The mechanism of dynamic equilibrium of ascorbate redox status mediated by PbrDHAR5 during scald development in pear fruit
Xu Zhang, Junpeng Niu, Yanmin Du, Lin Guo, Lichao Chen, Min Ma, Xin Qiao, Weiqi Luo, Chunlu Qian, Guodong Wang, Wenhui Wang, Zhen Zhang, Xinli Geng, Qiuqin Zhang, Lanqing Li, Libin Wang, Shaoling Zhang
TL;DR
This study explores how PbrDHAR5 helps maintain ascorbate redox balance in pear fruit during scald development, affecting chilling tolerance.
Contribution
The study identifies PbrDHAR5 and PbrWRKY83 as key regulators in ascorbate redox status during pear scald development.
Findings
PbrDHAR5 catalyzes DHA to AsA, improving fruit chilling tolerance.
PbrWRKY83 activates PbrDHAR5 expression via W-box elements in its promoter.
H2O2-mediated S-sulfenylation of PbrDHAR5 reduces its activity during scald.
Abstract
Ascorbate (AsA) redox status participated in the scald development of Pyrus bretschneideri Rehd. fruit as a cellular redox sensor. By a conjoint analysis of metabolites, enzyme activities and gene expression profiles in AsA-GSH cycle of the chilled pear, PbrDHAR5 was characterized as the candidate gene involved in this process. PbrDHAR5, located in cytosol and nucleus, catalyzed DHA reduction into AsA in vitro and in vivo, elevating AsA redox status and thus fruit chilling tolerance; moreover, the catalytic Cys20 residue in PbrDHAR5 played critical role in this reaction. After analyzing the expression profiles of the differentially expressed TFs, PbrWRKY83 demonstrated higher correlation with PbrDHAR5 than others. PbrWRKY83, located in nucleus, could interact with the only two W-box elements in PbrDHAR5 promoter as monomer and then activate its expression, leading to the improvement of…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8- —the Natural Science Foundation of Guangxi
- —the Municipal Science and Technology Project of Alar (Xinjiang) in 2022
- —the China Postdoctoral Science Foundation
- —the National Natural Science Foundation of China
- —the Fundamental Research Funds for the Central Universities
- —the Seed Industry Promotion Project of Jiangsu
- —the Guidance Foundation of the Hainan Institute of Nanjing Agricultural University
- —the Jiangsu Agriculture Science and Technology Innovation Fund
- —the Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the Earmarked Fund for China Agriculture Research System
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPostharvest Quality and Shelf Life Management · Plant Stress Responses and Tolerance · Plant responses to water stress
Core
The H_2_O_2_-mediated S-sulfenylation of Cys^20^ residue in PbrDHAR5 weakens the role of PbrWRKY83-PbrDHAR5 module, which positively regulates ascorbate (AsA) redox status during scald development in fruit (Pyrus bretschneideri Rehd.).
Gene and accession numbers
Sequence data in this article can be found in the database of the pear genome database (http://peargenome.njau.edu.cn/) under the accession numbers: PbrGR1 (Pbr030956.1), PbrGR2 (Pbr019547.1), PbrGR3(Pbr025294.1), PbrGR4 (Pbr009065.1), PbrDHAR1(Pbr026456.1), PbrDHAR2(Pbr032314.1), PbrDHAR3(Pbr016672.1), PbrDHAR4(Pbr024548.1), PbrDHAR5 (Pbr038163.1), PbrAPX1(Pbr023311.1), PbrAPX2(Pbr033934.1), PbrAPX3 (Pbr014180.2), PbrAPX4(Pbr020588.1), PbrAPX5(Pbr020590.1), PbrAPX6 (Pbr020734.1), PbrAPX7(Pbr020725.1), PbrAPX8(Pbr008291.1), PbrAPX9 (Pbr034821.1), PbrAPX10(Pbr027845.1), PbrAPX11(Pbr042913.1), PbrAPX12 (Pbr027136.1), PbrAPX13(Pbr027137.1), PbrAPX14(Pbr000988.3), PbrMDHAR1 (Pbr024844.1), PbrMDHAR2(Pbr024419.1), PbrMDHAR3(Pbr033247.1), PbrMDHAR4 (Pbr033249.1), PbrMDHAR5(Pbr017802.1), PbrMDHAR6 (Pbr007556.1), PbrMDHAR7 (Pbr017964.1), PbrMDHAR8(Pbr027037.1), PbrWRKY83 (Pbr008639.1), PbrWRKY84 (Pbr002398.1), PbrTub(Pbr028019.1), and PbrGapdh(Pbr036263.1).
Introduction
In order to meet the year-round market, pear (Pyrus spp.) fruits after harvest are usually stored at cold temperature for several months; however, such handling practice would trigger superficial scald in several cultivars, manifesting as brown or black patches on the epidermis (Hui et al. 2016; Giné-Bordonaba et al. 2020). Over the past years, efforts have been made to screen the effective handling practices to mitigate scald symptom, such as 1-methylcyclopropene (1-MCP) fumigation, diphenylamine (DPA) dipping, proline immersion, controlled atmosphere (CA) storage, etc. (Dias et al. 2020; Zhang et al. 2024). Nevertheless, these methods could not completely resolve the problem.
Superficial scald is proposed to be the outcome of the chilling-induced oxidant/antioxidant imbalance (Gong et al. 2021; Qian et al. 2021). The chilling-induced impairment of the cytochrome-dependent electron transport causes the accumulation of superoxide free radicals (O_2_∙^¯^) and hydrogen peroxide (H_2_O_2_); and the latter two could interact to form hydroxyl radical (∙OH) for α-farnesene oxidation into conjugated trienols (CTols) and 6-methyl-5-hepten-2-one (MHO), a scald trigger (Gong et al. 2021). During cold storage of Pyrus bretschneideri Rehd. cv. ‘Dangshansuli’ and ‘Yali’ fruits, reactive oxygen species (ROS) and CTols accumulated, with an MHO burst when scald took place (Hui et al. 2016; Feng et al. 2018; Zhang et al. 2024). Consistently, exogenous MHO fumigation promoted scald development in ‘Dangshansuli’ pear, while a opposite phenomenon was observed in the 1-MCP/DPA-treated fruit with lower MHO level than that in the control fruit (Hui et al. 2016; Niu et al. 2024). Accordingly, the cellular redox homeostasis plays a key role in scald development of pear (Qian et al. 2021).
As a highly abundant metabolite, ascorbate (AsA) and its redox status (ratio of AsA and dehydroascorbate (DHA)) take part in the plant response to abiotic stresses (Hossain et al. 2017). In addition to its antioxidative role, AsA attends a complex and well-orchestrated adaption to environmental stresses via redox signaling, transcriptional regulation, and protein function modification (Hossain et al. 2017). Furthermore, AsA redox status functions as a reliable sensor which perceives and coordinates their actions, including ROS scavenging and signal transduction, based on the cellular redox state in plant (Hossain et al. 2017). Previously, Larrigaudière et al. (2016) reported that scald resistance was positively associated with AsA level in the epidermal tissue of Pyrus communis L. cv. ‘Beurré D’Anjou’ and ‘Packham Triumph’ fruit; moreover, exogenous AsA treatment could mitigate scald development in fruit (Chellew and Little 1995). Overall, the abovementioned results implicate the involvement of AsA and its redox status in scald development.
AsA level and its redox status in plants depend on the balance of AsA biosynthesis and recycling (Hossain et al. 2017). In higher plants, the dominant route for AsA accumulation is the d-mannose/l-galactose (D-Man/L-Gal) pathway, with GDP-L-galactose phosphorylase (GGP) as the rate-limiting enzyme (Wang et al. 2021). After formation in mitochondria and then transportation into other subcellular organelles, AsA accepts electrons from a wide range of free radicals, producing its oxidized forms-DHA and monodehydroascorbate (MDHA) (Mellidou and Kanellis 2017). The latter two can be reduced via the ascorbate–glutathione (AsA-GSH) cycle into AsA by DHA reductase (DHAR) and MDHA reductase (MDHAR), respectively (Mellidou and Kanellis 2017; Wang et al. 2021). Heterogenous overexpression of the chloroplastic PbrDHAR2 (gene ID: Pbr016672.1) from Pyrus sinkiangensis would elevate AsA level and its redox status in tomato, alleviating the deleterious impact of saline and chilling conditions (Qin et al. 2015). Similar phenomenon was observed in the AtDHAR1-overexpressing potato with the upregulated tolerance to herbicide, drought and salt stresses (Eltayeb et al. 2011).
For horticultural fruits, the expression of the structural gene is transcriptionally under the control of TFs (Jia et al. 2023). A bunch of TFs in WRKY, MYB, bZIP, NAC, and CBF families have been characterized to participate in the plant accommodation to abiotic environment (Liu et al. 2015; Song et al. 2022). For example, PbrWRKY62 from ‘Dangshansuli’ could bind to the W-box element in arginine decarboxylase 1 (PbrADC1) promoter and then initiate its expression, improving chilling tolerance of fruit (Zhang et al. 2024). Similarly, PbrMYB5 alleviated chilling sensitivity of Pyrus betulaefolia via upregulating PbrDHAR2 transcription and thus AsA redox status (Xing et al. 2019). However, the molecular mechanism on the role of AsA and its redox status in scald development has not been completely clarified until recently.
In addition to the transcriptional regulation, a bunch of the posttranslational modifications of proteins, creating a complex landscape of protein diversity and function, have been discovered in plant response to abiotic condition (Waszczak et al. 2014; Guerra et al. 2015; Matamoros and Becana 2021). For example, H_2_O_2_ could oxidize the proteinaceous cysteinyl thiols to sulfenic acid, known as S-sulfenylation (Cys-SOH), thereby affecting protein stability, conformation, activity, interaction, and subcellular localization (Waszczak et al. 2014; Huang et al. 2019). A total of 1,537 S-sulfenylated sites had been mapped from 1,394 proteins in Arabidopsis (Huang et al. 2019); further experiments uncovered that the H_2_O_2_-mediated S-sulfenylation of the catalytic Cys^20^ residue in AtDHAR2 weakened its function (Waszczak et al. 2014; Do et al. 2016). Besides Cys^20^ residue in AtDHAR2, several other S-sulfenylazed sites have been identified from proteins in D-Man/L-Gal pathway and AsA-GSH cycle of Arabidopsis, such as glucose-6-phosphate isomerase 1 (GPI1), galactose dehydrogenase (GalDH), ascorbate peroxidases (APXs), etc. (Huang et al. 2019). Taking account of H_2_O_2_ accumulation during scald development in ‘Wujiuxiang’ (Pyrus bretschneideri Rehd. × Pyrus communis L.) and ‘Yali’ fruits as well as the role of Cys-SOH in the oxidative stress signal transduction, S-sulfenylation of the AsA-metabolism-related proteins might play a critical role in this process (Waszczak et al. 2014; Gao et al. 2015; Wang et al. 2018).
In this study, we firstly characterized the role of AsA redox status in scald development of white pear. A further study revealed that PbrDHAR5 and its upstream regulator PbrWRKY83 participated in fruit response to chilling stress via regulating AsA redox status. Finally, we explored that the H_2_O_2_-mediated S-sulfenylation of Cys^20^ residue in PbrDHAR5 weakened its function, thus attenuating the effect of PbrWRKY83-PbrDHAR5 module upon scald development in fruit.
