Integrative multi-omics analysis unravels chitosan-mediated delay of mango fruit ripening through hormonal and metabolic reprogramming

Abstract

Mango (Mangifera indica L.), a climacteric fruit of significant global economic value, suffers from rapid postharvest deterioration driven by complex hormonal and transcriptional networks. While chitosan coating is known to extend mango shelf-life, the precise molecular mechanisms, particularly its interplay with key ripening hormones like abscisic acid and ethylene, and the role of transcriptional regulators, remain poorly understood. We hypothesized that chitosan delays ripening by reprogramming abscisic acid and ethylene signaling pathways and associated transcriptional networks. To test this, we employed an integrative multi-omics approach combining transcriptomics and metabolomics to elucidate the molecular basis of chitosan-mediated ripening delay in mango fruit. Our findings reveal that chitosan orchestrates a dual hormonal attenuation associated with suppression of abscisic acid biosynthesis (NCED3, ABA2) and ethylene signaling (ETR1, EIN3/EIN4) while enhancing abscisic acid catabolism (CYP707A1/4). Crucially, γ-aminobutyric acid (GABA) accumulation antagonized abscisic acid to uncouple starch-to-sugar conversion from ripening progression, correlating with retained firmness and chlorophyll levels. Furthermore, we identified key transcription factors, WRKY53 and bZIP/RF2b, likely acts as central hubs modulating abscisic acid-ethylene crosstalk and cell wall integrity. This knowledge provides novel mechanistic insights into chitosan action and offers a foundation for developing targeted strategies to optimize postharvest preservation in climacteric fruits by synchronously modulating Abscisic acid-ethylene crosstalk, GABA-mediated metabolic reprogramming, and key TFs.

Highlights

• •Chitosan delayed ripening and was associated with retained firmness and chlorophyll.
• •GABA and citrate accumulation stabilized organic acid homeostasis.
• •Transcriptomics revealed 2290 DEGs linked to starch/cell wall dynamics.
• •Chitosan was associated with suppressed ABA biosynthesis and ethylene signaling.
• •WRKY53 and bZIP/RF2b emerged as hubs for ABA-ethylene crosstalk.

1. Introduction

Mango (Mangifera indica L.) is a vital tropical fruit crop, yet it faces significant postharvest losses due to rapid ripening and senescence (FAO, 2023). As a climacteric fruit with unique ripening traits, where abscisic acid initiates early maturation and ethylene amplifies late-stage senescence, it serves as an ideal model for studying the interplay between hormonal networks and postharvest physiology (Wu, Shan, et al., 2022). Fruit ripening and senescence involve coordinated biochemical shifts, including chlorophyll degradation, cell wall disassembly, starch-to-sugar conversion, and flavor metabolite dynamics (Li et al., 2021). While extensive studies have focused on mango ripening markers such as firmness and peel color (Pornchan Jongsri et al., 2016), critical aspects like organic acid homeostasis, aroma volatile synthesis, and crucially, the transcriptional regulation underlying these processes, remain underexplored. Current postharvest strategies often rely on single-parameter assessments (e.g., respiration rate), neglecting systemic metabolic profiling. Thus, integrating multi-omics approaches to map the molecular landscape of senescence is imperative for developing targeted preservation technologies.

Abscisic acid plays a pivotal role in coordinating mango fruit ripening by regulating sugar accumulation, carotenoid biosynthesis, and cell wall metabolism, with transcriptomic evidence indicating its dominance over ethylene in driving ripening-associated processes (Wu, Wu, et al., 2022). During the climacteric phase, endogenous abscisic acid peaks synchronously with ethylene production, suggesting their synergistic regulation of ripening initiation (Zaharah et al., 2012). Abscisic acid interacts with ethylene to modulate cell wall degradation genes and coordinates with other hormones to balance softening and antioxidant responses (Gupta et al., 2022). This hormonal crosstalk is mediated through core signaling components including NCEDs for abscisic acid biosynthesis, CYP707As for catabolism, and PYL-PP2C-SnRK2 modules for signal transduction (Leng et al., 2013). While exogenous abscisic acid enhances color development and flavor compounds, its overaccumulation may exacerbate physiological disorders like soft nose through disrupted ethylene signaling (Kumar et al., 2020). These findings highlight abscisic acid's functionality as both a ripening promoter and quality modulator. Despite these advances, the molecular mechanisms underlying abscisic acid's systemic coordination of metabolic and, importantly, the transcriptional networks it governs, remain poorly resolved.

Emerging evidence positions WRKY transcription factors as central integrators of hormonal signaling and cell-wall remodeling during climacteric fruit ripening (Cheng et al., 2016; Wang et al., 2024), yet their mango-specific targets and responsiveness to post-harvest interventions remain undefined. In peach, PpWRKY14 directly activates ethylene biosynthetic genes (PpACS1, PpACO1) via promoter binding (Liu et al., 2024), while in strawberry, FaWRKY48 controls pectin degradation by regulating pectate lyase (Zhang et al., 2022). These cross-species mechanisms, likely conserved in mango, indicate WRKYs could modulate ripening. Similarly, banana MaWRKY31/33/60/71 enhance cold tolerance partly by inducing MaNCED1/2-mediated abscisic acid synthesis (Luo et al., 2017), and bZIP TFs coordinate sugar metabolism and abscisic acid signaling (Wu, Wu, et al., 2022). However, how WRKY and bZIP networks govern abscisic acid-ethylene crosstalk in mango, especially under post-harvest treatments, remains to be elucidated.

