Non-climacteric fruit development and ripening regulation: ‘the phytohormones show’

Abstract Fruit ripening involves numerous physiological, structural, and metabolic changes that result in the formation of edible fruits. This process is controlled at different molecular levels, with essential roles for phytohormones, transcription factors, and epigenetic modifications. Fleshy fruits are classified as either climacteric or non-climacteric species. Climacteric fruits are characterized by a burst in respiration and ethylene production at the onset of ripening, while regulation of non-climacteric fruit ripening has been commonly attributed to abscisic acid (ABA). However, there is controversy as to whether mechanisms regulating fruit ripening are shared between non-climacteric species, and to what extent other hormones contribute alongside ABA. In this review, we summarize classic and recent studies on the accumulation profile and role of ABA and other important hormones in the regulation of non-climacteric fruit development and ripening, as well as their crosstalk, paying special attention to the two main non-climacteric plant models, strawberry and grape. We highlight both the common and different roles of these regulators in these two crops, and discuss the importance of the transcriptional and environmental regulation of fruit ripening, as well as the need to optimize genetic transformation methodologies to facilitate gene functional analyses.


Introduction
Fleshy fruit ripening is a complex and tightly regulated process that involves biochemical and physiological changes including: (i) alterations in color due to the disassembly of the photosynthetic system and the replacement of chlorophylls with anthocyanins and/or carotenoids; (ii) changes in texture as a result of cell wall degradation and loss of cell turgor; and (iii) an alteration of the aroma and flavor due to the accumulation of sugars, organic acids, and volatile compounds.These changes make the fruit more appealing for consumption, allowing dispersal of seeds once they are mature.Fleshy fruit species are traditionally classified as climacteric and non-climacteric based on the hormonal mechanisms that control the ripening process.Climacteric fruits, such as tomatoes, apples, or bananas, show a burst of respiration and a rapid increase in ethylene production at the onset of ripening, which acts as a key signal for the initiation and coordination of the process.On the other hand, in At white stage Changes in cell wall F. × ananassa cv.'Toyonoka'.Villarreal et al. (2016) non-climacteric fruits, such as strawberry, grape, sweet cherry, and lychee, among others, ethylene seems to play a minor role and does not necessarily initiate the fruit ripening.Instead, abscisic acid (ABA) is considered the key hormone regulating this process (Chai et al., 2011;Jia et al., 2011).However, other hormones, including brassinosteroids (BRs) and methyl   (Chervin et al., 2004;Symons et al., 2006;Chai et al., 2013;Villarreal et al., 2016;García-Pastor et al., 2019;Zuñiga et al., 2020).Although it will not be the focus of this review, there are various other important players and processes involved in the regulation of the non-climacteric ripening process besides hormones, including transcription factors (TFs) (Sánchez-Gómez et al., 2022), sucrose signaling (Jia et al., 2013a), DNA methylation (Shangguan et al., 2020;Martínez-Rivas et al., 2022), and changes in osmotic potential (Jia et al., 2020), among others, as well as external stimuli such as light and temperature (Abeysinghe et al., 2019;Twitchen et al., 2021).Two plant species have been used as the main experimental systems in the study of non-climacteric fruit ripening, namely strawberry [both the cultivated species Fragaria × ananassa (octoploid) and the woodland strawberry F. vesca (diploid)] and grape (Vitis vinifera), which are totally different in their fruit structure.Grapes develop ovary-derived berries, which are composed of seeds and three major tissues: skin or exocarp, pulp/flesh or mesocarp, and endocarp (Hardie et al., 1996).In contrast, strawberry fruit is an aggregate fruit (achenetum), where the fleshy part is originated from the enlargement of the floral receptacle, which is connected to the actual fruits, called achenes, located on its surface (Liu et al., 2020).For convenience, however, strawberries will be called 'fruits' throughout the text.
These two fruit crops are of great economic importance worldwide.Deciphering the molecular and environmental regulation of their fruit ripening is therefore a high priority to understand the biochemical changes that take place during this process and how they are regulated, and facilitate the control and improvement of fruit-related quality traits.Although there is no available mutant collection for either of these crops, gene functional analysis can be performed by either transient or stable transformation, although it is faster and easier in strawberry than in grapevine.The majority of studies in grape involve the application of different treatments or hormones at different stages of fruit development, followed by exhaustive sampling and subsequent transcriptomic analyses to investigate their effect on ripening (Zillioto et al., 2012;Chai et al., 2014;Cheng et al., 2015;Pilati et al., 2017), an approach that has also been commonly followed in strawberry (summarized in Table 1).Furthermore, transcriptomic and metabolomic analyses throughout strawberry and grape fruit development and ripening have been essential to decipher the molecular changes that occur during these processes, allowing the identification of related changes in gene expression or levels of metabolites (Kang et al., 2013;Härtl et al., 2017;Massonnet et al., 2017;Sánchez-Sevilla et al., 2017;Wang et al., 2017;Fasoli et al., 2018;Gu et al., 2019).Here we review classic and more recent studies that have shed light on the hormonal regulation of non-climacteric fruit, focusing mainly on these two model species.
Fruit development and ripening: stages and consequences.'The set list' Strawberry and grape development and ripening are classified into stages that differ between these two species.In the case of strawberry, there are four main stages defined by the color of the receptacle: green (which can also be subdivided into small, middle, and large green stages), white, turning, and red.Cell division and expansion occur in the first two stages, while the latter two are the actual ripening stages (Fig. 1A) (Symons et al., 2012).In the case of grape, fruit development and ripening can be separated into three stages: two sigmoidal phases separated by a lag phase (Fig. 1B).The first stage (phase I) comprises fruit set until the formation of green hard berries.During this time, rapid cellular division and enlargement take place.Then, a lag phase (phase II) is distinguished by a pause in berry growth.Then the last stage (phase III) starts with the socalled véraison, corresponding to the onset of ripening (Conde et al., 2007).