Results
Exploration of AsA redox status’ role during scald development in pear fruit
As shown in Fig. 1a-i and S1a-c, scald symptom occurred in the epidermal tissue after 120-d chilling exposure of ‘Dangshansuli’ fruit, and then expanded to all fruit on day 180, which was associated with the accumulation of H_2_O_2_ and CTols. Exogenous 1-MCP and DPA treatments inhibited H_2_O_2_ and then CTols formation, thus mitigating scald development. Meanwhile, the strong positive associations were detected among H_2_O_2_, CTols, scald incidence, and scald index (correlation coefficient ≥ 0.6; Fig. S1d).Fig. 1. Exploration of AsA redox status’ role during scald development in pear fruit. a Impact of 1-MCP and DPA treatments on AsA metabolism in ‘Dangshansuli’ fruit. (a-i) Visual quality. (a-ii) Impact of 1-MCP and DPA treatments on the AsA-related metabolites. Color scale represents normalized log_2_-transformed (mean value of three biological replicates + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. (a-iii) Correlations among attributes. Pearson correlation between different attributes is visualized as a heatmap, where red line or dot demonstrates positive association, while blue line or dot indicates negative correlation. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. b Alternation of AsA metabolism upon scald development in ‘Yali’ fruit. (b-i) Visual quality. (b-ii) AsA redox status and H_2_O_2_ content. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature. c AsA metabolism in ‘Yali’ fruit of different scald grades. (c-i) Visual quality. (c-ii) H_2_O_2_ content and AsA redox status. ‘Yali’ fruit of different scald grades were sampled for after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature. d Impact of exogenous AsA application on scald development in ‘Yali’ fruit. (d-i) Scald incidence and index. (d-ii) H_2_O_2_ content and AsA redox status. ‘Yali’ fruit were randomly divided into three groups, including 0.0 (control), 1.0, or 2.0 g L^−1^ AsA immersion, before sampling after 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. e Impact of exogenous DHA treatment on scald development in ‘Yali’ fruit. (e-i) Scald incidence and index. (e-ii) AsA redox status and H_2_O_2_ content. ‘Yali’ fruit, which were treated with 0.0 (control) and 1.0 g L.^−1^ DHA, were sampled after a 120-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples at the same sampling time (p < 0.05)
Due to its function in H_2_O_2_ scavenging (Hossain et al. 2017), we then assayed the dynamic change of AsA metabolism. As illustrated in Fig. 1a-ii, AsA fluctuated with the prolonged chilling exposure, while total AsA (T-AsA) and DHA increased in accompany with the reduction of AsA redox status; however, these alternations were inhibited by pre-storage 1-MCP and DPA treatments. Further analyses uncovered that AsA redox status was strongly negatively correlated with H_2_O_2_ and scald development (correlation coefficient ≤ −0.6; Fig. 1a-iii).
To detect if such phenomenon existed in other Pyrus bretschneideri Rehd. cultivars, ‘Yali’ fruit was used as the material. As displayed Fig. S2, several programmed cell death-related morphological alterations occurred in the epidermal tissue of the scalded fruit, including plasmolysis, cell shrinkage, cytosolic condensation, vacuolar collapse, subcellular organelle swelling, and DNA fragmentation. Moreover, when compared with the unscalded fruit, higher H_2_O_2_ and lower AsA redox status were detected in the scalded ‘Yali’ pear (Fig. 1b), and these changes became more severe with the increment of scald grade (Fig. 1c). These results implicated the involvement of H_2_O_2_ and AsA redox status in scald development of pear fruit.
As the role of H_2_O_2_ in scald development has previously been explored (Hu and Zhao 1999), we then assayed the involvement of AsA redox status in this process. As shown in Fig. 1d and S3a, exogenous AsA immersion of ‘Yali’ fruit elevated AsA redox status, suppressed H_2_O_2_ accumulation and morphological alterations, thereby mitigating scald development; and such impact was correlated with AsA concentration. On the other hand, an opposite phenomenon was observed in the DHA-treated fruit, which possessed lower AsA redox status, higher H_2_O_2_ level, and more severe morphological alterations than those in the control (Fig. 1e and S3b).
Subsequently, we detected the relationship between H_2_O_2_ and AsA redox status during scald development. As shown in Fig. S4, postharvest H_2_O_2_ treatment of ‘Yali’ fruit inhibited AsA redox status, and thus promote scald development. Besides, we also transformed fruit with a cytAPX gene from pear, whose homologue from pea demonstrated the potential to inhibit endogenous H_2_O_2_ accumulation in the transgenic plant (Diaz-Vivancos et al. 2013), to validate the abovementioned phenomenon.
In agreement with the phylogenetic result (Fig. S5a-i), PbrAPX2, without any signal peptide and transmembrane domain (Fig. S6a-i and S6b-i), was located in the cytosol of Arabidopsis protoplast (Fig. S5a-ii). Transient overexpression of PbrAPX2 in the epidermal tissue of ‘Dangshansuli’ fruit upregulated cytAPX activity in accompany with the reduced H_2_O_2_ accumulation, causing the elevation of AsA redox status (Fig. S5b-i); nevertheless, a reverse change was observed in the PbrAPX2-silenced fruit (Fig. S5b-ii). Similar result was observed in the PbrAPX2-overexpressing tomato fruit with the enhanced chilling resistance (Fig. S5c). Taken together, our results suggested that AsA redox status functions in scald development as a cellular redox sensor.
Characterization of PbrDHAR5 as the candidate gene regulating AsA redox status during scald development
Subsequently, we intended to characterize the candidate genes responsible for the alternation of AsA redox status, which was proposed to be under the control of AsA-GSH cycle (Fig. 2a) (Hossain et al. 2017; Mellidou and Kanellis 2017; Wang et al. 2021).Fig. 2. Characterization of PbrDHAR5 as the candidate gene regulating AsA redox status during scald development. a AsA-metabolic pathway in plants. The schematic model was drawn based on the results of previous report (Wang et al. 2021). b Impact of 1-MCP and DPA treatments on enzyme activities and gene expression profiles in AsA-GSH cycle during cold storage of ‘Dangshansuli’ fruit. (b-i) cytDHAR and cytMDHAR activities. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples at the same sampling time (p < 0.05). (b-ii) Gene expression profiles and their correlation with quality and physio-biochemical attributes. Data, adapted from transcriptome assay, represent the mean value of three biological replicates. Color scale represents normalized log_2_-transformed (mean FPKM + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. Pearson correlation between different attributes is visualized as a heatmap, where extremely strong (or strong) positive association is connected with red (or light red) line, while extremely strong (or strong) negative correlation with blue (or light blue) line (Long et al. 2014). ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. c Alternation of gene expression profiles in AsA-GSH cycle upon scald development in ‘Yali’ fruit. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature. Data, adapted from transcriptome assay, represent the mean value of two biological replicates. Color scale represents normalized log_2_-transformed (mean FPKM + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. d Expression profiles of genes in AsA-GSH cycle in ‘Yali’ fruit of different scald grades. (d-i) Gene expression profiles. (d-ii) Correlations among attributes. ‘Yali’ fruit of different scald grades were sampled for after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature. Data, adapted from transcriptome assay, represent the value of one biological replicate. Color scale represents normalized log_2_-transformed (FPKM + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. Pearson correlation between attributes is visualized as a heatmap, where red color demonstrates positive association, while blue indicates negative correlation
As illustrated in Fig. 2b-i, both cytMDHAR and cytDHAR activities steadily increased with the prolonged chilling exposure of ‘Dangshansuli’ fruit; exogenous 1-MCP and DPA treatments suppressed cytDHAR activity, but enhanced cytMDHAR activity. Based on the transcriptome result, 24 out of 25 genes in AsA-GSH cycle were expressed during low-temperature storage of ‘Dangshansuli’ fruit (Wang et al. 2021), with diverse expression patterns (Fig. 2b-ii and Table S3). The mRNA abundances of PbrDHAR5 and PbrAPX2/14, which were inhibited by 1-MCP and DPA treatments, gradually increased (Fig. 2b-ii and Table S3); and they were strongly positively correlated with H_2_O_2_ content, cytDHAR activity, and scald incidence/index, but negatively with AsA redox status (absolute correlation coefficient ≥ 0.6; Fig. 2b-ii and Table S3). However, a reverse phenomenon was observed for PbrAPX10 and PbrMDHAR7/8 (absolute correlation coefficient ≥ 0.6; Fig. 2b-ii and Table S3). Quantitative real-time polymerase chain reaction (qRT-PCR) validated the accuracy of transcriptome results on the expression patterns of several randomly selected genes (Fig. 2b-ii, S7a and Table S3). As plant DHARs directly reduce DHA into AsA and thus regulate AsA redox status (Hossain et al. 2017; Mellidou and Kanellis 2017; Wang et al. 2021), then PbrDHAR5 was paid more attention in further study.
PbrDHAR5, whose protein sequences were identical in two cultivars (Fig. S8a), plays a similar role in scald development of ‘Yali’ fruit as well. As shown in Fig. 2c, S7b-i and Table S4, PbrDHAR5 expression was elevated upon scald development (Experiment II: fold change ≥ 1.5 and FDR < 0.05). Additionally, it was strongly positively associated with H_2_O_2_ level, but negatively with AsA redox status in fruit of different scald grades (Experiment III: absolute correlation coefficient ≥ 0.6, Fig. 2d, S7b-ii and Table S5).
Based on the abovementioned results, we further validated the roles of H_2_O_2_ and AsA redox status in regulating PbrDHAR5 transcription with the aid of exogenous chemical treatments. As shown in Fig. S9, the expression of PbrDHAR5 was upregulated by H_2_O_2_ and DHA treatments (lower AsA redox status and higher H_2_O_2_ level than those in the control), but downregulated by AsA treatment (higher AsA redox status and lower H_2_O_2_ level than those in the control). Hossain et al. (2017) reported that overexpression of DHAR or MDHAR gene, whose expression was upregulated in association with ROS accumulation, could improve AsA redox status and thus plant resistance to oxidative stress. Therefore, PbrDHAR5 probably participated in scald development of pear fruit via regulating AsA redox status, and then it was selected for further study. The similar method was also used by Qin et al. (2015) and Xing et al. (2019) for the selection of PbrDHAR2 and its upstream regulator PbrMYB5 in pear response to abiotic stresses.
Functional validation of PbrDHAR5 in vitro and in vivo
By referring to the previous result (Do et al. 2016), PbrDHAR5, whose protein sequence was highly identical with AtDHAR2 from Arabidopsis and OsDHAR1 from O. sativa (Fig. 3a-i and Table S6), might catalyze the reduction of DHA into AsA by a bi-uni-uni-uni-ping-pong mechanism, with K_m_ and V_max_ of 1.63 mmol L^−1^ and 32.39 mmol s^−1^ kg^−1^ protein, respectively (Fig. 3a-ii and b). Accordingly, we performed molecular docking and molecular dynamic (MD) simulation of PbrDHAR5 and DHA. As displayed in Fig. 3c and S10, both results implicated the involvement of the catalytic residue Cys^20^ in DHA binding; consistently, substitution of the proposed catalytic Cys^20^ residue with Ala completed diminished its function (Fig. 3d). Similar phenomenon was observed for OsDHAR from rice (Do et al. 2016). Besides Cys^20^ residue, Lys^8^ and Lys^210^ residues were also proposed to participate in the reduction of DHA into AsA based on the results of previous report (they were proposed to bind with GSH) (Do et al. 2016) and molecular docking (Fig. 3c). In agreement with this, substitution of Lys^8^ or Lys^210^ residues with Ala just partly inhibited the catalytic reaction (Fig. 3d). However, no interaction between DHA and Lys^8^/Lys^210^ residue was detected during MD simulation; thus, further study was needed to clarified such phenomenon.Fig. 3. Characteristics of PbrDHAR5. a Possible reaction scheme for PbrDHAR5. (a-i) Alignment of PbrDHAR5 with AtDHAR2 and OsDHAR1. Sequence alignment was performed by the Jalview software (Waterhouse et al. 2009). Information on AtDHAR2 from Arabidopsis, PbrDHAR5 from Pyrus bretschneideri Rehd., and OsDHAR1 from O. sativa were summarized in Table S6. The catalytic residue (for example, Cys^20^ residue in PbrDHAR5) of plant DHARs were highlighted in the boxes (Do et al. 2016). (a-ii) Possible reaction scheme for PbrDHAR5. The reaction scheme was proposed based on the result of OsDHAR1 (Do et al. 2016). b Michaelis–Menten curve for DHA reduction into AsA by PbrDHAR5 in vitro. c Three-dimensional binding models of DHA and PbrDHAR5. Amino residues, which form van der Waals interaction with DHA, are highlighted with circles; of these, Cys^20^, Lys^8^, and Lys^210^ formed hydrogen bonds with DHA, which were highlighted with the red dotted lines in the figure. d Alternation of DHAR activity after mutation of Cys^20^, Lys^8^, and Lys^210^ residues in PbrDHAR5. Structures of the wide-type PbrDHAR5 and its mutants (including PbrDHAR5^Cys20Ala^, PbrDHAR5^Lys8Ala^, and PbrDHAR5.^Lys210Ala^) were predicted by AlphaFold2 (Cramer 2021). The residual activity was expressed as a percentage of the wide-type PbrDHAR5, whose activity was set as 1.0. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05)
Subsequently, we analyzed the evolution of plant DHARs. As shown in Fig. S11 and Table S6, a total of 79 DHARs were identified from 26 plants, with diverse physio-chemical properties, gene structures, and motif distributions; nevertheless, the catalytic residue (for example, Cys^20^ residue in PbrDHAR5) as well as Lys^8^ & Lys^210^ residues existed in most plant DHARs (Fig. S12), implying that the bi-uni-uni-uni-ping-pong mechanism was conserved during plant evolution.
