP82 served as a loading control. 1264 **b,** Confocal microscopy of roots from 7-day-old seedlings carrying both _GSNOR1-GFP_ and 1265 _IDH1-mCherry_ transgenes. The samples were treated with DMSO or ConA (1 µM) for 12 h 1266 prior to imaging. Line scans on the right side of the images indicate co-localization between 1267 GSNOR1-GFP (green) and IDH1-mCherry (red). Scale bar, 5 µm. 1268 **c,** Confocal microscopy of roots from 5-day-old _GSNOR1-GFP_ transgenic seedlings treated 1269 with 1 µM ConA for 12 h, followed by staining with MitoTracker Red for 10 min. Line 1270 scans below the images indicate co-localization between GSNOR1-GFP (green) and 55 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1271 MitoTracker Red (red). White arrows indicate punctate co-localization signals. Scale bar, 10 1272  m. 1273 56 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1274 1275 1276 **Extended Data** **Fig. 3: GSNOR1 interacts with ATG8 via its AIM motifs.** 1277 **a,** Putative AIM motifs of GSNOR1. The search was performed using the hfAIM tool as 1278 described(Xie et al., 2016). 1279 **b,** Analysis of the interaction of His-ATG8e and GST-GSNOR1 recombinant proteins by a 1280 GST pull-down assay. GST protein was used as a negative control. 1281 **c,** Confocal microscopy of Arabidopsis protoplasts transformed with _GSNOR1-mCherry_ and 1282 _eGFP-ATG8e_ and then cultured for 24 h prior to imaging. Scale bar, 10 µm. 57 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1283 **d,** Six-week-old plants with the indicated genotypes grown under long-day condition (16-h 1284 light/8-h dark) in a growth chamber. See also Fig. 3e. Same plants of the control group, i.e., 1285 wild type (WT), _gsnor1_, _GSNOR1-GFP_ (#87), and _GSNOR1-mCherry_ (#13), were used for 1286 photographing in Fig. 3e and here. Scale bar, 5 cm. 1287 58 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1288 1289 1290 **Extended Data** **Fig. 4: Alignment of VDAC proteins.** 1291 Protein sequence alignment of VDAC family proteins. Protein sequences used for alignment 1292 with the following accession numbers: _Arabidopsis thaliana_ VDAC1: NP_186777; 1293 _Arabidopsis thaliana_ VDAC2: NP_201551; _Arabidopsis thaliana_ VDAC3: NP_001190314; 1294 _Arabidopsis thaliana_ VDAC4: NP_200557; _Oryza sativa_ : NP_001390674; _Physcomitrium_ 1295 _patens_ : XP_024397563; _Caenorhabditis elegans_, NP_501211; _Danio rerio_ : NP_001001404; 1296 _Drosophila melanogaster_ : NP_001033899; _Mus musculus_ : NP_001387916; _Homo sapiens_ : 1297 NP_001387937. Sequence alignment was performed using Clustal X2, and the results were 1298 visualized in Genedoc. 1299 59 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1300 1301 1302 **Extended Data** **Fig. 5: Analysis of interactions between GSNOR1 and VDAC proteins.** 1303 **a,b,** Analysis of the interaction of GSNOR1 with VDAC1 through VDAC4 proteins ( **a** ) or 1304 truncated VDAC3 mutant proteins ( **b** ) by the yeast two-hybrid assay. GSNOR1 is fused to 1305 the GAL4 DNA-binding domain (BD), and VDAC proteins are fused to the GAL4 activation 1306 domain (AD). The paired combinations as indicated are transformed and expressed in yeast 1307 strain Y2HGold. Protein interactions were assessed by activation of the reporter genes _HIS3_, 60 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1308 _ADE2_, _AUR1-C_, or _MEL1_ . The p53-SV40 interaction and lamin-SV40 interaction served as 1309 positive and negative controls, respectively. pGBKT7 and pGADT7 were used as the bait 1310 and prey vectors, respectively. 1311 61 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1312 1313 1314 **Extended Data** **Fig. 6: Functional characterization of VDAC proteins.** 1315 **a,** Analysis of the interaction of ATG8a and VDAC proteins by a bimolecular fluorescence 1316 complementation assay in tobacco leaves infiltrated with _Agrobacterium_ transformed with 1317 various combinations of transgenes as indicated. Confocal microscopy was performed 4 days 1318 post-infiltration. Co-localization signals are shown in the boxed regions. Scale bar, 20 µm. 1319 **b,** Positions and nature of mutations in UDS (UIM-docking site) and LDS (LIR/AIM 1320 docking site) motifs of ATG8a. 1321 **c,d,** Analysis of the interaction of GST-ATG8a and MBP-VDAC2/VDAC3 recombinant 1322 proteins shown in ( **b** ) by pull-down assay. 1323 **e,** Confocal microscopy of Arabidopsis protoplasts transiently expressing eGFP, eGFP-D3, 62 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.28.758855; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.28.758855) preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1324 eGFP-D4, or eGFP-D5. Protoplasts were stained with 200 nM MitoTracker Red for 15 min 1325 prior to imaging. Scale bars, 10 µm. 63