Chitosan, a biodegradable polysaccharide, extends shelf life in mango by suppressing respiration and pathogen growth (Pornchan Jongsri et al., 2016) and delaying softening (Cosme Silva et al., 2017). Despite these documented physiological effects, the molecular mechanisms underpinning chitosan action, particularly its interplay with key ripening hormones (abscisic acid, ethylene) and transcriptional regulators (e.g., WRKY, bZIP), remain elusive. We hypothesized that chitosan delays mango ripening by reprogramming abscisic acid and ethylene signaling pathways and associated transcriptional networks. As chitosan concentrations exceeding 1 % negatively impact mango firmness and sensory quality (Cosme Silva et al., 2017; Pornchan Jongsri et al., 2016), we selected a 1 % (w/v) chitosan solution. This study employed an integrative multi-omics approach to dissect chitosan-triggered hormonal and metabolic reprogramming, offering insights for novel postharvest strategies to enhance quality and prolong shelf life.

2. Materials and methods

2.1. Preparation of control and chitosan solutions

We prepared the chitosan solution following Poverenov et al. (2014). Food-grade chitosan (Solarbo Co., Ltd., Beijing, China) was dissolved in 2 % aqueous acetic acid at 60 °C with continuous stirring to a final concentration of 1 % (w/v), adjusting pH to 5.6 using 1 M NaOH. The control was prepared by heating 2 % aqueous acetic acid at 60 °C for an equivalent duration without chitosan, with pH similarly adjusted to 5.6.

2.2. Plant material and experimental design

Mature green “Tainong” mangoes (Mangifera indica L.) were harvested from a commercial orchard in Yulin, Guangxi Province, China. Fruits of uniform size, color, and absence of physical damage were selected, packed in ventilated plastic crates with cushioning, and transported to the laboratory under ambient conditions. After overnight equilibration at 25 °C (40–60 % RH), fruits were rinsed, air-dried, and randomly assigned to control or chitosan-coated groups using Excel-generated random numbers (RAND() function), allocating sequentially to alternating groups (150 fruit per group). Groups were immersed in respective solutions (control: 2 % acetic acid; chitosan: 1 % chitosan in 2 % acetic acid) for 5 min, followed by air-drying. Fruits were stored at 25 °C (80–95 % RH) for 12 days.

Pulp tissues were sampled at 4 days after treatment (4 DAP) for transcriptomic and metabolomic analyses. This timepoint was selected because: (1) control fruits exhibited pronounced physiological changes (43 % reduction in firmness, 67 % decline in chlorophyll content, 62 % yellowing rate), while chitosan-treated fruits showed significantly delayed ripening (Fig. 1B–D); and (2) fruits remained free from visible decay (Fig. 1C), ensuring molecular responses reflected chitosan-specific effects rather than microbial spoilage.

Fig. 1.

Impact of chitosan treatment on mango postharvest quality. (A) Visual comparison of mango fruit quality between control and chitosan groups over 12 days. (B–D) Impact of chitosan on mango firmness, cell wall components (WSP, CSP, HSP, cellulose), yellowing rate, chlorophyll, carotenoid content, soluble solids, and titratable acidity during storage. Data represent means of three independent experiments (5 fruits each); error bars indicate SD.

For each treatment (control or chitosan), three independent biological groups were established (n = 3). All five fruits per group were individually assessed for firmness, color, and decay; these non-destructive data are reported as means ± SD across 15 fruits (3 groups × 5 fruits). For biochemical assays (pectin content, chlorophyll quantification, omics analyses), the five fruits per group were pooled, homogenized, and analyzed; reported values are means ± SD of three biological composite replicates.

2.3. Sensory quality evaluation

2.3.1. Color transition assessment

Mango yellowing degree was visually categorized: green, light yellow, yellow, dark yellow. Fruits exhibiting light yellow, yellow, or dark yellow were classified as yellowed (Shao Lihuan et al., 2017; Shi et al., 2022). Yellowed percentage was calculated as (number of yellowed fruits / total fruits) × 100 %.

2.3.2. Firmness and decay assessment

Fruit firmness was quantified using a digital penetrometer GY-4 (Tuopu Instrument Co., Ltd., Zhejiang, China) with a 2 mm cylindrical probe penetrating the equatorial region at 1 mm/s to 10 mm depth; peak resistance (N) was recorded. Measurements were taken at three equidistant equatorial positions per fruit, with three biological replicates (five fruits per replicate) per treatment.

Decay progression was monitored visually at 2-day intervals (days 0–12 post-treatment). Fruits exhibiting visible lesions, tissue collapse, or abnormal softening in ≥50 % surface area were categorized as decayed (Shi et al., 2022). Decay incidence (%) was calculated as: (Number of decayed fruit/Total number of fruit) × 100 %, with three independent replicates (five fruits per replicate) per treatment.

2.4. Physicochemical characterization

2.4.1. Chlorophyll and carotenoid quantification

Pigment analysis used UV–Vis spectroscopy Shimadzu UV-1800 (Shimadzu Corp., Kyoto, Japan) adapted from Chen et al. (2022). Briefly, 0.5 g peel tissue was homogenized in 80 % acetone, triple-extracted, centrifuged (12,000 ×g, 30 min), supernatants pooled, diluted to 10 mL with 80 % acetone, and absorbance measured at 470, 645, and 663 nm for concentration calculations.