Based on the abovementioned outcomes, the function of PbrDHAR5 in vivo was detected. In consistent with the result of phylogenetic analysis (Fig. 4a-i), PbrDHAR5, without any signal peptide and transmembrane domain (Fig. S6a-ii and S6b-ii), displayed the same subcellular localization as its homologue from Arabidopsis (AtDHAR2) (Rahantaniaina et al. 2017) and nuclear marker AtH2B-mcherry (Jia et al. 2023), implying that it is located in both cytosol and nucleus (Fig. 4a-ii and S13). Moreover, when compared with the control, the transient overexpression of PbrDHAR5 in the epidermal tissue of ‘Dangshansuli’ fruit elevated cytDHAR activity and then AsA redox status, but suppressed H_2_O_2_ accumulation (Fig. 4b-i); nevertheless, the PbrDHAR5-silenced fruit showed a reverse phenomenon (Fig. 4b-ii). Subsequently, the *PbrDHAR5-*overexpressing tomato fruit and pear calli, with stable inheritance, were generated to validate the result mentioned above. As shown in Fig. 4c and S14a, PbrDHAR5 expression level and then cytDHAR activity was upregulated in the transgenic lines with enhanced chilling tolerance, causing higher AsA redox status but lower H_2_O_2_ content than those in the control calli and the wide-type tomato fruit.Fig. 4. Functional validation of *PbrDHAR5 *in vivo. a Subcellular localization of PbrDHAR5. (a-i) Phylogenetic analyses of plant DHARs. Information on AtDHARs from Arabidopsis, PbrDHARs from Pyrus bretschneideri Rehd., and OsDHAR1 from O. sativa were summarized in Table S6. Phylogenetic tree was constructed via MEGA7.0 software, using neighbor-joining (NJ) method, with a bootstrap analysis of 1000 replicates (Kumar et al. 2016). (a-ii) Subcellular localization of PbrDHAR5. AtDHAR2-mcherry was used as a marker (Rahantaniaina et al. 2017). b Impact of the transient genetic transformation of ‘Dangshansuli’ fruit on AsA redox status. (b-i) Transient overexpression of PbrDHAR5. ‘Dangshansuli’ fruit transformed with the empty pCAMBIA1300 vector with a GFP tag was used as the control for the overexpressing (OE) pear. (b-ii) Transient silence of PbrDHAR5. ‘Dangshansuli’ fruit co-transformed with the empty pTRV2 and pTRV1 was used as the control for the silenced (SE) pear. The expression level of PbrDHAR5 in the control fruit was set as 1.0 for qRT-PCR assay. c Impact of overexpressing PbrDHAR5 in tomato on AsA redox status. (c-i) Visual quality change and chilling injury index. (c-ii) PbrDHAR5 expression level and cytDHAR activity. (c-iii) AsA redox status and H_2_O_2_ content. Tomato fruit at 35 DAFB, including the wide-type (control) and overexpressing (OE) lines, were harvested and then exposed to 4 °C for 10 d followed by a 7-d shelf life at 20 °C. The expression level of PbrDHAR5 in the OE-5 fruit was set as 1.0 for qRT-PCR assay. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05)
Identification of the possible upstream regulators of PbrDHAR5
The transcription of the structural gene in horticultural fruit was mediated by TFs (Jia et al. 2023). In this study, 33 out of 38 differentially expressed TFs, whose transcription were consistently downregulated after 1-MCP and DPA treatments, showed strong positive correlations with PbrDHAR5 during cold storage of ‘Dangshansuli’ fruit based on transcriptome result (fold change ≥ 1.5 and FDR < 0.05; correlation coefficient ≥ 0.6; Fig. S15a and Table S7). Of these, the expression of six members were upregulated upon scald development in ‘Yali’ fruit (Experiment II: fold change ≥ 1.5 and FDR < 0.05; Fig. S15b and Table S8) and displayed strong positive associations with PbrDHAR5 transcripts in fruit of different scald grades (Experiment III: correlation coefficient ≥ 0.6; Fig. S15c and Table S9), including Pbr008639.1, Pbr002398.1, Pbr016145.1, Pbr019902.1, Pbr037846.1, and Pbr025141.1. These six TFs might act as the possible upstream regulators of PbrDHAR5.
Pbr008639.1, named as PbrWRKY83 (Huang et al. 2015), was then selected for further study due to its relative high coefficient with PbrDHAR5 (Fig. S15a-ii and S15c). qRT-PCR validated transcriptome result on its expression profile upon scald development in pear (Fig. S7b-c, S15a-c and Table S7-S9). Consistent with PbrDHAR5, the expression of PbrWRKY83 was upregulated in fruit with lower AsA redox status and higher H_2_O_2_ level (H_2_O_2_ and DHA treatments), but downregulated in fruit with higher AsA redox status and lower H_2_O_2_ level (AsA treatment) (Fig. S9).
With the aid of the PlantRegMap database (Tian et al. 2020), PbrWRKY83, whose protein sequences were identical in two Pyrus bretschneideri Rehd. cultivars (‘Yali’ and ‘Dangshansuli’) (Fig. S8b), might interact with the only two W-box elements in PbrDHAR5 promoter (1502-bp sequences upstream of the transcriptional start site (ATG), Fig. S15d) (Lescot et al. 2002).
Confirmation of PbrWRKY83 as the upstream regulator of PbrDHAR5
Through the instrumentality of various plant biotechnologies, the function of PbrWRKY83 in regulating PbrDHAR5 expression was measured. As shown in the yeast one-hybrid (Y1H) assay (Fig. 5a), PbrWRKY83 illustrated the transcriptional activation capacity. When compared with the control, about 3.5-fold increase of luciferase/renilla (LUC/REN) ratio was observed in N. benthamiana leaves co-transformed with PbrWRKY83 and reporter containing PbrDHAR5 promoter; and such increment was positively correlated with the number of the W-box elements (Fig. 5b). However, it disappeared after the mutation of two binding sites (Fig. 5b). Consistently, the expression of PbrDHAR5 was upregulated in the PbrWRKY83-overexpressing calli (Fig. S16). In the Y1H assay, the positive control (AD-p53 & p53-AbAi), negative controls (AD & PbrDHAR5pro^S1^-pAbAi, AD & PbrDHAR5pro^S2^-pAbAi), bait-prey co-transformants (AD-PbrWRKY83 & PbrDHAR5pro^S1^-pAbAi, AD-PbrWRKY83 & PbrDHAR5pro^S2^-pAbAi), and co-transformants containing the mutated elements (AD & PbrDHAR5pro^S1mut^-pAbAi, AD & PbrDHAR5pro^S2mut^-pAbAi, AD-PbrWRKY83 & PbrDHAR5pro^S1mut^-pAbAi, AD-PbrWRKY83 & PbrDHAR5pro^S2mut^-pAbAi) grew normally on SD/-Leu medium (Fig. 5c); when aureobasidin A (AbA) was added, the growth of the negative control as well as transformants possessing the mutated binding elements was inhibited, without any influence on the positive control and bait-prey co-transformants (Fig. 5c). Chromatin immunoprecipitation (ChIP) combined with qPCR analyses showed an enrichment of more than three-fold on the PbrDHAR5 promoter fragments harboring the binding elements in comparison with the controls (Fig. 5d). In vitro electrophoretic mobility shift assay (EMSA) indicated that the protein-DNA complexes were formed when His-PbrWRKY83 was incubated with the labeled probes, and such bindings were gradually inhibited with the increased abundances of the unlabeled competitor probes, whereas these complexes disappeared after the mutation of the W-box element (Fig. 5e).Fig. 5. Confirmation of PbrWRKY83 as the upstream regulator of PbrDHAR5. a Transcriptional activation determination. Yeast cell harboring empty BD vector was used as the negative control. b Dual-LUC assay. Transformants containing the empty pSAK277 and each reporter were used as the controls. c Y1H assay. Yeast cell co-transformed with AD-p53 & p53-AbAi was used as the positive control, while yeast cell co-transformed with the empty AD vector and each bait as the negative control. d ChIP-PCR assay. Calli overexpressing the empty vector was used as a negative control. e EMSA assay. The unlabeled PbrDHAR5 promoter fragments possessing the W-box elements were used as competitor probes. The presence and absence of His protein, His-PbrWRKY83 protein, labeled probe, or competitor probe were indicated by “ + ” and “ − ”, respectively. Competitor probe concentrations were 50-fold (50 ×) and 100-fold (100 ×) those of the labeled probe. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05). P1 and P2 in the red circle were wide-type W-box elements for PbrWRKY83 binding (the color of bases was also marked with red), while P1 and P2 in the blue circle were the mutated elements (the color of bases was marked with blue as well)
In combination, these results implied that PbrWRKY83 can interact with two W-box elements in PbrDHAR5 promoter and then activate its expression.
Functional validation of PbrWRKY83 in vivo
Subsequently, the role of PbrWRKY83 in vivo was validated. PbrWRKY83, without any signal peptide and transmembrane domain (Fig. S6a-iii and S6b-iii), accumulated in the nucleus of Arabidopsis protoplast and N. benthamiana leaves as monomer (Fig. 6a-b). Transient overexpression of PbrWRKY83 in ‘Dangshansuli’ fruit substantially upregulated PbrDHAR5 mRNAs and thus AsA redox status, resulting in a lower H_2_O_2_ accumulation than in the control (Fig. 6c-i); whereas, a reverse phenomenon was observed for PbrWRKY83-silenced fruit (Fig. 6c-ii). A similar result was observed after transformation of tomato fruit and pear calli: a considerable upregulation of PbrDHAR5 mRNAs and/or cytDHAR activity was observed in the overexpressing lines, which possessed the elevated AsA redox status and chilling tolerance but the decreased H_2_O_2_ level (Fig. 6d and S14b).Fig. 6. Functional validation of PbrWRKY83. a Subcellular localization of PbrWRKY83. AtH2B-mcherry was used as nuclear indicator (Jia et al. 2023). b PbrWRKY83 self-interaction assay. (b-i) BiFC assay. Transformants containing YFP^N^ & YFP^C^, YFP^N^ & PbrWRKY83-YFP^C^, and PbrWRKY83-YFP^N^ & YFP.^C^ were used as the controls. (b-ii) Y2H assay. Transformants containing AD-T & BD-53, AD-T & BD-Lam, AD & BD, AD & BD-PbrWRKY83, and AD-PbrWRKY83 & BD were used as the controls. c Impact of the transient genetic transformation of ‘Dangshansuli’ fruit on AsA redox status. (c-i) Transient overexpression of PbrWRKY83. ‘Dangshansuli’ fruit transformed with the empty pCAMBIA1300 vector with a GFP tag was used as the control for the overexpressing (OE) pear. (c-ii) Transient silence of PbrWRKY83. ‘Dangshansuli’ fruit co-transformed with the empty pTRV2 and pTRV1 was used as the control for the silenced (SE) pear. The expression level of PbrWRKY83 in the control fruit was set as 1.0 for qRT-PCR. d Impact of overexpressing PbrWRKY83 in tomato on AsA redox status. (d-i) Visual quality change and chilling injury index. (d-ii) PbrWRKY83 expression level and cytDHAR activity. (d-iii) AsA redox status and H_2_O_2_ content. Tomato fruit at 35 DAFB, including the wide-type (control) and overexpressing (OE) lines, were harvested and then exposed to 4 °C for 10 d followed by a 7-d shelf life at 20 °C. The expression level of PbrWRKY83 in the OE-2 fruit was set as 1.0 for qRT-PCR assay. Data represented the mean value of three biological replicates; and vertical bars labeled with the same small letter were not significantly different between samples (p < 0.05)
Detection of the S-sulfenylated PbrDHAR5 in vivo
In this study, DHA content, PbrDHAR5 mRNA abundance, and cytDHAR activity accumulated during chilling exposure of ‘Dangshansuli’ fruit (Figs. 1a-ii and 2b and Table S3). Thus, to clarify the reason for the reduction of AsA redox status, we proposed a new conception: DHA reduction frequency (it was defined as the ratio of DHAR activity to DHA content). As shown in Fig. 1a, S1a and S17a, DHA reduction frequency displayed a falloff trend, which was strongly positively associated with AsA redox status, but negatively with scald development (absolute correlation coefficient ≥ 0.6). A similar phenomenon was observed upon scald development in ‘Yali’ fruit: in spite of the upregulation of PbrDHAR5 transcription upon scald development, DHA reduction frequency gradually decreased in association with the increment of scald grade and the reduction of AsA redox status (Figs. 1b-c and 2c-d, S17b-c and Table S4-S5). Conclusively, these results implied that DHA reduction frequency could explain the reason for the alternation of AsA redox status upon scald development in pear, and the relatively lower increment of cytDHAR activity than that of DHA level might be responsible for the falloff of DHA reduction frequency and thus AsA redox status. Then, we tried to explore if some factors could impact the increment of cytDHAR activity during scald development.
