agronomy

Article

Iron Biofortification of Red and Green PigmentedLettuce in Closed Soilless Cultivation ImpactsCrop Performance and Modulates Mineral andBioactive Composition

Maria Giordano 1dagger Christophe El-Nakhel 1dagger Antonio Pannico 1 Marios C Kyriacou 2 Silvia Rita Stazi 3 Stefania De Pascale 1 and Youssef Rouphael 1

1 Department of Agricultural Sciences University of Naples Federico II 80055 Portici Italymariagiordanouninait (MG) nakhel_christophehotmailcom (CE-N) antoniopannicouninait (AP)depascaluninait (SDP)

2 Department of Vegetable Crops Agricultural Research Institute 1516 Nicosia Cyprusmkyriacouarigovcy

3 Department for Innovation in Biological Agro-food and Forest Systems University of Tuscia 01100 ViterboItaly srstaziunitusit

Correspondence youssefrouphaeluninait Tel +39-(081)-2539134dagger Authors of equal contribution

Received 18 March 2019 Accepted 3 June 2019 Published 5 June 2019

Abstract Consumer demand for vegetables of fortified mineral and bioactive content is on the risedriven by the growing interest of society in fresh products of premium nutritional and functionalquality Biofortification of leafy vegetables with essential micronutrients such as iron (Fe) is an efficientmeans to address the human micronutrient deficiency known as hidden hunger Morphometric analysislipophilic and hydrophilic antioxidant capacities of green and red butterhead lettuce cultivars inresponse to Fe concentration in the nutrient solution (0015 control 05 10 or 20 mM Fe) were assessedThe experiment was carried out in a controlled-environment growth chamber using a closed soillesssystem (nutrient film technique) The percentage of yield reduction in comparison to the controltreatment was 57 135 and 253 at 05 10 and 20 mM Fe respectively Irrespective of the cultivarthe addition of 10 mM or 20 mM Fe in the nutrient solution induced an increase in the Fe concentrationof lettuce leaves by 205 and 537 respectively No significant effects of Fe application on phenolicacids and carotenoid profiles were observed in green Salanova Increasing Fe concentration in thenutrient solution to 05 mM triggered a spike in chlorogenic acid and total phenolics in red Salanovalettuce by 1101 and 291 compared with the control treatment respectively moreover higheraccumulation of caffeoyl meso tartaric phenolic acid by 314 at 10 mM Fe and of carotenoidsviolaxanthin neoxanthin and β-carotene by 370 at 20 mM Fe were also observed in red Salanovacompared with the control (0015 mM Fe) treatment Red Salanova exhibited higher yield P and Kcontents ascorbic acid phenolic acids and carotenoid compounds than green Salanova The wokshows how nutrient solution management in soilless culture could serve as effective cultural practicesfor producing Fe-enriched lettuce of premium quality notwithstanding cultivar selection being acritical underlying factor for obtaining high quality products

Keywords Ascorbic acid carotenoids profile hydroponics Lactuca sativa L mineral compositionnutrient solution management phenolic acids

Agronomy 2019 9 290 doi103390agronomy9060290 wwwmdpicomjournalagronomy

Agronomy 2019 9 290 2 of 21

1 Introduction

Food obligations and unbalanced diets lead to malnutrition that causes up to 3 million childrendeaths each year [1] This phenomenon known as hidden hunger is affecting both industrial anddeveloping countries In fact the cultivation in poor soils or where nutrients are not phytoavailablenegatively affect human health causing deficiencies in vitamins and in essential andor beneficialmicronutrients [1ndash3] Iron (Fe) is one of the indispensable microelements for life and although the earthcrust is rich in it Fe forms insoluble compounds [2] and its phytoavailable concentration (10minus17 M)does not reach the optimal range for plant growth (10minus9ndash10minus4 M) [4]

Fe is involved in very important processes in both plants and humans such as respirationphotosynthesis and oxygen transport [45] In the human body it exists in two different forms of hemecomplexes such as hemoglobin and myoglobin or in non-heme forms such as ironndashsulfur clustersand other prosthetic groups Two billion people are anemic worldwide and according to the WorldHealth Organization (WHO) the main cause is Fe deficient human diet [6] Fe deficiency is also amongthe most responsible factors for illnesses worldwide [7] Fe absorption in the human intestine canbe inhibited by various factors such as phytic acid (2ndash10 mg per meal) and polyphenols (ie tannicacid from 12 to 55 mg) while it is promoted by molecules such as ascorbic acid (50 mg per meal) andβ-carotene that can reduce or chelate Fe leading to more bioavailable complexes [89]