Auxins and gibberellins: 'the backstage crew'
Auxins are known to induce fruit growth in many different plant species, including climacteric fruits, such as tomato and apple (Serrani et al., 2008;Devoghalaere et al., 2012), and non-climacteric fruits, including both strawberries (Nitsch, 1950;Kang et al., 2013;Liao et al., 2018) and grapes (Godoy et al., 2021).In strawberries, auxins are synthesized primarily within the achenes and play a central role in the early stages of development.This was supported by classic experiments in F. × ananassa where removal of the achenes impaired receptacle development, which was restored upon exogenous application of the synthetic auxin, naphthoxy-3-acetic acid (NOA) (Nitsch, 1950).Furthermore, treatment of F. vesca open flowers with another synthetic auxin, naphthalene-1-acetic acid (NAA), promoted fruit growth in both width and length (Liao et al., 2018).An opposite role of auxin at later stages is also well established, since it inhibits non-climacteric ripening (Given et al., 1988;Davies et al., 1997).More recently, deachenization in fully developed strawberry fruits was found to promote ripening of the receptacle, but not when lanolin paste with indole-3-acetic acid (IAA), the most abundant natural auxin in strawberry, was applied, supporting the inhibitory role of auxin in this process (Gu et al., 2019).
Different patterns of IAA accumulation have been observed in strawberry fruits.A maximum in the level of IAA was found at the early green stages of entire F. × ananassa and F. vesca fruits, which declined as fruit growth and ripening progressed (Symons et al., 2012;Liao et al., 2018;Upadhyay et al., 2023).IAA patterns differed when achenes and receptacles of F. vesca fruits were analyzed separately.Using this approach, Gu et al. (2019) found much higher IAA levels in achenes than in receptacles, and that they remained constant throughout achene development and ripening.In contrast, IAA levels in receptacles were very low (Fig. 1A).
The most abundant natural auxin in grapes is also IAA, with an important role in enhancing a high rate of cell division and enlargement in the early phases of grape development.IAA has been detected in both flesh and seeds, increasing during the initial times of the vegetative phase, but gradually declining to very low levels during the middle of stage II, with a relatively low concentration at véraison (Fig. 1B) (Gouthu and Deluc, 2015).The reduced IAA levels are maintained in further stages due to an auxin inactivation mechanism, in which GH3 genes (GH3-1 and GH3-2), coding for IAA-amido synthetases, which conjugate IAA to amino acids, increase their expression in both the skin and flesh of grape fruits at the beginning of véraison (Böttcher et al., 2011;Ziliotto et al., 2012;Dal Santo et al., 2020).The detection of IAA-amino conjugates together with low levels of free IAA at the onset of maturation, and immediately afterwards, has also been described in strawberry (Archbold and Dennis, 1984;Upadhyay et al., 2023), and in other climacteric fruits, such as tomato (Böttcher et al., 2010), banana (Purgatto et al., 2002), and muskmelon (Dunlap et al., 1996), suggesting that this is a conserved ripening-related process in non-climacteric and climacteric crops.
The inhibitory role of auxins in ripening has also been widely demonstrated in grapevine.For example, treatment of grape berries cv.'Shiraz' with the synthetic auxin-like compound benzothiazole-2-oxyacetic acid (BTOA) delayed the beginning of ripening by 2 weeks and decreased anthocyanin accumulation due to the down-regulation of structural genes of the pathway, such as chalcone synthase (CHS) and UDPglucose:flavonoid 3-O-glucosyl transferase (UFGT), as well as a reduction in sugar accumulation, berry softening, and ABA levels (Davies et al., 1997).This delay has also been observed in other grapevine cultivars treated with NAA, such as 'Shiraz', 'Riesling', and 'Merlot', (Böttcher et al., 2011(Böttcher et al., , 2012;;Ziliotto et al., 2012).Interestingly, this alteration of the progress of ripening can be exploited to modify the quality of grape berries.Thus, significant differences were detected in the properties and volatile compounds of wine made from control and NAAtreated fruit, indicating that this kind of treatment might be used to manipulate grape and wine composition (Böttcher et al., 2012).
Like auxins, gibberellic acid (GA) promotes fruit growth in both climacteric and non-climacteric fruits (Serrani et al., 2008;Kang et al., 2013;Chai et al., 2014;Liao et al., 2018).In strawberry, however, GA has a different role from that of auxins during the first stages of development.Treatments with GA 3 of F. vesca fruits mainly promoted fruit elongation, but not widening as in the case of auxin, suggesting a more general role of the latter in receptacle development than that of GAs (Liao et al., 2018).This is also supported by the results of treatments in which the GA biosynthesis inhibitor paclobutrazol (PAC) was included, showing that auxin promotes fruit growth in both a GA-dependent and -independent manner (Liao et al., 2018).GA follows a trend of accumulation in entire F. × ananassa fruits that is similar to auxins, but is displaced slightly later in time, peaking at the large green stage (Fig. 1A).This is consistent with the role of auxin in promoting GA biosynthesis, not only in strawberries but also in other species (Serrani et al., 2008;Dorcey et al., 2009).Supporting the role of auxin promoting GA, fruits treated with NAA showed an increased expression of genes involved in GA synthesis, including several GA20ox and GA3ox genes, as well as a suppression of those with a negative role in GA signaling, such as GAI, RGA1, and RGL3 (Liao et al., 2018).
The connection between achenes and receptacles has been shown to be of great importance for fruit development based on transcriptome data.Thus, genes involved in auxin biosynthesis such as YUC5, YUC11, and TAR1, as well as GA biosynthesis genes, including GA20ox3, and GA3ox3, 4, 5, and 6 have been found to be mainly expressed in achenes but not in receptacles (Kang et al., 2013).However, the signaling components for auxin and GA were more highly and specifically expressed in receptacles, which supports a role for the achenes as the source of these phytohormones, and the receptacle as the responding tissue (Kang et al., 2013).