By a combination of transcriptome and proteome assay, a contradictory phenomenon was observed between PbrDHAR5 expression level and its protein abundance upon scald development in ‘Yali’ fruit in association with the elevated H_2_O_2_ level: scald development promoted PbrDHAR5 transcription without any impact on its protein abundance (Figs. 1b, 2c, S18 and Table S4); moreover, exogenous H_2_O_2_ suppressed cytDHAR activity in the scalded ‘Yali’ fruit, while DTT treatment demonstrated a opposite function (Fig. S19). By considering the characteristic of its homologue from Arabidopsis (AtDHAR2) (Waszczak et al. 2014) as well as H_2_O_2_ difference between the (un)scalded ‘Yali’ fruit (Fig. 1b), these outcomes implicated the existence of the S-sulfenylated PbrDHAR5 upon scald development in pear.
Based on the analytic result from pCysMod database (Li et al. 2021) as well as S-sulfenyl proteome (unpublished data), the only one cysteine residue in PbrDHAR5 (the catalytic Cys^20^ residue) would be S-sulfenylated by H_2_O_2_ (Fig. 7a), probably weakening its activity. Such hypothesis was validated based on the method used for AtDHAR2 (Waszczak et al. 2014). As shown in Fig. 7b, the H_2_O_2_-mediated S-sulfenylation of PbrDHAR5 in vitro would suppress it activity; and its impact was positively associated with H_2_O_2_ concentration and incubation time (absolute correlation coefficient ≥ 0.6), although extra addition of dithiothreitol (DTT) could relieve such negative effect (Fig. 7b). In vitro study, we just incubated PbrDHAR5 and H_2_O_2_ for a relative short time to the detect the impact of S-sulfenylation on protein function; however, the wide-type PbrDHAR5 and H_2_O_2_ continuously existed in vivo (in cytosol of epidermal tissue) for a relative long time, which increased the possibility of S-sulfenylation. Moreover, although in vitro H₂O₂ treatment likely exceed physiological levels in fruit tissue, H_2_O_2_ level in vivo (about 0.05 mmol L^−1^) were in the range of this study (from 0.0 to 2.0 mmol L^−1^).Fig. 7. Detection of the S-sulfenylated of PbrDHAR5 in vitro and in vivo. a Prediction of the possible S-sulfenylated Cys residues in PbrDHAR5 with the aid of pCysMod database (Li et al. 2021). b Impact of exogenous H_2_O_2_ and DTT treatments on PbrDHAR5 activity. (b-i) Alternation of PbrDHAR5 activity after H_2_O_2_ and/or DTT incubation. The residual activity was expressed as a percentage of the control (H_2_O treatment), whose activity was set as 1.0. (b-ii) Impact of H_2_O_2_ concentration on PbrDHAR5 activity. The residual activity was expressed as a percentage of the control (0.0 mmol L.^−1^ H_2_O_2_). (b-iii) Impact of H_2_O_2_ incubation time. The residual activity was expressed as a percentage of the initial (0 min). c Alternation in the abundance of the S-sulfenylated PbrDHAR5 during cold storage of ‘Dangshansuli’ fruit with(out) the pre-storage chemical treatments. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. d Alternation in the abundance of the S-sulfenylated type upon scald development in ‘Yali’ fruit. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature. For quantification of protein abundance, Image J software was used and signals from three independent experiments were quantified. Relative abundance of the S-sulfenylated DHAR5 was calculated as the ratio of the S-sulfenylated PbrDHAR5 and loading control (PbrDHAR5) in the sample. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05)
Afterwards, the S-sulfenylated PbrDHAR5 in vivo was further detected. As shown in Fig. 7c, the sulfenylated PbrDHAR5 accumulated with the prolonged cold storage of ‘Dangshansuli’ pear; meanwhile, 1-MCP and DPA treatments inhibited the formation of H_2_O_2_ and then the sulfenylation of PbrDHAR5 in fruit (Fig. 7c). Similarly, a higher level of the sulfenylated PbrDHAR5 was detected in the scalded ‘Yali’ pear than that in the unscalded fruit (Fig. 7d).
Taken together, our study implied that the H_2_O_2_-mediated S-sulfenylation of Cys^20^ residue in PbrDHAR5 occurres in vivo, partly responsible for the reduction of AsA redox status and thus scald development.
Discussion
Superficial scald, triggered by the oxidation products of α-farnesene, is a major physiological disorder occurred after chilling exposure of some pear cultivars (Giné-Bordonaba et al. 2020). Consistent with the previous result (Hui et al. 2016), scald developed and expanded in the epidermal tissue during cold storage of ‘Dangshansuli’ and ‘Yali’ in association with the accumulation of H_2_O_2_ and/or CTols but the reduction of AsA redox status (Fig. 1a-c and S1); nevertheless, exogenous 1-MCP and DPA treatments could maintain a relatively high AsA redox status, inhibit H_2_O_2_ and CTols formation, thereby compromising the deleterious effect of chilling stress (Fig. 1a and S1a-c). The positive role of 1-MCP fumigation on maintaining AsA redox status has been uncovered in apple fruit as well (Lv et al. 2024). Although H_2_O_2_ was validated to participate in scald development of pear (Hu and Zhao 1999) (Fig. S4), the involvement of AsA redox status in this physiological disorder was still un-fully clarified.
AsA redox status functions in plant response to abiotic stresses as a redox sensor (Hossain et al. 2017). Exogenous AsA treatment illustrated a potential to increase AsA redox status and thus mitigate scald development in ‘Beurré D’Anjou’ and ‘Packham Triumph’ fruit (Asghar et al. 2023; Chellew and Little 1995), while H_2_O_2_ exerted a opposite impact on AsA redox status in plant (Asghar et al. 2023); additionally, application of DHA promoted H_2_O_2_ accumulation in Meloidogyne graminicola (Chavan et al. 2022). Besides chemical treatments, overexpression of cytAPX could suppressed H_2_O_2_ accumulation in plant as well (Diaz-Vivancos et al. 2013; Guo et al. 2016). In this study, with the aid of the exogenous chemical (AsA, DHA, and H_2_O_2_) treatments and the PbrAPX2-transgenic pear and tomato, AsA redox status was confirmed to play a crucial role in scald development of pear as a reliable redox sensor (Fig. 1d-e and S4-S5).
AsA redox status in plants is mainly under the control of AsA regeneration ability (Hossain et al. 2017; Wang et al. 2021). Overexpression of DHAR or MDHAR gene, whose transcription was upregulated by ROS, could improve AsA redox status and thus plant tolerance to oxidative stress (Hossain et al. 2017). By a conjoint analysis of metabolites, enzyme activities and gene expression profiles in AsA-GSH cycle followed by experimental validation PbrDHAR5, located in both cytosol and nucleus, catalyzed the reduction of DHA in vitro and in vivo, improving AsA redox status and thus chilling tolerance of pear and tomato (Figs. 1, 2, 3 and 4, S13-S14 and Table S3-S5); and the catalytic Cys^20^ residue in PbrDHAR5 was of great importance for this reaction (Fig. 3d). These phenomena agreed with the function of its homologues from Arabidopsis (Eltayeb et al. 2006; 2011; Rahantaniaina et al. 2017) and O. sativa (Do et al. 2016).
TFs in horticultural fruits regulate the expression of the downstream structural genes (Jia et al. 2023). Until recently, a bunch of TFs, such as CsWRKY46, OsWRKY71, PbrWRKY62, AchnABF1, PbrMYB5, etc., have been explored to participate in plant response to chilling stress (Kim et al. 2016; Zhang et al. 2016, 2024; Xing et al. 2019; Jin et al. 2021). In this study, six TFs were characterized as the positive regulators of PbrDHAR5 (Fig. S15 and Table S7-S9); of these, the nucleus-located PbrWRKY83 was validated to directly interact with two W-box elements in PbrDHAR5 promoter as monomer and then activate its expression, thereby upregulating AsA redox status and thus chilling tolerance of the transgenic pear and tomato (Figs. 5 and 6, S14b, S16 and Table S8-S9). Consistently, the PbrMYB5-PbrDHAR2 and PbrWRKY62-PbrADC1 modules functioned in pear adaption to low-temperature environment as well (Xing et al. 2019; Zhang et al. 2024).
Although PbrDHAR5 expression was elevated upon scald development in ‘Yali’ fruit, there was no difference in its protein abundance between samples, implying the existence of the posttranslational modification (Fig. 2c, S18 and Table S4). By considering the reduction of DHA reduction frequency (Fig. S17), the characteristic of its homologue from Arabidopsis (AtDHAR2) (Waszczak et al. 2014), H_2_O_2_ difference between unscalded and scalded ‘Yali’ fruit (Fig. 1b) as well as cytDHAR activity change after H_2_O_2_ and DTT treatments (Fig. S19), the sulfenylation of Cys^20^ residue in PbrDHAR5 by H_2_O_2_, which would weaken its function, was validated in vitro and in vivo (Fig. 7). And the increment of this posttranslational modification during scald development might be partly responsible for the decrement of AsA redox status and then scald development in the epidermal tissue of pear fruit (Figs. 1a-c and 7c-d). Previously, an increment of S-sulfenylation on the cysteine residues of proteins, such as cytosolic DHAR2, mitochondrial citrate synthase 4, and mitochondrial aconitase 3, etc., were detected in Arabidopsis cell after H_2_O_2_ treatment (Waszczak et al. 2014; Huang et al. 2019).