Biofortification is a way to address hidden hunger by increasing the nutritional content of plantsedible parts Recently Finklestein et al [10] have shown that biofortification with Fe in staple foodcrops (beans cassava maize pearl millet rice sweet potatoes and wheat) can increase Fe status (serumferritin concentrations and total body Fe) zinc and provitamin A carotenoids in populations at riskas in the Philippines India and Rwanda According to their work the beneficial effect has beendemonstrated not only in Fe deficient youngsters or baseline adults but also among individuals whoare not at risk [11]

Biofortification can be achieved through mineral fertilization breeding or biotechnologicalapproaches [15] However each of these solutions have limitations For instance the excessiveuse of fertilizers contributes to soil pollution or turns minerals into insoluble forms Fertilizationalso requires a frequent supply of the element and consequently raises production cost Moreoverhigh mineral concentrations can turn into stress conditions for plants Excessive amounts of Fe canlead to phytotoxicity and growth inhibition as demonstrated in many cultures such as rice [12ndash14]potato [15] wheat [1617] and tea [18] On the other hand the spread of biofortified transgeniccrops in many countries must undergo procrastinated procedures before its legal distribution to thepublic [1] Hydroponic cultivation systems eliminate or reduce problems of nutrient phyto-availabilityThey have long been seen as an answer to the urgent need to produce food for an increasing globalpopulation [1920] since they allow the management of plant nutritional status during growth througheffective control of water and nutrient supply In fact manipulating the nutrient solution in terms ofconcentration or composition demonstrably improved the yield or quality of zucchini squash [21]cucumber [22] lettuce [2324] artichoke [25] cardoon [26] and tomato [27]

Lettuce (Lactuca sativa L) is one of the most cultivated and consumed leafy vegetables in the worldappreciated for its organoleptic properties and it is a good source of minerals vitamins terpenoidsas well as carotenoids phenolic acids and flavonoids [28ndash30] Among the different pre-harvest factors(ie agricultural practices developmental stages climatic control) the genetic factor is consideredthe major determinant of variation in nutraceutical properties [31ndash36] The response of lettuce to Febiofortification was investigated only in terms of yield and Fe status under soil cultivation [3738]Essentially nothing is known about Fe biofortification under closed soilless cultivation (ie nutrientfilm technique NFT) where the constant exposure of the root system to Fe fortified nutrient solutioncould maximize Fe uptake translocation and accumulation in edible parts In addition the efficiencyof biofortification may depend upon several interacting parameters such as cultivar and applicationrate [19203940] To our knowledge no information is available on how biofortification with an

Agronomy 2019 9 290 3 of 21

essential micronutrient such as Fe could differentially modulate the nutritional and functional qualityof lettuce accounting for potential interaction with tested cultivars

In view of this background our aim was to assess the effect of different Fe application rates withinthe nutrient solution on growth parameters fresh yield mineral composition antioxidant activitiesnitrate and ascorbic acid contents as well as on phenolics and carotenoids profiles of green and redpigmented butterhead lettuce grown in NFT system under controlled environment The obtainedinformation will assist the scientific community as well as growers of leafy vegetables in identifyingoptimum cultivar-application rate combinations for achieving high nutritional and functional valueand in understanding the boundary between biofortification and Fe toxicity in lettuce

2 Materials and Methods

21 Growth Chamber Conditions Lettuce Cultivars and Experimental Design

The experiment was carried out in a 28 m2 controlled-environment growth chamber (70 times 21 mtimes 40 m W timesH times D) located at the experimental station of the Department of Agricultural SciencesUniversity of Naples Federico II Italy Artificial light was provided by high pressure sodium lampswith an intensity of 420 plusmn 5 micromol mminus2 sminus1 (165 cm from the top of the canopy) according to a lightdarkregime of 1212h Temperature was set at 2418 C (lightdark) and relative humidity was 60ndash80the latter being maintained by a fog system The experiment was carried out at ambient carbon dioxideconcentration (370ndash410 ppm) and air exchange was performed by means of an air extractor

Two cultivars of lettuce (Lactuca sativa L var capitata) green Salanovareg and red Salanovareg

(Rijk Zwaan Der Lier The Netherlands) were grown in a closed soilless system based on the nutrientfilm technique (NFT) The nutrient solution being collected in polypropylene reservoir tanks of 25 Land recirculated with a constant flow of 15 L minminus1 by submerged pumps The troughs were 200 cmlong 145 cm wide and 8 cm deep with a 1 slope Seeds of lettuce were germinated in vermiculiteThe lettuce seedlings were transplanted 15 days after sowing at the two-true leaf stage in rockwoolcubes (7 times 7 times 7 cm) (Delta Grodan Roermond The Netherlands) Lettuce seedlings were spaced15 cm apart between rockwool cubes and 30 cm apart between troughs giving a plant density of22 plants per square meter Each trough was covered with propylene taps to avoid evaporation of thenutrient solution