GAs also have an important role in the early phases of grape berry development, enhancing the rate of cell division and enlargement.Seeds are the main source of GAs (Pérez et al., 2000), so GA is commonly applied to increase berry size and cluster length, especially in seedless table grape and raisin production (Acheampong et al., 2017).This enlargement of the berries following GA treatment may be due to GA-dependent changes in the expression of genes involved in cell wall modification and the cytoskeleton (Chai et al., 2014).GAs have a similar accumulation trend to auxin during grape development, increasing at the initial developmental stages and declining throughout development and ripening (Conde et al., 2007;Wang et al., 2020).Consistent with this accumulation pattern, a transcriptome analysis during grape development showed that the GA biosynthetic genes VvGA20ox1-1, VvGA3ox4-1, and VvGA2ox1-1 and the GA receptor gene VvGID1B were highly expressed at the green stage, while declining in subsequent stages (Wang et al., 2020).Interestingly, the expression of genes coding for the GA repressor DELLA proteins also peaked at this stage.Since the GA biosynthetic genes are targets of DELLA proteins (Zentella et al., 2007), this co-expression pattern suggests a negative feedback to regulate the GA content during berry development (Wang et al., 2020).However, and in contrast to strawberries, GA levels show a slight increase at mid stages of ripening (Fig. 1B) (Pérez et al., 2000).Interestingly, GA 3 treatment of grapevines before flowering promotes an earlier berry coloration, as well as an increase in reducing sugars, total phenolic, tannin, and total anthocyanin contents (Moreno et al., 2011;Cheng et al., 2015;Xie et al., 2022).Together, these data may support a putative role for GAs in the last steps of the ripening process.Whether this accelerated ripening upon GA treatment is the consequence of a faster berry development or a direct effect on ripening-related processes is not clear.Therefore, further studies are needed to understand the role of the increase in GA levels at the late stages of ripening.
Of special interest is the fleshless berry (flb) mutation.This mutation impairs the pericarp development of the grape fruit, reducing its size at ripening due to an impaired division and differentiation of cells in the inner mesocarp, without any effect on fertility and seed size and number.Strikingly, treatments with GA, the synthetic cytokinin 6-benzyl-aminopurine (6-BAP), and IAA failed to reverse this phenotype, suggesting an impaired hormone signal reception or transduction in this mutant (Fernandez et al., 2006).However, genes encoding TFs, namely ATHB13, PISTILATTA, and YABBY2, were differentially expressed, suggesting that they perform an important role in early grape fruit development (Fernandez et al., 2007).
Brassinosteroids: 'a support artist' with differing roles in strawberry and grapevine ripening BRs are steroidal plant hormones that are essential for normal plant growth, related to cell division and elongation, vascular differentiation, reproductive development, and pathogen and abiotic tolerance (Nolan et al., 2020).It is known that exogenous BRs stimulate ripening in climacteric fruits such as tomato (Vardhini and Rao, 2002), mango (Zaharah et al., 2012), and persimmon (He et al., 2018), increasing the ethylene levels.The same positive effect has been described in grapes.Thus, the exogenous application of the most bioactive BR, 24-epibrassinolide (EBR) or the BR biosynthesis inhibitor brassinazole (BZ) before véraison significantly stimulated or delayed ripening by increasing or decreasing the concentration of total soluble solids (TSS), respectively (Symons et al., 2006).EBR treatment has been shown to promote grape berry ripening, modifying the expression of sugar transporters and invertases, as well as the activity of sugar metabolic enzymes, enhancing their TSS and reducing the sugar content (Xu et al., 2015).This treatment also reduced the titratable acid content in the juices and increased the average berry weight (Xi et al., 2013;Xu et al., 2015).Enhanced activities of enzymes involved in the biosynthesis of anthocyanins, such as phenylalanine ammonialyase (PAL) and UFGT, have also been reported upon exogenous EBR application, resulting in an increased accumulation of the total anthocyanins in the berry skin (Xi et al., 2013).Consistent with these effects of exogenous BR treatments, a rise in the levels of the endogenous bioactive BR castasterone (CS) and its direct precursor 6-deoxocastasterone has been detected at the onset of ripening, which coincides with a higher expression of BR biosynthesis (VvDWFl) and receptor (VvBRI1) genes (Symons et al., 2006).All these data support an important positive contribution of BRs at the onset of grape fruit ripening.
Changes in BR content during F. × ananassa fruit development have also been described; however, the function of this hormone in strawberry is still poorly explored.CS levels showed a peak in flowers followed by a dramatic drop during fruit development and ripening (Fig. 1A), suggesting a possible role during the early developmental stages but not during the ripening process (Symons et al., 2012).A similar pattern in which BRs are more abundant at earlier stages in deachened receptacles was found shortly afterwards by Chai et al. (2013).Nevertheless, this work showed that, as found in grapevine, EBR and BZ application to strawberries accelerated and delayed the fruit coloration, respectively.Furthermore, the down-regulation of the BR receptor gene FaBRI1 by virus-induced gene silencing (VIGS) resulted in a lack of fruit coloration, suggesting a positive role for BRs in promoting strawberry fruit ripening (Chai et al., 2013).Therefore, further studies need to be performed to better understand the role of BRs in strawberry fruit development and ripening.
