Nca

INLa

b

Qc

Nd

B

a

ARRA

KAACLPV

1

wH

gct

h0

Scientia Horticulturae 175 (2014) 111–120

Contents lists available at ScienceDirect

Scientia Horticulturae

journa l h om epage: www.elsev ier .com/ locate /sc ihor t i

ear-isogenic lines enhancing ascorbic acid, anthocyanin andarotenoid content in tomato (Solanum lycopersicum L. cv Micro-Tom)s a tool to produce nutrient-rich fruits

van Sestari a,1, Agustín Zsögönb,1, Gabriel Garcia Rehderb, Luciane de Lira Teixeirac,euza Mariko Aymoto Hassimottoc, Eduardo Purgattoc, Vagner Augusto Beneditod,ázaro Eustáquio Pereira Peresb,∗

Universidade Federal de Santa Catarina, (UFSC), Campus Curitibanos, Rodovia Ulysses Gaboardi, Km 03, CP 101, 89520-000, Curitibanos, SC, BrazilLaboratório de Controle Hormonal do Desenvolvimento Vegetal, Departamento de Ciências Biológicas (LCB), Escola Superior de Agricultura Luiz deueiroz (ESALQ), Universidade de São Paulo (USP), Av. Pádua Dias, 11, CP 09, 13418-900, Piracicaba, SP, BrazilLaboratório de Quı′mica, Bioquı′mica e Biologia Molecular de Alimentos, Departamento de Alimentos e Nutric ão Experimental (FCF) and NAPAN, Food andutrition Research Center, USP, Av. Prof. Lineu Prestes 580, Bloco 14, 05508-900, São Paulo, SP, BrazilGenetics and Developmental Biology Program, Division of Plant & Soil Sciences, 2090 Agricultural Sciences Building, West Virginia University (WVU), P.O.ox 6108, Morgantown, WV 26506-6108, USA

r t i c l e i n f o

rticle history:eceived 26 March 2014eceived in revised form 9 May 2014ccepted 2 June 2014

eywords:nthocyaninsntioxidantsarotenoidsycopeneurple tomato

a b s t r a c t

A great emphasis of plant research is being placed on developing crops with increased nutritional value.Generating plant materials suitable for controlled studies as potential tools for pre-breeding is still themain hurdle. Improvements are needed in generating different allele combinations to stack various nutri-ents into a single genotype, without losses in fruit yield or quality, and in testing the specific effects ofnutrients in their original matrix, avoiding the noise caused by the characteristic mix of compounds.An elegant approach in both pre-breeding and diet supplementation tests is the use of near-isogeniclines (NILs). Here, we tap on the large pool of monogenic mutants and natural genetic variation avail-able in tomato to create a series of NILs in the genetic background of the cultivar Micro-Tom (MT). Wedescribe the introgression of the mutations Anthocyanin fruit (Aft), atroviolacium (atv), Aubergine (Abg),Beta-carotene (B), old-gold crimson (og) and high pigment 1 and 2 (hp1, hp2) and characterize their fruit

itamin C metabolic profiles in single, double and triple mutant combinations. We show that Brix can be raisedwithout yield penalty, along with increases in lycopene, �-carotene and ascorbic acid, and a concomi-tant enhancement of anti-oxidant capacity. As proof-of-concept of the suitability of stacking alleles forbreeding nutrient-rich tomatoes, we introduce three mutations leading to uniformly purple fruits andenhanced nutrient contents from MT into a commercial cherry tomato cultivar.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

There is considerable evidence associating healthy eating habitsith a reduced risk of chronic disease and obesity (WHO, 2003).igh consumption of fruits and vegetables, in particular, is strongly

Abbreviations: AA, ascorbic acid; DPPH, 1,1-diphenyl-2-picrylhydrazyl; GAE,allic acid equivalents; GM, genetically modified; HPLC, high-performance liquidhromatography; MT, Micro-Tom; NILs, near-isogenic lines; SSC, soluble solids con-ent.∗ Corresponding author. Tel.: +55 19 34294052; fax: +55 19 34224855.

E-mail address: lazaro.peres@usp.br (L.E.P. Peres).1 Contributed equally.

ttp://dx.doi.org/10.1016/j.scienta.2014.06.010304-4238/© 2014 Elsevier B.V. All rights reserved.

correlated with significant reductions in cardiovascular disease andcancer (Hung et al., 2004). As public awareness of the importanceof switching to plant-based diets grows, a greater emphasis ofplant research will be placed on developing crops with increasednutritional value (De Sa and Lock, 2008). The development ofnutrient-rich vegetables and fruits requires improvement in both(i) the capacity to test different combinations of alleles (pre-breeding) in order to stack different kinds of nutrients into a singlegenotype, ideally, with no loss in fruit yield or quality, and (ii)the capacity to test the effectiveness of the genotypes produced.

The latter is usually achieved through diet experiments in eitheranimal models or humans, which have the downside that identi-fying and characterizing the contributions of any single compoundin a mixture of nutrients is difficult. The alternative approach of

112 I. Sestari et al. / Scientia Horticulturae 175 (2014) 111–120

Table 1Mutations introgressed into the cv. Micro-Tom background.

Mutant Gene function Origin Reference

Beta carotene (B) Chromoplast-specific lycopene �-cyclase (Cyc-B),which converts most fruit lycopene into �-carotene

LA1401 (S. galapagense) Lincoln and Porter (1950) and Pecker et al. (1996)

crimson (og) Defective for a chromoplast-specific lycopene�-cyclase (Cyc-B), which abolishes the conversion oflycopene into �-carotene; a null allele of B

cv. Ourovelho Mustilli et al. (1999) and Mohr (1979)

atroviolacium (atv) Natural variation from S. cheesmaniae, probably anon-functional allele of a negative regulator ofphotomorphogenesis

LA0797 Hybrid Kendrick et al. (1994)

high pigment1 (hp1) Defective for a gene homologous to DDB1A ofArabidopsis, which codes for a protein interacting withDET1 (HP2), a repressor of photomorphogenesis

LA3004 cv. Rutgers Lieberman et al. (2004) and Peters et al. (1989)

high pigment 2 (hp2) Defective for a gene homologous to DET1 ofArabidopsis, a negative repressor ofphotomorphogenesis

LA2451 cv. Manapal Mustilli et al. (1999)

Anthocyanin fruit (Aft) Natural variation from S. chilense, probably coding for LA1996 Jones et al. (2003) and Schreiber et al. (2012)

LA

tttc2oiaaNc

bsaStpfmela1bva(2

t(1MraWcpdtwcfet

an R2R3 MYB transcription factorAubergine (Abg) Natural variation from S. lycopersicoides, probably

allelic to Aft

esting purified compounds in diets, on the other hand, hampershe study of molecules in the particular chemical forms in whichhey are most commonly consumed, and in their original biochemi-al matrix, which has a large impact on bioavailability (Martin et al.,011). An elegant alternative to overcome these hurdles is the usef near-isogenic lines (NILs) (Martin et al., 2011). NILs are genet-cally identical lines, except for one or a few loci, and represent

powerful tool to carry out detailed analyses of the physiologicalnd molecular basis of key traits within a fixed genetic background.ILs have been used extensively as a pre-breeding tool for variousrops (Edmeades et al., 2004).