The growth chamber experiment was designed as a factorial combination of two butterheadlettuce cultivars (red and green pigmented) and four concentrations of Fe in the nutrient solution(0015 mM control treatment and three concentrations of 05 10 and 20 mM Fe) The basic nutrientsolution nutrient was a modified Hoagland and Arnon formulation The composition of the basicnutrient solution was 80 mM N-NO3

minus 15 mM S 10 mM P 30 mM K 30 mM Ca 10 mM Mg10 mM NH4

+ 15 microM Fe 9 microM Mn 03 microM Cu 16 microM Zn 20 microM B and 03 microM Mo with an electricalconductivity (EC) of 14 dS mminus1 and a pH of 58 plusmn 02 The biofortified Fe nutrient solution had thesame basic nutrient composition plus an additional 05 10 and 20 mM Fe Fe biofortification wasinitiated three days after transplanting (DAT) Fe was added as Fe chelate EDDHA 6 ortho-ortho(Revive Total Italpollina Spa Rivoli Veronese Italy)

Eight treatments derived from the factorial combinations of two butterhead lettuce cultivars(red and green Salanova) and four Fe concentrations in the nutrient solution (0015-control 05 10 or20 mM) Treatments were arranged in a randomized complete-block design amounting to a total of24 experimental units with twelve plants each (288 green and red Salanova plants in total)

22 Growth Analysis Biomass Determination and Radiation Use Efficiency

19 DAT all green and red Salanova lettuce plants were harvested The number of leaves per plantwas determined and the total area was measured by a LI-COR 3100C area meter (Biosciences LincolnNE USA) Leaf tissues were dried at 80 C for 72 h until they reached a constant weight and weighedagain to determine the corresponding shoot dry biomass The leaf dry matter percentage was also

Agronomy 2019 9 290 4 of 21

calculated Finally radiation use efficiency (RUE) was expressed as the shoot dry biomass divided bycumulative daily intercepted photosynthetically active radiation (PAR)

23 Collection of Samples for Mineral and Nutritional Quality Analyses

Part of the dried leaf tissue of green and red Salanova plants was used for macro-mineral andFe analyses For the identification and quantification of total ascorbic acid lipophilic antioxidantactivity (LAA) phenolic acids and carotenoid compounds by spectophotometry and HPLC-DADfresh samples of three plants per experimental unit were instantly frozen in liquid nitrogen and storedat minus80 C before lyophilizing them in a Christ Alpha 1-4 (Osterode Germany) freeze drier

24 Mineral Analysis by Ion Chromatography and ICP-OES

For mineral analysis 250 mg of dried green and red butterhead lettuce leaves were ground at05 mm in a Wiley Mill and then suspended in 50 mL of ultrapure water (Milli-Q Merck MilliporeDarmstadt Germany) and shaken in water bath (ShakeTemp SW22 Julabo Seelbach Germany) at80 C for 10 min The solution was centrifuged at 6000 rpm for 10 min (R-10 M Remi ElektrotechnikLimited India) then filtered through a 045 microm nylon syringe filter (Phenomenex Torrance CAUSA) and analyzed by ion chromatography (ICS-3000 Dionex Sunnyvale CA USA) coupled to aconductivity detector An IonPac CG12A (4 times 250 mm Dionex Corporation) guard column and IonPacCS12A (4 times 250 mm Dionex Corporation) analytical column were used for the K Ca and Mg analysiswhile for nitrate and P determination an IonPac AG11-HC guard (4 times 50 mm) column and IonPacAS11-HC analytical column (4 times 250 mm) were adopted as detailed in Rouphael et al [41] Nitratewas expressed as mg kgminus1 fresh weight (fw) on the basis of each samplersquos original dry weight (dw)while P K Ca and Mg were expressed as g kgminus1 dw