Ethylene: 'the surprise act' While ethylene is the major hormone controlling fruit ripening in climacteric fruits, no such central role has been described in non-climacteric fruits, although several studies suggest that it also plays a positive role in their ripening.In the octoploid strawberry fruits, ethylene content increases at the green stage, followed by a reduction at the white stage and increasing again at the last stages of ripening, which, interestingly, is accompanied by an enhanced respiration rate, as occurs in climacteric fruits at the onset of ripening (Fig. 1A) (Iannetta et al., 2006).A similar increase from the turning to the ripe stage was also found for the ethylene precursor 1-aminocyclopropane-1-carboxylate (ACC) in F. × ananassa complete fruits (Upadhyay et al., 2023), and in receptacles but not in achenes of F. vesca fruits (Gu et al., 2019).This pattern is consistent with the expression of ACC synthase (ACS) and ACC oxidase (ACO) genes, which encode the enzymes involved in the last two steps of the ethylene biosynthesis pathway, respectively.Thus, the rise in ethylene content at early stages may be explained by the expression pattern of FaACS1 and FaACS2 in receptacles, and FaACS3, FaACS4, FaACO2, and FaACO3 in achenes (Merchante et al., 2013).The increase found in the ripening stages correlates with the up-regulation of FaACO1 in F. × ananassa and FveACO4b in F. vesca receptacles (Merchante et al., 2013;Gu et al., 2019).Recently, Upadhyay et al., (2023) found 17 ACS-encoding genes in F. × ananassa with a ripening-related expression pattern, suggesting an important role for these genes in the regulation of the ethylene content during ripening.
The application of ethylene to strawberry fruits has not resulted in clear and consistent effects on ripening.However, it has been widely reported that ethylene treatment affects the expression of cell wall-related genes, such as those encoding β-galactosidase, pectin methylesterase, pectate lyase, β-xylosidase, and endoglucanase (Trainotti et al., 2001;Castillejo et al., 2004;Bustamante et al., 2009;Villarreal et al., 2016).The resulting changes in cell wall composition increase susceptibility to pathogens such as Botrytis cinerea and Rhizopus stolonifera (Villarreal et al., 2016).A recent report has shown that ethylene treatment of F. chiloensis fruits at the large green developmental stage inhibited anthocyanin biosynthesis, but increased lignin content, which was consistent with the downregulation of anthocyanin-related genes (FcANS and FcUFGT) and the up-regulation of FcPOD27, involved in lignin biosynthesis (Figueroa et al., 2021a).In contrast, VIGS-mediated silencing of the product of the S-adenosyl-l-methionine (SAM) synthase-encoding gene FaSAMS1, which catalyzes the first committed step of the ethylene biosynthetic pathway, and the ethylene signaling gene FaCTR1 in F. × ananassa receptacles resulted in an inhibition of fruit coloring, which was partially rescued by the synthetic ethylene ethephon (Sun et al., 2013).Additionally, Merchante et al. (2013) generated stable transgenic lines overexpressing the Arabidopsis dominant negative allele of the ethylene receptor ETR (etr1-1).The insensitivity to ethylene resulted in an opposite trend in the expression of anthocyanin-related genes.Thus, FaPAL and FaCHS were slightly induced in receptacles, but significantly down-regulated in achenes, resulting in lighter achenes.These studies suggest that ethylene regulates fruit softening during ripening.However, further investigations are required to clarify the opposing effects of ethylene on anthocyanin accumulation that have been found to date, as well as to confirm the possible tissue-specific role of ethylene in anthocyanin biosynthesis.
In grape berries, only weak variations in endogenous ethylene levels occur around véraison (Fig. 1B) (Coombe and Hale, 1973;Chervin et al., 2004).Thus, a brief increase of endogenous ethylene has been detected at the onset of grape ripening, which coincided with higher activity and transcripts levels of VvACO1 (Chervin et al., 2004), and the signal transductionrelated genes Ethylene Response 2 (VvETR2) and Constitutive Triple Response 1 (VvCTR1) (Sun et al., 2010).When the ethylene receptors were blocked with 1-methylcyclopropene (1-MCP) around the ethylene peak, a delay in the increase of berry diameter, a higher acidity level, and a transient inhibition of anthocyanin accumulation in berry skins occurred (Chervin et al., 2004;Sun et al., 2010).Furthermore, if grapes were treated with low doses of ethylene gas for 24 h just before véraison, the fruit diameter increased.Even after 1 h of application, early responses to ethylene are detected as the up-regulation of cell wall-related genes and water exchange genes mainly in the berry skins, including expansin genes (EXP1/2), polygalacturonase (PG), cellulose synthase (CS), xyloglucan endotransglucosylase (XTH), and aquaporin genes (AQUA1-AQUA4) (Chervin et al., 2008).In contrast, treatment of berries at pre-véraison with high concentrations of either ethephon or the ethylene biosynthesis inhibitor AVG (aminoethoxyvinylglycine) results in a delayed or a faster beginning of ripening, respectively (Böttcher et al., 2013b), suggesting that ethylene dosage affects whether it has a positive or negative role in the regulation of ripening.In summary, all these studies confirm that ethylene does play a role in non-climacteric ripening, mainly by regulating the biosynthesis of phenolic compounds and the modification of cell wall structure, and therefore the fruit softening process.

ABA metabolism: 'the dancing queen headlines the show'
ABA is commonly considered the key hormone controlling fruit ripening in non-climacteric fruits.However, it has a negative effect on the initial developmental stages (Liao et al., 2018).
The ABA content in fruits is controlled through the balance between biosynthesis and catabolism (Bai et al., 2021;Figueroa et al., 2021b).The early steps of ABA biosynthesis take place in the plastid as part of the MEP and carotenoid pathway.The first committed step in ABA biosynthesis is catalyzed by the 9-cisepoxycarotenoid dioxygenase enzyme, encoded by NCED, which cleaves both 9ʹ-cis-violaxanthin and 9ʹ-cis-neoxanthin substrates to finally produce ABA (Fig. 2A).Two different ABA catabolism pathways have been described and are known to play a role in ripening regulation: (i) oxidative degradation and (ii) conjugation.The oxidative degradation of ABA is catalyzed by members of the cytochrome P450 monooxygenase (CYP707A) family.The predominant catabolic pathway involves the 8ʹ-hydroxylation of ABA, resulting in an unstable intermediate, 8ʹ-hydroxy-ABA.This is isomerized to phaseic acid (PA), which in turn is reduced to form dihydrophaseic acid (DPA) (Fig. 2A) (Rattanakon et al., 2016;Figueroa et al., 2021b).Oxidation of C-9ʹ can also occur, which is isomerized to neophaseic acid (neoPA) (Zhou et al., 2004).The second catabolic pathway is the ABA conjugation.This pathway is controlled by ABA-glucosyltransferase (UGTs), producing the main conjugate of ABA, the ABA-glucose ester (ABA-GE).ABA-GE is considered an inactive product and the form in which ABA is transported and stored in planta (Figueroa et al., 2021b).This pathway is reversible, as ABA-GE can be hydrolyzed to produce free and active ABA molecules through β-glucosidases (BGs) (Fig. 2A).
