“Tainong 17” pineapple was used in this research, which was planted in the Danzhou experiment field of Tropical Crops Genetic Resources Institute, Chinese Academy of Tropical Agricultural Sciences. All chemical reagents were purchased from the Sigma-Aldrich Company (Shanghai, China).

Plasmids pYLsgRNA-OsU3m (Plasmid#66193) and pYLCRISPR/Cas9Pubi-H (Plasmid#66187) were purchased from Addgene (USA). The transformation vector was constructed according to a published protocol (Ma et al., 2015). According to the genomic DNA sequence of AcoACS1 (GenBank accession number NC_033623.1), 5′-gaggattcgccttactttga-3′ was selected as the target sequence. Oligo A (5′-CTTGaggattcgccttactttgaGTTTTAGAGCTAGAAATAGC-3′) and oligo B (5′-AAACtcaaagtaaggcgaatcctCAAGC-3′) were synthesized. For annealing, 1 μL of Oligo A (100 μM), 1 μL of Oligo B (100 μM), 1 μL of 10× T4 ligase buffer, and 7 μL of nuclease-free water were mixed. The mixture was incubated at 95°C for 5 min, then slowly cooled to 25°C (at a rate of 0.1°C/min), and finally held at 4°C. The annealed double-stranded oligonucleotide was inserted into pYLsgRNA-OsU3m. The reaction mixture contained 50–100 ng of pYLsgRNA-OsU3m plasmid, 1 μL of the diluted annealed oligonucleotide (diluted 10-fold before use), 1 μL of BsaI-HFv2 restriction enzyme, 1 μL of T4 DNA ligase, and 2 μL of 10× T4 DNA ligase buffer, made up to a final volume of 20 μL with sterile water. The reaction conditions were as follows: 20 cycles of 37°C for 5 min and 16°C for 10 min, followed by 37°C for 30 min and 80°C for 10 min. The reaction product was transformed into competent Escherichia coli (DH5α) cells, which were then spread on LB plates containing 100 µg/mL of spectinomycin. Single colonies were picked and cultured, and plasmids were extracted. Sequencing was performed using the OsU3 promoter universal primer. The verified plasmid was designated as pYLsgRNA-A.

Subsequently, a second round of Golden Gate Assembly (a one-step cloning technique that utilizes type IIS restriction endonucleases and DNA ligase to directionally assemble multiple DNA fragments efficiently and seamlessly within a single reaction system) was performed to assemble the sgRNA expression cassette into the destination vector. The reaction mixture contained 100 ng of pYLCRISPR/Cas9Pubi-H plasmid, 50 ng of pYLsgRNA-A plasmid, 1 μL of BsaI-HFv2 restriction enzyme, 1 μL of T4 DNA ligase, 2 μL of 10× T4 DNA ligase buffer, and sterile water added to a final volume of 20 μL. The reaction conditions were as follows: 20 cycles of 37°C for 5 min and 16°C for 10 min, followed by 37°C for 30 min and 80°C for 10 min. This step excised the entire sgRNA (single-guide RNA) expression cassette from pYLsgRNA-A and ligated it into pYLCRISPR/Cas9Pubi-H. The product was transformed into E. coli (DH5α). The transformed E. coli mixture was spread on LB plates containing 75 µg/mL of kanamycin. Single colonies were picked, and plasmids were extracted. PCR was performed using primers P1 (5′-TGCCTGATCCCTTCAGGAG-3′) and P2 (5′-GCGGTAGCCAATTATACCTT-3′) to confirm correct assembly. The verified plasmid was named pYLCRISPR/Cas9Pubi-H-A. This plasmid was introduced into Agrobacterium tumefaciens-competent cells (EHA105) via electroporation. Embryogenic callus was induced and transformed following the protocol in a published paper (Shu et al., 2024). Kanamycin (20 mg/L) was used to select transformed callus. To confirm the transformation of pineapple plants, PCR was performed using primers Cas9-F (5′-GAGGGCAAGAATGGGAATAG-3′) and Cas9-R (5′-GTGGCTCGACGTCTACTTCA-3′) with genomic DNA extracted from putative transgenic plants as templates. Successful transformation was indicated by the presence of a 650-bp band on agarose gel. To identify mutations in the target sequence, PCR was performed using primers P3 (5′-tggtgattgctgatggggcg-3′) and P4 (5′-cggcaagccatggtagtctt-3′) with genomic DNA extracted from the transgenic pineapple plants as templates. The PCR products were sent to Sangon Biotech (Shanghai, China) for sequencing.