Isogenic lines with increased nutrient contents have alreadyeen produced in tomato by expressing the snapdragon tran-cription factors Delila (Del) and Rosea1 (Ros1), which led to theccumulation of anthocyanins in the fruit (Butelli et al., 2008a).upplementing the diet of Trp53−/− mice with anthocyanin-richomatoes increased their lifespan (Butelli et al., 2008a) when com-ared to the isogenic control. In that work the authors profitedrom the short life cycle, reduced adult size and genetic transfor-

ation capability of the tomato cultivar Micro-Tom (MT) (Meissnert al., 1997). The genetic resources available in tomato, such as aarge collection of monogenic mutants (http://tgrc.ucdavis.edu/)nd a wide pool of natural genetic variation (Stevens and Rick,986), have also been exploited in MT through conventionalreeding (Campos et al., 2010). In our laboratory, we have pre-iously used NILs in MT, differing only in one or a few alleles, toddress diverse questions of plant physiology and developmentCampos et al., 2009; Carvalho et al., 2011; Lombardi-Crestana et al.,012).

Here, we aimed at creating nutrient-rich tomato NILs carryinghe mutations Anthocyanin fruit (Aft), atroviolacium (atv), AubergineAbg), Beta-carotene (B), old-gold crimsom (og) and high pigment

and 2 (hp1, hp2) from their original genetic backgrounds intoT. We characterized the metabolic profiles of the fruits in the

esulting NILs in single, double and triple mutant combinations,nd compared their yield and total soluble solids content (Brix).e suggest that such a collection in a single genetic background

ould be useful for diet supplementation studies, and also as are-breeding tool to test allele combinations with the aim of pro-ucing nutrient-rich commercial varieties. As proof-of-concept ofhe latter, we combined three mutations to produce a genotypeith nutritionally valuable traits, such as increased levels of antho-

yanins, �-carotene, lycopene and vitamin C. Lastly, we prove theeasibility of producing a commercial crop with added health ben-fits by introducing all three mutations into a commercial cherryomato cultivar (VFNT).

3668 Kerckhoffs et al. (1997) and Rick et al. (1994)

2. Material and methods

2.1. Plant material and growth conditions

The near-isogenic lines (NILs) in Solanum lycopersicum L. cv.Micro-Tom (MT) carrying the alleles of interest (Table 1) were pro-duced as described in Fig. 1. Briefly, S. lycopersicum cv Micro-Tom(MT) and the cultivar carrying the mutation (in Fig. 1 exemplifiedby atv) were crossed, using the former as the female parent. TheF1 was backcrossed (BC) using MT as a recurrent female parentand the resulting BC1 seedlings were screened for characteristicdwarfism of MT (Martí et al., 2006). Selected plants were selfed,producing BC1F2 seeds. The process was repeated until the theo-retical proportion of MT genome was >99% (Stam and Zeven, 1981).All experiments were performed in BC6F4 homozygous plants.

To generate the double mutants Aft/hp2, Aft/hp1 and Abg/hp2,homozygous single mutants were crossed and screened for phe-notype in the F2 generation. The triple mutant Aft/atv/hp2 wasgenerated by crossing the double mutant Aft/hp2 with the homozy-gous mutant atv. The double and triple mutants, once identified,were self-fertilized to obtain the respective populations. Plantswere grown in 8 L rectangular plastic pots containing a 1:1 mix-ture of commercial substrate (Plantmax HT, Eucatex, São Paulo,Brazil) and expanded vermiculite, supplemented with 1 g L−1 ofNPK 10:10:10 and 4 g L−1 of dolomite limestone (MgCO3 + CaCO3).Plants were kept in a greenhouse at an average mean tempera-ture of 28 ◦C; 11.5/13 h (winter/summer) photoperiod, and exposedto 250–350 �mol photons m−2 s−1 photosynthetically active radi-ation (PAR) by natural radiation reduced with a reflecting mesh(Aluminet-Polysack Industrias Ltda, Leme, SP, Brazil). At floweringtime (∼35 days after sowing), plants were sprayed twice at 14-dayintervals with 1 g L−1 Peters 20-20-20 fertilizer. Yield (total fruitweight per plant) was calculated using 12 plants. Uniformly ripefruits, at the corresponding ripening stage, were harvested and partof a sample was immediately used to determine agronomic param-eters and carotenoids content. The remainder of the sample wasused to determine phenolics, flavonoids, ascorbic acid and antiox-idant activity in different fractions of the fruit (flesh and peel).Flesh and peel were handled separately, thoroughly homogenizedby powdering in liquid N2, and stored at −70 ◦C until analysis.

2.2. Soluble solids content (SSC)

SSC was measured in the juice of 10 ripe fruits of each genotypeusing a digital refractometer with automatic temperature compen-sation (Atago PR-101, Bellevue, WA).

I. Sestari et al. / Scientia Horticulturae 175 (2014) 111–120 113

Fig. 1. Introgression of mutations into the tomato Micro-Tom (MT) cultivar. The recessive atroviolacium (atv) maps to chromosome 7 and is found in cultivar VF36 (accessionLA0797). MT was used as a pollen receptor in all crosses. In F2, recombinants with MT small size, determinate growth habit (homozygous for the self-pruning, sp, mutant allele)and atv phenotype (increased anthocyanin in stems, leaves and fruits) were selected and back-crossed to the MT parental. Plants were selfed after every second back-cross( a singe even

2

posmmsLll�aae

2

e

BC) and screened for the atv phenotype in the subsequent F2. In BC6F2 we selectedach generation is illustrated in the right-hand side bar, as calculated in Stam and Z

.3. Quantification of carotenoids

Lycopene and �-carotene extraction and quantification wereerformed as described (Nagata and Yamashita, 1992). The pericarpf fresh tomatoes was finely ground in liquid N2 using a stainlessteel mill (IKA A11 Basic). Ten milliliters of acetone–hexaneixture (4:6 v/v) was added to the tomato homogenate (1 g) andixed in a test tube. The absorbance of the supernatant was mea-

ured at 453, 505, 645 and 663 nm on a spectrophotometer (UV Visibra S22, Biochrom, Cambridge, UK). Contents of �-carotene andycopene were calculated according to the following equations:ycopene (mg/100 ml) = −0.0458A663 + 0.372A505 − 0.0806A453;-carotene (mg/100 ml) = 0.216A663 − 0.304A505 + 0.452A453. Thessays were carried out in triplicate, using nine fruits per extract,nd the results are expressed in �g mL−1 (mean values ± standardrror).

.4. Quantification of total phenolics

Powdered samples were processed using a Brinkman homog-nizer (Polytron-Kinematica GmbH, Kriens-Lucern, Switzerland)

le plant for each mutant introgressed. The theoretical proportion of MT genome in(1981).

with 70% methanol plus 5% acetic acid, stirred for 30 minat 300 rpm/4 ◦C and filtered. Total phenolics were determinedusing the Folin–Ciocalteau method (Zielinski and Kozlowska,2000). The sample absorbance was read using a spectropho-tometer set at 765 nm. Total phenolics were expressed asgallic acid equivalents (GAE). The assays were carried out intriplicate.

