The promoter sequences (2000-bp upstream of the translation start codon of ERFs) were obtained from the EnsemblPlants database (https://plants.ensembl.org/). All cis-acting elements were assessed by PlantCARE (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed on 1 June 2022) (Lescot et al., 2002).

Cherry tomato (S. lycopersicum L. “Betatini”) fruits at mature-green stages were harvested during summer in Jungyeum, South Korea. Disease-free and intact fruits were sterilized with 2% sodium hypochlorite solution and washed with tap water twice. Following air drying at room temperature and removal of the pedicels, the fruits were divided into four groups and then treated with ET, BR, AUX, or GA. For ET and AUX treatments, the fruits were dipped in 1 mM ethephon solution (Inbio Corp., Jecheon, South Korea) and 0.45 mM 2,4-dichlorophenoxyacetic acid (Sigma-Aldrich, St. Louis, MO, USA), respectively, under a vacuum at 30 kPa for 5 min. For the BR treatment, the fruits were immersed in 6 μM brassinolide solution (Cayman Chemical Company, MI, USA) for 15 min. For the GA treatment, the fruits were dipped in a 0.5 mM GA3 solution (prepared in ethanol/distilled water [1:1000, v/v] containing 0.1% [v/v] Tween-20; Sigma-Aldrich) for 15 min. The fruit dipped with distilled water for 15 min was used as the control. Following the treatments, the fruits were kept in the dark at 20 ± 2°C (room temperature) with 90 ± 5% relativity humidity (RH) for 10 d or 4°C (cold storage) for 14 d followed by 2 d at 20 ± 2°C (retailer conditions).

Fifteen fruits were sampled per treatment to assess the fruit color. Skin color was monitored using a color difference meter (CR-400; Konica Minolta, Japan) and was reported based on Hunter’s scale: light (L*), red (a*), and yellow (b*).

qRT-PCR was performed on a CFX96 TouchTM Real-Time PCR detection system (Bio-Rad Laboratories, Hercules, CA, USA) as described by Park et al. (2018). The transcripts were amplified using the iQTM SYBR Green Supermix (Bio-Rad Laboratories) with specific primers ( Table S1 ). qRT-PCR was performed under the following conditions: 95°C for 30 s, followed by 40 cycles of 95°C for 10 s and 55–58°C for 40 s. The relative gene expression level was calculated using the 2^-ΔΔCt method (Schmittgen and Livak, 2008) and normalized using the expression levels of the housekeeping gene actin (solyc11g005330). qRT-PCR analysis was performed using at least three biological replicates and two technical replicates.

Values are presented as the mean ± standard error. Samples were subjected to analysis of variance, and significant differences were determined using Duncan’s multiple range test. All analyses were conducted using SAS v.9.2 (SAS Institute, Cary, NC, USA). The ERF expression data and fruit color (a* values) were normalized, scaled, and used for the hierarchical clustering analysis and pattern correlation analysis (Pearson correlation coefficient) in the MetaboAnalyst 3.0 software (www.metaboanalyst.ca).

To explore the hormonal regulation of ERFs, we first analyzed the 2000 bp region upstream of the translation start codon of the 18 genes and identified the cis-elements related to ET, AUX, and GA ( Figure 1 ). All the ERFs contained ET-related cis-elements. Additionally, all the genes contained cis-elements related to more than one hormone ( Figure 1 ). Then, tomatoes at the mature-green stage were treated with ET, BR, AUX, or GA to determine the hormonal effect on ripening progression and the expression of ERFs during storage at room temperature for 10 d and/or cold storage for 14 d followed by 2 d at retailer conditions. The effect of hormones on fruit color was observed on day 3 at room temperature ( Figure 2A ). ET and BR treatments accelerated fruit reddening, with consistently higher a* (redness, Hunter scale) values than those of the control for 3 d at room temperature ( Figure 2B ). The AUX- and GA-treated fruits showed delayed color transition, as evidenced by consistently lower a* values than those of the control, ET-, and BR-treated fruits throughout storage at room temperature ( Figures 2A, B ). During cold storage, ripening was delayed in all treatment groups, and a visible color break was observed on day 14. ET-treated fruits showed the highest a* values, while GA-treated fruits showed the lowest values. BR- and AUX-treated fruits had values lower than those of the control fruits ( Figures 2C, D ). Under retailer conditions, the ripening process was accelerated in all treatment groups. However, ET- and GA-treated fruits still had the highest and lowest a* values, respectively. AUX-treated fruits had a* values in between those of control and GA-treated fruits, while BR-treated fruits did not significantly differ from control fruits ( Figures 2C, D ). Additionally, fruits treated with AUX and GA tended to have higher L* values than those of ET and BR-treated fruits on days 3–7 at room temperature and under retailer conditions ( Figures S1A, C ). However, there was no specific trend in the changes of the b* value between control and hormone-treated fruits at both storage conditions ( Figures S1B, D ).