Plasmids pYLsgRNA-OsU3m (Plasmid#66193) and pYLCRISPR/Cas9Pubi-H (Plasmid#66187) were purchased from Addgene (USA). The vector for transformation was constructed according to a published protocol (Ma et al., 2015). According to the genomic DNA sequence of AcoACS1 (NC_033623.1), 5′-cgctggagttaatacgtgag-3′ was selected as the target sequence. Oligo A (5′-CTTGgctggagttaatacgtgagGTTTTAGAGCTAGAAATAGC-3′) and oligo B (5′-AAACctcacgtattaactccagcCAAGC-3′) were synthesized. The annealing of double-stranded oligonucleotides, cloning into pYLsgRNA-OsU3m, and subcloning into pYLCRISPR/Cas9Pubi-H were performed as described in Section 2.2. The transformed E. coli mixture was spread on LB plates containing 75 µg/mL of kanamycin. Single colonies were picked, and plasmids were extracted. PCR was performed using primers P1 (5′-TGCCTGATCCCTTCAGGAG-3′) and P2 (5′-GCGGTAGCCAATTATACCTT-3′). The PCR products were sent to Sangon Biotech (Shanghai, China) for sequencing. The verified plasmid was named pYLCRISPR/Cas9Pubi-H-ABRE. Subsequent steps were performed, including the transformation of pYLCRISPR/Cas9Pubi-H-ABRE into A. tumefaciens (EHA105), embryogenic callus induction, infection, and the identification of transgenic plants and AcoACS1-ABRE-knockout mutants, as described in Section 2.2.

Ethylene production in pineapple fruits was measured according to the protocol described previously (Zeng et al., 2020). A GC2010 gas chromatograph (Shimadzu, Kyoto, Japan) equipped with a flame ionization detector was used. Measurements were performed in triplicate for each sample, with each sample consisting of five to eight fruits.

The respiration rate was measured using the static alkaline absorption method (Ye et al., 2014). Two glass desiccators of identical volume were prepared. A Petri dish containing 20 mL of 0.4 mol/L NaOH solution was placed inside each desiccator. A perforated platform was then placed above the dish, and four pineapple fruits (with pre-measured weight and volume) were placed on the platform in one desiccator. The other desiccator served as a blank control and was filled with an equal volume of cooled distilled water. After 1 h of incubation, the Petri dishes were removed, and their contents were transferred into conical flasks. Saturated BaCl₂ solution and phenolphthalein indicator were added, followed by titration with 0.1 mol/L of standard oxalic acid solution. The blank sample was titrated in the same manner.

Quantitative real-time polymerase chain reaction (qRT-PCR) was performed according to a published protocol (Shu et al., 2024). Pineapple fruits were collected and placed in a room at 25°C with 85% relative humidity. Total RNA was extracted using TRIzol reagent (Invitrogen, USA). RNA was reverse-transcribed into cDNA using the PrimeScript™ RT reagent kit (TaKaRa, Dalian, China). The housekeeping gene AcoTubulin (Aco010170.1) was used as the internal control. Each real-time PCR reaction contained SYBR^® Premix Ex Taq II (Tli RNaseH Plus), forward and reverse primers, cDNA template, and ddH₂O. All PCR reactions were performed in triplicate. The relative mRNA expression levels of each gene were analyzed using the 2^-ΔΔCt method and normalized to the reference gene AcoTubulin.

Fruit firmness was measured according to a published method (Zhang et al., 2022). A TMS-Pro food property analyzer (FTC, USA) was used. Measurements were taken at three positions: the basal end (base), the middle section, and the apical end (far end) of the fruit. The average value of these measurements was calculated. Six fruits were measured for each treatment.