ABA biosynthesis has been described as taking place mainly in the receptacles of F. vesca, being much less present in achenes (Gu et al., 2019;Figueroa et al., 2021b).However, a recent work has shown that both achenes and receptacles have similar ABA levels in F. × ananassa fruits (Li et al., 2022).Nevertheless, this difference might be explained by the use of fresh instead of dry weight in Li's work.In any case, ABA content is consistently low at the green stages and continuously increases during strawberry development and ripening.
Three or four members of NCED and three or five CYP707A genes are encoded in both F. vesca and F. × ananassa genomes, depending on the study (Kang et al., 2013;Liao et al., 2018;Li et al., 2022).Furthermore, different nomenclatures for NCED genes has been followed, contributing to confusion.In 2011, Jia and collaborators showed that silencing the expression of FaNCED1 [FvH4_3g16730; named FveNCED3 in F. vesca (Liao et al., 2018)] by VIGS in F. × ananassa fruits resulted in a pronounced reduction in the ABA content and delayed fruit ripening (Jia et al., 2011).This phenotype was rescued when the fruits were treated with exogenous ABA, providing evidence for the first time of the promoting role of ABA in the ripening process.However, shortly after, it was shown that the expression of FaNCED2 [FvH4_3g05440; named FveNCED5 in F. vesca (Liao et al., 2018)] was much higher than that of FaNCED1 (Ji et al., 2012).This higher expression was confirmed in further transcriptome studies (Sánchez-Sevilla et al., 2017;Martín-Pizarro et al., 2021), suggesting a major role for this gene in ABA biosynthesis in strawberry fruits.
ABA biosynthesis in strawberry receptacles is known to be a tightly controlled process in which a crosstalk between ABA, auxin, and GAs in achenes and receptacles is essential.Expression data showed that, at early fruit developmental stages, auxin and GA induce the expression of FveCYP707A4a in receptacles, ensuring that the endogenous ABA content is kept low during the growth phase (Liao et al., 2018).Once the fruit has reached the precise size, auxin and GA content decline and the expression of FveCYP707A4a is reduced, releasing the activation of ABA biosynthesis, which in turn inhibits IAA biosynthesis (Liao et al., 2018;Li et al., 2022).Thus, when FveCYP707A4a was transiently silenced at early stages, the expression of NCED genes was found to be upregulated, and ABA levels increased in immature fruits (Liao et al., 2018).These fruits displayed lower fruit firmness than the control, which is explained by the high expression of genes responsible for cell wall disassembly, such as pectate lyase (FvePL) and endo-β-1,4-glucanase (FveCEL2).Furthermore, ABA treatment repressed FveCYP707A4a and promoted NCED gene expression, resulting in a regulatory feedforward loop to control ABA content in receptacles (Liao et al., 2018).
It has been recently suggested that the decline in auxin biosynthesis in achenes at the onset of ripening might be due to an autocatalytic ABA biosynthesis in this tissue.The application of ABA to detached achenes and receptacles increased ABA levels and reduced that of IAA in both tissues; however, in achenes, ABA biosynthesis was significantly promoted, supporting an ABA regulatory feedforward loop in the latter (Li et al., 2022).This study also suggested a more important role for ABA accumulation in achenes than in receptacles for strawberry fruit ripening regulation.Li et al. (2022) found ripening induction after ABA treatments via stalk feeding, rather than receptacle injection.Since ABA seems to be accumulated mainly in achenes or in receptacles with each method, respectively, they concluded that ABA in achenes plays a major role in ripening regulation.However, ABA treatment via injection in receptacles has been widely used with positive results (Table 1) (Jia et al., 2011;Li et al., 2019;Luo et al., 2020;Martín-Pizarro et al., 2021).Therefore, although the role of ABA in achenes seems to be important, the reasons for this different effect on ripening via ABA infiltration are not clear and might be explained by methodological differences.
Members of the ABA conjugation catabolic pathway regulating strawberry ripening have also been studied.In particular, two BGs have been characterized, FaBG3 (Li et al., 2013) and FaBG1 (Zhang et al., 2014).When their expression was silenced by RNAi, a reduction in the ABA content and an inhibition in the ripening process occurred, demonstrating the importance of ABA-GE hydrolysis, to release active ABA in the receptacle.However, differences in the levels of ABA-GE conjugate were found when comparing F. vesca accessions and one F. × ananassa cultivar.ABA-GE content decreased only in the octoploid accession, and increased in all the diploid accessions, suggesting differences in ABA catabolism that might be caused by the polyploid nature of F. × ananassa (Figueroa et al., 2021b).Furthermore, they found that FvUGT71A49 was able to glycosylate ABA more efficiently than other UGTs and that the overexpression of this gene in fruits produced a reduction in free ABA but not in ABA-GE content in receptacles.Similarly, ABA-GE levels did not change significantly when another UGT gene, FaUGT71W2, was down-regulated, which might be explained by compensation due to the redundancy of the ABA UGTs.Taken together, all these results are evidence that the ripening process in strawberries is controlled by endogenous ABA content, which at the same time is controlled by a tight balance between biosynthesis and catabolism.
