transcriptional activity of the mads boxarlequin … · per family that regulates cutin...

Transcriptional Activity of the MADS Box ARLEQUIN/TOMATO AGAMOUS-LIKE1Gene Is Required for CuticleDevelopment of Tomato Fruit1

Estela Giménez, Eva Dominguez, Benito Pineda, Antonio Heredia, Vicente Moreno,Rafael Lozano*, and Trinidad Angosto

Centro de Investigación en Biotecnología Agroalimentaria, Universidad de Almería, 04120 Almeria, Spain(E.G., R.L., T.A.); Instituto de Hortofruticultura Subtropical y Mediterránea La Mayora, Universidad deMálaga-Consejo Superior de Investigaciones Científicas, 29750 Algarrobo-Costa, Malaga, Spain (E.D., A.H.);and Instituto de Biología Molecular y Celular de Plantas, Universidad Politécnica de Valencia-Consejo Superiorde Investigaciones Científicas, 46022 Valencia, Spain (B.P., V.M.)

ORCID ID: 0000-0001-5458-2075 (R.L.).

Fruit development and ripening entail key biological and agronomic events, which ensure the appropriate formation and dispersalof seeds and determine productivity and yield quality traits. The MADS box gene ARLEQUIN/TOMATO AGAMOUS-LIKE1(hereafter referred to as TAGL1) was reported as a key regulator of tomato (Solanum lycopersicum) reproductive development,mainly involved in flower development, early fruit development, and ripening. It is shown here that silencing of the TAGL1 gene(RNA interference lines) promotes significant changes affecting cuticle development, mainly a reduction of thickness and stiffness,as well as a significant decrease in the content of cuticle components (cutin, waxes, polysaccharides, and phenolic compounds).Accordingly, overexpression of TAGL1 significantly increased the amount of cuticle and most of its components while rendering amechanically weak cuticle. Expression of the genes involved in cuticle biosynthesis agreed with the biochemical and biomechanicalfeatures of cuticles isolated from transgenic fruits; it also indicated that TAGL1 participates in the transcriptional control of cuticledevelopment mediating the biosynthesis of cuticle components. Furthermore, cell morphology and the arrangement of epidermalcell layers, on whose activity cuticle formation depends, were altered when TAGL1 was either silenced or constitutively expressed,indicating that this transcription factor regulates cuticle development, probably through the biosynthetic activity of epidermal cells.Our results also support cuticle development as an integrated event in the fruit expansion and ripening processes that characterizefleshy-fruited species such as tomato.

Among the angiosperm species yielding fleshy fruits,tomato (Solanum lycopersicum) has provided the mostrelevant discoveries of the last decades about the geneticand physiological mechanisms underlying fruit ripening(Giovannoni, 2004; Seymour et al., 2008, 2013; Ariizumiet al., 2013). Hence, tomato is currently considered the

model species for the study of this complex develop-mental process, which ensures the formation, matura-tion, and dispersal of seeds and makes the fruit healthyfor human consumption. As with other climactericfruits, ripening of tomato involves biochemical andphysiological changes affecting texture, pigmentation,nutritional, and organoleptic traits, most of them coor-dinated and synchronized by ethylene (for review, seeKlee and Giovannoni, 2011). However, ethylene is knownto be insufficient to regulate tomato ripening; thus, dif-ferent regulatory factors must be properly coordinatedwith ethylene synthesis as part of the genetic networkthat controls fruit ripening (for review, see Seymouret al., 2013). Characterization of several tomato mutantsaffected in fruit ripening has allowed for the cloningand functional analysis of some of the components inthe ripening pathway. It has also highlighted the crucialrole of the MADS box transcription factor encoded byRIPENING INHIBITOR (RIN) as a main regulator of thisprocess (Vrebalov et al., 2002). Recent evidence has in-dicated that RIN directly controls the expression of tar-get genes involved in a wide range of ripening-relatedevents (Fujisawa et al., 2012; Qin et al., 2012). This tran-scriptional activity also depends on COLORLESS NON-RIPENING, a SQUAMOSA Promoter Binding Protein-Like

1 Thisworkwas supported by theMinisterio de Ciencia e Innovación(grant nos. BIO2009–11484, AGL2012–32613, AGL2012–40150–C03–01,and AGL2012–40150–C03–02) and by the European Commissionthrough the Junta para la Ampliación de Estudios-Doc program ofthe Consejo Superior de Investigaciones Científicas (to B.P.).

* Address correspondence to [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:Rafael Lozano ([email protected]).

E.G. conducted the experiments, assisted with data interpretation,and drafted the article; E.D. collaborated on the experiments andreviewed the writing; B.P. and V.M. generated the transgenic plantsand collaborated on the genetic analyses; A.H. contributed a criticalreview of the article; T.A. assisted with the data analysis and re-viewed the article; R.L. planned the research work, assisted withthe data interpretation, and edited the article; all authors read andapproved the final article.

www.plantphysiol.org/cgi/doi/10.1104/pp.15.00469

1036 Plant Physiology�, July 2015, Vol. 168, pp. 1036–1048, www.plantphysiol.org � 2015 American Society of Plant Biologists. All Rights Reserved. www.plantphysiol.orgon October 6, 2020 - Published by Downloaded from

Copyright © 2015 American Society of Plant Biologists. All rights reserved.

http://orcid.org/0000-0001-5458-2075

mailto:[email protected]

http://www.plantphysiol.org

mailto:[email protected]

http://www.plantphysiol.org/cgi/doi/10.1104/pp.15.00469


box gene (Manning et al., 2006; Martel et al., 2011).Other ripening factors, such as NONRIPENING (NOR;GenBank accession no. AY573802; Tigchelaar et al., 1973),the tomato homeodomain leucine zipper homeoboxprotein LeHB1 (Lin et al., 2008), Tomato APETALA2(SlAP2a; Karlova et al., 2011), and the MADS boxTOMATO AGAMOUS-LIKE1 (TAGL1; Vrebalov et al.,2009; Giménez et al., 2010), exert their regulatory func-tions by interacting with RIN (Fujisawa et al., 2012; Qinet al., 2012; Seymour et al., 2013).Cuticle formation of tomato fruit has been considered

an integral part of the fruit-ripening program (Saladiéet al., 2007). Therefore, the composition and mechanicalperformance of cuticle have been suggested to play im-portant roles during fruit ripening (Domínguez et al.,2012). The cuticle is synthesized by the epidermal celllayer of the fruit pericarp. It is composed of a cutinpolymer matrix and cuticular waxes (Pollard et al.,2008). Phenolic compounds, mostly cinnamic acid de-rivatives and flavonoids, are also present (Hunt andBaker, 1980). In addition, a significant amount of poly-saccharides can be found, which represent the portionof epidermal cell wall to which the cuticle is attached(Jeffree, 2006). During fruit development, the cuticleundergoes several biochemical changes and enlargesconsiderably, surrounding the epidermal cells. The de-gree of invagination and the relative contribution ofeach cuticle component affect its biomechanical behav-ior as well as the physiological properties of the whole-fruit pericarp (Matas et al., 2004; López-Casado et al.,2007; Domínguez et al., 2011; España et al., 2014b).Thus, waxes and flavonoids have been shown to act asfillers stiffening the cuticle (Petracek and Bukovac, 1995;Domínguez et al., 2009). On the other hand, cutin is re-sponsible for the viscoelastic behavior and polysac-charides contribute to the elastic phase (López-Casadoet al., 2007, Domínguez et al., 2011). Molecular modelsfor cuticle biosynthetic pathways have been proposedmostly in Arabidopsis (Arabidopsis thaliana); however,there is little information about the genetic and molecularbases that regulate fruit cuticle development in tomato.Some genes have been identified by the characterizationof tomato mutants altered in fruit cuticle generation.Among them, the CUTIN DEFICIENT2 (CD2) gene en-codes a member of the class IV homeodomain-Leu zip-per family that regulates cutin biosynthesis (Isaacsonet al., 2009, Nadakuduti et al., 2012). LeCER6 encodes ab-ketoacyl-CoA synthase involved in cuticular waxcomposition during fruit development (Vogg et al., 2004;Leide et al., 2007). Like the Arabidopsis AP2 tran-scription factor SHINE1 (SHN1; Aharoni et al., 2004;Kannangara et al., 2007), the tomato SlSHN1, SlSHN2,and SlSHN3 orthologs have been described as tran-scriptional regulators of cutin and wax biosyntheticpathways (Mintz-Oron et al., 2008; Shi et al., 2013).Similarly, some key components of the flavonoid path-way are involved in tomato cuticle development. That isthe case of the transcription factor encoded by SlMYB12(Adato et al., 2009; Ballester et al., 2010); this regulatesflavonoid biosynthesis by controlling CHALCONE

SYNTHASE1 (CHS1) and CHS2 enzymes, the first stepin the flavonoid pathway (O’Neill et al., 1990). In addi-tion, a recent screening of a tomato ethyl methanesul-fonate collection for mutants displaying glossy and dullfruits has facilitated the isolation of novel genes relatedto cuticle formation (Petit et al., 2014).