Taken together, our results explored the mechanism of dynamic equilibrium of ascorbate redox status mediated by PbrDHAR5 during scald development in pear fruit. As shown in Fig. 8, low-temperature exposure promotes H_2_O_2_ accumulation and then the reduction of cellular redox sensor (AsA redox status) in Pyrus bretschneideri Rehd. fruit. Then, the latter induces the expression of PbrWRKY83, whose encoding protein could bind to two W-box elements in PbrDHAR5 promoter as monomer and then activates its transcription. After translation in ribosome and then import into cytosol/nucleus, PbrDHAR5 catalyzes the reduction of DHA into AsA for the maintenance of AsA redox status. However, the chilling-induced accumulation of H_2_O_2_ could S-sulfenylate Cys^20^ residue in PbrDHAR5 and thus attenuates its function, which is partly responsible for the decrement of AsA redox status and thus scald development in the epidermal tissue of pear fruit.Fig. 8. Schematic model on the mechanism of dynamic equilibrium of AsA redox status mediated by PbrDHAR5 during scald development in pear fruit. Low-temperature exposure promotes H_2_O_2_ accumulation and then the reduction of cellular redox sensor (AsA redox status) in Pyrus bretschneideri Rehd. fruit. Then, the latter induces the expression of PbrWRKY83, whose encoding protein could bind to two W-box elements in PbrDHAR5 promoter as monomer and then activates its transcription. After translation in ribosome and then import into cytosol/nucleus, PbrDHAR5 catalyzes the reduction of DHA into AsA for the maintenance of AsA redox status. However, the chilling-induced accumulation of H_2_O_2_ could S-sulfenylate Cys^20^ residue in PbrDHAR5 and thus attenuates its function, which is partly responsible for the decrement of AsA redox status and thus scald development in the epidermal tissue of pear fruit
Conclusion
AsA redox status acted as a cellular redox sensor during scald development in pear. The cytosolic/nuclear PbrDHAR5, which catalyzed the reduction of DHA, played a critical role in scald development through regulating AsA redox status; and the catalytic Cys^20^ residue in PbrDHAR5 was of great importance for its function. Additionally, PbrWRKY83, located in nucleus, directly mediated PbrDHAR5 expression as monomer, upregulated cytDHAR activity and AsA redox status, consequently mitigated chilling injury of pear and tomato. Further study uncovered that the H_2_O_2_-mediated S-sulfenylation of Cys^20^ residue in PbrDHAR5 accumulated upon scald development, and then weaken its function, partly accounting for the decrement of AsA redox status. In combination, our results suggest that PbrWRKY83 and its target gene PbrDHAR5 participate in pear scald development via regulating AsA redox status; nevertheless, their function is attenuated by the H_2_O_2_-mediated S-sulfenylation of Cys^20^ residue in PbrDHAR5.
Materials and methods
Experimental materials and treatments
Experiment I
Uniform and defect-free Pyrus bretschneideri Rehd. cv. ‘Dangshansuli’ fruit, after harvest from homogeneous trees from an experimental orchard in Weinan city, were immediately transported to the laboratory and then randomly divided into three treatments before storage at 0.5 °C: H_2_O (control) or 2.0 mg L^−1^ DPA dipping for 120 s, 1.0 μL L^−1^ 1-MCP fumigation for 24 h. Fruit were sampled every 60 d followed by a 7-d shelf life at ambient temperature.
Experiment II
Pyrus bretschneideri Rehd. cv. ‘Yali’ fruit, uniform and defect-free, were harvested from homogeneous trees in an experimental orchard in Xinji city. Fruit, with and without scald symptom, were sampled after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature.
Experiment III
‘Yali’ fruit, uniform and defect-free, were harvested from an orchard in Xinji city. Fruit of different scald grades were sampled after a 180-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature (Hui et al. 2016).
Experiment IV
Uniform and defect-free ‘Yali’ fruit from an orchard in Xinji city were randomly divided into three treatments: 0.0 (control), 1.0, or 2.0 g L^−1^ AsA immersion for 30 min before storage at −0.5 °C for 180 d followed by a 7-d shelf life at ambient temperature.
Experiment V
Uniform and defect-free ‘Yali’ fruit from an orchard in Xinji city were randomly divided into two treatments according to a previous method (Hu and Zhao 1999): 0.0 (control) and 5.0 mmol L^−1^ H_2_O_2_ immersion. Fruit was sampled after a 120-d storage at −0.5 °C followed by a 7-day shelf life at ambient temperature.
Experiment VI
Uniform and defect-free ‘Yali’ fruit from an orchard in Xinji city were randomly divided into three treatments: 0.0 (control) and 1.0 g L^−1^ DHA immersion for 30 min before storage at −0.5 °C for 120 d followed by a 7-d shelf life at ambient temperature.
Scald incidence and index analyses
Based on the area of scald symptom in the epidermal tissue, scald grades were assigned as following: Grade 0: no scald symptom; Grade 1: scald area < 25%; Grade 2: 25% ≤ scald area < 50%; Grade 3: scald area ≥ 50% (Hui et al. 2016). Scald incidence and index were calculated following the formulas of the previous study (Zhang et al. 2024):
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Scald}\;\mathrm{incidence}\:=\:\mathrm{Number}\;\mathrm{of}\;\mathrm{scalded}\;\mathrm{fruit}/\mathrm{Total}\;\mathrm{number}\;\mathrm{of}\;\mathrm{fruit}\:\times\:100\%$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Scald}\;\mathrm{index}\:=\:\sum\lbrack\mathrm{Number}\;\mathrm{of}\;\mathrm{fruit}\:\times\:\mathrm{grade}\rbrack/\lbrack\mathrm{Total}\;\mathrm{number}\;\mathrm{of}\;\mathrm{fruit}\:\times\:3\rbrack.$$\end{document}Measurement of the cytosolic metabolites in plant tissue
According to a previous report, the supernatant of the crude extract from epidermal tissue, which was obtained after homogenization and then centrifugation at 50, 000* g* for 15 min, was designated as ‘cytosol’ (Matamoros et al. 2013).
α-Farnesene and CTols in cytosol were measured based on the methods as described previously (Qian et al. 2021). Briefly, 1.0 g tissue was homogenized with 3.0 mL hexane; after incubation in the dark for 20 min, the sample was centrifuged at 50, 000* g* for 15 min before analyses of the supernatant at 232 nm (α-farnesene) and 281–290 nm (CTols).
Cytosolic H_2_O_2_ was measured following the manual of the correspondent assay kit (H_2_O_2_−1-Y, Suzhou Comin Biotechnology Co., Ltd., Suzhou, China). Briefly, sample was homogenized with the extraction buffer, and then centrifugated at 50, 000* g* for 15 min prior to supernatant collection for H_2_O_2_ determination.
AsA, DHA, and T-AsA in cytosol of plant tissue were analyzed based on the previous protocol (Wang et al. 2021). Briefly, 0.5 g tissue was homogenized in 2.0 mL oxalic acid (w/v, 0.1%), and then centrifuged at 50, 000* g* for 15 min at 4 °C. Subsequently, the supernatant was filtered through a 0.45 μm filter before AsA assay by an Ultimate 3000 high-performance liquid chromatography (HPLC) system (Thermo Fisher Scientific Inc., Massachusetts, USA) equipped with a Acquity UPLC HSS T_3_ column (Waters Corp., Massachusetts, USA) and a photodiode array (PDA) detector. T-AsA was measured after addition of DTT into the supernatant, and DHA content was calculated as the difference between the T-AsA and AsA.
Morphological assay
Paraffin sectioning
The epidermal tissue was fixed with the aid of formaldehyde-acetic acid–ethanol (FAA) fixative for 24 h. After the successive immersion in 75–100% ethanol and xylene/ethanol (v:v = 1:1) solutions, sample was waxed, and then transferred to a paper box for slicing.
Safranin O-fast green staining
Safranin O-fast green staining was performed according to the protocol of Schuller and Ludwig‐Müller (2016). Briefly, the paraffin-embedded sample was orderly dewaxed, stained with safranin O solution, and then immersed in ethanol. Subsequently, sample was stained with fast green solution followed by the successive treatments with xylene, xylene-ethanol, and ethanol. After sealing with gum, microscopic analyses were carried out with the aid of a Zeiss Axiolab 5 polarized light microscope (Carl Zeiss AG, Jena, Germany).
Toluidine blue O staining
Toluidine blue O staining was conducted following the previous method (Schuller and Ludwig‐Müller 2016). In brief, after dewaxing and staining with toluidine blue O stain, the paraffin-embedded tissue was sequentially cleaned, dehydrated, blocked with neutral gum, and then dried at 40 °C for 24 h, prior to image capturing by a Zeiss Axiolab 5 polarized light microscope (Carl Zeiss AG, Jena, Germany).
Terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) staining
TUNEL staining was performed according to the previous method (Vizcay-Barrena and Wilson 2006). After dewaxing and rehydration, the paraffin-embedded tissue was immersed in the proteinase K solution followed by washing with phosphate-buffered saline (PBS). Subsequently, sample was dropwise treated with a film-breaking solution, incubated at room temperature, and then added with a mixture of TUNEL enzyme (terminal deoxynucleotidyl transferase, TdT) and TUNEL tag (dUTP) (ratio 1:9) before 4’, 6-diamidino-2-phenylindole (DAPI) staining. Image was captured with the aid of an ortho-fluorescence microscope (DS-Ri2, Nikon Corp., Tokyo, Japan).
Transmission electron microscope (TEM) analysis
TEM analysis was carried out based on the protocol of Vizcay-Barrena and Wilson (2006). Briefly, the epidermal tissue was fixed with the fixative for electron microscopy, incubated at room temperature, rinsed with 0.1 mol L^−1^ pH 7.4 PBS, and then post-fixed with 1% OsO_4_ before dehydration. Afterwards, 2% uranyl acetate and 2.6% lead citrate were used for staining prior to image capture by a transmission electron microscopy (Hitachi HT7800, Hitachi High-Tech Corp., Tokyo, Japan).
Purification of the cytosol and enzyme activity assay
Purification of the cytosol from plant tissue was conducted following the abovementioned procedure. In brief, 0.2 g sample was homogenized with the correspondent extraction buffer in the assay kit (APX-1-W for APX assay, DHAR-1-W for DHAR assay, while MDHAR-1-W for MDHAR assay, Suzhou Comin Biotechnology Co., Ltd., Suzhou, China), prior to 50, 000* g* centrifugation for 15 min. After collection of the supernatant, the cytosolic APX (cytAPX), DHAR (cytDHAR), or MDHAR (cytMDHAR) activity was determined based on the protocol of the related assay kit (Wang et al. 2021). Protein concentration in the supernatant was measured, using the bicinchoninic acid protein assay kit (A045-4, Nanjing Jiancheng Bioengineering Institute, Nanjing, China).
Transcriptome and qRT-PCR analyses
Transcriptome assay was conducted based on the previous protocols (Wang et al. 2018; Guo et al. 2023), with some modifications. Briefly, total RNA was extracted with the aid of the RNAprep Pure Plant Kit (Tiangen Biotech (Beijing) Co., Ltd., Beijing, China). Then, 5.0 μg RNA was used for construction of the complementary DNA (cDNA) library before sequencing in the BGISEQ-500 platform (BGI Life Tech Co., Ltd., Shenzhen, China) (Experiment I and II) or the Illumina HiSeq 2500 platform (Novogene Technology Co., Ltd., Beijing, China) (Experiment III). After quality assessment and data filtering, clean reads were mapped to the Pyrus bretschneideri Rehd. genome database (Wu et al. 2013). Fragments Per Kilobase Million (FPKM) was used to calculate gene expression; and the differentially expressed genes (DEGs) were identified with the NOISeq software, in accordance with the following criteria: fold change ≥ 1.5 and FDR < 0.05.
For qRT-PCR assay, the gene-specific primers were designed by Primer 5.0 software (Table S1). Then, total RNA was isolated using TRizol™ Reagent (Invitrogen Corp., California, USA) followed by RNase-free DNase treatment (Qiagen Inc., California, USA). After analyses of RNA purity and integrity as well as first-strand cDNA synthesis, qRT-PCR assay was performed with the aid of One Step SYBR® PrimeScript™ RT-PCR Kit (Perfect Real Time) (Takara Bio Inc., Shiga, Japan). PbrTub and PbrGapdh were used as the internal reference genes for the gene-overexpressing/silenced pear fruit (Ma et al. 2020; Li et al. 2023), while SlActin-51 and SlCAC were used as the housekeeping genes for the gene-overexpressing tomato fruit (Zhang et al. 2018; Liu et al. 2020). The relative gene expression was calculated based on the 2^−ΔΔCT^ method (Wang et al. 2021).