2.4.2. Soluble solids content (SSC)

SSC was assessed through digital refractometric analysis PAL-1 (Atago Co., Ltd., Tokyo, Japan). Pulp homogenates were filtered, and measurements were performed on composite samples derived from five fruits per replicate, with triplicate technical repetitions to ensure precision.

2.4.3. Titratable acidity (TA)

TA was evaluated using an acid-base titration protocol. Homogenized pulp (10 g) was diluted in 90 mL distilled water and titrated with 0.1 M NaOH to pH 8.2 under pH meter monitoring. Results were expressed as citric acid percentage equivalents, with triplicate analyses per sample to confirm reproducibility.

2.4.4. Pectin fraction isolation and quantification

Mango pulp samples (10 g) were ground in 95 % ethanol (100 mL) for 10 min to precipitate pectin. The resulting residue underwent two washes with 75 % ethanol to isolate alcohol-insoluble solids (AIS). For water-soluble pectin (WSP) extraction, 0.5 g AIS was suspended in deionized water and agitated at 25 °C for 2 h, followed by centrifugation (4662×g, 15 min, 4 °C). The supernatant was collected and freeze-dried. Subsequent extractions utilized 0.05 M CDTA (pH 5.2) containing 0.05 M potassium acetate to recover chelator-soluble pectin (CSP) from the pellet, with three sequential extractions. The remaining residue was treated with 0.5 M H₂SO₄ at 85 °C for 1 h to isolate acid-soluble pectin (HSP), followed by centrifugation. All fractions were quantified via the Blumenkrantz assay, calibrated against D-galacturonic acid standards.

2.5. Metabolomic profiling

Freeze-dried mango pulp was ground for 1.5 min at 30 Hz with a zirconia bead; 0.1 g powder was extracted in 1.2 mL 70 % methanol, vortexed 30 s every 30 min for six cycles, kept at 4 °C overnight, and centrifuged (12,000×g, 10 min). The supernatant was filtered (0.22 μm) before UPLC-MS/MS analysis on a Waters ACQUITY UPLC HSS T3 C18 column (1.8 μm, 2.1 × 100 mm) at 40 °C with a 0.4 mL min⁻¹ gradient of water/0.1 % formic acid and acetonitrile/0.1 % formic acid. A QTRAP® 4500+ system (AB Sciex Pte. Ltd., Framingham, MA, USA) equipped with an ESI Turbo Ion-Spray interface operated in positive and negative modes under Analyst 1.6.3 control. Metabolites were identified by matching MS¹ and MS² spectra against an in-house library (≥70 % spectral similarity, ≤5 ppm mass error) and the GNPS public library (≥70 % cosine score); peaks present in blank controls at ≥30 % of QC intensity were discarded.

2.6. Transcriptomic sequencing

Total RNA was isolated using a Plant RNA Kit (Omega Bio-Tek, Norcross, GA, USA) followed by DNase I digestion, and RNA integrity was confirmed by electrophoresis and a 260/280 nm ratio ≥1.9. Libraries were constructed with the NEBNext Ultra II RNA Kit and sequenced on an Illumina NovaSeq 6000 platform (150 bp paired-end). Raw reads were trimmed and filtered with fastp (v0.19.7) (Chen et al., 2018) (−q 20 −u 50 −n 15) to remove adapters, low-quality bases, and ambiguous reads. Clean reads were aligned to the Mangifera indica genome v2.0 (GCF_011075055.1) with HISAT2 (v2.2.1) (Kim et al., 2019); gene-level counts were converted to TPM after first deriving RPKM and scaling the total to 1,000,000. Differential expression was determined with Deseq2 (v. 1.34.0) (Love et al., 2014) (|log2FC| > 1, FDR <0.05) and GO/KEGG enrichment tested by Fisher's exact test (p < 0.05).

2.7. Quantitative real-time PCR (qRT-PCR) validation

qRT-PCR was conducted as described by Ren et al. (2017). Total RNA was isolated with TRIzol reagent (Invitrogen Corp., Carlsbad, CA, USA) and then treated with DNase I (Thermo Fisher Scientific, Waltham, Massachusetts, USA). First-strand cDNA was generated from 2 μg of RNA using the Fermentas Reverse Transcription Kit. The expression levels of target genes were determined using ACT as an internal control. Each reaction was performed in triplicate, and the primer sequences are provided in supplementary Table S1. Results are shown as means ± SE.

3. Results

3.1. Impact of chitosan treatment on postharvest physiological and biochemical characteristics

Chitosan treatment was associated with mitigation of ripening and senescence, with differences evident between chitosan and control groups from day 2 to 12 (Fig. 1A). Chitosan treatment was associated with maintained fruit firmness. While control firmness declined sharply from 28.60 ± 8.18 N (2 DAP) to 2.68 ± 1.53 N (12 DAP), chitosan-treated fruit retained higher firmness (34.42 ± 3.13 N to 9.62 ± 0.56 N; Fig. 1B). Cell wall analysis revealed chitosan suppressed WSP accumulation (12.7 ± 3.1 % to 60.8 ± 3.4 % vs control: 13.9 ± 3.2 % to 65.1 ± 3.0 %) and preserved structural pectins CSP/HSP and cellulose (Fig. 1B), indicating chitosan stabilizes cell wall architecture by inhibiting pectin solubilization and maintaining rigidity-linked components.