There is increasing knowledge about the geneticcontrol of cuticle biosynthesis; yet, the upstream keygenes and regulatory mechanisms that integrate cuticledevelopment in the fruit-ripening program are still to becharacterized. Previous studies have reported significantdifferences in the fruit cuticle of nonripening mutants,suggesting new roles of RIN and NOR in the geneticcontrol of fruit ripening with a direct effect on cuticledevelopment (Kosma et al., 2010). Similarly, transcrip-tion factors that participate in fruit development andripening regulation, such as those belonging to the AP2/ETHYLENE-RESPONSIVE ELEMENT BINDING PRO-TEIN, MADS box, MYB, and homeodomain-Leu zipperfamilies, have recently been shown to be involved indifferent processes of cuticle formation in tomato fruit(for review, see Hen-Avivi et al., 2014). It is interestingthat tomato fruits yielded by TAGL1-silencing plants(RNA interference [RNAi] lines) were unable to com-plete ripening and showed greater firmness than wild-type fruit. In addition, cuticle of RNAi fruits displayedreduced thickness and invagination degree (Giménezet al., 2010). These results suggested a connection be-tween cuticle formation and the function of TAGL1 as atranscriptional regulator of fruit ripening. Furthermore,Matas et al. (2011) have shown that the TAGL1 gene isexpressed at high levels in fruit outer epidermis, sug-gesting its role in the development of fruit cuticle. Togain insight into the functional role of TAGL1 in cuticledevelopment, structural, biomechanical, and gene ex-pression analyses have been performed in peels iso-lated from TAGL1-silencing and overexpression lines.The results show that TAGL1 plays a key role in tomatofruit cuticle development through the genetic control ofcuticular component biosynthesis. These results willprovide a better understanding of the genetic basesof cuticle formation and of TAGL1’s contribution to thetomato reproductive program, where it is a crucialregulator.

RESULTS

Previous works have demonstrated the crucial role ofthe TAGL1 gene as a transcriptional regulator of flowerdevelopment and fruit ripening (Itkin et al., 2009;Vrebalov et al., 2009; Giménez et al., 2010). Cuticle for-mation is currently considered part of the developmen-tal program leading to fruit ripening. Therefore, thisarticle is aimed at studying the contribution of TAGL1to the development of fruit cuticle in the tomato modelspecies. With this purpose, the composition and bio-mechanical properties of cuticles have been analyzedin tomato fruits in which TAGL1 was either silencedby an RNAi approach or overexpressed (OE) under the

Plant Physiol. Vol. 168, 2015 1037

TAGL1 Regulates Cuticle Development of Tomato

www.plantphysiol.orgon October 6, 2020 - Published by Downloaded from Copyright © 2015 American Society of Plant Biologists. All rights reserved.


control of a 35S-cauliflower mosaic virus constitutivepromoter. Compared with wild-type fruits of cv Mon-eymaker, TAGL1-RNAi fruits displayed a yellow-orangecolor and a glossy phenotype; they were also unable tocomplete normal ripening (Fig. 1A). Yet, fruits yielded byOE lines showed an accelerated ripening and were ac-companied by succulent sepals, which, in turn, devel-oped and ripened like a normal tomato fruit (Fig. 1A).

The expression of TAGL1 was analyzed in tomatopeels, a fruit tissue that includes the epidermis, which,

in turn, is responsible for cuticle biosynthesis. TAGL1transcripts were detected in wild-type fruit peels at theimmature green (IG) stage; later, they were accumulatedduring the ripening process and reached a maximumexpression level at the red ripe (RR) stage (Fig. 1B). Thesame expression pattern was found in TAGL1-OE peels,although the TAGL1 transcript level was higher than inwild-type ones. On the contrary, no TAGL1 expressionwas observed in peels isolated from TAGL1-RNAi fruits(Fig. 1B).

Figure 1. Phenotypic studies and expression analyses in TAGL1-silencing and -overexpresing fruits. A, Fruits, isolated cuticles, andcuticle sections stainedwith Sudan IV to visualize the cuticle ofwild-type (WT), TAGL1-RNAi, and TAGL1-OE tomato plants. Bars = 20mm.B, Relative expression of the TAGL1 gene in peel fromwild-type, TAGL1-RNAi, and TAGL1-OE tomato fruits at IG,mature green (MG),breaker (BR), and RR stages. Data are means 6 SE. Values followed by the same letter are not statistically significant (P , 0.05).

1038 Plant Physiol. Vol. 168, 2015

Giménez et al.



TAGL1 Transcript Levels Affect Cuticle Development

A glossy phenotype such as that shown by tomatofruits silencing TAGL1 (Fig. 1A) was recently associatedwith alterations in cuticle development and epidermalpatterning (Petit et al., 2014). Therefore, developmentalchanges in cuticle formation promoted by either si-lencing or overexpression of the TAGL1 gene were an-alyzed. The results indicated that, once isolated, cuticlesfrom TAGL1-RNAi fruits showed a visible pale yellow-orange color together with noncolored patches randomlydistributed over their surface; this differed from the ho-mogenous and intense orange color of cuticles from wild-type fruits (Fig. 1A). Furthermore, pericarp cross sectionsof wild-type fruits stained with Sudan IV displayed thecharacteristic cutinization of epidermal cell walls, almostcompletely surrounded by cuticle material, which is alsodeposited on the radial walls of the first collenchymacell layer (Fig. 1A). In contrast, RNAi fruits developed a3-fold thinner cuticle than the wild-type ones (Table I)and a complete absence of epidermal invaginations(Fig. 1A). It is noteworthy that the thinnest regions ofRNAi cuticles coincided with the noncolored patchesmentioned above. The amount of cuticle and its com-ponents (i.e. cutin, waxes, polysaccharides, and pheno-lics) decreased significantly in tomato cuticles isolatedfrom TAGL1-RNAi fruits (Table I). Changes affectingcuticular content and thickness also suggested differ-ences in the mechanical resistance of TAGL1-RNAi cu-ticles. Therefore, several parameters characterizing thebiomechanical properties of fruit cuticle were measured(Table I). Data normalized per cuticle thickness unitshowed that only Young’s modulus, an indicator ofcuticle stiffness, decreased significantly when TAGL1was silenced, while mean values of breaking stressand maximum strain did not differ with respect to thewild type (Table I). However, nonnormalized analyticalresults showed that cuticles from TAGL1 silencing fruitswithstood only 40 to 50 g before breaking, 3-fold lessthan control cuticles (130 g), indicating that both thick-ness and composition influence the mechanical proper-ties of cuticles.Contrarily, cuticles isolated from tomato fruits con-

stitutively overexpressing TAGL1 showed a more severe

cutinization level than those of wild-type fruits. This isbecause the cuticle material substantially encased notonly the epidermis but also several collenchyma layers,three in the most extreme cases (Fig. 1A). This high levelof cutinization was accompanied by the development ofepidermal ridges on the cell walls, a feature not observedin wild-type cuticles (Fig. 1A). Furthermore, cuticlethickness was 2-fold increased, on average, in TAGL1-OE fruits (Table II). As expected, the total amount ofcuticle was also significantly higher than in wild-typefruits, most likely due to the increased content of cutinand waxes. However, the amounts of polysaccharidesand phenolics were similar to wild-type ones (Table II).In addition to differences related to biochemical com-position, thickness, and invagination degree of OE fruitcuticles, changes in their biomechanical properties werefound. Thus, significant decreases in Young’s modulusand breaking stress were observed (Table II), indicatingthat constitutive expression of TAGL1 modified themechanical behavior of fruit cuticle. Normalized datashowed that OE cuticles are mechanically different fromwild-type ones (Table II); however, the higher thicknessof the former allowed them to withstand a breakingstress (about 120 g) similar to the latter.