2.5. Ascorbic acid, dehydroascorbic acid and total vitamin C

Ascorbic acid (AA) content was determined as described byPasternak et al. (2005) with modifications. Samples were extractedwith 5% metaphosphoric acid (0.3% w/v) and analyzed by HPLC ina Hewlett-Packard 1100 system with an autosampler and a qua-ternary pump coupled with a diode array detector. The columnused was a �-Bondapack (300 mm × 3.9 mm i.d., Waters, Milford,MA), and elution (flow rate of 1.5 mL/min) was performed in iso-

cratic conditions with 2 mM KCl, pH 2.5, and monitored at 262 nm.Total AA was estimated after reduction of dehydroascorbic acid(DHA) with 10 mM dithiothreitol. The assays were carried out intriplicate.

114 I. Sestari et al. / Scientia Horticulturae 175 (2014) 111–120

Fig. 2. Uniform size and shape of fruits in introgressed mutants of tomato. Whole (a) and cross-section (b) of ripe tomato fruit and its corresponding extractable juice (c)f lines

( tion o

2

s(ApawtrTt

2

iwnhtmr1tpctred0pwo

rom selected genotypes. From left to right: Micro-Tom (MT) and the near-isogenicAft/hp2) and Anthocyanin fruit/atroviolacium/high pigment 2 (Aft/atv/hp2). A descrip

.6. Antioxidant capacity

The antioxidant capacity was measured by DPPH free radical-cavenging activity following the method of Brand-Williams et al.1995) with modifications (Maurício Duarte-Almeida et al., 2006).

0.1 mM solution of DPPH (1,1-diphenyl-2-picrylhydrazyl) wasrepared in methanol. An aliquot of 40 �L of the sample wasdded to 200 �L of this solution. The decrease in absorbanceas determined at 517 nm using a microplate spectrophotome-

er (Benchmark Plus, Biorad, Hercules, CA) when the reactioneached a plateau (after 20 min). Results were expressed as �molrolox equivalent/100 g sample. The assays were carried out inriplicate.

.7. Flavonoid content

The extraction of flavonoids was performed in duplicate accord-ng to the method of Hassimotto et al. (2007). The sample (∼5 g)

as extracted three times (100 mL the first time and 50 mL theext two times) in methanol/water/acetic acid (70:30:5 v/v). Theomogenate was filtered under reduced pressure through fil-er paper (Whatman No. 6). The extract was concentrated until

ethanol elimination under vacuum at 40 ◦C on a rotary evapo-ator (Rotavapor RE 120, Büchi, Flawil, Sweden) and made up to00 mL with distilled water for application to solid-phase extrac-ion (SPE) columns. An aliquot of the extract was passed througholyamide SC6 (Macherey-Nagel Gmbh and Co., Duren, Germany)olumns (1 g/6 mL) previously conditioned with methanol and dis-illed water. Impurities were washed out with distilled water andetained flavonoids were eluted with 0.1% HCl in methanol. Theluate was evaporated to dryness under reduced pressure at 40 ◦C,issolved in methanol: acetic acid (99:5 v/v), and filtered through a

.45 �m tetrafluoroethylene (PTFE) filter (Millipore, Bedford, MA)rior to quantification by high performance liquid chromatographyith a diode array detector (HPLC-DAD). The assays were carried

ut in triplicate.

Beta carotene (B), crimson (og), atroviolacium (atv), Anthocyanin fruit/high pigment 2f the mutants can be found in Table 1.

2.8. HPLC analysis

Identification and quantification of flavonoids were achievedusing analytical reversed-phase HPLC in a Hewlett-Packard 1260Infinity system with an autosampler and quaternary pump cou-pled with a 1100 diode array detector (Hewlett-Packard, Palo Alto,CA). The column used was a Prodigy 5 �m ODS3 (250 mm × 4.6 mmi.d., Phenomenex, Torrance, CA) and the elution solvents were: (A)0.5% formic acid and (B) acetonitrile acidified with 0.5% formic acid.The solvent gradient consisted of 10% B at the beginning, 10% at5 min, 20% at 15 min, 25% at 25 min, 35% at 33 min, 50% at 38 min,and 90% at 43 min, 90% at 45 min (run time, 45 min). Eluates weremonitored at 270, 370 and 525 nm. The flow rate was 1 mL/min,column temperature was 25 ◦C and injection volume was 5–20 �L.Calibration was performed by injecting the standards three timesat five different concentrations (r2 > 0.999). The standards werecyanidin-3-glucoside, delphinidin, naringenin, quercetin, luteolin(Extrasyntese, Genay, France), chlorogenic acid and ellagic acid(Sigma–Aldrich). Peak identification was performed by compar-ison of retention times and diode array spectral characteristicswith the standards and the library spectra. Co-chromatography wasused when necessary. Samples were injected in duplicate. Petuni-din derivates were expressed as cyanidin-3-glucoside, and otherflavonoids were expressed according to respective aglycone moi-eties. The assays were carried out in triplicate.

2.9. Liquid chromatography (LC)–electrospray ionization(ESI)-mass spectrometry (MS)

Identification of anthocyanins was carried out using anEsquires-LC mass spectrometer (MS) (Bruker Daltonics, Billerica,MA) with an electrospray ionization (ESI) interface. HPLC runconditions were the same as described above. Mass spectrome-

ter operating conditions were as follows: a capillary temperatureof 275 ◦C, source voltage of 3.5 kV and positive mode for antho-cyanins. Analyses were carried out using full scan from m/z 100 to2000.

I. Sestari et al. / Scientia Horticulturae 175 (2014) 111–120 115

Fig. 3. Fruit quality in near-isogenic mutants of tomato. Agronomic parameters and sugars content (glucose, fructose and sucrose) in the flesh and peel of ripe fruits fromd (b) aw highe

2

vi

3

3g

acTdnw(dpby

ifferent tomato genotypes. Total fruit weight per plant (a), soluble solids contenteight, n = 10 for Brix and n = 3 for sugars). Asterisks (*) denote values significantly

.10. Statistical analyses

Statistical analyses were performed by a one-way analysis ofariance (ANOVA) and Tukey’s test was applied to establish signif-cant differences between means at P ≤ 0.01.

. Results and discussion

.1. A collection of nutrient-rich mutants in the Micro-Tom (MT)enetic background

Following the scheme described in Fig. 1, we have introgressed series of mutations affecting accumulation of anthocyanins andarotenoids in the fruit (Table 1) into the tomato cultivar Micro-om (MT). The outcome is a collection of lines with fruits of highiversity in pigment composition, but otherwise very homoge-eous in attributes such as size and shape (Fig. 2). Among others,e combine the mutations Aft, atv and hp2 into a single genotype

forthwith referred to as “triple mutant”), resulting in a uniformly

ark purple fruit (Fig. 2). The effort to produce non-transgenicurple tomato fruits has already been conducted by conventionalreeding (Li et al., 2011; Povero et al., 2011; Patil and Patil, 1988),et the strength and uniformity of pigmentation achieved here are

nd sugars in the flesh (c) and in the peel (d). Data are means ± SE (n = 12 for fruitr than Micro-Tom (MT) according to Tukey’s test (P ≤ 0.01).

more reminiscent of a previously produced deep purple transgenictomato cultivar (Butelli et al., 2008b). The characteristically highirradiance required for anthocyanin production seems to have beenovercome by the photomorphogenetic effect of the hp2 mutation(Mustilli et al., 1999), as plants were grown under limited irradiance(∼300 �M photons m−2 s−1) in a glasshouse. Due to the difficulty ofintrogressing mutations into a single genetic background, in pre-vious works they had to be compared between multiple tomatocultivars (Mes et al., 2008; Sapir et al., 2008). Here, the use of MTas a common background allowed us to draw inferences about theimpact of stacking nutrients on other important parameters, suchas yield and total soluble solids (Brix) in the fruit.