Chitosan delayed mango fruit yellowing, as evidenced by high chlorophyll retention (0.06 ± 0.006 mg/g vs control 0.02 ± 0.004 mg/g at 12 DAP), and modulated carotenoid dynamics, sustaining gradual accumulation (0.013–0.030 mg/g) versus control's transient peak (0.032 mg/g at 6 DAP) followed by decline (Fig. 1C), suggesting chitosan mitigates senescence by stabilizing chlorophyll and regulating carotenoid turnover.

Chitosan also modulated flavor-related metabolism in mango fruit, as evidenced by delayed soluble solids content (SSC) elevation and attenuated organic acid degradation. While control SSC increased from 21.33 % to 27.50 % by 12 DAP, chitosan-treated fruit exhibited slower sugar accumulation, reaching 27.83 % at 12 DAP (Fig. 1D). Titratable acidity (TA) declined sharply in control (1.12 % to 0.12 %), but chitosan maintained higher TA levels (0.29 % at 12 DAP), retaining 112.5 % more acidity than control (Fig. 1D).

Furthermore, chitosan suppressed climacteric respiratory peaks (145.80 vs control 313.47 mL CO₂·kg⁻¹·h⁻¹), delayed decay onset (80 % vs control 100 % at 12 DAP), and reduced cumulative weight loss by 21 % (Supplementary Fig. 1A–C). These results demonstrate chitosan’s dual role in retarding ripening-associated metabolic shifts and preserving quality through suppressed respiration, pathogen resistance, and moisture retention.

3.2. Chitosan treatment modulates metabolite profiles

Metabolomic profiling identified 166 differentially abundant metabolites (DAMs) in response to chitosan: 81 upregulated chitosan, 85 downregulated (Fig. 2B; Table S1). Significant changes occurred in organic acids, phenolic acids, amino acids, and derivatives (Fig. 2C). chitosan upregulated 21 of 27 organic acids and 17 of 23 phenolic acids. Key upregulated metabolites included glutaric acid, citric acid, GABA, p-coumaric acid, and sinapyl alcohol; abscisic acid levels were significantly reduced (Fig. 2D). These results indicate chitosan preserves postharvest quality by attenuating organic acid degradation, potentially via enhanced phenolic biosynthesis and suppressed abscisic acid-mediated senescence.

Fig. 2.

Metabolomic changes post-chitosan treatment. (A) PCA plot distinguishing metabolite profiles. (B) Volcano plot identifying upregulated (red) and downregulated (green) metabolites. (C) Distribution of DAMs across compound classes. (D) Heatmap of DAMs; color indicates relative content (red-high, blue-low). (E) Metabolite interaction network; upregulated (red), downregulated (blue). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Nine chitosan-responsive metabolites were prioritized for network analysis based on dual criteria: (1) statistical significance (VIP > 1.5, |log₂FC| > 0.7); (2) functional relevance to three key mechanisms, energy metabolism (citric acid, cis-aconitic acid, oxoadipic acid and malic acid), phenylpropanoid defense (shikimic acid, p-coumaric acid, sinapyl alcohol), and hormonal regulation (abscisic acid, GABA). Metabolite interaction networks were then constructed among these nine prioritized metabolites by Spearman correlation analysis (p < 0.01, |r| > 0.8), yielding seven significant edges that resolved into three distinct clusters. Cluster 1 included abscisic acid, GABA, and citric acid, with GABA and citric acid showing strong positive correlation (r = 0.92, P < 0.001), while abscisic acid was negatively correlated with both GABA (r = −0.85, P < 0.01) and citric acid (r = −0.80, P < 0.05). Cluster 2 contained cis-aconitic acid, shikimic acid, and sinapyl alcohol, all positively correlated. Cluster 3 included p-coumaric acid and oxoadipic acid, which were positively correlated (Fig. 2E).

3.3. Transcriptomic insights into chitosan-mediated delay of mango ripening and senescence

Transcriptomic analysis (RNA-seq) identified candidate DEGs responsive to chitosan treatment. PCA confirmed treatment reproducibility before differential expression analysis (Fig. 3A). Comparative analysis (chitosan vs control at 4 DAP) revealed 2290 DEGs (|log2FC| > 1, padj <0.05): 1378 downregulated, 912 upregulated (Fig. 3B; Table S2). KEGG enrichment analysis highlighted significant DEG enrichment in plant hormone signal transduction, starch and sucrose metabolism, photosynthesis-antenna proteins, flavonoid biosynthesis, and carotenoid biosynthesis (Fig. 3C; Table S3), indicating chitosan modulates ripening via coordinated regulation of hormonal signaling and carbon metabolism.

Fig. 3.

Transcriptome analysis of chitosan vs control. (A) Transcriptome PCA. (B) Volcano plot: downregulated (green), upregulated (red), non-significant (orange). (C) KEGG pathway enrichment for DEGs; x-axis: pathway categories, y-axis: -log10(padj); bar numbers indicate DEG counts per pathway. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Transcriptomic analysis (RNA-seq) identified candidate DEGs responsive to chitosan treatment. Principal component analysis (PCA) confirmed treatment reproducibility before differential expression analysis (Fig. 3A). Comparative analysis (chitosan vs control at 4 DAP) revealed 2290 DEGs(|log2FC| > 1 and padj <0.05), comprising 1378 downregulated and 912 upregulated genes (Fig. 3B; Supplementary Table S2). KEGG enrichment analysis highlighted significant DEGs enrichment in plant hormone signal transduction, starch and sucrose metabolism, photosynthesis-antenna proteins, flavonoid biosynthesis, and carotenoid biosynthesis (Fig. 3C; Supplementary Table S3), indicating chitosan modulates ripening via coordinated regulation of hormonal signaling and carbon metabolism.