Compositional and Biomechanical Properties of Cuticlesfrom Fruit-Like Sepals Developed by TAGL1-OE Plants

It was reported previously that the ectopic expressionof TAGL1 was responsible for the conversion of sepalsinto succulent organs, which developed and ripened ina fruit fashion (Fig. 1A; Giménez et al., 2010). In thiswork, the morphological and biochemical features ofthe cuticle isolated from TAGL1-OE sepals were ana-lyzed with the aim to check the contribution of TAGL1to the homeotic conversion of sepals into fruit-like struc-tures and, hence, to gain insight into the functional roleof TAGL1 as a key regulator of fruit development.The results showed that cuticles isolated from fruit-likesepals overexpressing TAGL1 were extremely delicate,thinner than wild-type fruit cuticle (Fig. 2A), andtherefore hard to isolate. Their morphology and thick-ness resembled those of cuticles covering wild-type se-pals (Fig. 2A). It is interesting that succulent sepals didnot develop characteristic epidermal cells (Fig. 2B).Their cuticles had 4 times less material than those ofwild-type and TAGL1-OE tomato fruits, despite theirfruit-like appearance (Table II). Nevertheless, the cellanatomy of succulent sepals was more similar to thatof a wild-type fruit pericarp than to a wild-type sepal,since large oval parenchyma cells, which are charac-teristic of fruit pericarp, were observed instead of theround parenchyma cells typical of wild-type sepals (Fig.2B). These results indicated that the constitutive ex-pression of TAGL1 in sepals promotes similar devel-opmental changes to those occurring in wild-type tomatofruits during the ripening process. These changes mainlyaffected cell identity and biochemical composition(i.e. the content of sugars, carotenoid, lycopene, and

Table I. Cuticle components and biomechanical parameters of wild-type (cv Moneymaker) and TAGL1-RNAi fruits at the RR stage

Values are means 6 SE. The statistical significance of mean differ-ences was analyzed using Student’s t test with *, P , 0.05 and **, P ,0.01.

Parameter Wild-Type Fruits TAGL1-RNAi Fruits

Thickness (mm) 11.46 6 0.24 4.02 6 0.15**Cuticle (mg cm22) 1,962.0 6 72.3 1,024.3 6 46.0**Cutin (mg cm22) 1,383.1 6 51.0 742.1 6 33.3**Waxes (mg cm22) 69.7 6 2.6 41.2 6 1.8**Polysaccharides (mg cm22) 509.2 6 18.8 236.7 6 10.6**Phenolics (mg cm22) 109.7 6 8.0 25.4 6 5.7**Young’s modulus (MPa) 535.0 6 25.3 457.8 6 1.3*Breaking stress (MPa) 34.2 6 1.4 36.6 6 4.1Maximum strain (%) 27.1 6 4.1 23.4 6 1.60





ethylene). They were almost identical to those observedin the fruits and sepals of tomato Arlequin (Alq) mutant,which also overexpressed the TAGL1 gene (Giménezet al., 2010; Pineda et al., 2010).

Pericarp Development and Cuticle Formation ofTAGL1-RNAi and TAGL1-OE Fruits

It is known that fruit cuticle develops from the activityof epidermal cells (Javelle et al., 2011) and that TAGL1 isexpressed in this cell layer of fruit pericarp (Fig. 1B;Matas et al., 2011). Therefore, we analyzed whethercuticular changes promoted by the silencing of TAGL1might be correlated with alterations in the epidermalcells of tomato fruit. Pericarp of RR wild-type fruitsconsisted of a single epidermal layer composed ofelongated cells evenly distributed, followed by two tothree hypodermal layers of elongated collenchyma cellsand several layers of large parenchyma cells (Fig. 3A).This tissue-patterning distribution is clearly distin-guishable from the MG stage. However, pericarp tissuedevelopment is initiated earlier from 10 to 12 cell layersforming the carpels, right after the ovary fertilization offlowers on anthesis day (Fig. 3A). As reported (Gillaspyet al., 1993), carpel cells undergo an active period of cell

division, which ceases to give way to an expansion cellperiod, both leading to the full development of fruitpericarp (Fig. 3A). The results obtained here showedthat repression of TAGL1 in fruit pericarp resulted ina significant decrease of cell number per surface unit inboth epidermis and collenchyma tissues; hence, cell sizein TAGL1-silencing tissues was greater than in wild-type ones (Fig. 3D; Table III). In addition, the number ofcollenchyma layers diminished from 3.2 6 0.4 in thewild type to 1.3 6 0.5 in TAGL1-RNAi pericarps, whilethe epidermis was not affected (Fig. 3D). All thesechanges would explain the decreased pericarp thicknessobserved previously in TAGL1-silencing lines (Vrebalovet al., 2009; Giménez et al., 2010). On the contrary, con-stitutive expression of TAGL1 doubled the cell numberper surface unit in the parenchyma tissue with respectto wild-type pericarp, indicating that cell size wasclearly smaller in OE lines (Fig. 3A; Table III). Thus, thenumber of parenchyma layers increased from 19.44 60.8 in the wild type to 26.2 6 1.84 in TAGL1-OE fruits,while the cell layer number of epidermis and collen-chyma tissues did not change with respect to the wildtype (Fig. 3D). Additionally, differentiation of pericarptissues was anticipated in OE fruits with respect towild-type fruits. Thus, the degree of epidermis andcollenchyma cell differentiation of OE pericarp at the IG

Table II. Components and biomechanical parameters of cuticles isolated from wild-type (cv Money-maker) fruits and their comparison with cuticles from fruits and fruit-like sepals overexpressing the TAGL1gene

Values are means6 SE. Statistical significance of mean differences is indicated either by letters (Tukey’s btest) with P, 0.05 or by asterisks (Student’s t test) with *, P, 0.05 and **, P, 0.01. n.d., Not determined.

Parameter Wild-Type Fruits TAGL1-OE Fruits TAGL1-OE Sepals

Thickness (mm) 6.7 6 0.1 11.9 6 0.2** n.d.Cuticle (mg cm22) 1,460.2 6 31.9 b 1,742.3 6 53.8 a 417.0 6 72.0 cCutin (mg cm22) 970.4 6 21.2 b 1,209.7 6 37.4 a 115.9 6 20.0 cWaxes (mg cm22) 49.6 6 1.1 b 59.8 6 1.8 a 31.7 6 5.5 cPolysaccharides (mg cm22) 440.2 6 9.6 a 472.7 6 14.6 a 269.3 6 46.5 bPhenolics (mg cm22) 64.7 6 5.1 a 56.1 6 4.7 a 6.4 6 1.7 bYoung’s modulus (MPa) 488.1 6 51.7 306.6 6 42.1* n.d.Breaking stress (MPa) 36.6 6 2.4 17.8 6 1.4** n.d.Maximum strain (%) 12.1 6 1.6 8.2 6 1.6 n.d.

Figure 2. Morphological analysis from TAGL1-OE sepals. A, Sections of sepal and fruit pericarpfrom wild-type (WT) and TAGL1-OE plantsharvested at the RR stage stained with Aura-mine O fluorescence stain to visualize thecuticle with a light microscope equipped witha fluorescein isothiocyanate filter. Bars = 25 mm.B, Sections of sepal and fruit pericarp fromwild-type and TAGL1-OE plants harvested at the RRstage stained with Toluidine Blue for light mi-croscopy observation. c, Collenchyma; e, epi-dermis; p, parenchyma. Bars = 50 mm.


Giménez et al.



stage resembled that observed in wild-type pericarp at alater developmental stage such as the MG stage (Fig. 3,B and C). Similarly, OE pericarp at the MG stage dis-played the characteristic tissue patterning of wild-typepericarp at the RR stage (Fig. 3, C and D). It is interestingthat epidermis and collenchyma tissues were not dif-ferentiated correctly in TAGL1-RNAi fruits (Fig. 3, Cand D).The results described above supported the fact that

the transcriptional activity of TAGL1 influences the tis-sue patterning of fruit pericarp. Therefore, it was rea-sonable to think that cuticle formation could also bealtered in tomato fruits where TAGL1 expression wasmodified. As occurred in wild-type fruits, cuticle for-mation was visible at the IG stage of fruit developmentin both RNAi and OE plants. Later, cuticle depositionincreased at the MG stage; cuticle acquired its maximumthickness and invagination degree at the RR stage inwild-type and OE fruits (Fig. 4). However, cuticle

formation was affected at initial stages in RNAi fruits,and it was unable to evolve normally during the fruit-ripening process. Consequently, TAGL1 pericarp of RRfruits displayed a thin cuticle without invaginations,quite similar to that observed in wild-type fruits at suchan early stage of development as the IG stage (Fig. 4).