Pineapple fruits at 80% maturity were collected. An Agilent 6890–5975 gas chromatography-mass spectrometry (GC-MS) system (Agilent Technologies, USA), a manual SPME sampler (Supelco, USA), and a 65-μm PDMS solid-phase microextraction fiber (Supelco, USA) were used. Pineapple pulp was homogenized, and 6.5 g was placed into a sample vial, which was sealed with a PTFE-lined butyl rubber septum; an empty vial served as a control. Headspace solid-phase microextraction (HS-SPME) was used to extract aroma components for injection. The sample vial was incubated at 50°C for 40 min. After extraction, the fiber was inserted into the GC injection port and desorbed at 250°C for 2.5 min. Chromatographic separation was performed on an HP-5 column (30 m × 0.32 mm × 0.25 μm). The injection port temperature was 200°C, and splitless injection was used. The column temperature program was initiated at 35°C (held for 2 min), increased to 180°C at a rate of 8°C/min (held for 4 min), and finally increased to 230°C at 15°C/min (held for 1 min). The mass spectrometry conditions were as follows: helium was used as the carrier gas at a flow rate of 1.06 mL/min; electron ionization (EI) was employed with an electron energy of 70 eV; the ion source temperature was 230°C; and mass spectra were acquired in full scan mode with a scan range of m/z 45–450. Compounds were identified by comparing their mass spectra with the NIST05 library and through manual interpretation. The relative content of each component was calculated using the peak area normalization method. Quantification was performed using the internal standard method with n-tetradecane (3.85 μg) added as the internal standard. Data were subjected to statistical analysis using SPSS software, with P <0.05 considered statistically significant.

Polygalacturonase (PG) activity was determined according to a published protocol (Ye et al., 2014). Five grams of frozen pineapple pulp was homogenized in 50 mmol/L of sodium phosphate buffer (pH 6.0) containing 2.4 mol/L of NaCl, 0.5 g of PVP, and a small amount of quartz sand. The mixture was ground at 4°C for 2 min and then centrifuged at 10,000 rpm for 30 min at 4°C. The supernatant was transferred to a dialysis bag (12,800 Da MW cutoff cellulose membrane). Dialysis was performed against 50 mmol/L of sodium phosphate buffer (pH 4.5) containing 150 of mmol/L NaCl for 24 h, and the buffer was changed once during the process. The dialyzed supernatant was used as the crude PG enzyme extract. For the reaction, 0.1 mL of crude PG enzyme was added to 0.4 mL of 0.2% pectin solution. The reaction mixture was incubated in a water bath at 40°C for 1 h. Then, 1.5 mL of 3,5-dinitrosalicylic acid (DNS) reagent was added, and the mixture was incubated in a boiling water bath for 10 min. After cooling to room temperature, the mixture was diluted 5-fold. The absorbance was measured at 540 nm. A standard curve was constructed using galacturonic acid as the standard.

Pectin methylesterase (PME) was extracted, and the activity was measured according to a published method (Ye et al., 2014). Twenty grams of frozen pulp was homogenized for 2 min at 4°C with 40.0 mL of a pre-cooled mixture containing 2 mol/L of NaCl, 0.50 g of PVP, and a small amount of quartz sand. The homogenate was centrifuged at 10,000 rpm for 30 min at 4°C. The supernatant was dialyzed against 0.1 mol/L of sodium phosphate buffer (pH 7.5) for 24 h. The dialyzed supernatant was used as the crude PME enzyme extract. For the activity assay, 0.3 mL of 0.5% pectin solution (dissolved in 0.1 mol/L of sodium acetate buffer, pH 5.5) was mixed with 0.1 mL of PME extract and 0.6 mL of 0.1 mol/L sodium acetate buffer (pH 5.5). The reaction mixture was incubated at 30°C for 1 h. After incubation, the solution was cooled to room temperature, and 2 mL of potassium permanganate–phosphoric acid solution was added. The mixture was allowed to stand at room temperature for 90 min. Then, 2 mL of 10% oxalic acid solution was added, and the solution was heated until the purple color disappeared. Finally, 2 mL of acetylacetone solution was added, and the mixture was heated in a boiling water bath for 5 min. After cooling to room temperature under running tap water, the absorbance was measured at 418 nm. A boiled enzyme solution was used as a blank control. A standard curve was constructed using anhydrous methanol.

β-Galactosidase (β-Gal) activity in pineapple was measured according to a published protocol (Ye et al., 2014). Five grams of frozen pulp was homogenized for 2 min at 4°C with 20.0 mL of 1.4 mol/L NaCl solution (pH 6.0), 0.50 g of PVP, and a small amount of quartz sand. The mixture was stirred for 1 h at 4°C, followed by centrifugation at 10,000 rpm for 30 min at 4°C. The supernatant was dialyzed against 0.1 mol/L of citrate buffer (pH 4.0) for 24 h. The resulting supernatant served as the crude β-Gal enzyme extract. For the activity assay, 0.1 mL of the crude extract was mixed with 0.4 mL of bovine serum albumin (BSA) solution, 0.5 mL of 0.1 mol/L citrate buffer (pH 4.0), and 0.4 mL of 16 mmol/L p-nitrophenyl-β-D-galactopyranoside (PNPG) dissolved in 0.1 mol/L of citrate buffer. The reaction mixture was incubated at 37°C for 5 min. The reaction was terminated immediately by adding 2 mL of 0.2 mol/L Na₂CO₃. The absorbance of the released p-nitrophenol was measured at 400 nm. A blank control was prepared by adding 2 mL of 0.2 mol/L Na₂CO₃ to the enzyme mixture prior to the addition of the substrate. A standard curve was constructed using p-nitrophenol.