On day 1 at room temperature, ET treatment led to the expression of ERF.B2, B6, E2, and F1, while BR treatment increased the expression of ERF.E5, F2, and F3. However, AUX treatment reduced the expression of both ET- and BR-responsive ERFs, including ERF.C1, which was induced by all hormone treatments ( Figure 3 ). BR was found to inhibit ET-responsive ERFs, except ERF.E2. GA treatment had no significant effect on most ET- and BR-responsive genes, except that it suppressed the ET-responsive ERF.B2. The expression of ERF.B3 was enhanced by treatment with all hormones except AUX, and ERF.F4, F5, and D2 were specifically reduced by AUX and BR treatments. Furthermore, ERF.E4 was downregulated upon treatment with all hormones, except ET ( Figure 3 ).

On day 3 at room temperature, ET treatment induced the expression of ERF.B1, B6, F4, and F5 up to 2.5-fold, and these ERFs were downregulated by both AUX and BR treatments. ET-responsive ERFs were not affected by GA treatment, except ERF.F5, which was induced by GA treatment. Moreover, ERF.C1 transcription was exclusively increased in GA-treated fruits. The level of ERF.B2 was reduced in all fruits except those treated with ET. Hormone treatments had no significant effect on ERF.B1 (day 1), ERF.E2, F1, E5, F2, F3, B3, D2, and E4 (day 3), and ERF.A3, E1, F6, and H7 (days 1 and 3) ( Figure 3 ).

On the first day of cold storage, ET treatment increased the expression of ERF. C1, D2, E1, F4, and H7, while GA treatment also increased the expression of these genes except ERF.D2. AUX treatment decreased the expression of ERF.D2 and C1 and moderately increased the expression of ERF.E1, F4, and H7. BR treatment had varied effects, as it could trigger ERF.F4 while decreasing ERF.C1 and H7. The expression of ERF.F1 was found to be most responsive to BR treatment but was reduced in other hormone treatment groups. ERF.B3, E5, and F5 showed the highest level of responsiveness to AUX treatment, while ERF.E5 was exclusively expressed in AUX-treated fruits. The transcript levels of ERF.B3 and F5 were considerably decreased in other hormone-treated fruits. Multiple hormones led to a decrease in ERF.A3, B6, and E4. However, on the first day of cold storage, the hormone treatments did not have a significant effect on the expression of ERF.B1, B2, E2, F2, F3, and F6 ( Figure 4 ).

On day 14 at cold storage, the hormone treatments had a significant impact on six ERFs (ERF.B2, D2, E2, E4, E5, and F3). ERF.D2 and B2 were most responsive to ET and BR treatments, respectively. BR treatment was found to suppress ERF.D2, while ET treatment had no effect on ERF.B2. In contrast, AUX and GA treatments could reduce the expression of ERF.B2 but not D2. The response of ERF.E4 and E5 to AUX and GA treatments was the opposite, while ET and BR treatments had little effect on these genes. Additionally, ERF.E2 and F3 expression was found to be suppressed in AUX- and GA-treated fruits. ERF. A3, B1, B3, B6, C1, E1, F4, F1, F2, F5, F6, and H7 were not significantly affected by hormone treatments after 14 d ( Figure 4 ).

Two days after transferring the fruits from cold storage to retailer conditions, twelve ERF genes were found to be significantly affected by hormone treatments. Interestingly, all these genes were most responsive to either ET or BR treatments. Specifically, ERF.B3, E2, E4, E5, and F2 were highly responsive to ET treatment, with ERF.B3 and E2 were exclusively activated in ET-treated fruits. However, the activity of ERF.E5 and F2 could be reduced by GA treatment, while BR and AUX treatments showed moderate activation. BR treatment resulted in an increased expression of ERF.B1, B2, B6, F1, F4, F5, and E1. Notably, ET treatment was found to decrease the expression levels of some of the BR-responsive genes (ERF.B1, B2, and B6). AUX treatment had an inductive effect on some of the ET-responsive (ERF.F2) and BR-responsive (ERF.B2, B6, F1, and F4) genes, while its suppressive effect was limited to ERF.E1. Meanwhile, GA treatment was found to reduce the expression of ET-responsive (ERF.E4 and F2) and BR-responsive genes (ERF.B2 and E1). However, GA treatment increased the expression of the ET-responsive gene ERF.E4 and BR-responsive genes ERF.F1, F4, and F5. The expression of ERF.A3, F3, C1, D2, F6, and H7 were not significantly affected by treatment with any of the hormones ( Figure 4 ).