In addition to macro-minerals analysis the Fe content was also measured in green and red Salanovaleaf tissue Each sample was subjected to a first phase of acid digestion performed using a commercialhigh-pressure laboratory microwave oven (Mars plus CEM Italy) operating at an energy output of1800 W Approximately 300 mg of each dry sample was inserted directly into a microwave-closed vesselTwo mL of 30 (mm) H2O2 05 mL of 37 HCl and 75 mL of HNO3 69 solution were added toeach vessel The heating program was performed in one step Temperature was ramped linearly from25 to 180 C in 37 min then held at 180 C for 15 min After the digestion procedure and subsequentcooling samples were transferred into a Teflon beaker and total volume was made up to 25 mL withMilli-Q water The digest solution was then filtered (DISMIC 25HP PTFE syringe filter pore size045 microm Toyo Roshi Kaisha Ltd Japan) and stored in a screw cap plastic tube (Nalgene New YorkNY USA) Blanks were prepared in each lot of samples The reagents of super-pure grade used forthe microwave-assisted digestions were Hydrochloric acid (36 HCl) nitric acid (69 HNO3) andhydrogen peroxide (30 H2O2) (Merck Darmstadt Germany) High-purity water (18 MΩ cmminus1)from a Milli-Q water purification system (Millipore Bedford MA USA) was used for the dilutionof the standards for preparing samples throughout the chemical process and for final rinsing of theacid-cleaned vessels glasses and plastic utensils For this work tomato leaves (SRM 1573a) were usedas external certified reference material The calculated concentration for Fe was very close (within ca2) to the expected one 368 microg gminus1 certified value versus 37175 microg gminus1 determined value

Fe quantification was performed using an inductively coupled plasma optical emissionspectrometer (ICP-OES) with an axially viewed configuration (8000 DV PerkinElmer Shelton CTUSA) equipped with an ultrasonic nebulizer To assess Fe concentration calibration standards wereprepared treated equally to samples before dilution For detection we have chosen the frequency withthe lowest interferences high analytical signal and background ratio line at 2599 nm

Agronomy 2019 9 290 5 of 21

25 Total Ascorbic Acid Analysis

The total ascorbic acid defined as dehydroascorbic (DHA) and ascorbic acid (AA) was determinedby UVndashVis spectrophotometry (Hach DR 2000 Hach Co Loveland CO USA) as described byKampfenkel et al [42] Briefly 400 mg of sample fresh plant tissues were extracted with trichloroaceticacid (TCA) 6 200 microL of the extract was solubilized with 22-dipyridyl (14454 Sigma-Aldrich StLouis MO USA) The assay is based on the reduction of Fe3+ to Fe2+ by total ascorbic acid andthe spectrophotometric detection of Fe2+ complexes with 22-dipyridyl DHA is reduced to ASA bypre-incubation of the sample with dithiothreitol (DTT) The absorbance of the solution was measuredat 525 nm and data were expressed as mg AA per 100 g fw

26 Lipophilic Antioxidant Activity Analysis

The lipophilic antioxidant activity (LAA) was extracted from freeze-dried butterhead lettuceleaves (200 mg) with methanol and the antioxidant activity of this extract was measured with the22prime-azinobis 3-ethylbenzothiazoline-6-sulfonic acid ABTS method [43] Similarly to the total ascorbicacid LAA was determined by UVndashVis spectrophotometry The absorbance of the solutions wasmeasured at 734 nm LAA fraction was expressed as mmol Trolox (6-hydroxy-2578-tetramethylchroman-2-carboxylic acid) per 100 g dw

27 Phenolic Acids and Anthocyanins Identification and Quantification

400 mg of lyophilized samples was solubilized in a solution of methanolwaterformic acid(50455 vvv 12 mL) as described by Llorach et al [28] to determine phenolic acids as hydroxycinnamicderivatives The suspensions were sonicated for 30 min and then centrifuged (2500 g for 30 min at4 C) After a second centrifugation of supernatants at 21100 g for 15 min at 4 C samples were filteredthrough 022 microm cellulose filters (Phenomenex) A reversed phase C18 column (Prodigy 250 times 46 mm5 microm Phenomenex Torrance CA) equipped with a C18 security guard (40 times 30 mm Phenomenex)was used for the separation of hydroxycinnamic derivatives and anthocyanins 20microL of each extract wasinjected and the following mobile phases was used (A) water formic acid (955 vv) and (B) methanolthrough the following gradient of solvent B (t in [min][B]) (05) (2540) (3240) The flow ratewas 1 mL minminus1 LC column was installed onto a binary system (LC-10AD Shimadzu Kyoto Japan)equipped with a DAD (SPD-M10A Shimadzu Kyoto Japan) and a Series 200 autosampler (PerkinElmer Waltham MA) Chlorogenic and chicoric acids at 330 nm were used for the calibration curves ofhydroxycinnamic derivatives Identification of caffeoyl-meso-tartaric acid and caffeoyl-tartaric acidwas performed by LC-MSMS experiments