ABA also plays a key role in the regulation of grape berry ripening (Pilati et al., 2017).As in strawberry, the ABA level rises before véraison, during ripening, but in contrast to strawberry, it declines during the last ripening stages (Fig. 1B) (Zhang et al., 2009;Zhang et al., 2013).Exogenous ABA application to different grape cultivars at véraison hastens the ripe state.This is the result of an increase in the expression of genes involved in the synthesis of anthocyanins, which color the skin berry rapidly after the treatment, as well as in the total phenolic content, and the concentration of flavonoids and antioxidants, while organic acids decrease and berry weight and volume do not change (Ban et al., 2003;Jeong et al., 2004;Peppi et al., 2006Peppi et al., , 2007;;Cantín et al., 2007;Giribaldi et al., 2010;Koyama et al., 2010;Wang et al., 2016).Nevertheless, treatment with exogenous ABA before véraison (at the hard green stage) results in long-term effects, since the ABA content of the treated berries remains higher not only immediately after its application, but also nearly 40 d later (Villalobos-González et al., 2016), modulating significantly several genes involved in ABA biosynthesis, perception, and signaling (Wheeler et al., 2009;Pilati et al., 2017).
As in strawberries, ABA content is controlled by the rate of biosynthesis and catabolism of this hormone in grape fruits.Three NCED isoforms in grapevine have been found (Young et al., 2012), which differ in their expression pattern and, as in strawberry, have been named differently depending on the study.However, there is a consensus that, in most grape cultivars, the VvNCED1 isoform is associated with the beginning of ripening.The expression levels of VvNCED1 (VIT_219s0093g00550; named VvNCED1 in Wheeler et al., 2009;Zhang et al., 2009;Sun et al., 2010;Zhang et al., 2013;Pilati et al., 2017;and VvNCED3 in Young et al., 2012;Wong et al., 2016) were low at the early developmental stages but increased at the onset of ripening, mainly in the seeds and skin, but also in the pulp, and declined in later ripening stages, except in the skin (Zhang et al., 2013), which is consistent with the timing and pattern of ABA accumulation.In contrast, VvNCED2 expression seems to be up-regulated at the later stages of véraison (Lund et al., 2008;Young et al., 2012), and also 44 h post-ABA treatment (Pilati et al., 2017), as well as expression of VvNCED4, which is also modulated during post-harvest withering (Wong et al., 2016;Pilati et al., 2017).Finally, the NCED isoform VvCCD4b appears to be induced during the whole berry ripening in all tissues, although slightly more in the pulp (Pilati et al., 2017).
a more important role than VvNCED1 in ABA content regulation at late developmental stages, since VvBG1 expression remained at high levels longer than that of VvNCED1 (Sun et al., 2010).In contrast, transcript levels of VvBG3 were high during early grape growth and decreased gradually throughout development, suggesting that it might influence ABA content in the early stages (Sun et al., 2010;Zhang et al., 2013).
ABA signaling: 'the dancing queen gets the crowd to sing along' The ABA signal transduction pathway is essential for this phytohormone to play its regulatory role.Two different signaling pathways exist, the first one constituted by three major components: PYR/PYL/RCAR (ABA receptors), clade A protein phosphatase 2C (PP2Cs, negative regulators), and sucrose-nonfermenting kinase 1-related protein kinase 2 (SnRK2s, positive regulators) (Fig. 2B).In the presence of ABA, the PYR/PYL/ RCAR cytosolic receptors change their conformation and bind to PP2C, inhibiting its activity and releasing SnRK2s.In turn, SnRK2 is phosphorylated, activating TFs of the ABAresponsive element-binding factor subgroup (AREB/ABF) to mediate transcription of ABA-responsive genes (Fig. 2B) (Wang et al., 2013).Various studies have been performed to understand the molecular mechanism underlying ABA perception and signaling (Chen et al., 2020).In strawberries, nine members of both FaPYR/PYL and FaPP2C gene families have been identified (Hou et al., 2021).Transient up-and downregulation of different members of these families have been performed to study their function.Firstly, down-regulation of the ABA receptor FaPYR in F. × ananassa promoted a loss in receptacle color, which could not be restored by exogenous ABA treatment (Chai et al., 2011).In contrast, silencing of the PP2C1-encoding gene ABSCISIC ACID INSENSITVE 1 (FaABI1) promoted fruit ripening, while its overexpression had the opposite effect (Jia et al., 2013b).These data support a positive and a negative role for FaPYR and FaABI1, respectively, in the regulation of strawberry fruit ripening.It has been recently found that FaABI1 also interacts with FaPYL2, which might be relevant in the control of ripening due to the ripening-induced expression pattern of FaPYL2 (Hou et al., 2021).Furthermore, FaABI1 interacts with the FERONIA/ FER-like receptor kinase FaMRKL47.Overexpression and RNAi-mediated silencing of FaMRKL47 delayed and accelerated strawberry fruit ripening, respectively, supporting a negative role for FaMRKL47 in ABA signaling (Jia et al., 2017).Another component of this pathway is FaSnRK2.6.RNAi silencing of FaSNRK2.6 significantly promoted ripening, while its overexpression arrested it, supporting a negative role in the regulation of this process (Han et al., 2015).A direct interaction was also found between FaSnRK2.6 and FaABI1, although it is hypothesized that the mechanism by which FaSnRK2.6 regulates ABA-mediated strawberry fruit ripening is at the transcriptional and not at the post-transcriptional level, since the ABA/FaPYR1/FaBI1 signaling cascade would be expected to activate rather than deactivate FaSnRK2.6.This hypothesis is supported by the decrease in the transcript levels of FaSNRK2.6 that has been observed during development and ripening, and after ABA treatment (Han et al., 2015).