3.4. Co-expression analysis identify genes accounting for changes in several metabolites

Correlation analysis between 2290 DEGs and 166 DAMs (|correlation coefficient| >0.8, P < 0.01) identified 448 DEGs (241 upregulated, 207 downregulated) significantly associated with 64 KEGG pathways co-enriched in transcriptomic and metabolomic datasets (Fig. 4A–D; Table S4). Linear correlation network analysis highlighted hub genes governing organic acid dynamics: citrate and GABA accumulation positively correlated with PDHE1-B, GDH, and FBA2 but inversely linked to starch degradation genes (PHS2, BAM7, TPPA) (Fig. 4E–F). abscisic acid displayed antagonistic regulatory patterns. Shikimate pathway intermediates (shikimic acid, cis-aconitic acid) showed positive associations with GDH, UGT73C4, ZOG1 but negative correlations with SS III, TPPA. These findings underscore chitosan ‘s role in redirecting metabolic flux by coordinating transcriptional and metabolic networks to delay ripening.

Fig. 4.

Multi-omics integrative network of chitosan-mediated ripening regulation. (A) Nine-quadrant correlation matrix linking DMs and DEGs. (B) Venn diagram: intersection of DEGs and DMs sharing 64 KEGG pathways. (C) Heatmap of gene expression profiles. (D) Volcano plot of DEGs: upregulated (red), downregulated (green). (E) Bar chart: genes associated with key metabolites. (F) Network diagram: hub gene-metabolite interactions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.5. Chitosan impact on starch, sucrose, and glycolysis/gluconeogenesis pathways

Chitosan significantly influenced starch and sucrose metabolism. Transcriptomic data revealed coordinated suppression of starch degradation genes (BAM1/BAM7, PHSH, PHSL) alongside upregulation of starch biosynthesis genes (AGPL1, AGPS, AGPL2), indicating a shift toward starch stabilization rather than sugar conversion (Supplementary Fig. S2A). Concurrently, glycolysis-related genes, including PFK and PK, were broadly upregulated under chitosan treatment (Supplementary Fig. S2B), likely compensating for reduced hexose availability by enhancing pyruvate production. This increased pyruvate flux may prioritize citrate synthesis via mitochondrial pathways, as suggested by co-upregulated GDH and ALDH (Fig. 4F). In contrast, the gluconeogenesis gene FBP was downregulated, reflecting a suppression of carbon recycling from non-carbohydrate precursors.

3.6. Chitosan adjusts the flavonoid and carotenoid pathways in mango

Given the significant enrichment of flavonoid and carotenoid biosynthesis, we examined chitosan modulation. In flavonoid pathway, chitosan downregulated genesd C4H (mango002178) and CHS1 (mango002800) relative to control (Fig. 5), suggesting diminished supply of precursors supply. Conversely, genes integral to anthocyanin production F3H (mango003261) and DFR (mango024421) were significantly upregulated. Additionally, upregulation of LDOX (mango001587) and LAR (mango015597) implied enhanced anthocyanins and proanthocyanidin biosynthesis, indicating chitosan modulates flavonoid metabolism, potentially delaying color development while enhancing antioxidant capacity.

Fig. 5.

Expression profiles of genes in flavonoid and carotenoid pathways. (A) Heatmap: flavonoid pathway gene expression (chitosan vs control). (B) Heatmap: carotenoid pathway gene expression (chitosan vs control). Color scale: red (higher in chitosan), blue (lower in chitosan). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

In carotenoid biosynthesis, chitosan triggered distinct regulatory effects (Fig. 5). Key carotenoid genes (PSY, NCED1) were upregulated, enhancing precursor flux toward carotenoid synthesis. Concurrently, abscisic acid catabolism genes CYP707A1 and CYP707A4 were strongly activated, promoting abscisic acid degradation. In contrast, abscisic acid biosynthesis genes NCED3, ABA2, and AAO1/AAO3 were downregulated, suppressing abscisic acid production. This dual regulation, upregulating abscisic acid degradation while inhibiting its synthesis, suggests chitosan delays ripening by reducing abscisic acid accumulation, aligning with observed metabolic shifts that stabilize postharvest quality.

3.7. Chitosan regulates chlorophyll metabolism and cell wall dynamics

To elucidate chitosan preservation of chlorophyll, we analyzed regulatory effects on chlorophyll metabolism genes (Fig. 6A). Chitosan dynamically balanced synthesis and degradation by upregulating POR and CPO1 (enhancing synthesis) while elevating CLH1 and SGR (promoting turnover). Downregulation of HY2, NYC1, and CHLP suppressed phytochrome activity and chlorophyll b recycling, collectively stabilizing chlorophyll pools and delaying degreening.

Fig. 6.