Proteome assay
The isobaric tags (IBT)-based quantitative proteome was conducted according the previous procedure (Wang et al. 2019). Briefly, total protein was extracted from the epidermal tissue, and then digested by trypsin Gold (Promega Corp., Madison, Wisconsin, USA). Following desalination with a Strata-X C18 column (Phenomenex Inc.), IBT labeling reagents were added into samples, prior to separation using a Gemini C18 column (4.6 mm × 250 mm, 5 μm particles, Phenomenex Inc., California, USA). Finally, the separated peptides were analyzed by tandem mass spectrometry (MS) on Q-Exactive Plus hybrid quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific Inc., California, USA).
Subcellular localization assay
The coding sequences (CDSs) of PbrAPX2, PbrDHAR5 and PbrWRKY83, without the stop codon, was amplified from ‘Dangshansuli’ fruit (Table S1), and then inserted into the pBI221 vector with a green fuorescent protein (GFP) tag. Subsequently, the recombinant plasmid was transformed, together with the correspondent marker, into Arabidopsis protoplasts and/or N. benthamiana leaves prior to the fluorescence signal detection by a TCS SP5 confocal microscope (Leica Microsystems, Wetzlar, Germany) (Jia et al. 2023).
AtH2B-mcherry as nuclear indicator for PbrWRKY83 (Jia et al. 2023), while AtDHAR2-mcherry and AtH2B-mcherry were used as the markers for PbrDHAR5 (Jia et al. 2023; Rahantaniaina et al. 2017).
Functional validation of PbrDHAR5 and its mutants in vitro
PbrDHAR5 CDS was amplified from ‘Dangshansuli’ fruit (Table S1), inserted into the pCold-TF plasmid, and then introduced into E. coli BL21 (DE3) to express the His-tagged fusion protein (Jia et al. 2023). After purification through a Ni–NTA His Bind resin column (Sangong Biotech Inc., Shanghai, China), the recombinant protein was used for further assay (Ma et al. 2020).
To introduce the Cys20Ala (substitution of Cys^20^ residue in PbrDHAR5 with Ala, PbrDHAR5^Cys20Ala^), Lys8Ala (substitution of Lys^8^ residue with Ala, PbrDHAR5^Lys8Ala^), or Lys210Ala (substitution of Lys^210^ residue with Ala, PbrDHAR5^Lys210Ala^) point mutation into PbrDHAR5, the pCold-TF::PbrDHAR5 plasmid was used as a template, and the mutation was achieved by overlapping PCR (Table S1) (Zhang et al. 2024). After confirmation of the mutation by DNA sequencing, the constructed plasmid was transformed into E. coli BL21 (DE3) for protein expression and then purified based on the protocol as PbrDHAR5 (Ma et al. 2020).
In vitro PbrDHAR5, PbrDHAR5^Cys20Ala^, PbrDHAR5^Lys8Ala^, and PbrDHAR5^Lys210Ala^ activities were determined with the aid of the DHAR assay kit (DHAR-1-W; Suzhou Comin Biotechnology Co., Ltd., Suzhou, China) (Wang et al. 2021).
Impact of exogenous H2O2 and DTT treatments on PbrDHAR5 and cytDHAR activities
Impact of exogenous H_2_O_2_ and DTT treatments on PbrDHAR5 activity was conducted based on the previous protocols (Waszczak et al. 2014; Byrne et al. 2020). Briefly, the His-tagged PbrDHAR5 was incubated with H_2_O_2_ of different concentrations (0.0 (control), 0.2, 0.5, 1.0, or 2.0 mmol L^−1^) for 0 (initial), 5, 10, 20, 30, 60, or 90 min. After removement of the excess H_2_O_2_ using the Micro Bio-Spin P-6 gel column (Bio-Rad Laboratories, Marnes-la-Coquette, France) (Waszczak et al. 2014), sample was incubated with(out) DTT before the residual activity measurement (Wang et al. 2021).
Similarly, after purification of the cytosol from the epidermal tissue of the scalded ‘Yali’ fruit followed by filtration through the Micro Bio-Spin P-6 gel column (Bio-Rad Laboratories, Marnes-la-Coquette, France), the supernatant was incubation in the presence of H_2_O_2_ or DTT, prior to the residual cytDHAR activity determination (Wang et al. 2021). H_2_O treatment was used as the control.
Gene function validation in vivo
Transient transformation of pear fruit
(a) Transient gene overexpression The CDSs of PbrAPX2, PbrDHAR5 and PbrWRKY83 were cloned from ‘Dangshansuli’ fruit (Table S1), inserted into the pCAMBIA1300 vector with a GFP tag, and then transformed into A. tumefaciens strain GV3101 before incubation at 28 °C until OD_660_ = 1.0. After resuspension of the bacterial strain in the infiltration buffer (10 mmol L^−1^ MgCl_2_, 10 mmol L^−1^ 2-(4-morpholino)-ethane sulfonic acid (MES), and 150 μmol L^−1^ acetosyringone), 5.0 μL of the solution was slowly and vertically injected into the epidermal tissue of the ripe ‘Dangshansuli’ fruit (Zhang et al. 2024). After 3-d storage in the dark at 4 °C, the epidermis from each injection site was sampled. Fruit infiltrated with the empty vector were used as the control. (b) Transient gene silence Approximately 200-bp fragments of PbrAPX2, PbrDHAR5 and PbrWRKY83 CDSs were amplified from the ‘Dangshansuli’ fruit and then inserted into the pTRV2 vector (Table S1). The recombinant plasmid and pTRV1 were transformed into A. tumefaciens strain GV3101, respectively. Subsequently, the bacterial resuspensions containing the recombinant pTRV2 and pTRV1 were mixed in a 1:1 ratio prior to injection into the epidermal tissue of ‘Dangshansuli’ fruit (Zhang et al. 2024). After 3-d storage in the dark at 4 °C, the epidermis from each injection site was sampled. Fruit co-injected with the empty pTRV2 vector and pTRV1 were used as the control. There were three biological replicates per treatment, with eight fruit per biological replicate.
Transformation of tomato fruit
After construction of the overexpressing vectors as mentioned above, S. lycopersicum cv. MicroTom transformation was performed, using leaf disk or epicotyl as explant (Zhang et al. 2024). The positive transgenic lines were screened on 100 mg L^−1^ kanamycin-containing medium, and then confirmed at the DNA level by PCR and RNA level by qRT-PCR. All plants were grown in a greenhouse (25 °C light for 18 h/18 °C dark for 6 h, 60% relative humidity). Subsequently, tomato fruit from the wild-type (control) and transgenic homozygous lines (T2 generation) were harvested at 35 DAFB and then stored at 4 °C for 10 d followed by a 7-d shelf life at 20 °C. Chilling injury index of tomato fruit was calculated based on the level of uneven ripening (a five-point scale based on the ripening stage for each criterion: 0 = red, 1 = orange, 2 = yellow, 3 = breaker, 4 = mature green) (Zhang et al. 2024).
Transformation of pear calli
The overexpressing vectors, after construction as mentioned above, were introduced into calli, which was induced from fruitlets of P. communis cv. ‘Clapp’s Favorite’ pear (Jia et al. 2023). These calli were then cultivated on hygromycin B (Hyg)-containing Murashige and Skoog (MS) medium with sucrose as the carbon source for 1.5–2 months, prior to PCR analyses to identify the positive lines. Subsequently, the transgenic lines were transferred to Hyg-containing MS medium with sorbitol/sucrose (w:w = 1:1, 15 g L^−1^) as the carbon source. After growth at 20 °C for a period of time, the transgenic calli and control calli (calli transformed with the empty pCAMBIA1300 vector with a GFP tag) were exposure to 4 °C for 15 d. Calli transformed with the empty vector was used as the control.
DNA and protein interaction
Transcriptional activation determination
PbrWRKY83 CDS was clone from ‘Dangshansuli’ fruit (Table S1), and then inserted into the pGBKT7 (BD) vector to generate the BD-PbrWRKY83 vector. After introduction into the yeast two-hybrid (Y2H) Gold strain, yeast was dripped on to SD/-Trp medium in the presence or absence of AbA (Jia et al. 2023). Yeast transformed with the empty BD vector was used as the negative control.
Dual-LUC assay
PbrWRKY83 CDS was amplified and then introduced into the pSAK277 vector (Table S1); on the other hand, PbrDHAR5 promoter fragments which contained diverse numbers of the wide-type W-box element (core motif: TTGACC/T; PbrDHAR5pro, PbrDHAR5pro^frag1^, and PbrDHAR5pro^frag2^) or the mutated elements (TTGACC/T → TTTAGC/T, PbrDHAR5pro^mut^), were inserted into the pGreen 0800-LUC vector (Table S1), producing various reporters. Subsequently, a mixture of A. tumefaciens containing pSAK277-PbrWRKY83 vector and each reporter was infiltrated into N. benthamiana leaves. LUC and REN activities were determined using a dual-LUC reporter assay system (Promega Corp., Madison, Wisconsin, USA) (Zhang et al. 2024). Transformant containing the empty pSAK277 vector and each reporter was used as the control.
Y1H assay
PbrWRKY83 CDS was introduced into the prey vector pGADT7 (AD); on the other hand, about 200-bp fragments of PbrDHAR5 promoter, containing the wide-type W-box element (core motif: TTGACC/T; PbrDHAR5pro^S1^ and PbrDHAR5pro^S2^) or the mutated element (TTGACC/T → TTTAGC/T, PbrDHAR5pro^S1mut^ and PbrDHAR5pro^S2mut^), were inserted into the bait vector pAbAi (Table S1). Matchmaker Gold Yeast One-Hybrid Library Screening System (Shanghai Weidi Industrial Co., Ltd., Shanghai, China) was applied to perform the Y1H assay (Jia et al. 2023). SD/-Ura medium supplemented with AbA was used to check the self-activation of PbrDHAR5pro^S1^, PbrDHAR5pro^S2^, PbrDHAR5pro^S1mut^, and PbrDHAR5pro^S2mut^, and then select the appropriate AbA concentration. Yeast cell co-transformed with AD-p53 & p53-AbAi was used as the positive control, while yeast co-transformed with the empty AD vector and each bait as the negative control.
ChIP-qPCR analyses
The PbrWRKY83-overexpressing calli and the control (empty vector), which were obtained following the protocol as described above, were used for the DNA–protein cross-link. After homogenization and cell lysis, chromatin was obtained and then sonicated to get soluble sheared chromatin with an average DNA length of 200–500 bp. One part was served as input DNA, while the other was used for immunoprecipitation with anti-GFP antibody (Ab290, Abcam) (Zhang et al. 2024). The enrichment of PbrDHAR5 promoter fragments was evaluated via qPCR assay (Table S1).
EMSA assay
The His-tagged recombinant PbrWRKY83 protein was obtained according to the procedure as described above (Table S1). On the other hand, about 30-bp biotin-labeled DNA probe carrying either the wide-type W-box element (core motif: TTGACC/T; PbrDHAR5pro^S1^ and PbrDHAR5pro^S2^) or the mutated element (TTGACC/T → TTTAGC/T; PbrDHAR5pro^S1mut^ and PbrDHAR5pro^S2mut^), as well as unlabeled competitor probes, were synthesized by Sangong Biotech, Co., Ltd. (Shanghai, China) (Table S1). EMSA assay was conducted based on the instruction provided by the Chemiluminescent EMSA Kit (Beyotime Inc., Shanghai, China) (Jia et al. 2023).
Protein self-interaction
Y2H assay
PbrWRKY83 CDS was amplified from ‘Dangshansuli’ fruit (Table S1), and then introduced into the AD and BD vectors. Afterwards, they were co-transformed into S. cerevisiae AH109, and dripped on the synthetic dropout nutrient mediums, including SD/-Leu/-Trp and SD/-Leu/-Trp/-His/-Ade (Ma et al. 2020). Transformants containing AD-T & BD-53, AD-T & BD-Lam, AD & BD, AD & BD-PbrWRKY83, and AD-PbrWRKY83 & BD were used as the controls.