3.2. Fruit yield and brix in tomato mutants

No yield penalty was observed when comparing each mutantto the MT control, and in some cases there was a trend towardsincreased yield (Fig. 3a). The amount of total soluble solids in thefruit (measured in Brix units), a parameter of great agronomic

importance in tomato, was significantly increased in lines carry-ing the hp2 mutation (Kerckhoffs et al., 1997), except when pairedwith the Abg mutation, in which case there was no difference fromthe MT control (Fig. 3b). This could be due to the presence of other

1 rticulturae 175 (2014) 111–120

gowipitttsB

3

f(am(tctaiAatftb

3

c1Ltti(wBoplIsclgatctst

shGtima

Fig. 4. Quantification of phytonutrients in the fruits of mutants. Content ofcarotenoids (lycopene and �-carotene) in the whole fruit (a) and ascorbic acid (AsA),dehydroascorbic acid (DHA) and total vitamin C in the flesh (b) and peel (c) of ripe

16 I. Sestari et al. / Scientia Ho

enes by linkage drag in Abg/hp2, as S. lycopersicoides, the donorf Abg, carries a paracentric inversion in the chromosome regionhere Abg is located (Canady et al., 2005), which precludes the

ntrogression of a small stable chromosome segment. Both highigment mutants tend to accumulate large amounts of chlorophylln their unripe fruit (Kerckhoffs et al., 1997), so it is possible thathe higher Brix in hp2 was due to photosynthesis taking place inhe fruit, as recently proposed for other dark fruit tomato geno-ypes (Powell et al., 2012). We next assessed the composition ofugars and quantified them in order to dissect the differences inrix observed among genotypes.

.3. Sugar composition is altered in the fruit of different mutants

Tomato normally accumulates hexoses and not sucrose in theruit (Chetelat et al., 1993), which may increase its nutritional valueMilton, 1999). We quantified sugars in flesh and peel and observed

reduction in the amount of hexoses in both Aft/hp2 and the tripleutant (Fig. 3c and d). Since these genotypes presented a high Brix

Fig. 3a), a possible interpretation is that Brix in both Aft/hp2 andhe triple mutant is due to an increase in organic acids or otheromponents and not hexoses in the fruit. A marginal, though sta-istically significant, increase in sucrose is observed in both the fleshnd peel of B and hp2 mutants (Fig. 3c and d). As for the hexoses,ncreased glucose contents were observed in the peel of B, hp2 andft/hp1. Enhanced levels of fructose were found in the flesh of hp2nd in the peel of og. Interestingly, a significantly inverse correla-ion (r2 = 0.5280) was observed between the fructose content in theruit flesh (Fig. 3c) and yield (Fig. 3a). This correlation was strongerhan the correlation between yield and Brix (r2 = 0.4584, Fig. 3a and), which is well-known in tomato.

.4. Increased lycopene, ˇ-carotene and vitamin C in mutants

The tomato fruit is rich in the carotenoids �-carotene (a pre-ursor of vitamin A and powerful antioxidant (Paiva and Russell,999)) and lycopene (from which it derives its red color (Shi ande Maguer, 2000)). Consumption of either is associated with mul-iple health benefits (Hadley et al., 2003; Boileau, 2003). Wild-typeomatoes normally convert a small amount of lycopene in the fruitnto �-carotene through a chromoplast-specific lycopene �-cyclaseCyc-B). The B mutant showed a three-fold increase in �-carotenehen compared to MT (Fig. 4a), which was expected, given that the

allele leads to overexpression of Cyc-B (Ronen et al., 2000). Theg allele, on the other hand, encodes a non-functional �-cyclase,recluding the conversion to �-carotene and leading to enhanced

evels of lycopene in the fruit (Long et al., 2006; Araújo et al., 2002).n this work, we probably introgressed a weak mutant allele of og, asuggested by both the absence of significant differences in lycopeneontent and the fruit color compared to MT (Figs. 2 and 4a). Atved to an accumulation of both lycopene and �-carotene, but thereatest increase was shown by the triple mutant, possibly due to

synergistic effect of the atv and hp2 mutations. Indeed, all geno-ypes harboring hp1 or hp2 enhanced carotenoid content, whichan be attributed to the hypersensitive response of both genotypeso the light (Kerckhoffs et al., 1997; Kendrick et al., 1994), with aubsequent increase in the accumulation of antioxidants as well ashe formation of chloroplasts (and later, chromoplasts) in the fruit.

Ascorbic acid protects against lipid peroxidation by acting as acavenger of reactive-oxygen species (ROS) and by reducing lipidydroperoxyl radicals via the vitamin E redox cycle (Halliwell andutteridge, 2007). It is well known that both hp1 and hp2 lead

o higher vitamin C contents in the fruit than their correspond-ng wild-type alleles (Mochizuki et al., 1988; Bino et al., 2005). We

easured ascorbic acid (AsA), its oxidized form dehydroascorbiccid (DHA) and total vitamin C contents in fruit tissues of tomato

fruits from different tomato genotypes. Data are represented as means ± SE (n = 3for carotenoids and ascorbic acid). Asterisks (*) denote values significantly higherthan Micro-Tom (MT) according to Tukey’s test (P ≤ 0.01).

and confirmed an increase in all genotypes carrying either hp1 orhp2 (Fig. 4b and c). Interestingly, the B mutation also led to a con-siderable (∼30%) increase in vitamin C in the fruit flesh, but notin the peel, as did the hp2 mutation by itself (Fig. 4b and c). In thetriple mutant, the increase in vitamin C in the fruit flesh was almosttwo-fold compared to the MT control line. This result, coupledwith its increased levels of carotenoids (Fig. 4a), makes the triplemutant Aft/atv/hp2 a promising combination as a general accumu-lator of relevant phytonutrients. We next assessed whether the

intense purple color of the triple mutant results from increased lev-els of flavonoids, a third group of antioxidants. Whereas carotenoidsare lipophilic and accumulate in plastids, flavonoids are water-soluble phenylpropanoids with antioxidant capacity that use to

rticulturae 175 (2014) 111–120 117

aacsca

3m

psvc2fiiWacp(aafOBFwdmdt

Fig. 5. Increased anthocyanin contents in the fruits of the triple mutant. HPLC-DAD chromatogram from peel sections of ripe fruits from Micro-Tom (MT) (a) andAnthocyanin fruit/atroviolacium/high pigment 2 (Aft/atv/hp2) (b) genotypes. Peaks: 1,

TP

Vo

TP

Vo

I. Sestari et al. / Scientia Ho

ccumulate in vacuoles. The differences in chemical compositionnd subcellular accumulation between carotenoids and antho-yanins suggest that they are non-competitive nutrients that can betacked in the same fruit. Foods rich in both groups of antioxidantompounds most likely offer the best protection against diseasend aging (Yeum et al., 2004).