Chitosan influence on chlorophyll and cell wall genes. (A) Heatmap: chlorophyll metabolism gene expression; red (upregulated), blue (downregulated). (B) Heatmap: cell wall metabolism gene expression; red (upregulated), blue (downregulated). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Beyond chlorophyll dynamics, chitosan significantly delayed fruit softening. Transcriptomic analysis revealed coordinated regulation of cell wall metabolism genes linked to firmness retention (Fig. 6B). Upregulation of cellulose synthase genes CSLE6 and CSLD2 suggested enhanced cell wall reinforcement, while elevated PMEI1.2 isoforms implied tighter pectin crosslinking. Concurrently, XTH23 induction and suppressed PG expression balanced cell wall loosening and degradation. These transcriptional adjustments synergistically maintained structural integrity, aligning with prolonged firmness and extended shelf life.

3.8. Chitosan modulates plant hormone signal transduction

To elucidate the hormonal mechanisms underlying chitosan-mediated delay of mango ripening, we focused on plant hormone signal transduction pathways, which emerged as the most significantly enriched KEGG category. Transcriptomic analysis revealed chitosan profoundly regulated abscisic acid and ethylene signaling (Fig. 3C). These pathways are critical drivers of fruit ripening and senescence, and their modulation by chitosan explains the delayed deterioration of mango quality.

Chitosan treatment was associated with suppression of abscisic acid biosynthesis, evidenced by downregulation of ABA2 and upregulation of catabolic gene CYP707A1/A4 (Fig. 7A), aligning with reduced abscisic acid levels (Fig. 2D). Concurrently, the abscisic acid receptor PYL4 and phosphatase PP2CA were downregulated, attenuating abscisic acid signal transduction.

Fig. 7.

Differential regulation of hormone signaling genes. (A) Expression profiles of abscisic acid-related genes. (B) Ethylene signaling genes. Values represent log2 fold change (chitosan vs. control); color intensity: expression levels (red: upregulation; blue: downregulation). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Ethylene signaling showed attenuated activity (Fig. 7B), coinciding with downregulation of the ethylene receptor ETR1 and transcription factors EIN3/EIN4， while upregulation of negative regulators EBF1/EBF2. Broad suppression of ethylene-responsive transcription factors, including ERF3, ERF4, ERF109 and ERF118, further delayed ethylene-driven processes such as cell wall softening and chlorophyll degradation.

3.9. Transcription factor

Chitosan differentially modulated stress- and ripening-associated TFs. Pronounced suppression occurred across WRKY family members (WRKY7, WRKY17, WRKY57, WRKY53; Fig. 8). bZIP TFs exhibited subtype-specific regulation: bZIP/TGA7 and bZIP/CPRF1 were downregulated, while bZIP/ATHB-6 and bZIP/RF2b were significantly upregulated (Fig. 8). These results indicate chitosan selectively represses senescence-promoting TFs (e.g., WRKY53) while activating bZIP isoforms associated with metabolic stabilization.

Fig. 8.

Differential expression of transcription factors (chitosan vs control). Heatmap: log2 fold changes; red (upregulated), blue (downregulated). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.10. Validation of gene expression by qRT-PCR

qRT-PCR analysis confirmed the transcriptomic trends for key genes associated with abscisic acid metabolism (NCED3 downregulation, CYP707A1 upregulation), ethylene signaling (ETR1 suppression), cell wall dynamics (PG inhibition, CSLD2 activation), starch metabolism (AGPL2 upregulation, BAM7 repression), and transcriptional regulation (WRKY53 and bZIP/RF2b downregulation) (Supplementary Fig. S3). The expression patterns aligned closely with RNA-seq data, demonstrating consistency and reinforcing chitosan's multi-targeted modulation of ripening-related pathways in mango fruit.

4. Discussion

This study provides the first integrated multi-omics evidence that chitosan delays mango ripening by synchronously reprogramming hormonal crosstalk (abscisic acid-ethylene), metabolic flux, and transcriptional networks. Crucially, we identified WRKY53 and bZIP/RF2b likely act as central hubs coordinating this process—a mechanism previously unreported in climacteric fruits.

4.1. Chitosan reprograms carbon metabolism to sustain organic acids and delay postharvest senescence in mango

Our integrative analyses revealed that chitosan reprograms carbon metabolism, correlating with delayed mango ripening through a newly identified dual mechanism. Notably, chitosan-induced GABA accumulation tightly corresponds with retained citric acid (r = 0.92, P < 0.001) and was accompanied by heightened expression of glycolytic genes PFK and PK (Supplementary Fig. S2) together with glutamate dehydrogenase GDH (Fig. 4F), thereby rerouting pyruvate toward GABA and citrate synthesis (Yan et al., 2024). Simultaneously, chitosan downregulated starch-degrading genes BAM7 and PHS2 and upregulated biosynthetic genes AGPS and AGPL2 (Supplementary Fig. S2), limiting hexose release and conserving both energy reserves and structural integrity (Cosme Silva et al., 2017).

GABA levels showed a strong negative correlation with abscisic acid content (r = −0.85, P < 0.01), reflecting transcriptional suppression of abscisic acid biosynthesis gene NCED3 and activation of catabolic gene CYP707A (Figs. 2D, 5B). This GABA–abscisic acid antagonism reveals a previously undocumented regulatory axis for chitosan-mediated ripening delay. Although GABA-mediated inhibition of starch degradation has been reported in kiwifruit (Yan et al., 2024), our findings are the first to demonstrate its coordination with abscisic acid suppression in mango, effectively uncoupling sugar accumulation from ripening. The outcome is sustained acidity, preserved firmness and an extended retail shelf life, offering a molecular template for postharvest technologies that exploit GABA-shunt pathways in climacteric fruits.