Bimolecular fluorescence complementation (BiFC) analyses
PbrWRKY83 CDS encoding the mature protein were cloned and then inserted into 35S-pSPYNE-YFP^N^ and 35S-pSPYCE-YFP^C^ vectors (Table S1), producing PbrWRKY83-YFP^N^ and PbrWRKY83-YFP^C^. Subsequently, the recombinant constructs were co-transformed into A. tumefaciens strain GV3101 before infiltration into the epidermal cells of N. benthamiana leaves (Ma et al. 2020). Transformants containing YFP^N^ & YFP^C^, YFP^N^ & PbrWRKY83-YFP^C^, and PbrWRKY83-YFP^N^ & YFP^C^ were used as the controls.
Analyses of the S-sulfenylated PbrDHAR5 in vivo
The S-sulfenylated PbrDHAR5 in the epidermal tissue of pear was detected based on the method as described previously (Wang et al. 2017), with some modifications. Briefly, epidermis of the (un)scalded ‘Yali’ fruit was homogenized with lysis buffer, and then centrifugated at 50, 000 g for 10 min prior to incubation of the supernatant with 1 mmol L^−1^ DCP-Bio1 probe (Millipore Corp., Massachusetts, USA) for 2 h. After incubation with the high-capacity streptavidin-agarose beads (Thermo Fisher Scientific Inc., Massachusetts, USA), sample was subjected to a series of stringent wash, using 1% sodium dodecyl sulfate (SDS), 4 mol L^−1^ urea, 1 mol L^−1^ NaCl, 10 mmol L^−1^ DTT, 100 µmol L^−1^ ammonium bicarbonate, and ultrapure water, prior to 10-min boil (Wang et al. 2017).
PbrDHAR5 and the S-sulfenylated PbrDHAR5 in the sample was detected by the immunoblot with a specific antibody according to the method of Zhang et al. (2025). Due to its extreme high similarity in protein sequence with AtDHAR2 from A. thaliana, AtDHAR2 Antibody was used in this study (Species/Host: Rabbit; Catalog Number: orb787290; Biorbyt Ltd. (UK); https://www.biorbyt.com/dhar2-antibody-orb787290.html). For quantification of protein abundance, Image J software was used and signals from three independent experiments were quantified. Relative abundance of the S-sulfenylated DHAR5 was calculated as the ratio of the S-sulfenylated PbrDHAR5 and loading control (PbrDHAR5) in the sample.
In silico analyses
The coding sequences (CDSs) of PbrDHAR5 and PbrWRKY83 were cloned from ‘Dangshansuli’ and ‘Yali’ fruits (Table S1), prior to deduction of the correspondent protein sequences by the Translate tool (https://web.expasy.org/translate/) (Jia et al. 2023). Then, sequence alignment was conducted with the aid of the Jalview software (Waterhouse et al. 2009).
Arabidopsis DHARs were used as a query in BLASTP of plant genome database (https://phytozome-next.jgi.doe.gov/ and http://peargenome.njau.edu.cn/) (Goodstein et al. 2012; Wu et al. 2013), prior to confirmation of the conserved domain in each candidate by Pfam (https://pfam.xfam.org/) and SMART (http://smart.embl-heidelberg.de/) databases (Mistry et al. 2021; Letunic et al. 2021). Their physio-chemical properties were calculated by ProtParam tool (https://web.expasy.org/protparam/) (Gasteiger et al. 2003).
Timescale tree was drawn by TIMETREE (http://www.timetree.org/) (Kumar et al. 2017); and phylogenetic trees were constructed by MEGA7.0 software, using the NJ method, with a bootstrap analysis of 1000 replicates (Kumar et al. 2016). Gene structure was visualized by the Gene Structure Display Server 2.0 (http://gsds.cbi.pku.edu.cn/) (Hu et al. 2015), the conserved motif was identified with the aid of the MEME Suite tool (https://meme-suite.org/meme/index.html) (Bailey et al. 2015), while cis-acting element in the promoter was characterized through the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) (Lescot et al. 2002). Signal peptide was predicted, using the SignalP 5.0 database (https://services.healthtech.dtu.dk/services/SignalP-5.0/) (Almagro Armenteros et al. 2019), and transmembrane helix was analyzed by the TMHMM-2.0 server (https://services.healthtech.dtu.dk/services/TMHMM-2.0/). The binding sites of TF in PbrDHAR5 promoter was predicted through the PlantRegMap database (http://plantregmap.gao-lab.org/) (Tian et al. 2020). On the other hand, the possible S-sulfenylated Cys residues in PbrDHAR5 were predicted via the pCysMod database (http://pcysmod.omicsbio.info/webserver.php) (Li et al. 2021).
Homology modeling and molecular docking
Structures of PbrDHAR5 and its mutants (including PbrDHAR5^Cys20Ala^, PbrDHAR5^Lys8Ala^, and PbrDHAR5^Lys210Ala^) were predicted by AlphaFold2 (Cramer 2021); on the other hand, DHA was converted into 3D configuration in the MOE v2015 through energy minimization (Zhang et al. 2021). MOE-Dock was used for the molecular docking of PbrDHAR5 & DHA (Wadood et al. 2013), in association with definition of the binding pocket around the catalytic Cys^20^ residue. Result with the highest binding score was selected and visualized by PyMOL (Zhang et al. 2021).
MD simulation
MD simulations were performed using the GROMACS 2022 software (Hess et al. 2008). General Amber Force Field (GAFF) (Wang et al. 2004) and Amber99SB-ILDN force field (Lindorff-Larsen et al. 2010) were selected for ligands and the receptor, respectively. The proteins were solvated in cubic boxes of the transferable interatomic potential with three points model (TIP3P) (Jorgensen et al. 1983) water molecules situated at a minimal distance of 1.2 nm from the box edges. The box size is 80 Å × 80 Å × 80 Å. Three positive (Na^+^) ions were added to the system to achieve neutralization. The cutoffs of the van der Waals interactions were set to be 1 nm, and the long-range electrostatic interactions were calculated by the particle mesh Ewald (PMF) algorithm (Essmann et al. 1995) with a mesh spaced 0.16 nm. The neighbor list for the nonbonded interactions was updated every 10 steps. Hydrogen bond was constrained using the linear constraint solver (LINCS) algorithm (Hess 2008) and the time step for all simulations was 2 fs. All simulations were under periodic boundary conditions. The steepest descent method was applied to energy minimization, followed by a 100 ps canonical ensemble (NVT, constant number of particles, volume, and temperature) equilibration, a 100 ps constant-pressure and constant-temperature (NPT, constant number of particles, pressure, and temperature) equilibration. The temperature was maintained at 300 K with Berendsen thermostat (Berendsen et al. 1984), and the pressure was set at 1 bar with Parrinello-Rahman barostat (Parrinello and Rahman 1981) throughout the duration of each simulation. Subsequently, a 100-ns production run was conducted, and the root mean square deviation (RMSD), root mean square function (RMSF), radius of gyration (Rg), and hydrogen bonding (H-bond) were calculated to assess the stability of the compound-protein complexes. Additionally, the MOLECULAR MECHANICS POISSON-BOLTZMANN SURFACE AREA (MMPBSA) (Valdés-Tresanco et al. 2021) analysis was performed to evaluate the free energy of binding between the protein and the molecular ligand.
Statistical analyses
Data represented the mean value of three biological replicates, except for transcriptome and proteome assay of the (un)scalded ‘Yali’ fruit (Experiment II: two replicate) as well as ‘Yali’ fruit of different scald grades (Experiment III: one replicate). SAS version 9.3 (SAS Institute, Cary, NC) was used for the data analysis, especially the analysis of variance (PROC ANOVA) with multi-comparison correction. Mean separation was determined by Duncan’s multiple range test at the 0.05 level. R package was used to calculate the Pearson correlation coefficient between different attributes, where the extremely strong correlation was in the range of 0.8–1.0, and the strong correlation in the range of 0.6–0.8 (Long et al. 2014).
Supplementary Information
Supplementary Material 1: Fig. S1 Impact of exogenous 1-MCP and DAP treatments on quality and physiological attributes during cold storage of ‘Dangshansuli’ fruit. (a) Scald incidence and index. (b) α-Farnesene-related metabolites. (c) H_2_O_2_. (d) Correlations among attributes. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples at the same sampling time (p < 0.05). Pearson correlation between attributes is visualized as a heatmap, where red color demonstrates positive association, while blue indicates negative correlation. Fig. S2 Morphological alterations upon scald development in‘Yali’ fruit. (a) Safranin O-fast green staining and toluidine blue O staining. Abbreviations: FW, fruit wax; P, plasmolysis; SC, shrunken cytosol. (b) TUNEL staining. Abbreviations: FW, fruit wax; N, nucleus. (c) TEM analyses. Abbreviations: C, chloroplast; CW, cell wall; ER, endoplasmic reticulum; M, mitochondrion; P, plasmolysis; V, vacuole. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. Fig. S3 Impact of exogenous AsA and DHA treatments on morphological alterations in ‘Yali’ fruit. (a) AsA treatment. ‘Yali’ fruit were treated with 0.0 (control) and 2.0 g L^-1^ AsA before sampling after 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. (b) DHA treatment. ‘Yali’ fruit were treated with 0.0 (control) and 1.0 g L^-1^ DHA before sampling after 120-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. Abbreviations: FW, fruit wax; N, nucleus; P, plasmolysis; SC, shrunken cytosol. Fig. S4. Impact of exogenous H_2_O_2_ treatment on scald development in ‘Yali’ fruit. (a) Scald incidence and index. (b) H_2_O_2_ content and AsA redox status. ‘Yali’ fruit were treated with 0.0 (control) and 5.0 mmol L^-1^ H_2_Obefore sampling after a 120-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05). Fig. S5 Alternation of AsA redox status after the genetic transformation of pear and tomato. (a) Subcellular localization of PbrAPX2. (a-i) Phylogenetic analyses of plant APXs. Information on APXs from Arabidopsis (AtAPXs) and *Pyrus bretschneideri *Rehd. (PbrAPXs) were summarized in Table S2. Phylogenetic tree was constructed via MEGA7.0 software, using NJ method, with a bootstrap analysis of 1000 replicates (Kumar et al. 2016). (a-ii) Subcellular localization of PbrAPX2. The fluorescence signal was observed by a confocal microscope (Leica Microsystems, Germany). (b) Impact of the transient genetic transformation of‘Dangshansuli’ fruit on AsA redox status. (b-i) Transient overexpression of PbrAPX2.‘Dangshansuli’ fruit transformed with the empty pCAMBIA1300 vector with a GFP tag was used as the control for the overexpressing (OE) pear. (b-ii) Transient silence of PbrAPX2.‘Dangshansuli’ fruit co-transformed with the empty pTRV2 and pTRV1 was used as the control for the silenced (SE) pear. The expression level of PbrAPX2 in control fruit was set as 1.0 for qRT-PCR assay. (c) Impact of overexpressing PbrAPX2 in tomato on AsA redox status. (c-i) Visual quality change and chilling injury index. (c-ii) PbrAPX2 expression level and cytAPX activity. (c-iii) H_2_O_2_ content and AsA redox status. Tomato fruit at 35 DAFB, including the wide-type (control) and overexpressing (OE) lines, were harvested and then exposed to 4 °C for 10 d followed by a 7-d shelf life at 20 °C. The expression level of PbrAPX2 in the OE-7 fruit was set as 1.0 for qRT-PCR assay. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05). Fig. S6 Signal peptide and transmembrane helix assay. (a) Signal peptide in PbrAPX2 (a-i), PbrDHAR5 (a-ii), and PbrWRKY83 (a-iii). Signal peptide was predicted by the SignalP 5.0 database (Almagro Armenteros et al. 2019). (b) Transmembrane helix in PbrAPX2 (b-i), PbrDHAR5 (b-ii), and PbrWRKY83 (b-iii). Transmembrane helix was analyzed, using the TMHMM-2.0 server. Fig. S7 qRT-PCR validation of transcriptome result on gene expression profiles in pear fruit. (a) Expression profiles of several genes in AsA-GSH cycle during cold storage of ‘Dangshansuli’ fruit. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. The expression level of *PbrDHAR5 *in the control fruit on day 0 were set as 1.0. (b) Expression profiles of PbrDHAR5 and PbrWRKY83 in the (un)scalded ‘Yali’ pear (b-i) and fruit of different scald grades (b-ii). The (un)scalded ‘Yali’ pear (b-i) and fruit of different scald grades (b-ii) were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. The expression level of each gene in the unscalded fruit or fruit with scald grade 0 was set as 1.0. (c) Expression profiles of two differentially expressed TFsduring cold storage of ‘Dangshansuli’ fruit. The expression level of Pbr002398.1 in the control fruit on day 0 were set as 1.0. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples at the same sampling time (p< 0.05). Fig. S8 Alignment of protein sequences from‘Dangshansuli’ and ‘Yali’ fruits. (a) PbrDHAR5 sequences. (b) PbrWRKY83 sequences. Sequence alignment was performed by the Jalview software (Waterhouse, et al., 2009). Fig. S9 Impact of exogenous H_2_O_2_, AsA, and DHA treatments on the expression profiles of PbrDHAR5 and PbrWRKY83 in pear fruit. (a) H_2_O_2_.‘Yali’ fruit were treated with 0.0 (control) and 5.0 mmol L^-1^ H_2_Obefore sampling after a 120-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. (b) AsA. ‘Yali’ fruit were treated with 0.0 (control) and 2.0 g L^-1^ AsA before sampling after 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. (c) DHA. ‘Yali’ fruit were treated with 0.0 (control) and 1.0 g L^-1^ DHA before sampling after 120-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. The expression level of each gene in the control fruit was set as 1.0. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05). Fig. S10 Molecular docking analyses during MD simulation. (a) The root mean square deviation (RMSD). (b) The radius of gyration (Rg). (c) The root mean square fluctuation (RMSF). (d) The number of hydrogen bonds. (e) The energetic components. (f) Energy contribution per residue. Abbreviations: GGAS, total gas phase free energy; GSOLV, total solvation free energy; TOTAL, total free energy. Fig. S11 Characteristics of plant DHARs. (a) Identification of DHARs from plants. DHARs, which were identified from 26 plants, were summarized in Table S6. Timescale tree of 26 plants was drawn by TIMETREE; and phylogenetic tree was constructed via MEGA7.0 software, using NJ and ML methods, with a bootstrap analysis of 1000 replicates (Kumar, et al., 2016). (b) Gene structures and distributions of the conserved motifs. (b-i) Gene structures. Gene structure was visualized by the Gene Structure Display Server 2.0 (Hu, et al., 2015). (b-ii) Distributions of the conserved motifs. The conserved motif was identified with the aid of the MEME Suite tool (Bailey, et al., 2015). (c) Detailed information of the conserved motifs. Eight conserved motifs were characterized from 79 plant DHARs; Motif 1, 4 & 5 composed the conserved domain ‘GST_N_3’, while the domain ‘GST_C’ was consisted of Motif 2 & 8. Fig. S12 Alignment of protein sequences of 79 plant DHARs. 79 DHARs, which were identified from 26 plants, were summarized in Table S6. Sequence alignment was performed by the Jalview software (Waterhouse, et al., 2009). The catalytic residue (for example, Cys^20^ residue in PbrDHAR5) are highlighted in the boxes. Fig. S13 Validation of PbrDHAR5’s subcellular localization in nucleus. AtH2B-mcherry as nuclear indicator for PbrDHAR5 (Jia, et al., 2023). Fig. S14 Gene function validation in pear calli. (a) Function validation of PbrDHAR5. (a-i) Phenotypes of the calli. (a-ii) PbrDHAR5 expression level and cytDHAR activity. (a-iii) AsA redox status and H_2_O_2_ content. (b) Function validation of PbrWRKY83. (b-i) Phenotypes of the calli. (b-ii) PbrWRKY83 and PbrDHAR5 expression level. (b-iii) AsA redox status and H_2_O_2_ content. Pear calli transformed with the empty pCAMBIA1300 vector with a GFP tag was used as the control for the overexpressing (OE) lines; and the expression levels of PbrDHAR5 and PbrWRKY83 in the control calli were set as 1.0 for qRT-PCR assay. After growth at 20 °C for a period of time, pear calli was exposure to 4 °C for 15 d. Data represent the mean value of three biological replicates; and vertical bars labelled with the same small letter are not significantly different between samples (p < 0.05). Fig. S15 Identification of the possible upstream regulators of PbrDHAR5 gene. (a) Information on the differentially expressed TFs during cold storage of ‘Dangshansuli’fruit with(out) the pre-storage chemical treatments. (a-i) Number of the differentially expressed TFs. (a-ii) Expression profiles of the differentially expressed TFs and their correlations with PbrDHAR5 mRNA abundance. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. The differentially expressed TFs, whose mRNA abundances were consistently higher or lower in the control fruit than those in the 1-MCP/DPA-treated fruit during cold storage, were identified by DESeq2_EBSeq software based on the following criteria: fold change ≥ 1.5 and FDR < 0.05. Data, adapted from transcriptome assay, represent the mean value of three biological replicates. Color scale represents normalized log_2_-transformed (mean FPKM + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. Pearson correlation between different attributes is visualized as a heatmap, where extremely strong (or strong) positive association is connected with red (or light red) line, while extremely strong (or strong) negative correlation with blue (or light blue) line (Long et al.2014). (b) Expression profiles of TFs upon scald development in ‘Yali’ fruit.‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. TFs were summarized in Table S7. Data, adapted from transcriptome assay, represent the mean value of two biological replicates. Color scale represents normalized log_2_-transformed (mean FPKM + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. (c) Expression profiles of TFs in ‘Yali’ fruit of different scald grades and their correlations with PbrDHAR5 mRNA abundance. ‘Yali’ fruit of different scald grades were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. TFs were summarized in Table S7. Data, adapted from transcriptome assay, represent the value of one biological replicate. Color scale represents normalized log_2_-transformed (FPKM + 1), where red indicates a high level, blue presents a low level, and white demonstrates a medium level. Pearson correlation between different attributes is visualized as a heatmap, where extremely strong (or strong) positive association is connected with red (or light red) line, while extremely strong (or strong) negative correlation with blue (or light blue) line (Long, et al., 2014). (d) Detailed information of two W-box elements in PbrDHAR5 promoter. Two W-box element were characterized from PbrDHAR5 promoter with the aid of the PlantCARE database (Lescot, et al., 2002). Fig. S16 Impact of transformation of pear calli with PbrWRKY83 on PbrDHAR5 expression level. (a) Phenotype of pear calli. (b) PbrWRKY83 and PbrDHAR5 expression profiles. Pear calli transformed with the empty pCAMBIA1300 vector with a GFP tag was used as the control for the overexpressing (OE) lines; and the expression level of each gene in the control calli was set as 1.0 for qRT-PCR assay. Pear calli was grown at 20 °C until sampling. Data represent the mean value of three biological replicates; and vertical bars labelled with the same small letter are not significantly different between samples (p < 0.05). Fig. S17 Alternation of DHA reduction frequency during scald development in pear fruit with(out) the pre-storage chemical treatments. (a) Impact of 1-MCP and DAP treatments on DHA reduction frequency during cold storage of ‘Dangshansuli’ fruit. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. (b) Alternation of DHA reduction frequency upon scald development in ‘Yali’ fruit. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. (c) Impact of scald grade on DHA reduction frequency in ‘Yali’ fruit. ‘Yali’ fruit of different scald grades were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. DHA reduction frequency was defined as the ratio of cytDHAR activity to DHA content. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples at the same sampling time (p < 0.05). Fig. S18 Alternation of PbrDHAR5 abundance upon scald development in‘Yali’ fruit. (a) Proteome result. Data, adapted from proteome assay, represent the mean value of two biological replicates. (b) Western blot result. Relative abundance of the DHAR5 was calculated as the ratio of PbrDHAR5 and Actin in the sample. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. Data, adapted from proteome assay, represent the mean value of two biological replicates. Fig. S19 Impact of exogenous H_2_O_2_ or DTT treatment on cytDHARactivity in the epidermal tissue of the scalded ‘Yali’ fruit. The residual cytDHAR activity was expressed as a percentage of the control (H_2_O treatment), whose activity was set as 1.0. Data represent the mean value of three biological replicates; and vertical bars labeled with the same small letter are not significantly different between samples (p < 0.05).Supplementary Material 2: Table S1 Primers used in this study. Table S2 The physio-chemical properties of APXs from Arabidopsis and *Pyrus bretschneideri *Rehd. APXs from Arabidopsis (AtAPXs) were characterized from TAIR database (https://www.arabidopsis.org/), while APXs in *Pyrus bretschneideri *Rehd. genome (PbrAPXs) were previously identified (Wang, et al., 2021). The physio-chemical properties of plant APXs were calculated by ProtParam tool (Gasteiger, et al., 2003). Table S3 Expression profiles (FPKMs) of genes in AsA-GSH cycle during cold storage of ‘Dangshansuli’ fruit with(out) the pre-storage chemical treatments. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. Genes in AsA-GSH cycle were previously identified from *Pyrus bretschneideri *Rehd. genome (Wang, et al., 2021). Data, adapted from transcriptome assay, represent the mean value of three biological replicates. Table S4 Alternation in the expression profiles of genes (FPKMs) and proteins (ratio of the scalded pear to the unscaled fruit) in AsA-GSH cycle upon scald development in ‘Yali’ fruit. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. Genes and proteins in AsA-GSH cycle were previously identified from *Pyrus bretschneideri *Rehd. genome (Wang, et al., 2021). Data, adapted from transcriptome and proteome assay, represent the mean value of two biological replicates. Table S5 Expression profiles (FPKMs) of genes in AsA-GSH cycle in ‘Yali’ fruit of different scald grades. ‘Yali’ fruit of different scald grades were sampled after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. Genes in AsA-GSH cycle were previously identified from *Pyrus bretschneideri *Rehd. genome (Wang, et al., 2021). Data, adapted from transcriptome assay, represent the value of one biological replicate. Table S6 The physio-chemical characteristics of 79 plant DHARs. Arabidopsis DHARs were used as a query in BLASTP of plant genome database (Goodstein, et al., 2012; Wu, et al., 2013), prior to the confirmation of the conserved domain in each candidate by Pfam (Mistry, et al., 2021) and SMART (Letunic, et al., 2021) databases. The physio-chemical properties of plant DHARs were calculated by ProtParam tool (Gasteiger, et al., 2003)). Table S7 Expression profiles (FPKMs) of the differentially expressed TFs during cold storage of ‘Dangshansuli’ fruit with(out) the pre-storage chemical treatments. ‘Dangshansuli’ fruit were treated with H_2_O (control), DPA, and 1-MCP before sampling after 0-, 60-, 120-, 180-d storage at 0.5 °C followed by a 7-d shelf life at ambient temperature. The differentially expressed TFs, whose mRNA abundances were consistently higher or lower in the control fruit than those in the 1-MCP/DPA-treated fruit during cold storage, were identified by DESeq2_EBSeq software based on the following criteria: fold change ≥ 1.5 and FDR < 0.05. Data, adapted from transcriptome assay, represent the mean value of three biological replicates. Table S8 Alternation in the expression profiles (FPKMs) of TFs upon scald development in ‘Yali’ fruit. ‘Yali’ fruit, with and without scald symptom, were sampled for after a 180-d storage at -0.5 °C followed by a 7-day shelf life at ambient temperature. TFs are summarized in Table S7. Data, adapted from transcriptome assay, represent the mean value of two biological replicates. Table S9 Expression profiles (FPKMs) of TFs in ‘Yali’ fruit of different scald grades. ‘Yali’ fruit of different scald grades were sampled for after a 180-d storage at -0.5°C followed by a 7-day shelf life at ambient temperature. TFs are summarized in Table S7. Data, adapted from transcriptome assay, represent the value of one biological replicate.