.5. Increased anthocyanin content and antioxidant capacity inutants

Besides their well-known antioxidant capacity, anthocyaninsresent additional health benefits, such as prevention of obe-ity, diabetes and cardiovascular diseases, and improvement ofisual and brain functions through modulation of signaling cas-ades and gene expression (Tsuda, 2012; De Pascual-Teresa et al.,010; Jayaprakasam et al., 2005). Thus, we investigated the pro-les of the anthocyanin precursor naringenin and anthocyanidins

n tomato fruit tissues (exocarp and mesocarp) at full maturity.e also assessed the content of the phenolic compounds quercetin

nd luteolin along with the total antioxidant capacity. The antho-yanin composition in tomato mutants was characterized by theresence of two aglycones, delphinidin (m/z 303) and petunidinm/z 317), mainly in acylated form (Fig. 5 and Table S.1). Amongnthocyanidins, petunidin is not usually synthesized in vegetablesnd fruits, and little is known about its health benefits. It has beenound in blueberry (18–60 mg/100 g FW, mainly as petunidin 3--galactoside and petunidin 3-O-arabinoside) (Harnly et al., 2006;unea et al., 2013), in jambolan (Syzygium cumini) (68 mg/100 gW, mainly as petunidin 3,5-diglucoside) and at low levels in redine (Rivero-Pérez et al., 2008). Red wine fractions rich in petuni-

in 3-glucoside improved the growth rate of S. cerevisiae knockoututants (lacking genes involved in redox metabolism and stress

efense) in the presence of pro-oxidants. In this yeast model sys-em, petunidin induced a reversible nuclear translocation of the

able 2henolics contents and antioxidant capacity (DPPH) in the flesh of mutants in the cv. Mic

Genotypes Chlorogenic acid(mg 100 g−1 FW)

Quercetin(mg 100 g−1 FW)

MT 0.94 ± 0.06f 0.50 ± 0.05e

B 4.48 ± 0.20d 0.87 ± 0.05de

atv 7.32 ± 0.39a 0.97 ± 0.08de

hp1 6.57 ± 0.24b 1.84 ± 0.11bc

hp2 – 1.73 ± 0.06bc

Abg/hp1 7.96 ± 0.75a 5.03 ± 1.02a

Abg/hp2 5.14 ± 0.43c 1.42 ± 0.04 cd

Aft/hp1 6.83 ± 0.21b 1.82 ± 0.05bc

Aft/hp2 2.89 ± 0.09e 1.52 ± 0.07 cd

Aft/atv/hp2 – 2.24 ± 0.08b

alues shown are mean ± SD of three determinations. Different letters in the same columnly detected in trace quantities; “nd” means not detected.

able 3henolics contents and antioxidant capacity (DPPH) in the peel of mutants in the cv. Micr

Genotypes Chlorogenic acid(mg 100 g−1 FW)

Quercetin(mg 100 g−1 FW)

L(m

MT – 80.12 ± 5.61f 4B nd 71.25 ± 3.14f 4atv – 101.59 ± 7.84de 6hp1 8.58 ± 0.39a 7.25 ± 0.57h 6hp2 – 127.18 ± 10.81d 4Abg/hp1 8.43 ± 0.15a 82.25 ± 5.53f 5Abg/hp2 6.36 ± 0.60b 49.80 ± 1.24g 2Aft/hp1 nd 276.15 ± 15.10a 2Aft/hp2 – 207.15 ± 2.96b 1Aft/atv/hp2 – 168.50 ± 3.82c –

alues shown are mean ± SD of three determinations. Different letters in the same columnly detected in trace quantities; nd means not detected.

quercetin derivate; 2, naringenin; 3, luteolin; 4, petunidin (expressed as cyanidin-3-glucoside); 5, delphinidin. Insets: peel sections from MT and Aft/atv/hp2 fruit. Scalebar = 0.5 mm.

stress-responsive transcription factors Yap1-GFP and Msn2-GFP,

which participate in the cellular defense against oxidative stress(Rivero-Pérez et al., 2008).

HPLC-DAD quantification showed increased naringenin levelsin the fruit peel of atv, Abg/hp2 and Aft/hp1 (Fig. 6b). Petunidin was

ro-Tom (MT) background.

Luteolin(mg 100 g−1 FW)

Total phenolics(mg g−1 FW)

DPPH(�mol trolox g−1 FW)

nd 0.59 ± 0.00ef 2.22 ± 0.15d0.12 ± 0.00d 0.58 ± 0.02ef 0.38 ± 0.02fnd 0.51 ± 0.02f 6.96 ± 0.53bnd 0.72 ± 0.05 cd 3.72 ± 0.29cnd 0.67 ± 0.01de 0.22 ± 0.03f0.36 ± 0.02b 0.71 ± 0.03 cd 0.25 ± 0.02f0.13 ± 0.01d 0.64 ± 0.01de 1.69 ± 0.13de0.24 ± 0.01c 0.80 ± 0.01bc 0.28 ± 0.01f0.45 ± 0.01a 1.18 ± 0.06a 8.64 ± 0.45and 0.83 ± 0.03b 4.57 ± 0.18c

n indicate statistical differences according to theTukey’s test (P ≤ 0.01). “–“means

o-Tom (MT) background.

uteoling 100 g−1 FW)

Total phenolics(mg g−1 FW)

DPPH(�mol trolox g−1 FW)

.97 ± 0.59ef 1.37 ± 0.07e 2.42 ± 0.23e

.31 ± 0.13f 1.39 ± 0.03e 11.30 ± 1.26d

.71 ± 0.31c 1.30 ± 0.06e 2.30 ± 0.15e

.02 ± 0.13d 3.20 ± 0.12c 19.7 ± 0.67c

.44 ± 0.99f 2.21 ± 0.07d 0.54 ± 0.03e

.61 ± 0.08e 3.02 ± 0.07c 0.70 ± 0.04e

.65 ± 0.18g 1.50 ± 0.05e 3.29 ± 0.17e1.28 ± 1.07a 3.36 ± 0.16c 1.25 ± 0.08e5.84 ± 0.92b 5.95 ± 0.27a 38.12 ± 4.27a

4.90 ± 0.05b 34.00 ± 1.31b

n indicate statistical differences according to the Tukey’s test (P ≤ 0.01). “–“, means

118 I. Sestari et al. / Scientia Horticulturae 175 (2014) 111–120

Fig. 6. Flavonoid composition in the fruits of tomato mutants. Content of flavonoidsin the flesh (a) and peel (b) of ripe fruits from different tomato genotypes. PetunidinwAT

fowfbsc(lwwdgpc

am

TFf

as expressed as cyanidin-3-glucoside. Data are represented as means ± SE (n = 3).sterisks (*) denote values significantly higher than Micro-Tom (MT) according toukey’s test (P ≤ 0.01).

ound in considerable amounts (>60 mg/100 mg) in the fruit peelf the lines combining Aft and hp2. Although the synergistic effectith hp1 was previously described (Sapir et al., 2008), we observed

urther increases of flavonoid contents in combination with hp2,oth in the fruit peel and the flesh (Fig. 6a and b). The triple mutanthowed the highest amount of acylated petunidin (83% of all antho-yanins) followed by the Aft/hp2 double mutant in the fruit skinFig. 6b). Only the genotypes harboring the Aft mutation accumu-ated delphinidin, and then only in the fruit peel, except for Aft/hp2,

hich also accumulated it in the flesh. The flavanone naringenin,hich is one of the precursors of anthocyanidins, such as petuni-in and delphinidin, was not detected in Aft/atv/hp2 and Aft/hp2enotypes (Fig. 6b). These results suggest that one of the possibleathways for anthocyanidin accumulation in purple tomatoes is theonversion of the native naringenin into petunidin and delphinidin.