4.2. Multi-targeted regulation of Cell Wall integrity and pigment dynamics

Our integrative analysis revealed that chitosan's preservation of mango quality hinges on its synchronized regulation of cell wall architecture and pigment homeostasis—a mechanism advancing beyond prior fragmented reports. While earlier studies established chitosan's capacity to delay softening through physiological metrics (Jongsri et al., 2016), we demonstrate for the first time how it transcriptionally coordinates antagonistic pathways: upregulating cellulose synthases (CSLE6, CSLD2) and pectin methylesterase inhibitors (PMEI1.2) to fortify wall structure, while concurrently suppressing polygalacturonase (PG) and xyloglucan endotransglucosylase (XTH23) to limit disassembly (Fig. 6B). This dual-axis regulation, validated by retained CSP/HSP and reduced WSP accumulation (Fig. 1B), explains the prolonged firmness more comprehensively than single-target interventions. Notably, the suppression of PG/XTH23 mirrors their established role in cherry softening (Zhai et al., 2021), suggesting chitosan may exploit conserved cell wall remodeling nodes across climacteric fruits.

Parallelly, retained firmness and chlorophyll levels under chitosan treatment correlate with delayed ripening, likely through dynamic pigment turnover and cell wall stabilization. Contrary to antioxidants that merely inhibit chlorophyll breakdown (e.g., in broccoli; Tan et al., 2023)), chitosan simultaneously elevates synthesis genes (POR, CPO1) and degradation markers (CLH1, SGR) (Fig. 6A), creating a homeostasis that delays net chlorophyll loss—a strategy aligning with the ‘synthesis-degradation balance’ model (Hörtensteiner, 2006; Kräutler, 2008). This principle extends to other pigments via conserved transcriptional regulation (Rahman et al., 2018; Singh et al., 2020; Yang et al., 2024; Zheng et al., 2023), suggesting a conserved strategy by which chitosan orchestrates transcriptional reprogramming across pigment pathways. In flavonoid metabolism, chitosan upregulates key biosynthetic genes (PAL, F3H; Fig. 5A), mirroring its effect in strawberries where flavonoid content increases 2.6-fold (Rahman et al., 2018; Shi et al., 2024). Similarly, for carotenoids, chitosan sustains accumulation (Fig. 1C) by upregulating PSY while diverting metabolites from ABA biosynthesis via CYP707A1/4 activation (Fig. 5B). Collectively, this multi-pigment balancing strategy decouples visual quality retention from senescence progression, addressing a key limitation in ABA-targeted approaches where hormone depletion compromises stress tolerance (Kai et al., 2019).”

4.3. Involvement of abscisic acid and ethylene in postharvest fruit ripening and senescence

Our findings demonstrate a departure from ethylene-centric models by revealing that chitosan disrupts the synergistic yet phase-specific interplay between abscisic acid and ethylene. While ethylene traditionally dominates late-stage mango senescence (Tipu & Sherif, 2024), chitosan treatment was linked to suppression of abscisic acid biosynthesis (NCED3/ABA2 downregulation; Figs. 5B, 7A) delays early ripening events like starch degradation and carotenoid accumulation, consistent with abscisic acid's role as a ripening initiator in mango (Wu, Shan, et al., 2022) and strawberry (Kai et al., 2019). Crucially, chitosan attenuation of ethylene signaling through coordinated repression: downregulating the receptor ETR1 to limit signal perception while suppressing EIN3/EIN4 transcription factors to block downstream responses (Fig. 7B). This synchronized hormonal intervention effectively decouples abscisic acid-driven ripening initiation from ethylene-amplified senescence, a multi-target strategy distinct from single-pathway inhibitors.

Notably, we resolved an apparent paradox in abscisic acid signaling: despite reduced abscisic acid levels (Fig. 2D), chitosan upregulated the receptor PYL4 while downregulating phosphatase PP2CA (Fig. 7A). This mirrors mechanisms in Arabidopsis where PYL-PP2C complexes function as abscisic acid-independent stress sensors (Li et al., 2013), suggesting chitosan repurposes these components to maintain SnRK2-mediated stress resilience without triggering senescence. Such rewiring exemplifies how chitosan modulates hormone signaling through transcriptional reprogramming rather than merely suppressing pathways, an innovation over approaches like 1-MCP (ethylene blocker) that compromise stress adaptation.

4.4. WRKY and bZIP transcription factors as dual regulators of postharvest quality

Beyond direct hormonal modulation, transcriptional regulation provides a secondary tier of control. Chitosan-mediated transcriptional reprogramming extends beyond hormonal pathways to selective modulation of WRKY and bZIP transcription factors—a tier of regulation that fine-tunes the senescence delay. WRKY transcription factors are established regulators of plant development and stress responses, orchestrating processes from gamete formation to senescence through hormonal interactions (Khoso et al., 2022). While implicated in fruit ripening models, such as FaWRKY71 modulating anthocyanin biosynthesis in strawberry (Yue et al., 2022) and MaWRKY21 integrating stress signals in banana (Jia et al., 2022), their mechanistic roles in mango ripening remain largely unknown, with no studies linking specific isoforms to hormonal crosstalk during postharvest senescence. Our work addresses this gap by demonstrating chitosan significantly downregulates key WRKYs (WRKY53, WRKY7, WRKY17; Fig. 8), coinciding with reduced expression of ABA biosynthesis genes and lower ABA levels (Fig. 5B), and delayed ethylene-driven senescence.