TAGL1 Influences the Expression Pattern of Tomato GenesInvolved in Cuticle Development

Given the structural and compositional alterationsfound in cuticles of TAGL1-RNAi and TAGL1-OE fruits,the transcription levels of genes known to be involvedin cuticle development were analyzed by quantitative re-verse transcription-PCR in peels isolated from tomatofruits at the IG, MG, BR, and RR stages (Figs. 5 and 6).Comparedwithwild-type peels,CD2was down-regulatedin RNAi peels, particularly at the BR and RR stages,

Figure 3. Cell morphology and cell layer pat-terning of TAGL1-RNAi and TAGL1-OE fruitpericarp. A, Light microscopy images of peri-carp sections from wild-type (WT), TAGL1-RNAi, and TAGL1-OE fruit at anthesis day (AD),AD+10, IG, MG, and RR stages, stained withToluidine Blue. Bars = 50 mm. B, A 2.5-foldamplification of epidermis and collenchymacells within the rectangles shown at theIG stage in A. C, A 2.5-fold amplification ofepidermis and collenchyma cells within therectangles shown at the MG stage in A. D, A2.5-fold amplification of epidermis and col-lenchyma cells within the rectangles shown atthe RR stage in A. c, Collenchyma; e, epi-dermis; p, parenchyma.





while it was expressed at a similar level in OE peelsexcept at the end of the ripening process (RR stage),when it was suddenly repressed (Fig. 5). Among thegenes involved in wax biosynthesis, LeCER6 did notshow expression changes in RNAi or OE peels with re-spect to the wild type (Fig. 5). However, SlSHN1 wassignificantly up-regulated in RNAi from the MG stageup to full ripening of tomato fruits (Fig. 5), while theexpression of SlSHN3 was clearly induced in OE peelsat the early stage of fruit development (IG stage), andlater it drastically decreased during fruit-ripening stages(Fig. 5).

Regarding the genes regulating the phenylpropanoid/flavonoid pathway, the transcript levels of SlMYB12,CHS1, CINNAMYL ALCOHOL DEHYDROGENASE(CAD), and 4-COUMARATE CoA LIGASE (4CL) wereassessed (Fig. 6). In wild-type fruits, all these genes wereup-regulated in tomato peels during the first stagesof fruit ripening; they reached a maximum level oftranscripts at the BR stage and were subsequently down-regulated. The expression of SlMYB12 and CHS1 flavo-noid biosynthesis genes was significantly inhibited inRNAi lines and did not display the ripening-dependentincrease observed in the normal peels. However, noexpression alterations were observed for the SlMYB12gene in OE lines, while CHS1was down-regulated (Fig. 6).Regarding CAD and 4CL, these were up-regulated inpeels of RNAi fruits at the IG stage, when these geneswere not expressed in wild-type peels. Contrarily, tran-scripts of CAD and 4CL were down-regulated at earlystages of fruit development in OE lines, although theirexpression increased during ripening stages, particu-larly that of CAD (Fig. 6).

In addition to cuticle, epidermal cell walls, which arein direct contact with the cuticular lipids, contribute tothe mechanical properties of the tomato fruit exocarpand the resistance of the turgor-driven fruit growth intomato (Andrews et al., 2002; Bargel and Neinhuis, 2005).Therefore, the expression of POLYGALACTURONASE(PG) and PECTIN METHYLESTERASE (PME) genes,which encode major enzymes involved in cell wall deg-radation (Dellapenna et al., 1986; Smith et al., 1988;Harriman et al., 1991), were analyzed. Both PG and PMEwere down-regulated when TAGL1was silenced in fruitpeels. However, PME expression was up-regulated in

TAGL1-OE peels, while the PG transcript level wassimilar to that detected in wild-type peels (Fig. 6). Fur-thermore, the expression of PHYTOENE SYNTHASE(PSY), the major gene responsible for carotenoid bio-synthesis (Fray and Grierson, 1993), was significantlyrepressed in RNAi peels (Fig. 6), which is in accordancewith the colorless phenotype of the TAGL1-silencingfruits. Nevertheless, the PSY transcript levels were notaffected in OE peels (Fig. 6).

DISCUSSION

TAGL1 Regulates the Composition and BiomechanicalProperties of Fruit Cuticle

The TAGL1 MADS box gene is particularly relevantfor fruit ripening and has been reported as one of themajor regulators of fruit development (Itkin et al., 2009;Vrebalov et al., 2009; Giménez et al., 2010). This workdelves into the study of fruit development by analyzingthe functional role of TAGL1 in cuticle development.With this aim, tomato fruits either silencing or over-expressing TAGL1 have been analyzed at the biome-chanical, cellular, and molecular levels. The results showthat the transcriptional activity of TAGL1 affects thebiochemical composition and biomechanical propertiesof fruit cuticle. Thus, repression of TAGL1 causes a sig-nificant decrease in the amount of cuticle and all itscomponents. Reduction of phenolic compounds duringripening is responsible for the pale cuticle color dis-played in RNAi lines. Other transcription factors, suchas SlMYB12 and NOR, have been shown to alter cuticlecolor during ripening due to a decrease in phenolics(Bargel and Neinhuis, 2004; Adato et al., 2009), butcuticle deposition was only affected by SlMYB12 (Adatoet al., 2009). More recently, a significant decrease incuticle and most of its components, except for waxes,was reported in tomato fruits silencing CHS duringripening (España et al., 2014a). The significant reductionin cuticle components known to confer stiffness (i.e. poly-saccharides, phenolics, and waxes) would explain thelower Young’s modulus measured in TAGL1-repressedcuticles. This observation supports the results obtained inthe cuticles of nor mutants and CHS-silenced tomatoes(Bargel and Neinhuis, 2004; España et al., 2014a).

Overexpression of TAGL1 results in an expected op-posite behavior: an increase in cuticle components, ex-cept for phenolics and polysaccharides, which remainedunaltered in OE lines. A severe decrease in cuticle bio-mechanical properties, mainly Young’s modulus andbreaking stress, is observed after TAGL1 overexpres-sion. The reduced values of Young’s modulus in bothRNAi and OE lines would be explained by the decreasein the relative proportion of polysaccharides and phe-nolics with respect to waxes and cutin detected in thesetransgenic lines. However, the lower mechanical resis-tance of the OE cuticle is compensated by the increasein the cuticle thickness.

These results, together with the fact that the TAGL1gene is expressed in the fruit’s outer epidermis (Matas

Table III. Number of cells from pericarps of the wild type (cv Mon-eymaker) and silencing and overexpressing lines at the RR stage.

Values are means 6 SE. Statistical significance of mean differences isindicated by asterisks (LSD) with **, P , 0.01.

LineCell No.

Epidermisa Collenchymab Parenchymac

Wild type 9.3 6 0.9 27 6 2.6 24 6 4.3TAGL1-RNAi 7.6 6 1.5** 18.2 6 1.5** 24 6 3.1TAGL1-OE 8.5 6 1.1 27 6 2.5 59.2 6 6.5**

aCell no. from epidermis was measured on a line of 300 mm.bCell no. from collenchyma was measured on a square area of 50 3500 mm. cCell no. from parenchyma was measured on a squarearea of 1.25 3 1.25 mm.


Giménez et al.



et al., 2011), suggested that this MADS box gene par-ticipates in cuticle development as part of the geneticprogram leading to fruit formation (Vrebalov et al.,2009; Giménez et al., 2010).

TAGL1 Controls Cuticle Development by Regulating theBiosynthesis of Cuticle Components

Expression analyses of genes involved in cuticle bio-synthesis were performed in TAGL1-silencing and

-overexpressing peels, which have allowed for a betterunderstanding of the complexity of the genetic networkinvolved in cuticle formation and the role of TAGL1 aspart of this developmental program. Flavonoid genessuch as CHS1 and SlMYB12 were down-regulated inTAGL1-RNAi peels, a result that agreed with the lowlevels of phenolic compounds and suggested a role ofTAGL1 in flavonoid biosynthesis. Such a regulatoryfunction may occur through the SlMYB12 transcriptionfactor, which has indeed been proposed as a positive

Figure 4. Morphological analyses of TAGL1-RNAi and TAGL1-OE cuticles during carpeldevelopment. Fruit epidermis sections fromwild-type (WT), TAGL1-RNAi, and TAGL1-OElines at anthesis day (AD), AD+10, IG, MG,and RR stages of development were stainedwith Sudan III to visualize cuticle. Bars = 50 mm.