A second pathway has been described in strawberry (Fig. 2C).A putative ABA receptor gene, FaCHLH/ABAR (also named FaABAR in Hou et al., 2018), was identified.Its downregulation by RNAi resulted in an uncolored fruit phenotype, despite containing a higher ABA content than control fruits.Moreover, exogenous ABA application could not restore the fruit color, supporting a blockage of the ABA signaling pathway and a feedback effect on the ABA content when this pathway is repressed (Jia et al., 2011).These fruits exhibited lower sugar content, probably due to the up-regulation of Sigma factor E (FaSigE), a regulator of sugar catabolism, and α-amylase (FaAMY) (Jia et al., 2011).Consistently, the RNAi silencing of FaSigE in F. × ananassa produced firmer fruits than the controls, with a reduction in soluble solids, anthocyanins, and ABA content, indicating that FaSigE is a positive regulator of fruit ripening (Zhang et al., 2017).Interestingly, FaABAR expression and ABA content were down-regulated in the FaSigE RNAi fruits, and the interaction between FaABAR and FaSigE was confirmed.All these data suggested that FaSigE, FaABAR, sugars, and ABA are involved in the pathway controlling fruit ripening (Zhang et al., 2017).Another member of this pathway has been found with the identification of the protein interaction between FaABAR and Red-Initial Protein Kinase 1 (FaRIPK1) (Hou et al., 2018).FaRIPK1 function was analyzed by transient down-regulation in F. × ananassa, which resulted in an inhibition of strawberry fruit ripening and a reduced expression of genes involved in softening, sugar content, anthocyanin biosynthesis, and ABA biosynthesis and signaling, including FaNCED1 and the TF-encoding FaABI4 gene.Thus, in this model, ABA binds to the complex constituted by FaABAR and FaRIPK1, promoting fruit ripening through the modulation of FaABI4 gene expression (Hou et al., 2018).Whether FaSigE and FaRIPK1 are part of the same protein complex with FaABAR is yet to be elucidated.
In grapevine, nine ABA receptors of the VvPYL/PYR/ RCAR family, 85 VvPP2Cs, and eight SnRK2 kinases have been described, with organ-specific expression patterns (Boneh et al., 2012a, b;Liu et al., 2016;Zhang et al., 2021).An interesting correlation between gene expression and the interaction of ABA receptors with PP2Cs has been described (Gambetta et al., 2010;Boneh et al., 2012b;Pilati et al., 2017).In berry fruits, exogenously applied ABA at the hard green pre-véraison stages down-regulated the expression of VvRCAR5 (VIT_208s0058g00470) and VvRCAR7 (VIT_202s0012g01270) (named VlPYL3 and VlPYL1, respectively, in Li et al., 2011) (Pilati et al., 2017).VvRCAR7/ VlPYL1 is mainly expressed in leaves and fruit tissues.It increases during fruit development and declines before ripening, correlating with ABA levels (Gao et al., 2018).Furthermore, transient overexpression of VvRCAR7/VlPYL1 in grape berries enhanced color development and anthocyanin accumulation in the skin, supporting a positive role in this biosynthetic pathway (Gao et al., 2018).The expression pattern of the negative regulators, the VvPP2C genes, showed that VvPP2C3 and VvPP2C6 were up-regulated during the ripening stages, suggesting a role for these two genes in ABA signaling during grape berry ripening (Gambetta et al., 2010;Pilati et al., 2017).Moreover, the protein kinase VvSnRK2.1 is considered a candidate to be part of this ABA signaling pathway in grape berries, since its transcript levels increase after ABA exogenous application (Pilati et al., 2017).Among the downstream AREB/ABFs TFs, only two VvABFs have been identified in grapevine: VvbZIP08 and VvbZIP45/ ABF2 (Liu et al., 2014).In particular, the transcript levels of VvAREB/ZIP45/ABF2 increased from véraison until the end of the ripening phase, mainly in both seeds and skin, and after exogenous ABA treatment (Nicolas et al., 2014;Wong et al., 2016;Pilati et al., 2017).Moreover, the overexpression of VvAREB/ZIP45/ABF2 strongly increased the accumulation of stilbenes and activated processes related to fruit softening and ripening, indicating that this TF is the main phosphorylated intermediary between SnRK2 and downstream genes of the ABA signaling cascade such as VvNAC17 and Armadillolike, expression of which was also induced by ABA treatment and by VvAREB/ZIP45/ABF2 in trans-activation assays in tobacco leaves (Nicolas et al., 2014;Pilati et al., 2017).Finally, VviNAC060 has been identified as a positive regulator of grape ripening, as it regulates the ABA transduction pathway and promotes chlorophyll degradation and anthocyanin accumulation (D'Incà et al. 2023).
A crosstalk between ABA and auxins, GAs, and ethylene has been described in grapevine, where auxin and GA 3 application caused a decrease in the cellular ABA levels by inhibiting the synthesis via down-regulation of VvNCED genes (Zilioto et al., 2012;Chai et al., 2014).Furthermore, auxin and GA 3 application also modulated ABA signaling by decreasing the transcript levels of the VvPP2C genes and the TF-encoding VvABF gene (Zilioto et al., 2012;Chai et al., 2014).In contrast, ethylene and ABA are part of a feedforward loop promoting the expression of VvNCED1 and VvACO1, respectively (Sun et al., 2010).
The positive role of ABA has also been described in the regulation of ripening of other non-climacteric fruits, such as lychee, in which ABA treatment promotes anthocyanin accumulation mediated by several TFs such as LcABFs, LcMYB1, LcbHLH1, and LcNAC002 (Singh et al., 2014;Hu et al., 2019;Zou et al., 2023).The same role has also been identified in sweet cherry, in which ABA signaling is regulated by DOF TFs, such as PavDof2, and PavDof6/15, with opposite roles in the regulation of cell wall-related and ABA biosynthetic genes (Zhai et al., 2022), and PaMADS7 or PavNAC56, which promote fruit softening (Qi et al., 2020(Qi et al., , 2022)).The importance of ABA regulating fruit ripening and ABA-related TFs has also been reported in Citrus reticulata and Citrus sinensis (Zhu et al., 2020).In summary, all these data together confirm that, despite the different fruit types, ABA is a common and important regulator in the promotion of ripening in non-climacteric fruits.