Total phenolic contents were increased in both fruit fleshnd the skin of the double Aft/hp2 and the triple Aft/atv/hp2utants, with the flavonol quercetin, but not the flavone luteolin,

able 4resh weight (FW) and dry weight (DW) of the peel and flesha of Micro-Tom (MT)ruitb.

FW (g fruit−1) DW (g fruit−1)

Peel 0.311 ± 0.004 0.061 ± 0.000Flesh 4.09 ± 0.025 0.207 ± 0.002Peel/(Peel + Flesh)×100 7.01% 22.76%

a Without the locular content (mucilage and seeds).b Data are shown as mean ± SE, n = 40.

Fig. 7. Feasibility of introducing mutant allelic combinations into a commercial cul-tivar. Representative VFNT plant carrying the combination of mutations (Aft/atv/hp2)enhancing anthocyanin accumulation in the tomato fruit. Scale bar = 10 cm.

contributing in the latter (Tables 2 and 3). Phenolics contentis known to be widely variable depending on crop variety andcultivation conditions (Pinela et al., 2012). A concomitant increasein antioxidant capacity (DPPH) was observed in the triple mutant,although it seems to have been reduced by the presence of theatv mutation, since it was higher in both the flesh and skin of thedouble Aft/hp2 (Tables 2 and 3).

3.6. Stacking of mutant alleles in a commercial variety of cherrytomato

Anthocyanins accumulate mostly in the tomato fruit skin, whichhas been previously presented as a case for genetic engineering oftheir expression in the flesh (Butelli et al., 2008a). Here, however,

we show that in a small-fruited cultivar such as MT, dried fruitskin amounts to almost one-fourth of the fruit’s total dry weight(Table 4). The large contribution of the peel to the overall nutrientcontent confirms the suitability of cherry-type tomato cultivars as

I. Sestari et al. / Scientia Horticulturae 175 (2014) 111–120 119

Table S.1Mass spectra of anthocyanins detected in tomato peel and pulp (positive mode). All compounds were detected mainly in the peel, while trace amounts were detected in thepulp.

Compounds RT (min) [M]+ m/z MS2

Petunidin-caffeic acid + rutinoside + hexoside 15.4 949 787/625/479/317Delphinidin + rutinoside + hexoside 16.4 771 611/465/303

tamtemwTptlq

4

ticlSesgaiotufctgtp

A

CJDSBGotRt

R

A

Delphinidin + p-coumaroyl + rutinoside + hexoside 17.2Petunidin + p-coumaroyl + rutinoside + hexoside 18.7Delphinidinderivative 20.3

argets for conventional breeding of anthocyanin expression. Welso prove the feasibility of transferring the best combination ofutations (Aft, atv and hp2) into a commercial cultivar of cherry

omato (more sought after by a demanding market of fresh veg-tables). We used the elite cultivar VFNT, which is resistant toultiple pathogens (Oladiran and Iwu, 1992). In the F2 generation,e retrieved a promising line harboring all three mutations (Fig. 7).

his line is currently being bred to produce a new cultivar withurple fruits and improved nutritional qualities. All the parame-ers described in this work can eventually be assessed in such aine and further mutations later on introduced to improve the fruituality even further.

. Conclusion

In this work we describe the introgression of alleles related tohe accumulation of phytonutrients from various tomato cultivarsnto the genetic background of the cultivar Micro-Tom (MT). Weharacterized the nutrient content of the resulting near-isogenicines (NILs) in individual, double and triple mutant combinations.tacking different alleles into a single genotype brought about syn-rgistic effects in nutrient accumulation, which reinforces previousuggestions (Povero et al., 2011). The use of MT as genetic back-round sped up considerably the process of crossing and combininglleles and did not have a deleterious effect on fruit yield and qual-ty. We also showed that promising allele combinations can be latern transferred to commercial tomato varieties. This collection andhe information presented here thus highlight the advantages ofsing MT as a model for generating near-isogenic materials idealor diet-feeding studies, in which improved experimental controlsan be used. Our approach can be extended to add further mutantso the collection, generated either by conventional crossing or byenetic transformation. The outcome of this research can be appliedo the study of diet supplements or the specific effects of com-ounds present in natural diets in both humans and animal models.

cknowledgments

The financial support was provided by FAPESP (IS and AZ) andNPq (GGR, EP and LEPP). We thank Dr. Tania Shiga, Lucia Helena

ustinos da Silva and Jonata Freschi for technical assistance andr. Ernani Pinto Jr. and Dr. Felipe Augusto Dörr (Universidade deão Paulo) for mass spectrometry analyses. Dr. Elliot Cooper, Dr.enjamin Rae (Oxford University), Dr. Nicole Walczyk (Institutoulbenkian de Ciência) and Adrienne R. Washington (Universityf Pittsburgh) are acknowledged for valuable input in preparinghe manuscript. We thank Dr. Roger Chetelat (Tomato Geneticsesource Center, Davis, USA) for the donation of tomato seeds inheir original genetic backgrounds.

eferences

raújo, M.L. de, Maluf, W.R., Gomes, L.A.A., Oliveira, A.C.B., 2002. Intra and inter-locus interactions between alcobac a (alc), crimson (og c), and high pigment(hp) loci in tomato Lycopersicon esculentum Mill. Euphytica 125, 215–226,http://dx.doi.org/10.1023/A: 1015809526156.

919 757/595/465/303933 771/479/317773 465/345/303

Bino, R.J., Ric de Vos, C.H., Lieberman, M., et al., 2005. The light-hyperresponsivehigh pigment-2dg mutation of tomato: alterations in the fruit metabolome. NewPhytol. 166, 427–438, http://dx.doi.org/10.1111/j.1469-8137.2005.01362.x.

Boileau, T.W.-M., 2003. Prostate carcinogenesis in N-methyl-N-nitrosourea(NMU)-testosterone-treated rats fed tomato powder, lycopene, orenergy-restricted diets. Cancer Spectr. Knowl. Environ. 95, 1578–1586,http://dx.doi.org/10.1093/jnci/djg081.