Figure 5. Relative expression of genesrelated to biosynthetic pathways of cutin(CD2) and waxes (SlCER6, SlSHN1, andSlSHN3). The analyses were performed inpeel from wild-type control (WT), TAGL1-RNAi, and TAGL1-OE tomatoes at IG, MG,BR, and RR stages. Data are means 6 SE.Values followed by the same letter are notstatistically significant (P , 0.05).





regulator of flavonoid pathway genes such as CHS1(Adato et al., 2009; Ballester et al., 2010). Accordingly,increased CHS1 expression and high levels of the nar-ingenin chalcone flavonoid were detected in fruit-likesepals developed by tomato plants overexpressingTAGL1 (Itkin et al., 2009).

Previous studies have shown that phenylpropanoid/lignin pathway genes, such as PHENYLALANINEAMMONIA-LYASE5, 4CL, CAD, and CINNAMOYL-CoAREDUCTASE1 (Anterola and Lewis, 2002; Vanholmeet al., 2010; Wang et al., 2013), were up-regulated insilencing TAGL1 fruit pericarp, which agrees with theincreased lignin content observed in this tissue (Giménez

et al., 2010). Similarly, transcript levels of 4CL and CADgenes were induced in RNAi and repressed in OE peels,supporting the role of TAGL1 as a repressor of ligninbiosynthesis in tomato fruits. Such negative regula-tion could occur either by a direct inhibition of thephenylpropanoid/lignin pathway genes or throughan indirect control of the flavonoid pathway. Flavonoidgenes are down-regulated when TAGL1 is silenced intomato fruits; thus, the substrate used from both fla-vonoid and phenylpropanoid/lignin pathways (i.e.p-coumaric acid) could be preferentially used for ligninbiosynthesis in the detriment of flavonoid biosyn-thesis. If so, TAGL1 transcriptional factor would act by

Figure 6. Expression studies of genesrelated to the biosynthesis of phenyl-propanoids (4CL and CAD), flavonoids(CHS1 and SlMYB12), carotenoids (PSY),and cell wall degradation (PG and PME)in peel from wild-type (WT), TAGL1-RNAi,and TAGL1-OE tomatoes at IG, MG, BR,and RR stages. Data are means 6 SE.Values followed by the same letter arenot statistically significant (P , 0.05).


Giménez et al.



maintaining a balance between flavonoid and ligninpathways in tomato fruits. Redirection of the meta-bolic flux between both pathways had previously beenreported in silencing Arabidopsis mutants affected inhydroxycinnamoyl-CoA shikimate/quinate hydrox-ycinnamoyl transferase, a lignin biosynthetic gene(Besseau et al., 2007; Li et al., 2010), and also in CHSantisense plants of strawberry (Fragaria 3 ananassa;Lunkenbein et al., 2006), supporting a role of TAGL1mediating the flavonoid and lignin pathways.Decreased CD2 transcript level assessed in TAGL1

knockout peels was in accordance with the loweramounts of cutin, indicating that cuticle biosynthesis ismediated by TAGL1 transcriptional activity. However,other regulatory genes may be involved in cutin bio-synthesis, since constitutive expression of TAGL1 didnot affect CD2 expression, although it promoted in-creased cutin content. Among them, ripening tran-scription factors such as RIN and NOR have been shownto be involved in wax and cutin composition (Kosmaet al., 2010) and therefore could participate in regulatingcuticle development.Regarding wax biosynthesis, fruits silencing TAGL1

showed diminished wax content in their cuticles. Genesinvolved in wax biosynthesis, SlSHN1 and SlSHN3,displayed an opposite expression pattern during fruitdevelopment. While SlSHN3 was partially inhibitedbefore the MG stage, SlSHN1 was significantly up-regulated after the MG stage of RNAi fruits. Theseresults suggest that TAGL1 might regulate wax bio-synthesis during fruit development through a temporalexpression balance of SHN genes. According to thishypothesis, it is known that SHN1 influences wax ac-cumulation in Arabidopsis by modulating cutin pro-duction and changing the physical properties of cuticle(Kannangara et al., 2007). Similarly, LeCER6 mediateschemical modifications affecting the cuticular waxcomposition of tomato fruits (Vogg et al., 2004; Leideet al., 2007). Taking these observations and the geneexpression levels measured in RNAi peels into account,TAGL1 is likely to modulate the wax content of fruitcuticle through the control of cutin production ratherthan regulate the expression of wax genes. Together, theresults reported in this work support that the TAGL1transcription factor regulates the cuticle developmentof tomato fruit, mediating the biosynthesis of cuticlecompounds such as cutin, wax, and flavonoids.On the other hand, constitutive expression of TAGL1

promoted significant developmental changes affectingcell epidermal patterning and the biochemical com-position of sepals, which turn them into fleshy fruit-like organs (Giménez et al., 2010). Similar results werealso reported in the tomato Alq mutant, which alsooverexpressed the TAGL1 gene (Pineda et al., 2010).However, developmental features of cuticles from suc-culent TAGL1-OE sepals were more similar to cuticlescovering the leaf surface than to fruit cuticles. It is rea-sonable to think that cell signals that promote cuticleformation in TAGL1-OE sepals are sent before sepalcells are converted into carpel cells (Jeffree, 2006).

Alternatively, the transcriptional activity of other reg-ulatory genes specifically required for the developmentof tomato fruit cuticle might be absent from sepal or-gans, which would prevent sepals from developing afruit-like cuticle.

The Genetic Regulation of Fruit Cell Patterning Mediatedby TAGL1 Influences Cuticle Development

Previous results showed that TAGL1 expression be-gins at early stages of flower development and that thehighest accumulation of TAGL1 transcripts coincideswith flower anthesis and the RR stage of fruit ripening(Giménez et al., 2010). Furthermore, expression of TAGL1has been detected in collenchyma and parenchyma offruit pericarp (Matas et al., 2011). Its repression pro-moted changes in the size and distribution of cellsforming these tissues, indicating that TAGL1 is a crucialstep for the development and ripening of a fleshy fruitlike tomato (Vrebalov et al., 2009; Giménez et al., 2010).In this work, detailed analyses of cuticles isolated fromtomato fruits silencing or overexpressing TAGL1 pro-vide evidence that the transcriptional activity of TAGL1is also a key factor for epidermis and cuticle develop-ment of tomato fruits. The cuticle is synthesized by theepidermis (Javelle et al., 2011). Thus, the reduced cuticlethickness, and the alterations in the cuticle developmentshown by tomato fruits lacking TAGL1 expression,might be due to the abnormal biosynthetic activity ofepidermal cells, which in turn affects the amount andinvagination degree of cuticle. It is also noteworthy thatsuch compositional alterations of TAGL1-repressed cu-ticles are associated with significant changes in the size,morphology, and arrangement of epidermal cells. Thisindicates that the patterning and biosynthetic activity ofepidermal cells influence cuticle properties of tomatofruit. Indeed, a close relationship between epidermalcell differentiation and cuticle development has beenreported as coordinated in Arabidopsis by SHN1 andtwo MIXTA-like regulators of epidermal cell morphol-ogy (Oshima et al., 2013). Similarly, the SlSHN3 tran-scription factor has been put forward as a link factorbetween epidermal patterning and fruit cuticle forma-tion in tomato (Shi et al., 2013). In this work, it wasproved that TAGL1 controls the expression of the to-mato SlSHN1 and SlSHN3 genes. Therefore, the func-tion of TAGL1 in epidermis development could beexercised through regulating the expression patternof SlSHN1 and SlSHN3, which in turn might act in aredundant manner, as Shi et al. (2013) proposed. Takentogether, the results gathered here support a role ofTAGL1 as a transcriptional regulator of fruit cuticledevelopment, which links epidermal cell patterning andcuticle formation, most likely through the expression oftarget genes such as SlSHN homologs. Cuticle altera-tions promoted by TAGL1 silencing also agree with thechanges detected in fruit pericarp, particularly with al-tered structural properties and the reduction of thenumber of cell layers, which in turn correlated to the





decreased expression of cell cycle genes such asCYCLIN-DEPENDENT KINASE A (CDKA1; Vrebalovet al., 2009; Giménez et al., 2010). Recent studies haveshown that CDKA1-regulated cell division occurring infruit pericarp is required for cuticle development intomato (Czerednik et al., 2012). Therefore, cell divisionand expansion characterizing fruit pericarp should beaccompanied by an appropriate development of fruitcuticle. Hence, TAGL1 could coordinate both processesin a fleshy-fruited species such as tomato.