Conclusions
We have presented a comprehensive review on the hormonal regulation of non-climacteric fruit ripening, focusing on the two main model species, strawberry and grapevine.We have highlighted the importance of auxin and GAs in the early stages of fruit development and their role in inhibiting ripening.Furthermore, the data discussed here are evidence that ABA, our 'dancing queen', is the main hormone positively regulating non-climacteric fruit ripening.Recently, an interesting autocatalytic biosynthesis of ABA has been described in strawberry, in which achenes seem to play a much greater role than has been previously considered (Li et al., 2022).However, the differences found in some reports concerning ABA levels in achenes and receptacles, as well as in the ripening effect after ABA infiltration, need to be reconsidered before drawing further conclusions.In any case, the major role of ABA is undeniable.However, ABA should not 'take all the credit for the show' since other hormones such as ethylene and BRs also contribute to reaching the final ripe stage in non-climacteric fruits.
Here we have discussed the similarities and differences between the hormone accumulation pattern in strawberries and grapes.We have also described the effects of these hormones on fruit development and ripening, as well as the main genes involved in both biosynthesis and signaling.All these data highlight the different nature of strawberries and grape berries.Strawberry has become the main model for non-climacteric fruits, despite it developing false fruits derived from vegetative tissue in contrast to the ovary-derived grape berry.It is important to take this into account, and results obtained in strawberry fruit receptacles should not be directly extrapolated to other non-climacteric species with ovary-derived fruits.Furthermore, ethylene increases and a respiratory burst even occurs in the late stages of strawberry ripening (Iannetta et al., 2006), which should invite the scientific community to reconsider the classification of fruit ripening into more categories than just climacteric and non-climacteric.
Although not the focus of this review, it is worth mentioning that the role of TFs and environmental factors in the regulation of fruit development and ripening is essential.Several key TFs that are regulated by ABA and/or regulate ABA biosynthesis and signaling have been identified so far in non-climacteric fruits, such as the NAC TFs FaRIF and VviNAC060 (Martín-Pizarro et al., 2021;D'Incà et al., 2023), or FaGAMYB and VvAREB/ZIP45/ABF2 (Nicolas et al., 2014;Vallarino et al., 2015) in strawberry and grapevine, respectively, among others.Therefore, there is a chicken-and-egg situation here: what is the most upstream molecular signal to trigger fruit ripening?Is it actually an epigenetic modification instead?This open question is still unresolved and will be complicated to answer.
Besides the key role of ABA in promoting non-climacteric fruit ripening, this phytohormone is essential to deal with abiotic stresses.Therefore, it is not unexpected that salt and drought stresses induce the expression of ABA biosynthetic genes, and therefore a rise in ABA accumulation and, subsequently, in the anthocyanin content in F. × ananassa fruits (Galli et al., 2016;Perin et al., 2019).In grapes, water deficit alters the volatile composition of the berries (Buesa et al., 2021), and also promotes ripening, increasing the anthocyanin levels as the result of an up-regulation of genes of the anthocyanin biosynthetic pathway (Castellarin et al., 2007).Temperature is another environmental factor influencing fruit ripening.In strawberry, high temperature dramatically accelerates fruit ripening, which seems to be mediated by the ABA negative regulator FaSnRK2.6 (Han et al., 2015).In contrast, high temperatures negatively affect the maturation process in grapes by accelerating anthocyanin degradation and inhibiting anthocyanin biosynthesis in berry skin (Mori et al., 2005a(Mori et al., , b, 2007)), which is consistent with the reduction found in ABA levels at high temperatures (Ryu et al., 2020).Thus, it is clearly important to study further the molecular mechanisms underlying the response to the abiotic stresses associated with the current context of climate change, providing gene targets which when modulated will allow optimization of fruit set and ripening and improve fruit quality traits.However, this requires the development of efficient transformation protocols.
Strawberry has a great advantage over other non-climacteric species, since it allows gene functional analysis to be performed in a rather simple way by either transient transformation of the receptacles or stable plant transformation.Unfortunately, this is not easy to achieve, or is more time-consuming, in other non-climacteric species, as in the case of grapevine, and tree species such as citrus, which are recalcitrant for generation of stable transgenic plants.Nevertheless, stable transformation has been achieved in grapevine (Rinaldo et al., 2015;D'Incà et al., 2023), and new strategies are being developed in order to optimize the regeneration problems in this species (Campos et al., 2021).Importantly, although it is not a straightforward methodology and requires proper controls, transient transformation by agroinfiltration can be directly performed in the strawberry receptacles and grape berries.This facilitates studying the role of genes of interest for fruit ripening in situ and is much faster than obtaining stable transgenic plants (Carvalho et al., 2016;Gao et al., 2018;Zhao et al., 2019).In spite of the advantages of transient transformations, we believe that the development of efficient stable transformation methodologies will be essential to gain further and more accurate knowledge on the genetic mechanisms underlying non-climacteric fruit ripening, allowing optimization of the fruit quality traits acquired during this process, as well the resistance to pathogens, and thus obtaining more nutritious, flavorful, and longer lasting fruits in the near future for all these important crops.

Fig. 1 .
Fig. 1.Schematic model of strawberry (A) and grape berry (B) developmental and ripening stages and the changes in hormone content from green to ripe fruit.Dotted areas represent hormone levels at strawberry achenes (A) and grape berry seeds (B).Filled areas represent receptacle/pulp or entire fruits.

Table 1 .
Treatments with hormonal and inhibitor compounds in strawberry and grape fruits.

Table 1 .
Continuedjasmonate, have been shown to play important roles during ripening regulation in non-climacteric fruits