To investigate the correlation between fruit color and ERF expression in hormone-treated tomatoes, hierarchical clustering and pattern correlation analysis were performed by combining the gene expression data with a* values. The heat map of hierarchical clustering analysis showed that the profiles of all treatment groups, except that of AUX-treated fruits, were closely correlated on days 1 and 3 at room temperature ( Figure 5A ). The a* values were grouped with ERF.F3, F6, E5, F2, B3, E1, and A3, and these genes, except ERF.F6, E1, and A3, were most responsive to BR treatment ( Figure 5A ). In cold-stored fruits, a distinct separation was observed between the profiles of ET treatment and those of other treatments. ERF.B1, F4, F6, F2, F3, and B2 were clustered with a* values, and their profiles in ET-treated fruits were significantly different from those in fruits treated with other hormones ( Figure 5B ). ERF.E5, F5, B3, E1, and H7 were closely clustered, and their expression was induced by AUX ( Figure 5B ). When fruits were transferred from cold storage to retailer conditions, a distinct separation was observed between the profiles of ET- and BR-treated fruits and other hormone-treated fruits ( Figure 5C ). The a* values were found to be correlated with several ERFs, including ERF.E1, F3, H7, D2, E4, and F2, which were most responsive to ET and/or BR ( Figure 5C ).

Pattern analysis between a* values and gene expression revealed that certain ERFs (ERF.B2, B6, and E4) were positively associated with the a* values in fruits stored at room temperature, whereas other ERFs (ERF.F5 and E2) were negatively associated ( Figure 6A ). When stored in cold temperatures, a* values were positively correlated with ERF.A3, B2, B6, E4, and F4, but negatively correlated with ERF. B3, E5, and F1 ( Figure 6B ). Under retailer conditions, most ERFs were positively associated with color development, which suggests that ripening accelerated and quality changes occurred after cold storage regardless of hormone treatment ( Figure 6C ).

The effect of ET, BR, AUX, and GA treatment on the ripening process and ERF expression was evaluated during storage at room temperature, cold temperature, and retailer conditions. The ripening process was characterized by measuring fruit color. At room temperature, the ripening process of the fruits was influenced by hormone treatment, being accelerated by ET and BR and delayed by AUX and GA. This was reflected in the fruit color as shown in Figure 2A and was consistent with previous research (Dostal and Leopold, 1967; Li J. et al., 2017; Malka and Park, 2022). While the a* values of control and ET- and BR-treated fruits were not significantly different after 7 d at room temperature, AUX- and GA-treated fruits consistently showed lower values throughout the storage period ( Figure 2B ). Of note, low storage temperatures inhibit fruit ripening (Vincent et al., 2020). Thus, during cold storage, the ripening process of fruits in all treatment groups was delayed ( Figure 2C ). This was evidenced by a delayed color break on day 14, with the highest and lowest a* values observed in ET- and GA-treated fruits, respectively ( Figure 2D ). When the fruits were transferred from cold storage to retailer conditions, the ripening process was accelerated in all treatment groups. Nonetheless, the effects of the hormone treatments remained largely consistent with those observed at room temperature ( Figures 2C, D ).

The fruit quality of tomatoes rapidly changed at room temperature. ET is considered to be the primary regulator of fruit ripening, with BRs and AUX acting as modulators of ET signaling and biosynthesis (Kumar et al., 2014). In this study, ET treatment induced the expression of ERF.B1, B2, B6, E2, and F1 in fruits stored at room temperature. Meanwhile, the transcript levels of ERF.E5, F2, and F3 were increased by BR ( Figure 3 ). Strikingly, the ERFs that were responsive to ET and BR treatment were suppressed in AUX-treated tomatoes concomitant with its inhibitory effect on ripening ( Figure 3 ). The interaction between ET and AUX is well-established in various physiological processes of plant growth and development. In tomato fruit ripening, ERF.B3 and D7 play a critical role in integrating ET and AUX signaling (Liu et al., 2014; Gambhir et al., 2022). Furthermore, some of the ET-responsive ERFs were suppressed in BR-treated fruits ( Figure 3 ), suggesting that they may be involved in a feedback regulation between ET and BR signaling. Additionally, ET-responsive ERF.B2 and B6, along with AUX-suppressive ERF.E4, were positively correlated with color development ( Figure 6A ). ERF.E4 has been shown to participate in ripening and carotenoid accumulation by integrating both ET-dependent and ET-independent regulatory activities, thereby enabling precise signal output modulation (Lee et al., 2012). These results suggest that the ET- and BR-responsive ERFs identified in this study may play crucial roles in the interplay between ET, BR, and AUX during the transition to ripening in tomato at normal temperature.