Xylanase (XYL) activity in pineapple was measured according to a published method (Ye et al., 2014). Twenty-five grams of frozen pulp was homogenized for 2 min at 4°C in 50 mL of 0.05 mol/L sodium acetate buffer (pH 5.0) containing 0.1 mol/L of NaCl, 2% (v/v) dithiothreitol (DTT), and 5% (w/v) PVP. The homogenate was centrifuged at 8,000 rpm for 30 min at 4°C. The supernatant was collected as the crude XYL extract. For the activity assay, 0.5 mL of the crude extract was mixed with 1.3 mL of 0.1 mol/L sodium acetate buffer (pH 5.0) and 0.2 mL of 0.1% (w/v) xylan (prepared in 0.1 mol/L of sodium acetate buffer, pH 5.0). The reaction mixture was incubated at 37°C for 1 h. The released reducing sugars (xylose) were quantified using DNS, with boiled, heat-inactivated crude XYL extract serving as the blank control. A standard curve was constructed using xylose.

Cellulase (CX) activity in the pulp was measured according to a published method (Ye et al., 2014). Five grams of frozen pulp was homogenized for 2 min at 4°C in 20.0 mL of pre-cooled 0.04 mol/L sodium acetate buffer (pH 4.6) containing 6.8% NaCl and 1.8 mmol/L of EDTA. The mixture was stirred at 4°C for 1 h. Subsequently, the homogenate was centrifuged at 10,000 rpm for 15 min at 4°C, and the supernatant was dialyzed against 0.04 mol/L of sodium acetate buffer (pH 4.6) at 4°C for 24 h. The dialyzed supernatant served as the crude CX extract. For the activity assay, 0.5 mL of the crude extract was incubated with 0.5 mL of 1% carboxymethyl cellulose (CMC) solution at 40°C for 1 h. The reaction was terminated by adding 2.5 mL of DNS reagent, followed by boiling for 5 min. After cooling to room temperature, the absorbance was measured at 540 nm. A standard curve was constructed using glucose.

Potential off-target sites were predicted using Cas-OFFinder according to a published method (Bae et al., 2014). The target sequences were entered into the ‘Query Sequences’ box to identify potential off-target loci (http://www.rgenome.net/cas-offinder/). Primers flanking these sites were designed for PCR amplification and sequencing. The resulting sequences were aligned with the pineapple reference genome to identify off-target mutations.

Each treatment consisted of three replicates. Significant differences among the treatments were analyzed using analysis of variance (ANOVA) followed by Duncan’s multiple range test (DMRT). Statistical analysis was performed using SPSS software. Differences were considered significant at P <0.05. Data are presented as mean ± standard error (SE).

To investigate whether AcoACS1 is the key gene responsible for pineapple softening, pineapple AcoACS1 mutants were generated using CRISPR/Cas9. The target sequence was selected based on two criteria: first, proximity to the 5′ end to maximize the effects of gene disruption; and second, a low number of predicted off-target sites to facilitate screening. Based on these criteria, the sequence 5′-gaggattcgccttactttga-3′ was selected for sgRNA construction (Figure 1). Following the transformation and differentiation of embryogenic calli, 17 transgenic plants were obtained (Figure 2). PCR amplification and sequencing of the target regions revealed that five plants were successfully edited. The editing rate was 29%. These lines were designated ACS1–1 through ACS1-5 (Figures 1; Supplementary Figure S1). Among the five lines, ACS1–2 was found harboring a premature termination codon (Figure 1). Off-target analysis was performed using a published method (Bae et al., 2014), and no off-target mutations were detected in the pineapple genome. Consequently, ACS1–2 was selected for further experiments.