The suppression of WRKY53, a conserved activator of abscisic acid genes (NCED3) in Arabidopsis (Miao et al., 2004), suggests its parallel role in mango ripening initiation, potentially through transcriptional regulation of ethylene biosynthesis genes (ACS/ACO homologs) akin to PpWRKY14 in peach (Liu et al., 2024). Concurrent downregulation of defense-associated WRKY7/17 (Fuenzalida-Valdivia et al., 2024) may redirect resources from stress responses to quality preservation. This first report of WRKY-abscisic acid-ethylene coordination in mango positions WRKY53 as a critical senescence accelerator, with chitosan-mediated repression decoupling ripening initiation from deterioration. Future validation of WRKY53 binding to NCED3 and ethylene promoters should clarify its hub function.

In contrast to WRKYs, bZIP transcription factors, though well-documented in plant stress adaptation and secondary metabolism regulation (Lu et al., 2022) (Xiao et al., 2023), remain underexplored in fruit ripening contexts. Current knowledge is largely confined to models like banana, where phosphorylated MabZIP21 accelerates ripening by activating ethylene-dependent genes (Wu, Shan, et al., 2022), and tomato, where bZIPs participate in ethylene-mediated transcriptional networks (Li et al., 2022). Our study revealed chitosan differentially modulated bZIP subtypes in mango, with significant upregulation of bZIP/ATHB-6 and bZIP/RF2b alongside suppression of stress-linked isoforms (e.g., bZIP/TGA7; Fig. 8). This selective induction during chitosan treatment coincided with attenuation of abscisic acid biosynthesis (NCED3) and ethylene signaling (ETR1; Figs. 5B, 7B), suggesting potential crosstalk between these TFs and hormonal pathways.

Critically, the timely upregulation of RF2b/ATHB-6 correlates with delayed quality deterioration (retained firmness/organic acids; Fig. 1B, D), indicating their potential contribution to metabolic stabilization against senescence. While direct mechanistic links require validation, parallels exist: MabZIP21 integrates ethylene cues via phosphorylation (Wu, Shan, et al., 2022), and tomato bZIPs respond to ethylene flux (Li et al., 2022). We thus propose RF2b/ATHB-6 may similarly interface with hormonal signals—potentially by modulating abscisic acid catabolism or ethylene sensitivity genes—to decelerate ripening. It is imperative for future studies to elucidate the DNA-binding targets of RF2b/ATHB-6 and characterize their phosphorylation status under abscisic acid/ethylene perturbations.

While this study elucidates the chitosan-mediated regulatory network in mango ripening, certain aspects merit further validation. The findings are derived from a single mango cultivar (‘Tainong’), and thus the generalizability to other cultivars requires verification. Additionally, the fixed chitosan concentration (1 % w/v) may not capture dose-dependent effects. Future multi-variety studies and concentration gradients will strengthen the relevance of the proposed abscisic acid-ethylene-GABA axis.

5. Conclusions

This study demonstrates that chitosan treatment effectively delays mango fruit ripening and senescence by modulating metabolic, hormonal, and transcriptional networks. Chitosan preserves fruit quality by reducing firmness loss, chlorophyll degradation, and WSP accumulation, demonstrating its association with delayed ripening processes. Multi-omics analysis revealed that chitosan is associated with suppression of abscisic acid biosynthesis (NCED3, ABA2) and enhanced catabolism (CYP707A1/4), while downregulating ethylene signaling (ETR1, EIN3/EIN4). Key transcription factors (WRKY53, bZIP/RF2b) regulate abscisic acid-ethylene crosstalk and cell wall stability. Additionally, chitosan maintains organic acid levels and delays starch-to-sugar conversion. These findings highlight chitosan's potential as a sustainable postharvest treatment to extend shelf life by targeting ripening-associated pathways. Future research should explore chitosan's synergistic effects with other eco-friendly technologies for commercial applications.

CRediT authorship contribution statement

Zhiwei Wu: Software, Methodology, Formal analysis, Data curation, Conceptualization. Zhisheng Lin: Writing – original draft, Methodology, Investigation. Qinghua Qiao: Writing – original draft, Methodology, Investigation. Lieqin Shi: Writing – original draft, Methodology, Investigation. Zhenxin Ren: Writing – review & editing, Visualization, Validation, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Funding

This research was funded by the National Natural Science Foundation of China (grant number: 32260619) and the Guangxi Natural Science Foundation Project (grant number: 2023GXNSFAA026353).

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Zhenxin Ren reports financial support was provided by National Natural Science Foundation of China. Zhenxin Ren reports a relationship with Guangxi Natural Science Foundation Project that includes: funding grants. Has patent pending to. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material 1

Supplementary Figures S1–S3. Physiological, Metabolic, and Transcriptional Validation of Chitosan Effects on Mango Fruit Ripening

Supplementary material 2

Supplementary Table S1. qRT-PCR Primer Sequences for Key Ripening-Related Genes in Mango

Supplementary material 3

Supplementary Table S2. DEGs in Chitosan-Treated vs. Control Mango Fruit

Supplementary material 4

Supplementary Table S3. KEGG Pathway Enrichment Analysis of DEGs in Chitosan-Treated vs. Control Mango Fruit

Supplementary material 5

Supplementary Table S4. Integrated Multi-Omics Correlation Matrix of DEGs and DAMs in Chitosan-Treated vs. Control Mango Fruit

Data availability

Data will be made available on request.