Low temperatures can significantly affect ET biosynthesis and signaling pathways in tomato fruit, which can result in delayed ripening (Kumar et al., 2014). During cold storage, the ERFs that were most responsive to ET and BR at room temperature remained largely unaffected by these hormones, potentially contributing to the observed ripening inhibition ( Figure 4 ). While several ERFs responded specifically to ET during cold storage, they can also be co-regulated by other hormones, and the specific effect of each hormone may vary depending on the ERF ( Figure 4 ). This suggests that these hormones may interact with the ET signaling pathway to regulate cold acclimation at low temperature.

Several studies highlighted the role of ERFs in the regulation of cold stress. For instance, ERF105 has been found to be critical in freezing tolerance by operating in conjunction with CBF-regulon in Arabidopsis (Bolt et al., 2017). ERF108 and ERF9 from Poncirus trifoliata have been found to positively regulate cold tolerance by activating the raffinose biosynthesis gene and glutathione S-transferase gene, respectively (Khan et al., 2021; Zhang et al., 2022). Additionally, genes most responsive to BR (ERF.B2) and AUX (ERF.B3 and F5) were reported to be involved in stress regulation (Tournier et al., 2003; Chen et al., 2008; Pan et al., 2012; Klay et al., 2014). Interestingly, the expression of AUX-responsive ERF.B3 and E5 were also negatively correlated with a* values ( Figure 6C ). These results suggest that these ERFs are involved in complex hormone interactions that govern cold response and ripening in cold storage.

When fruits were transferred from cold storage to retailer conditions, the majority of these genes were activated by ET or BR ( Figure 4 ). For example, ERF.B3, E5, and F5 regulated by AUX on day 1 at cold storage were upregulated in ET- or BR-treated fruits at retailer conditions ( Figure 4 ). ERF.B3 responds to both ET and AUX, mediating salt and cold stress response in addition to its role in the regulation of ripening in tomato (Klay et al., 2014; Liu et al., 2014). Moreover, ERF.E2 was found to be responsive to ET at both room temperature and retailer conditions but was downregulated by all the hormones during cold storage ( Figures 3 , 4 ). ERF.E2 integrates multiple signaling pathways that are responsive to biotic and abiotic stresses, including ET (Zhang et al., 2004). This suggests that the ERFs responsive to ET or BR have a dual function: cold response at low temperatures and ripening progression at retailer conditions. They may also be involved in the control of chilling injury. ERF.E1, which was most responsive to ET on day 1 during cold storage, was induced by BR at retailer conditions. ERF.E1 has been found to be responsive to chilling injury in tomato (Bai et al., 2021).

Interestingly, certain ERFs most responsive to ET at room temperature were instead regulated by BR at retailer conditions; similarly, ERFs regulated by BR at room temperature were found to be most responsive to ET at retailer conditions ( Figures 3 , 4 ). This indicates that although both room temperature and retailer conditions favored ripening, the hormonal regulation of ERFs was specific and depended on the storage conditions. However, ERF.E2 was regulated by ET at both room temperature and retailer conditions ( Figures 3 , 4 ). Furthermore, ERF.F5 was negatively correlated with a* values at both room temperature and retailer conditions ( Figures 6A, C ), suggesting that these genes may play a significant role in the ripening and color development of tomatoes. Moreover, the inhibitory effect of AUX on ET- and BR-regulated ERFs, which was observed at room temperature, was largely absent under retailer conditions ( Figure 4 ). However, GA had the ability to inhibit particular ERFs that were responsive to ET (ERF.E5 and F2) and BR (ERF.B2 and E1) ( Figure 4 ). Furthermore, GA could co-regulate some ET- or BR-responsive genes at both room temperature and retailer conditions, suggesting complex interactions between these hormones in the harvested tomatoes.

Based on these results, we propose a model in which ERFs responsive to ET (ERF.B2, B6, E2, and F1) and BR (ERF.E5, F2, and F3) regulate fruit transition to ripening and its associated changes in harvested tomatoes at room temperature ( Figure 7 ). ERFs, such as B3, E1, and F5, participate in intricate hormone interactions contributing to both cold acclimation and ripening inhibition in cold storage. Upon transfer of fruits from cold storage to retailer conditions, ET and BR may induce several ERFs to recover from chilling injuries and progress in ripening. AUX and GA can fine-tune the transition to ripening in tomatoes at room temperature and after being transferred from cold storage to retailer conditions, respectively ( Figure 7 ).

In summary, we identified ERFs responsive to specific hormones at different storage temperatures. ERFs responsive to ET and BR were inhibited by AUX at room temperature and by GA at retailer conditions. Some ERFs may be involved in cold response at low temperatures and ripening progression and stress recovery after transfer from cold storage to retailer conditions. Overall, this study provides valuable insights into the complex hormonal regulation of ERFs associated with tomato postharvest ripening. These findings will be useful for further understanding dynamic hormonal interactions and for developing strategies to improve fruit quality and shelf life through regulation of hormone signaling pathways and ERF expression.