The ACS1-2 plants exhibited no significant differences in phenotypes compared to Tainong 17. The flowering time and fruit shape were also similar (Figure 3A). However, the ethylene content in ACS1–2 fruits was significantly lower than that in Tainong 17 (Figure 3B). ACS1–2 fruits remained firmer and greener than Tainong 17, staying in an unripe state for a long time (Figure 3C). The respiration rate of the ACS1–2 fruit was remarkably slower than that of Tainong 17 (Figure 3D). ACS1–2 has a significantly longer shelf life than Tainong 17. The firmness of the ACS1–2 fruit showed no significant change within 60 days, after which it decreased. However, the contents of many aroma compounds in ACS1–2 fruits were much lower than those in Tainong 17 (Figure 4). Its eating quality was less than that of Tainong 17. Before sale, ethephon application was required for ACS1–2 fruits, which increased cost and inconvenience.

In the genomic sequence of AcoACS1, 171 bp from the start codon, there is a 161-bp sequence containing an ABA-responsive element (ACGTG) (ABRE) and an AT-rich region, which are the characteristics of a typical enhancer (Figure 5; Supplementary Figure S2). This sequence may play an important role in regulating the transcription of AcoACS1 in pineapple fruit, as well as influencing ethylene content. ABREs often appear in the promoters of plant genes, where they mediate ABA responses (Mishra et al., 2014). Therefore, the ABRE may play a key role in regulating AcoACS1 expression. Knocking out the ABRE may reduce enhancer activity and ethylene content in pineapple fruits without destroying the function of AcoACS1. This would allow the fruits to maintain a certain amount of ethylene, achieving the goal of significantly reducing the softening rate without significantly decreasing the contents of aroma compounds or affecting edible quality.

To identify whether the ABRE in AcoACS1 determines ethylene content in pineapple fruit, pineapple ABRE-knockout mutants were generated using the CRISPR/Cas9 system. In generating the mutants, the target for deletion was ACGTG. According to the principle for designing sgRNAs, a PAM sequence should be located 3–4 bp downstream of the deletion target. No suitable PAM was identified immediately adjacent to the 3′ end of ACGTG, rendering the direct deletion of this specific sequence infeasible. Although a CGG site was identified 19 bp downstream, the distance between the target sequence and this PAM exceeded the optimal editing window (3–13 bp), suggesting a very low probability of successful deletion. However, by shortening the deletion target to the ACGT motif within the broader sequence 5′-agttaatacgtgagtggatcaag-3′, a TGG PAM site was identified 3 bp downstream of the target. Although the ACGTG ABRE site was not completely removed, the deletion of the ACGT segment is sufficient to disrupt the ABRE function. Therefore, in this study, ACGT within 5′-agttaatacgtgagtggatcaag-3′ was selected as the deletion target. 5′-cgctggagttaatacgtgag-3′ was selected as the guide sequence for constructing the sgRNA. After the embryogenic callus was transformed and differentiated, 21 transgenic pineapple plants were obtained. Target sequences were amplified and sequenced, revealing that four plants had been edited successfully. They were named ACS1AN-1 to ACS1AN-4. The editing rate was 19%. Among the four plants, the ACGT segment in the ABRE sequence in ACS1AN-4 was successfully knocked out (Figures 6, 7). Based on the method published for predicting off-target sites (Bae et al., 2014), no possible off-target sequence was found in the pineapple genome. ACS1AN-4 was chosen for subsequent experiments.

To verify whether knocking out the ABRE can reduce enhancer function, the transcription levels of AcoACS1 in ACS1AN-4 fruits were measured. Results showed that the transcript levels of AcoACS1 in ACS1AN-4 were significantly lower than those in pineapple fruits transformed with the empty vector (NV) and Tainong 17 (Figure 8). These results indicated that after the sequence ACGT in the ABRE of the AcoACS1 enhancer was deleted, the transcript levels of AcoACS1 were significantly reduced, demonstrating that this ABRE is a core element of the enhancer. Its deletion can significantly impair the function of the enhancer.

To study whether knocking out the ABRE in AcoACS1 affects the edible quality of pineapple, fruits at 80% maturity were collected, and the contents of the main aroma compounds were measured. Results showed that the contents of main aroma compounds measured in ACS1AN-4 fruits were similar to those in Tainong 17 fruits (Table 1). For example, the content of benzyl acetate in the Tainong 17 fruit at 80% maturity was 21.86 ± 3.21 µg/kg. This value in the ACS1AN-4 fruit was 20.04 ± 3.46 µg/kg. However, the content of benzyl acetate in the ACS1–2 fruit at the same stage was 5.32 ± 0.65 µg/kg (Table 1). These results demonstrated that knocking out the ABRE reduced the enhancer function and ethylene content in pineapple fruits without destroying the function of AcoACS1. This allowed ACS1AN-4 fruits to maintain a certain amount of ethylene, which maintained the contents of main aromatic substances in the ACS1AN-4 pineapple fruit, thereby not affecting its edible quality.