Brand-Williams, W., Cuvelier, M.E., Berset, C., 1995. Use of a free radicalmethod to evaluate antioxidant activity. LWT—Food Sci. Technol. 28, 25–30,http://dx.doi.org/10.1016/S0023-6438(95)80008-5.

Bunea, A., Rugina, D., Sconta, Z., et al., 2013. Anthocyanin determination in blue-berry extracts from various cultivars and their antiproliferative and apoptoticproperties in B16-F10 metastatic murine melanoma cells. Phytochemistry 95,436–444, http://dx.doi.org/10.1016/j.phytochem.2013.06.018.

Butelli, E., Titta, L., Giorgio, M., et al., 2008a. Enrichment of tomato fruit withhealth-promoting anthocyanins by expression of select transcription factors.Nat. Biotechnol. 26, 1301–1308, http://dx.doi.org/10.1038/nbt.1506.

Butelli, E., Titta, L., Giorgio, M., et al., 2008b. Enrichment of tomato fruit withhealth-promoting anthocyanins by expression of select transcription factors.Nat. Biotechnol. 26, 1301–1308.

Campos, M.L., de Almeida, M., Rossi, M., et al., 2009. Brassinosteroids interact neg-atively with jasmonates in the formation of anti-herbivory traits in tomato. J.Exp. Bot. 60, 4347–4361, http://dx.doi.org/10.1093/jxb/erp270.

Campos, M.L., Carvalho, R.F., Benedito, V.A., Peres, L.E.P., 2010. Small and remarkable.The Micro-Tom model system as a tool to discover novel hormonal functions andinteractions. Plant Signal. Behav. 5, 1–4.

Canady, M.A., Meglic, V., Chetelat, R.T., 2005. A library of Solanum lycop-ersicoides introgression lines in cultivated tomato. Genome 48, 685–697,http://dx.doi.org/10.1139/g05-032.

Carvalho, R.F., Campos, M.L., Pino, L.E., et al., 2011. Convergence of devel-opmental mutants into a single tomato model system: “Micro-Tom” asan effective toolkit for plant development research. Plant Methods 7, 18,http://dx.doi.org/10.1186/1746-4811-7-18.

Chetelat, R.T., Klann, E., DeVerna, J.W., et al., 1993. Inheritance and genetic mappingof fruit sucrose accumulation in Lycopersicon chmielewskii. Plant J. 4, 643–650.

De Pascual-Teresa, S., Moreno, D.A., García-Viguera, C., 2010. Flavanols and antho-cyanins in cardiovascular health: a review of current evidence. Int. J. Mol. Sci.11, 1679–1703, http://dx.doi.org/10.3390/ijms11041679.

De Sa, J., Lock, K., 2008. Will European agricultural policy for school fruit and vegeta-bles improve public health? A review of school fruit and vegetable programmes.Eur. J. Public Health 18, 558–568, http://dx.doi.org/10.1093/eurpub/ckn061.

Edmeades, G.O., McMaster, G.S., White, J.W., Campos, H., 2004. Genomics and thephysiologist: bridging the gap between genes and crop response. Field CropsRes. 90, 5–18, http://dx.doi.org/10.1016/j.fcr.2004.07.002.

Hadley, C.W., Clinton, S.K., Schwartz, S.J., 2003. The consumption of processedtomato products enhances plasma lycopene concentrations in associationwith a reduced lipoprotein sensitivity to oxidative damage. J. Nutr. 133,727–732.

Halliwell, B., Gutteridge, J., 2007. Free Radicals in Biology and Medicine, 4th ed., 888.Harnly, J.M., Doherty, R.F., Beecher, G.R., et al., 2006. Flavonoid content of

U.S. fruits, vegetables, and nuts. J. Agric. Food Chem. 54, 9966–9977,http://dx.doi.org/10.1021/jf061478a.

Hassimotto, N.M.A., Genovese, M.I., Lajolo, F.M., 2007. Identification and character-isation of anthocyanins from wild mulberry (Morus nigra L.) growing in Brazil.Food Sci. Technol. Int. 13, 17–25, http://dx.doi.org/10.1177/1082013207075602.

Hung, H.-C., Joshipura, K.J., Jiang, R., et al., 2004. Fruit and vegetable intakeand risk of major chronic disease. J. Natl. Cancer Inst. 96, 1577–1584,http://dx.doi.org/10.1093/jnci/djh296.

Jayaprakasam, B., Vareed, S.K., Olson, L.K., Nair, M.G., 2005. Insulin secretion bybioactive anthocyanins and anthocyanidins present in fruits. J. Agric. Food Chem.53, 28–31, http://dx.doi.org/10.1021/jf049018+.

Jones, C.M., Mes, P., Myers, J.R., 2003. Characterization and inheritance of the Antho-cyanin fruit (Aft) tomato. J. Hered. 94, 449–456.

Kendrick, R.E., Peters, J.L., Kerckhoffs, L.H., et al., 1994. Photomorphogenic mutantsof tomato. Biochem. Soc. Symp. 60, 249–256.

Kerckhoffs, L.H.J., De Groot, N., Van Tuinen, A., et al., 1997. Physiological character-ization of exaggerated-photoresponse mutants of tomato. J. Plant Physiol. 150,578–587.

Li, H., Deng, Z., Liu, R., et al., 2011. Characterization of phytochemicals and antioxi-dant activities of a purple tomato (Solanum lycopersicum L.). J. Agric. Food Chem.59, 11803–11811, http://dx.doi.org/10.1021/jf202364v.

Lieberman, M., Segev, O., Gilboa, N., et al., 2004. The tomato homolog of the geneencoding UV-damaged DNA binding protein 1 (DDB1) underlined as the gene

1 rticult

L

L

L

M

M

M

M

M

M

M

M

M

N

O

P

P

P

20 I. Sestari et al. / Scientia Ho

that causes the high pigment-1 mutant phenotype. Theor. Appl. Genet. 108,1574–1581.

incoln, R.E., Porter, J.W., 1950. Inheritance of beta-carotene in tomatoes. Genetics35, 206–211.

ombardi-Crestana, S., da Silva Azevedo, M., e Silva, G.F.F., et al., 2012. Thetomato (Solanum lycopersicum cv. Micro-Tom) natural genetic variation Rg1and the DELLA mutant procera control the competence necessary to formadventitious roots and shoots. J. Exp. Bot. 63, 5689–5703, http://dx.doi.org/10.1093/jxb/ers221.

ong, M., Millar, D.J., Kimura, Y., et al., 2006. Metabolite profiling of carotenoid andphenolic pathways in mutant and transgenic lines of tomato: identification of ahigh antioxidant fruit line. Phytochemistry 67, 1750–1757, http://dx.doi.org/10.1016/j.phytochem.2006.02.022.

artí, E., Gisbert, C., Bishop, G.J., et al., 2006. Genetic and physiologicalcharacterization of tomato cv Micro-Tom. J. Exp. Bot. 57, 2037–2047,http://dx.doi.org/10.1093/jxb/erj154.

artin, C., Butelli, E., Petroni, K., Tonelli, C., 2011. How can research onplants contribute to promoting human health? Plant Cell 23, 1685–1699,http://dx.doi.org/10.1105/tpc.111.083279.