Cuticle development has been suggested as part of thetomato development and ripening programs (Saladiéet al., 2007). Furthermore, the biochemical compositionand mechanical performance of fruit cuticle seem tohave a specific impact on the ripening process of thisfleshy fruit species (Domínguez et al., 2012). It wasreported that cuticles developed by TAGL1-silencedfruits showed alterations in their composition and bio-mechanical properties and that these changes are com-prehensively associated with those affecting fruitdevelopment and ripening. In addition, RIN and NORdo not only represent crucial factors in the genetic net-work controlling fruit ripening (for review, see Kleeand Giovannoni, 2011), but they are also likely to actas transcriptional regulators of early fruit development,with a direct effect on cuticle composition (Kosma et al.,2010). The MADS box transcription factors RIN andTAGL1 have been found to form heterodimers and toact together in the ripening process (Leseberg et al.,2008; Itkin et al., 2009; Vrebalov et al., 2009). Presum-ably, such transcriptional interaction might be nec-essary to control cuticle development during tomatoripening. Moreover, RIN interacts with the related rip-ening transcription factors FRUITFULL1 (FUL1) andFUL2 to regulate fruit ripening through a subset of RINtarget genes different from that controlled by theTAGL1-RIN complex (Bemer et al., 2012). Given thatFUL1/2 have also been involved in the production ofcuticle components (Bemer et al., 2012), TAGL1-RINand FUL1/2-RIN complexes might participate in dif-ferent genetic pathways regulating the development offruit cuticle in tomato.

Besides fruit pericarp, TAGL1 induces expressionchanges of PSY, PME, and PG ripening-related genes infruit peels, indicating that TAGL1 also influences ca-rotenoid biosynthesis and cell wall degradation in thisfruit compartment. These results provide new evidenceabout the relationships between cuticle formation andfruit ripening, two developmental events integratedinto a single developmental program mediated by thetranscriptional activity of the TAGL1 gene.

MATERIALS AND METHODS

Generation of TAGL1 Transgenic Tomato Plants

The binary plasmids generation mediating the RNAi approach (TAGL1-RNAi) and the cloning of the TAGL1 complete open reading frame underthe control of the constitutive promoter 35S (TAGL1-OE), to obtain tomatoTAGL1-silencing and overexpression lines, respectively, have been described

previously by Giménez et al. (2010). The generated binary plasmids wereelectroporated into Agrobacterium tumefaciens LBA 4404 strain, and the trans-formation of tomato (Solanum lycopersicum ‘Moneymaker’) cotyledon explantswas performed as described previously (Ellul et al., 2003).

The presence of the transgene was verified in the T0 generation by PCRusing kanamicin resistance gene-specific primers; homozygous lines were ob-tained from TAGL1-RNAi and TAGL1-OE transgenic plants to use for struc-tural, biochemical, and gene expression analyses. Wild-type tomato plants andTAGL1-RNAi and TAGL1-OE transgenic lines were grown under greenhouseconditions using standard culture practices with regular addition of fertilizers.The cv Moneymaker, used as genetic background, was provided by the C.M.Rick Tomato Genetics Resource Center (http://tgrc.ucdavis.edu/).

RNA Preparation and Gene Expression Analyses

Wild-type tomato peels were removed using a peeler; they normally includethe cuticle, a single layer of epidermal cells, several layers of collenchyma cells,and even parenchyma cell layers. Peels from transgenic fruits were removed as inthe wild type, which resulted in tissue sections of similar thickness. Total RNAfrom three biological replicates of tomato peels was isolated using the Trizolreagent (Invitrogen). A total of 1.2 mg of DNA-free RNA was used for com-plementary DNA synthesis using the First Strand cDNA Synthesis Kit (Fer-mentas) according to the manufacturer’s instructions. Expression analyses wereperformed by quantitative real-time PCR from three biological and two technicalreplicates for each sample. With this aim, the SYBR Green PCR Master Mix Kit(Applied Biosystems) and the 7300 Real-Time PCR System (Applied Biosystems)were used. System Sequence Detection Software version 1.2 (Applied Biosys-tems) was used to calculate gene-specific threshold cycles including the en-dogenous reference (UBIQUITIN3) for every sample. The DDCt calculationmethod was used to obtain the results, expressed in arbitrary units, in com-parison with a data point from the wild-type samples. Relative transcript levelswere calculated using UBIQUITIN3 controls. The sequences of the primers usedin the expression analyses are listed in Supplemental Table S1.

Tissue Sectioning and Cuticle Staining

Pericarp pieces of three different fruits at the RR stage as well as sepals fromthe wild-type control, TAGL1-RNAi, and TAGL1-OE were collected in twodifferent seasons (autumn and spring), fixed in a formaldehyde, acetic acid,and ethanol solution (1:1:18), dehydrated in an ethanol series (70%–95%), andembedded in commercial resin (Leica Historesin Embedding Kit) or ParaplastPlus (Sigma). Four- to 8-mm-thick cross sections were stained with Sudan III orIV or Auramine O in order to visualize the cuticle (Domínguez et al., 2008) andwith Toluidine Blue O to distinguish the general structure. A minimum of10 sections per sample were inspected with a light microscope (Nikon; EclipseE800).

Cuticle thickness was estimated from the previous cross sections using animage-capture analysis program (Visilog-Noesis 6.3) also used to calculatecuticle area from small flat pieces of cuticle following the protocol alreadydescribed by Domínguez et al. (2008).

Cuticle Isolation

Cuticles were enzymatically isolated with an aqueous solution of a mixtureof fungal cellulase (0.2% [w/v]; Sigma) and pectinase (2.0% [w/v]; Sigma) and1 mM NaN3 in sodium citrate buffer (50 mM, pH 3.7; Domínguez et al., 2008).Vacuum was used to facilitate enzyme penetration. Samples were incubated withcontinuous agitation at 35°C for at least 14 d and later separated from the epi-dermis, rinsed in distilled water, and stored under dry conditions. Isolated cuticles,which include cutin, waxes, phenolics, and a small amount of polysaccharides,were used to analyze their components and biomechanical properties.

Cuticle Components Analysis

Cuticle components (i.e. waxes, cutin, polysaccharides, and phenolics) weregravimetrically estimated from 10 samples per genotype after selective re-moval. Total waxes were extracted by heating at 50°C the isolated cuticles for2 h in chloroform:methanol (2:1, v/v). Polysaccharides were removed afterrefluxing dewaxed cuticles in a 6 M HCl solution for 12 h, hence leaving thecutin isolates intact (Domínguez et al., 2009). Phenolic compounds were extractedafter cutin depolymerization in 1% (w/v) KOH/methanol at 65°C for 16 h by


Giménez et al.


http://tgrc.ucdavis.edu/

http://www.plantphysiol.org/cgi/content/full/pp.15.00469/DC1


measuring the absorbance of the solution at 324 nm in a UV-visible light spec-trophotometer. The amount of phenolics was estimated from a calibration curve ofa naringenin solution dissolved in 1% (w/v) KOH/methanol (Domínguez et al.,2009).

Biomechanical Tests of Isolated Cuticles

Mechanical properties of fruit cuticles were measured following the pro-tocols already described by Matas et al. (2005). Isolated cuticles were inspectedwith a microscope to confirm the absence of cracks; afterward, rectangularsegments (3 mm 3 9 mm) were cut with the aid of a metal block. Thesesegments were fixed between the ends of two hollow stainless-steel needles sothat the cuticle formed a flat surface. Each sample was held inside the envi-ronmentally controlled extensometer chamber for at least 30 min to equilibratethe temperature and humidity before the extension test.

The cross-sectional area of the samples and the length of the exposed surfaceof the sample between the two supports were measured prior to the mechanicalextension tests. Mechanical tests were performed as a transient creep test todetermine changes in the length of a cuticle segment by maintaining samples inuniaxial tension, under a constant load, for 1,200 s, during which the longi-tudinal extension of each sample was recorded every 3 s by a computer system.Each sample was tested repeatedly using an ascending sequence of sustainedtensile forces (from 0.098 N to the breaking point by 0.098 N load increments)without recovery time (Matas et al., 2005). The tensile force exerted along thesample was divided by the cross-sectional area of the sample in order to de-termine the stresses. The stress-strain curve was obtained after plotting theapplied stress against the total change in length after 20 min. Young’s mo-dulus, a measure of stiffness, breaking stress, force needed to break the cuticle,maximum strain, and deformation attained by the cuticle before breakingwere determined for each sample. Strain time and the corresponding stress-strain curves were calculated for a set of five to seven samples of cuticles at25°C and 40% relative humidity.