The chromatographic profiles of reference curves and samples were recorded in multiple reactionmonitoring mode (MRM) by using an API 3000 triple quadrupole (ABSciex Carlsbad CA) Negativeelectrospray ionization was used for detection and source parameters were selected as follows sprayvoltage minus42 kV capillary temperature 400 C dwell time 100 ms nebulizer gas and cad gas were set to10 and 12 respectively (arbitrary units) Target compounds [M-H] were analyzed using mass transitionsgiven in parentheses Chicoric acid (mz 473rarr311 293) chlorogenic acid (mz 353rarr191) caffeoyltartaric acid (mz 311rarr179 149 retention time 158 min) caffeoyl-meso-tartaric acid (mz 311rarr179 149retention time 178 min) The concentration of phenolic acids was reported as mg 100 gminus1 of dw

Anthocyanins were also measured within the same LC-DAD chromatographic runs at 520 nmand the concentration calculated by using cyanidin as reference standard to calculate the concentrationThe results were reported as microg of cyanidin equivalent per g of samples

28 Carotenoids Identification and Quantification

One gram of lyophilized samples was used to determine carotenoids content following the methodof Vallverduacute-Queralt et al [44] with slight modifications Samples were solubilized in ethanolhexane(43 vv 25 mL) with 1 BHT vortexed at 22 C for 30 s and sonicated for 5 min in the dark Then

Agronomy 2019 9 290 6 of 21

the solution was centrifuged (2500 g 4 C 10 min) and filtered through 045 microm nylon syringe filters(Phenomenex Torrance CA USA) The extracts were dried in N and the dried extracts were dissolvedin 1 BHT in chloroform 20 microL of each sample was injected onto a C18 column (Prodigy 250 times 46mm 5 microm Phenomenex Torrance C A USA) with a C18 security guard (40 times 30 mm Phenomenex)Two mobile phases were used (A) acetonitrile hexane methanol and dichloromethane (4222vvvv) and (B) acetonitrile Carotenoids were eluted at 08 mL minminus1 through the following gradientof solvent B (t in [min][B]) (070) (2060) (3030) (402) Carotenoids were quantified by a binaryLC-10AD system connected to a DAD (SPD-M10A Shimadzu Kyoto Japan) equipped with a Series200 auto-sampler (Perkin Elmer Waltham MA USA) Violaxanthin neoxanthin β-cryptoxanthinlutein and β-carotene were used as reference standards Identification of the peaks was achieved bycomparison of UV-vis spectra and retention times of eluted compounds with pure standards at 450 nmThree separate sets of calibration curves were built each set was injected three times in the same day(intraday assay) and three times in three different days (interday assay) The accuracy was reported asthe discrepancies between the calibration curves performed intraday and interday and the results wereexpressed as relative standard deviation RSD () A recovery test was performed spiking two sampleswith two known amounts of carotenoids (50 and 100 microg mLminus1 final concentration) and taking intoaccount the overestimation due to the target analytes already present in the samples The concentrationof the target carotenoids was expressed as microg gminus1 dw

29 Statistics

The ShapirondashWilk and KolmororovndashSmirnov procedures were performed to verify that thedata had a normal distribution and the Levene OrsquoBrien and Bartlet tests were conducted to verifythe homogeneity of variances Then all morphometric nutritional and functional quality datawere subjected to analysis of variance (two-way ANOVA) using IBM SPSS 20 software package(wwwibmcomsoftwareanalyticsspss) The means were separated by Tukeyrsquos honestly significantdifference (HSD) test (significance level 005) Butterhead lettuce cultivar main effects were comparedby t-Test

3 Results and Discussion

31 Growth Response Fresh Yield Dry Matter and Radiation Use Efficiency

Inter and intra-specific genetic variability is among the most important preharvest factors whichinfluence lettucersquos phenotypic and biochemical traits Lettuce presents within the same speciesa variety of colors sizes textures and shapes [45ndash47] In our study two cultivars of green and redSalanova were evaluated from a productive and nutritional point of view in response to different Feconcentrations in the nutrient solution

For leaf number per plant marketable fresh yield and leaf dry matter percentage no significantinteraction between cultivar (C) and Fe nutrient solution concentration (I) was observed whereas leafarea dry biomass and radiation use efficiency (RUE) were significantly affected by the interaction ofthese two factors (Table 1) Irrespective of the Fe concentration in the nutrient solution the red Salanovahad higher marketable yield and percentage dry matter than those recorded in green Salanova plantsby 93 and 70 respectively (Table 1) Analogous genotypic variation in marketable fresh yieldand leaf dry matter content has been previously demonstrated over seven iceberg cultivars (lsquoEquinosrsquolsquoIce Castlersquo lsquoMetaliarsquo lsquoNum 189rsquo lsquoSilvinasrsquo lsquoOmbrinasrsquo and lsquoVanguardiarsquo [36])