aurício Duarte-Almeida, J., Novoa, A.V., Linares, A.F., et al., 2006. Antioxi-dant activity of phenolics compounds from sugar cane (Saccharum offici-narum L.) juice. Plant Foods Hum. Nutr. 61, 187–192, http://dx.doi.org/10.1007/s11130-006-0032-6.

eissner, R., Jacobson, Y., Melamed, S., et al., 1997. A new model systemfor tomato genetics. Plant J. 12, 1465–1472, http://dx.doi.org/10.1046/j.1365-313x.1997.12061465.x.

es, P.J., Boches, P., Myers, J.R., Durst, R., 2008. Characterization of tomatoes express-ing anthocyanin in the fruit. J. Am. Soc. Hortic. Sci. 133, 262–269.

ilton, K., 1999. Nutritional characteristics of wild primate foods: do the diets ofour closest living relatives have lessons for us? Nutrition 15, 488–498.

ochizuki, T., Ishiuchi, D., Kamimura, S., 1988. Early seedling growth, yield compo-nents and fruit chemical composition of high-pigment processing tomato lines.Acta Hortic. 220, 85–92.

ohr, W.P., 1979. Pigment bodies in fruits of crimson and high pigment lines oftomatoes. Ann. Bot. 44, 427–434.

ustilli, A., Fenzi, F., Ciliento, R., et al., 1999. Phenotype of the tomato high pigment-2mutant is caused by a mutation in the tomato homolog of DEETIOLATED1. PlantCell 11, 145–158, http://dx.doi.org/10.1105/tpc.11.2.145.

agata, M., Yamashita, I., 1992. Simple method for simultaneous determina-tion of chlorophyll and carotenoids in tomato fruit. J. Jpn. Soc. Food Sci. 39,925–928.

ladiran, A.O., Iwu, L.N., 1992. Changes in ascorbic acid and carbohydratecontents in tomato fruits infected with pathogens. Plant Foods Hum. Nutr. 42,373–382.

aiva, S.A.R., Russell, R.M., 1999. �-carotene and other carotenoids asantioxidants. J. Am. Coll. Nutr. 18, 426–433, http://dx.doi.org/10.1080/07315724.1999.10718880.

asternak, T., Potters, G., Caubergs, R., Jansen, M.A.K., 2005. Complementaryinteractions between oxidative stress and auxins control plant growth

responses at plant, organ, and cellular level. J. Exp. Bot. 56, 1991–2001,http://dx.doi.org/10.1093/jxb/eri196.

atil, A.A., Patil, S.S., 1988. Preliminary studies on combiners and combinations forquality traits in tomato (Lycopersicon esculentum Mill.). Plant Foods Hum. Nutr.38, 43–49, http://dx.doi.org/10.1007/BF01092309.

urae 175 (2014) 111–120

Pecker, I., Gabbay, R., Cunningham Jr., F.X., Hirschberg, J., 1996. Cloning and charac-terization of the cDNA for lycopene �-cyclase from tomato reveals decrease inits expression during fruit ripening. Plant Mol. Biol. 30, 807–819.

Peters, J.L., van Tuinen, A., Adamse, P., et al., 1989. High pigment mutants of tomatoexhibit high sensitivity for phytochrome action. J. Plant Physiol. 134, 661–666,http://dx.doi.org/10.1016/S0176-1617(89)80024-0.

Pinela, J., Barros, L., Carvalho, A.M., Ferreira, I.C.F.R., 2012. Nutritional composi-tion and antioxidant activity of four tomato (Lycopersicon esculentum L.) farmervarieties in Northeastern Portugal homegardens. Food Chem. Toxicol. 50, 829–834.

Povero, G., Gonzali, S., Bassolino, L., et al., 2011. Transcriptional analysis in high-anthocyanin tomatoes reveals synergistic effect of Aft and atv genes. J. PlantPhysiol. 168, 270–279, http://dx.doi.org/10.1016/j.jplph.2010.07.022.

Powell, A.L.T., Nguyen, C.V., Hill, T., et al., 2012. Uniform ripening encodes a Golden2-like transcription factor regulating tomato fruit chloroplast development. Sci-ence 336, 1711–1715, http://dx.doi.org/10.1126/science.1222218.

Rick, C.M., Cisneros, P., Chetelat, R.T., DeVerna, J.W., 1994. Abg—a gene on chromo-some 10 for purple fruit derived from S. lycopersicoides. Tomato Genet. Coop.Rep. 44, 29–30.

Rivero-Pérez, M.D., Muniz, P., González-Sanjosé, M.L., 2008. Contribution of antho-cyanin fraction to the antioxidant properties of wine. Food Chem. Toxicol. 46,2815–2822, http://dx.doi.org/10.1016/j.fct.2008.05.014.

Ronen, G., Carmel-Goren, L., Zamir, D., Hirschberg, J., 2000. An alternative pathway to�-carotene formation in plant chromoplasts discovered by map-based cloningof Beta and old-gold color mutations in tomato. Proc. Natl. Acad. Sci. U S A 97,11102–11107.

Sapir, M., Oren-Shamir, M., Ovadia, R., et al., 2008. Molecular aspects of Antho-cyanin fruit tomato in relation to high pigment-1. J. Hered. 99, 292–303,http://dx.doi.org/10.1093/jhered/esm128.

Schreiber, G., Reuveni, M., Evenor, D., et al., 2012. ANTHOCYANIN1 fromSolanum chilense is more efficient in accumulating anthocyanin metabolitesthan its Solanum lycopersicum counterpart in association with the ANTHO-CYANIN FRUIT phenotype of tomato. Theor. Appl. Genet. 124, 295–307,http://dx.doi.org/10.1007/s00122-011-1705-6.

Shi, J., Le Maguer, M., 2000. Lycopene in tomatoes: chemical and physicalproperties affected by food processing. Crit. Rev. Food Sci. Nutr. 40, 1–42,http://dx.doi.org/10.1080/10408690091189275.

Stam, P., Zeven, A.C., 1981. The theoretical proportion of the donor genome innear-isogenic lines of self-fertilizers bred by backcrossing. Euphytica 30, 227–238.

Stevens, M.A., Rick, C.M., 1986. Genetics and breeding. In: Atherton, J., Rudich, J.(Eds.), The Tomato Crop: A Scientific Basis for Improvement. Chapman & Hall,London, pp. 35–100.

Tsuda, T., 2012. Dietary anthocyanin-rich plants: biochemical basis and recentprogress in health benefits studies. Mol. Nutr. Food Res. 56, 159–170,http://dx.doi.org/10.1002/mnfr.201100526.

WHO, 2003. Joint WHO/FAO Expert Consultation on Diet, Nutrition and the Preven-tion of Chronic Diseases, vol. 916.

Yeum, K.-J., Russell, R.M., Krinsky, N.I., Aldini, G., 2004. Biomarkers of antioxidant

capacity in the hydrophilic and lipophilic compartments of human plasma. Arch.Biochem. Biophys. 430, 97–103, http://dx.doi.org/10.1016/j.abb.2004.03.006.

Zielinski, H., Kozlowska, H., 2000. Antioxidant activity and total phenolics in selectedcereal grains and their different morphological fractions. J. Agric. Food Chem. 48,2008–2016.