Statistics

Mean comparison was used to determine whether the measured charac-teristics of the cuticle varied significantly among genotypes. Analyses wereperformed using the SPSS software package (IBM SPSS 2010), and data arepresented as means 6 SE.

Supplemental Data

The following supplemental materials are available.

Supplemental Table S1. Primers used for real-time quantitative PCRassays.

ACKNOWLEDGMENTS

We thank the Campus de Excelencia Internacional Agroalimentario for theuse of equipment and facilities.

Received March 24, 2015; accepted May 27, 2015; published May 27, 2015.

LITERATURE CITED

Adato A, Mandel T, Mintz-Oron S, Venger I, Levy D, Yativ M, Domínguez E,Wang Z, De Vos RC, Jetter R, et al (2009) Fruit-surface flavonoid accumu-lation in tomato is controlled by a SlMYB12-regulated transcriptional net-work. PLoS Genet 5: e1000777

Aharoni A, Dixit S, Jetter R, Thoenes E, van Arkel G, Pereira A (2004) TheSHINE clade of AP2 domain transcription factors activates wax bio-synthesis, alters cuticle properties, and confers drought tolerance whenoverexpressed in Arabidopsis. Plant Cell 16: 2463–2480

Andrews J, Adams SR, Burton KS, Edmondson RN (2002) Partial purifi-cation of tomato fruit peroxidase and its effect on the mechanicalproperties of tomato fruit skin. J Exp Bot 53: 2393–2399

Anterola AM, Lewis NG (2002) Trends in lignin modification: a compre-hensive analysis of the effects of genetic manipulations/mutations onlignification and vascular integrity. Phytochemistry 61: 221–294

Ariizumi T, Shinozaki Y, Ezura H (2013) Genes that influence yield intomato. Breed Sci 63: 3–13

Ballester AR, Molthoff J, de Vos R, Hekkert B, Orzaez D, Fernández-Moreno JP, Tripodi P, Grandillo S, Martin C, Heldens J, et al (2010)Biochemical and molecular analysis of pink tomatoes: deregulated ex-pression of the gene encoding transcription factor SlMYB12 leads to pinktomato fruit color. Plant Physiol 152: 71–84

Bargel H, Neinhuis C (2004) Altered tomato (Lycopersicon esculentum Mill.)fruit cuticle biomechanics of a pleiotropic non ripening mutant. J PlantGrowth Regul 23: 61–75

Bargel H, Neinhuis C (2005) Tomato (Lycopersicon esculentum Mill.) fruitgrowth and ripening as related to the biomechanical properties of fruitskin and isolated cuticle. J Exp Bot 56: 1049–1060

Bemer M, Karlova R, Ballester AR, Tikunov YM, Bovy AG, Wolters-ArtsM, Rossetto PdeB, Angenent GC, de Maagd RA (2012) The tomatoFRUITFULL homologs TDR4/FUL1 and MBP7/FUL2 regulate ethylene-independent aspects of fruit ripening. Plant Cell 24: 4437–4451

Besseau S, Hoffmann L, Geoffroy P, Lapierre C, Pollet B, Legrand M(2007) Flavonoid accumulation in Arabidopsis repressed in lignin syn-thesis affects auxin transport and plant growth. Plant Cell 19: 148–162

Czerednik A, Busscher M, Bielen BA, Wolters-Arts M, de Maagd RA,Angenent GC (2012) Regulation of tomato fruit pericarp developmentby an interplay between CDKB and CDKA1 cell cycle genes. J Exp Bot 63:2605–2617

Dellapenna D, Alexander DC, Bennett AB (1986) Molecular cloning oftomato fruit polygalacturonase: analysis of polygalacturonase mRNAlevels during ripening. Proc Natl Acad Sci USA 83: 6420–6424

Domínguez E, Cuartero J, Heredia A (2011) An overview on plant cuticlebiomechanics. Plant Sci 181: 77–84

Domínguez E, España L, López-Casado G, Cuartero J, Heredia A (2009)Biomechanics of isolated tomato (Solanum lycopersicum) fruit cuticlesduring ripening: the role of flavonoids. Funct Plant Biol 36: 613–620

Domínguez E, Fernández MD, Hernández JC, Parra JP, España L, HerediaA, Cuartero J (2012) Tomato fruit continues growing while ripening,affecting cuticle properties and cracking. Physiol Plant 146: 473–486

Domínguez E, López-Casado G, Cuartero J, Heredia A (2008) Develop-ment of fruit cuticle in cherry tomato (Solanum lycopersicum). Funct PlantBiol 35: 403–411

Ellul P, Garcia-Sogo B, Pineda B, Ríos G, Roig LA, Moreno V (2003) Theploidy level of transgenic plants in Agrobacterium-mediated transfor-mation of tomato cotyledons (Lycopersicon esculentum Mill.) is geno-type and procedure dependent. Theor Appl Genet 106: 231–238

España L, Heredia-Guerrero JA, Reina-Pinto JJ, Fernández-Muñoz R,Heredia A, Domínguez E (2014a) Transient silencing of CHALCONESYNTHASE during fruit ripening modifies tomato epidermal cells andcuticle properties. Plant Physiol 166: 1371–1386

España L, Heredia-Guerrero JA, Segado P, Benítez JJ, Heredia A, Domínguez E(2014b) Biomechanical properties of tomato (Solanum lycopersicum) fruitcuticle during development are modulated by changes in the relativeamounts of its components. New Phytol 202: 790–802

Fray RG, Grierson D (1993) Identification and genetic analysis of normaland mutant phytoene synthase genes of tomato by sequencing, com-plementation and co-suppression. Plant Mol Biol 22: 589–602

Fujisawa M, Shima Y, Higuchi N, Nakano T, Koyama Y, Kasumi T, Ito Y(2012) Direct targets of the tomato-ripening regulator RIN identified bytranscriptome and chromatin immunoprecipitation analyses. Planta 235:1107–1122

Gillaspy G, Ben-David H, Gruissem W (1993) Fruits: a developmentperspective. Plant Cell 5: 1439–1451

Giménez E, Pineda B, Capel J, Antón MT, Atarés A, Pérez-Martín F,García-Sogo B, Angosto T, Moreno V, Lozano R (2010) Functionalanalysis of the Arlequin mutant corroborates the essential role of theArlequin/TAGL1 gene during reproductive development of tomato. PLoSONE 5: e14427

Giovannoni JJ (2004) Genetic regulation of fruit development and ripening.Plant Cell (Suppl) 16: S170–S180

Harriman RW, Tieman DM, Handa AK (1991) Molecular cloning of tomatopectin methylesterase gene and its expression in Rutgers, ripeninginhibitor, nonripening, and never ripe tomato fruits. Plant Physiol 97:80–87

Hen-Avivi S, Lashbrooke J, Costa F, Aharoni A (2014) Scratching thesurface: genetic regulation of cuticle assembly in fleshy fruit. J Exp Bot65: 4653–4664




http://www.plantphysiol.org/cgi/content/full/pp.15.00469/DC1


Hunt GM, Baker EA (1980) Phenolic constituents of tomato fruit cuticles.Phytochemistry 19: 1415–1419

Isaacson T, Kosma DK, Matas AJ, Buda GJ, He Y, Yu B, Pravitasari A,Batteas JD, Stark RE, Jenks MA, et al (2009) Cutin deficiency in thetomato fruit cuticle consistently affects resistance to microbial infectionand biomechanical properties, but not transpirational water loss. Plant J60: 363–377

Itkin M, Seybold H, Breitel D, Rogachev I, Meir S, Aharoni A (2009)TOMATO AGAMOUS-LIKE 1 is a component of the fruit ripeningregulatory network. Plant J 60: 1081–1095

Javelle M, Vernoud V, Rogowsky PM, Ingram GC (2011) Epidermis: theformation and functions of a fundamental plant tissue. New Phytol 189:17–39

Jeffree CE (2006) The fine structure of the cuticle. In M Riederer, C Müller, eds,Biology of the Plant Cuticle. Blackwell Publishing, Oxford, pp 11–125

Kannangara R, Branigan C, Liu Y, Penfield T, Rao V, Mouille G, Höfte H,Pauly M, Riechmann JL, Broun P (2007) The transcription factor WIN1/SHN1 regulates cutin biosynthesis in Arabidopsis thaliana. Plant Cell 19:1278–1294