Agronomy 2019 9 290 3 of 21

essential micronutrient such as Fe could differentially modulate the nutritional and functional qualityof lettuce accounting for potential interaction with tested cultivars

In view of this background our aim was to assess the effect of different Fe application rates withinthe nutrient solution on growth parameters fresh yield mineral composition antioxidant activitiesnitrate and ascorbic acid contents as well as on phenolics and carotenoids profiles of green and redpigmented butterhead lettuce grown in NFT system under controlled environment The obtainedinformation will assist the scientific community as well as growers of leafy vegetables in identifyingoptimum cultivar-application rate combinations for achieving high nutritional and functional valueand in understanding the boundary between biofortification and Fe toxicity in lettuce

2 Materials and Methods

21 Growth Chamber Conditions Lettuce Cultivars and Experimental Design

The experiment was carried out in a 28 m2 controlled-environment growth chamber (70 times 21 mtimes 40 m W timesH times D) located at the experimental station of the Department of Agricultural SciencesUniversity of Naples Federico II Italy Artificial light was provided by high pressure sodium lampswith an intensity of 420 plusmn 5 micromol mminus2 sminus1 (165 cm from the top of the canopy) according to a lightdarkregime of 1212h Temperature was set at 2418 C (lightdark) and relative humidity was 60ndash80the latter being maintained by a fog system The experiment was carried out at ambient carbon dioxideconcentration (370ndash410 ppm) and air exchange was performed by means of an air extractor

Two cultivars of lettuce (Lactuca sativa L var capitata) green Salanovareg and red Salanovareg

(Rijk Zwaan Der Lier The Netherlands) were grown in a closed soilless system based on the nutrientfilm technique (NFT) The nutrient solution being collected in polypropylene reservoir tanks of 25 Land recirculated with a constant flow of 15 L minminus1 by submerged pumps The troughs were 200 cmlong 145 cm wide and 8 cm deep with a 1 slope Seeds of lettuce were germinated in vermiculiteThe lettuce seedlings were transplanted 15 days after sowing at the two-true leaf stage in rockwoolcubes (7 times 7 times 7 cm) (Delta Grodan Roermond The Netherlands) Lettuce seedlings were spaced15 cm apart between rockwool cubes and 30 cm apart between troughs giving a plant density of22 plants per square meter Each trough was covered with propylene taps to avoid evaporation of thenutrient solution

The growth chamber experiment was designed as a factorial combination of two butterheadlettuce cultivars (red and green pigmented) and four concentrations of Fe in the nutrient solution(0015 mM control treatment and three concentrations of 05 10 and 20 mM Fe) The basic nutrientsolution nutrient was a modified Hoagland and Arnon formulation The composition of the basicnutrient solution was 80 mM N-NO3

minus 15 mM S 10 mM P 30 mM K 30 mM Ca 10 mM Mg10 mM NH4

+ 15 microM Fe 9 microM Mn 03 microM Cu 16 microM Zn 20 microM B and 03 microM Mo with an electricalconductivity (EC) of 14 dS mminus1 and a pH of 58 plusmn 02 The biofortified Fe nutrient solution had thesame basic nutrient composition plus an additional 05 10 and 20 mM Fe Fe biofortification wasinitiated three days after transplanting (DAT) Fe was added as Fe chelate EDDHA 6 ortho-ortho(Revive Total Italpollina Spa Rivoli Veronese Italy)

Eight treatments derived from the factorial combinations of two butterhead lettuce cultivars(red and green Salanova) and four Fe concentrations in the nutrient solution (0015-control 05 10 or20 mM) Treatments were arranged in a randomized complete-block design amounting to a total of24 experimental units with twelve plants each (288 green and red Salanova plants in total)

22 Growth Analysis Biomass Determination and Radiation Use Efficiency

19 DAT all green and red Salanova lettuce plants were harvested The number of leaves per plantwas determined and the total area was measured by a LI-COR 3100C area meter (Biosciences LincolnNE USA) Leaf tissues were dried at 80 C for 72 h until they reached a constant weight and weighedagain to determine the corresponding shoot dry biomass The leaf dry matter percentage was also

Agronomy 2019 9 290 4 of 21

calculated Finally radiation use efficiency (RUE) was expressed as the shoot dry biomass divided bycumulative daily intercepted photosynthetically active radiation (PAR)