To investigate the effect of knocking out the ABRE in the AcoACS1 enhancer on pineapple softening, fruit firmness was measured using a TMS-Pro food texture analyzer. The results showed that the firmness of the Tainong 17 fruit, harvested at 80% maturity, began to decline immediately after harvest and continued to decrease when stored at 25°C with 85% relative humidity. By the 20th day after harvest, the Tainong 17 fruit had completely rotted. In contrast, the firmness of the ACS1AN-4 fruit showed no significant change within 40 days after harvest (Figure 9A). After 40 days, the firmness decreased but remained significantly higher than that of Tainong 17. Similarly, the firmness of the ACS1–2 fruit showed no significant change within 60 days (Figure 9A), after which it began to decrease.

Ethylene content and respiration rate play decisive roles in fruit ripening and softening (Dos Santos et al., 2019). To investigate whether knocking out the ABRE affects ethylene content and respiration rate in pineapple fruit, these parameters were measured in fruits harvested at 80% maturity in this research. The results showed that ethylene content in the ACS1AN-4 fruit was significantly lower than that in the Tainong 17 fruit but higher than that in the ACS1–2 fruit (Figure 9B). A similar trend was observed for the respiration rate (Figure 9C). These findings indicated that knocking out the ABRE in the AcoACS1 enhancer reduced its activity, thereby decreasing ethylene synthesis. However, AcoACS1 remained functional, though its transcription rate is lower than that in Tainong 17.

Pineapple is a non-climacteric fruit, and ABA treatment can promote its softening. Knocking out the ABRE in AcoACS1 may block this promotional effect. To test this hypothesis, AcoACS1 expression levels were measured in pineapple fruits sprayed with 50 mL of 0.1 mol/L ABA at different stages. The results showed that after ABA treatment, AcoACS1 expression in Tainong 17 fruit increased significantly, whereas no significant difference was observed in ACS1AN-4 fruits sprayed with ABA compared to those sprayed with water (Figure 9D). Tainong 17 fruits sprayed with ABA softened faster than those treated with water. Moreover, Tainong 17 fruits treated with ABA softened faster than ACS1AN-4 fruits treated with ABA. The firmness of the ACS1AN-4 fruit treated with ABA was similar to that of fruit treated with water. These results demonstrate that ABA promotes AcoACS1 expression and accelerates fruit softening. Following the knockout of the ABRE sequence, AcoACS1 expression in the fruit was unaffected by ABA treatment. This indicates that ABA promotes pineapple softening via the ABRE in the AcoACS1 enhancer. ABA regulates pineapple softening through AcoACS1.

The composition and structure of flesh cell walls are the main factors in maintaining fruit firmness (Tucker et al., 2017). The degradation of cell walls by hydrolases is the primary cause of fruit softening. To investigate the mechanism by which the ABRE regulates pineapple softening, hydrolase activities in pineapple flesh were measured. Results showed that the activities of PG, PME, GAL, XYL, and CX in the ACS1AN-4 fruit were significantly lower than those in Tainong 17 (Table 2). However, the activities of these hydrolases in the ACS1AN-4 fruit were higher than those in ACS1-2 (Table 2). These results indicated that the ABRE promotes the activities of endogenous hydrolases and regulates the fruit softening process. Knocking out the ABRE in the AcoACS1 enhancer inhibits endogenous hydrolase activities, thereby delaying pineapple softening.

Based on the results described above, a mechanism by which ABA regulates pineapple softening through the ABRE in AcoACS1 was proposed (Figure 10). First, ABA activates ABRE-binding factors (ABFs) (Nakashima and Yamaguchi-Shinozaki, 2013). ABFs then bind to the ABRE, enhancing AcoACS1 expression, which subsequently increases ethylene synthesis. Increased ethylene levels in pineapple fruit stimulate the respiration rate (Dos Santos et al., 2019). This accelerates the production of ATP and NADPH, providing energy for the synthesis and activation of hydrolases. Consequently, hydrolase levels and activities increase, accelerating the degradation of flesh cell walls and leading to fruit softening.