Karlova R, Rosin FM, Busscher-Lange J, Parapunova V, Do PT, FernieAR, Fraser PD, Baxter C, Angenent GC, de Maagd RA (2011) Tran-scriptome and metabolite profiling show that APETALA2a is a majorregulator of tomato fruit ripening. Plant Cell 23: 923–941

Klee HJ, Giovannoni JJ (2011) Genetics and control of tomato fruit rip-ening and quality attributes. Annu Rev Genet 45: 41–59

Kosma DK, Parsons EP, Isaacson T, Lü S, Rose JKC, Jenks MA (2010)Fruit cuticle lipid composition during development in tomato ripeningmutants. Physiol Plant 139: 107–117

Leide J, Hildebrandt U, Reussing K, Riederer M, Vogg G (2007) The de-velopmental pattern of tomato fruit wax accumulation and its impact oncuticular transpiration barrier properties: effects of a deficiency in ab-ketoacyl-coenzyme A synthase (LeCER6). Plant Physiol 144: 1667–1679

Leseberg CH, Eissler CL, Wang X, Johns MA, Duvall MR, Mao L (2008)Interaction study of MADS-domain proteins in tomato. J Exp Bot 59:2253–2265

Li X, Bonawitz ND, Weng JK, Chapple C (2010) The growth reductionassociated with repressed lignin biosynthesis in Arabidopsis thaliana isindependent of flavonoids. Plant Cell 22: 1620–1632

Lin Z, Hong Y, Yin M, Li C, Zhang K, Grierson D (2008) A tomato HD-Ziphomeobox protein, LeHB-1, plays an important role in floral organo-genesis and ripening. Plant J 55: 301–310

López-Casado G, Matas AJ, Domínguez E, Cuartero J, Heredia A (2007)Biomechanics of isolated tomato (Solanum lycopersicum L.) fruit cuticles:the role of the cutin matrix and polysaccharides. J Exp Bot 58: 3875–3883

Lunkenbein S, Coiner H, de Vos CH, Schaart JG, Boone MJ, Krens FA,Schwab W, Salentijn EM (2006) Molecular characterization of a stableantisense chalcone synthase phenotype in strawberry (Fragaria 3 ana-nassa). J Agric Food Chem 54: 2145–2153

Manning K, Tör M, Poole M, Hong Y, Thompson AJ, King GJ, Giovannoni JJ,Seymour GB (2006) A naturally occurring epigenetic mutation in a geneencoding an SBP-box transcription factor inhibits tomato fruit ripening. NatGenet 38: 948–952

Martel C, Vrebalov J, Tafelmeyer P, Giovannoni JJ (2011) The tomatoMADS-box transcription factor RIPENING INHIBITOR interacts withpromoters involved in numerous ripening processes in a COLORLESSNONRIPENING-dependent manner. Plant Physiol 157: 1568–1579

Matas AJ, Cuartero J, Heredia A (2004) Phase transitions in the bio-polyester cutin isolated from tomato fruit cuticles. Thermochim Acta409: 165–168

Matas AJ, López-Casado G, Cuartero J, Heredia A (2005) Relative hu-midity and temperature modify the mechanical properties of isolatedtomato fruit cuticles. Am J Bot 92: 462–468

Matas AJ, Yeats TH, Buda GJ, Zheng Y, Chatterjee S, Tohge T, Ponnala L,Adato A, Aharoni A, Stark R, et al (2011) Tissue- and cell-type specifictranscriptome profiling of expanding tomato fruit provides insights into

metabolic and regulatory specialization and cuticle formation. Plant Cell23: 3893–3910

Mintz-Oron S, Mandel T, Rogachev I, Feldberg L, Lotan O, Yativ M,Wang Z, Jetter R, Venger I, Adato A, et al (2008) Gene expression andmetabolism in tomato fruit surface tissues. Plant Physiol 147: 823–851

Nadakuduti SS, Pollard M, Kosma DK, Allen C Jr, Ohlrogge JB, Barry CS(2012) Pleiotropic phenotypes of the sticky peel mutant provide new in-sight into the role of CUTIN DEFICIENT2 in epidermal cell function intomato. Plant Physiol 159: 945–960

O’Neill SD, Tong Y, Spörlein B, Forkmann G, Yoder JI (1990) Moleculargenetic analysis of chalcone synthase in Lycopersicon esculentum and ananthocyanin-deficient mutant. Mol Gen Genet 224: 279–288

Oshima Y, Shikata M, Koyama T, Ohtsubo N, Mitsuda N, Ohme-TakagiM (2013) MIXTA-like transcription factors and WAX INDUCER1/SHINE1 coordinately regulate cuticle development in Arabidopsis andTorenia fournieri. Plant Cell 25: 1609–1624

Petit J, Bres C, Just D, Garcia V, Mauxion JP, Marion D, Bakan B, JoubèsJ, Domergue F, Rothan C (2014) Analyses of tomato fruit brightnessmutants uncover both cutin-deficient and cutin-abundant mutants and anew hypomorphic allele of GDSL lipase. Plant Physiol 164: 888–906

Petracek PD, Bukovac MJ (1995) Rheological properties of enzymaticallyisolated tomato fruit cuticle. Plant Physiol 109: 675–679

Pineda B, Giménez-Caminero E, García-Sogo B, Antón MT, Atarés A,Capel J, Lozano R, Angosto T, Moreno V (2010) Genetic and physio-logical characterization of the arlequin insertional mutant reveals a keyregulator of reproductive development in tomato. Plant Cell Physiol 51:435–447

Pollard M, Beisson F, Li Y, Ohlrogge JB (2008) Building lipid barriers:biosynthesis of cutin and suberin. Trends Plant Sci 13: 236–246

Qin G, Wang Y, Cao B, Wang W, Tian S (2012) Unraveling the regulatorynetwork of the MADS box transcription factor RIN in fruit ripening.Plant J 70: 243–255

Saladié M, Matas AJ, Isaacson T, Jenks MA, Goodwin SM, Niklas KJ,Xiaolin R, Labavitch JM, Shackel KA, Fernie AR, et al (2007) Areevaluation of the key factors that influence tomato fruit softening andintegrity. Plant Physiol 144: 1012–1028

Seymour G, Poole M, Manning K, King GJ (2008) Genetics and epigeneticsof fruit development and ripening. Curr Opin Plant Biol 11: 58–63

Seymour GB, Chapman NH, Chew BL, Rose JK (2013) Regulation ofripening and opportunities for control in tomato and other fruits. PlantBiotechnol J 11: 269–278

Shi JX, Adato A, Alkan N, He Y, Lashbrooke J, Matas AJ, Meir S, Malitsky S,Isaacson T, Prusky D, et al (2013) The tomato SlSHINE3 transcription factorregulates fruit cuticle formation and epidermal patterning. New Phytol 197:468–480

Smith C, Watson C, Ray J, Bird C, Morris P (1988) Antisense RNA inhi-bition of polygalacturonase gene expression in transgenic tomatoes. Na-ture 334: 724–726

Tigchelaar EC, Tomes M, Kerr EA, Barman RJ (1973) A new fruit ripeningmutant, non-ripening (nor). Rep Tomato Genet Coop 23: 33–34

Vanholme R, Demedts B, Morreel K, Ralph J, Boerjan W (2010) Ligninbiosynthesis and structure. Plant Physiol 153: 895–905

Vogg G, Fischer S, Leide J, Emmanuel E, Jetter R, Levy AA, Riederer M(2004) Tomato fruit cuticular waxes and their effects on transpirationbarrier properties: functional characterization of a mutant deficient in avery-long-chain fatty acid b-ketoacyl-CoA synthase. J Exp Bot 55: 1401–1410

Vrebalov J, Pan IL, Arroyo AJ, McQuinn R, Chung M, Poole M, Rose J,Seymour G, Grandillo S, Giovannoni J, et al (2009) Fleshy fruit ex-pansion and ripening are regulated by the tomato SHATTERPROOFgene TAGL1. Plant Cell 21: 3041–3062

Vrebalov J, Ruezinsky D, Padmanabhan V, White R, Medrano D, DrakeR, Schuch W, Giovannoni J (2002) A MADS-box gene necessary for fruitripening at the tomato ripening-inhibitor (rin) locus. Science 296: 343–346

Wang Y, Chantreau M, Sibout R, Hawkins S (2013) Plant cell wall ligni-fication and monolignol metabolism. Front Plant Sci 4: 220


Giménez et al.



transcriptional activity of the mads boxarlequin … · per family that regulates cutin...

Documents