23 Collection of Samples for Mineral and Nutritional Quality Analyses

Part of the dried leaf tissue of green and red Salanova plants was used for macro-mineral andFe analyses For the identification and quantification of total ascorbic acid lipophilic antioxidantactivity (LAA) phenolic acids and carotenoid compounds by spectophotometry and HPLC-DADfresh samples of three plants per experimental unit were instantly frozen in liquid nitrogen and storedat minus80 C before lyophilizing them in a Christ Alpha 1-4 (Osterode Germany) freeze drier

24 Mineral Analysis by Ion Chromatography and ICP-OES

For mineral analysis 250 mg of dried green and red butterhead lettuce leaves were ground at05 mm in a Wiley Mill and then suspended in 50 mL of ultrapure water (Milli-Q Merck MilliporeDarmstadt Germany) and shaken in water bath (ShakeTemp SW22 Julabo Seelbach Germany) at80 C for 10 min The solution was centrifuged at 6000 rpm for 10 min (R-10 M Remi ElektrotechnikLimited India) then filtered through a 045 microm nylon syringe filter (Phenomenex Torrance CAUSA) and analyzed by ion chromatography (ICS-3000 Dionex Sunnyvale CA USA) coupled to aconductivity detector An IonPac CG12A (4 times 250 mm Dionex Corporation) guard column and IonPacCS12A (4 times 250 mm Dionex Corporation) analytical column were used for the K Ca and Mg analysiswhile for nitrate and P determination an IonPac AG11-HC guard (4 times 50 mm) column and IonPacAS11-HC analytical column (4 times 250 mm) were adopted as detailed in Rouphael et al [41] Nitratewas expressed as mg kgminus1 fresh weight (fw) on the basis of each samplersquos original dry weight (dw)while P K Ca and Mg were expressed as g kgminus1 dw

In addition to macro-minerals analysis the Fe content was also measured in green and red Salanovaleaf tissue Each sample was subjected to a first phase of acid digestion performed using a commercialhigh-pressure laboratory microwave oven (Mars plus CEM Italy) operating at an energy output of1800 W Approximately 300 mg of each dry sample was inserted directly into a microwave-closed vesselTwo mL of 30 (mm) H2O2 05 mL of 37 HCl and 75 mL of HNO3 69 solution were added toeach vessel The heating program was performed in one step Temperature was ramped linearly from25 to 180 C in 37 min then held at 180 C for 15 min After the digestion procedure and subsequentcooling samples were transferred into a Teflon beaker and total volume was made up to 25 mL withMilli-Q water The digest solution was then filtered (DISMIC 25HP PTFE syringe filter pore size045 microm Toyo Roshi Kaisha Ltd Japan) and stored in a screw cap plastic tube (Nalgene New YorkNY USA) Blanks were prepared in each lot of samples The reagents of super-pure grade used forthe microwave-assisted digestions were Hydrochloric acid (36 HCl) nitric acid (69 HNO3) andhydrogen peroxide (30 H2O2) (Merck Darmstadt Germany) High-purity water (18 MΩ cmminus1)from a Milli-Q water purification system (Millipore Bedford MA USA) was used for the dilutionof the standards for preparing samples throughout the chemical process and for final rinsing of theacid-cleaned vessels glasses and plastic utensils For this work tomato leaves (SRM 1573a) were usedas external certified reference material The calculated concentration for Fe was very close (within ca2) to the expected one 368 microg gminus1 certified value versus 37175 microg gminus1 determined value

Fe quantification was performed using an inductively coupled plasma optical emissionspectrometer (ICP-OES) with an axially viewed configuration (8000 DV PerkinElmer Shelton CTUSA) equipped with an ultrasonic nebulizer To assess Fe concentration calibration standards wereprepared treated equally to samples before dilution For detection we have chosen the frequency withthe lowest interferences high analytical signal and background ratio line at 2599 nm

Agronomy 2019 9 290 5 of 21

25 Total Ascorbic Acid Analysis

The total ascorbic acid defined as dehydroascorbic (DHA) and ascorbic acid (AA) was determinedby UVndashVis spectrophotometry (Hach DR 2000 Hach Co Loveland CO USA) as described byKampfenkel et al [42] Briefly 400 mg of sample fresh plant tissues were extracted with trichloroaceticacid (TCA) 6 200 microL of the extract was solubilized with 22-dipyridyl (14454 Sigma-Aldrich StLouis MO USA) The assay is based on the reduction of Fe3+ to Fe2+ by total ascorbic acid andthe spectrophotometric detection of Fe2+ complexes with 22-dipyridyl DHA is reduced to ASA bypre-incubation of the sample with dithiothreitol (DTT) The absorbance of the solution was measuredat 525 nm and data were expressed as mg AA per 100 g fw

26 Lipophilic Antioxidant Activity Analysis

The lipophilic antioxidant activity (LAA) was extracted from freeze-dried butterhead lettuceleaves (200 mg) with methanol and the antioxidant activity of this extract was measured with the22prime-azinobis 3-ethylbenzothiazoline-6-sulfonic acid ABTS method [43] Similarly to the total ascorbicacid LAA was determined by UVndashVis spectrophotometry The absorbance of the solutions wasmeasured at 734 nm LAA fraction was expressed as mmol Trolox (6-hydroxy-2578-tetramethylchroman-2-carboxylic acid) per 100 g dw

27 Phenolic Acids and Anthocyanins Identification and Quantification

400 mg of lyophilized samples was solubilized in a solution of methanolwaterformic acid(50455 vvv 12 mL) as described by Llorach et al [28] to determine phenolic acids as hydroxycinnamicderivatives The suspensions were sonicated for 30 min and then centrifuged (2500 g for 30 min at4 C) After a second centrifugation of supernatants at 21100 g for 15 min at 4 C samples were filteredthrough 022 microm cellulose filters (Phenomenex) A reversed phase C18 column (Prodigy 250 times 46 mm5 microm Phenomenex Torrance CA) equipped with a C18 security guard (40 times 30 mm Phenomenex)was used for the separation of hydroxycinnamic derivatives and anthocyanins 20microL of each extract wasinjected and the following mobile phases was used (A) water formic acid (955 vv) and (B) methanolthrough the following gradient of solvent B (t in [min][B]) (05) (2540) (3240) The flow ratewas 1 mL minminus1 LC column was installed onto a binary system (LC-10AD Shimadzu Kyoto Japan)equipped with a DAD (SPD-M10A Shimadzu Kyoto Japan) and a Series 200 autosampler (PerkinElmer Waltham MA) Chlorogenic and chicoric acids at 330 nm were used for the calibration curves ofhydroxycinnamic derivatives Identification of caffeoyl-meso-tartaric acid and caffeoyl-tartaric acidwas performed by LC-MSMS experiments

The chromatographic profiles of reference curves and samples were recorded in multiple reactionmonitoring mode (MRM) by using an API 3000 triple quadrupole (ABSciex Carlsbad CA) Negativeelectrospray ionization was used for detection and source parameters were selected as follows sprayvoltage minus42 kV capillary temperature 400 C dwell time 100 ms nebulizer gas and cad gas were set to10 and 12 respectively (arbitrary units) Target compounds [M-H] were analyzed using mass transitionsgiven in parentheses Chicoric acid (mz 473rarr311 293) chlorogenic acid (mz 353rarr191) caffeoyltartaric acid (mz 311rarr179 149 retention time 158 min) caffeoyl-meso-tartaric acid (mz 311rarr179 149retention time 178 min) The concentration of phenolic acids was reported as mg 100 gminus1 of dw

Anthocyanins were also measured within the same LC-DAD chromatographic runs at 520 nmand the concentration calculated by using cyanidin as reference standard to calculate the concentrationThe results were reported as microg of cyanidin equivalent per g of samples

28 Carotenoids Identification and Quantification

One gram of lyophilized samples was used to determine carotenoids content following the methodof Vallverduacute-Queralt et al [44] with slight modifications Samples were solubilized in ethanolhexane(43 vv 25 mL) with 1 BHT vortexed at 22 C for 30 s and sonicated for 5 min in the dark Then

Agronomy 2019 9 290 6 of 21

the solution was centrifuged (2500 g 4 C 10 min) and filtered through 045 microm nylon syringe filters(Phenomenex Torrance CA USA) The extracts were dried in N and the dried extracts were dissolvedin 1 BHT in chloroform 20 microL of each sample was injected onto a C18 column (Prodigy 250 times 46mm 5 microm Phenomenex Torrance C A USA) with a C18 security guard (40 times 30 mm Phenomenex)Two mobile phases were used (A) acetonitrile hexane methanol and dichloromethane (4222vvvv) and (B) acetonitrile Carotenoids were eluted at 08 mL minminus1 through the following gradientof solvent B (t in [min][B]) (070) (2060) (3030) (402) Carotenoids were quantified by a binaryLC-10AD system connected to a DAD (SPD-M10A Shimadzu Kyoto Japan) equipped with a Series200 auto-sampler (Perkin Elmer Waltham MA USA) Violaxanthin neoxanthin β-cryptoxanthinlutein and β-carotene were used as reference standards Identification of the peaks was achieved bycomparison of UV-vis spectra and retention times of eluted compounds with pure standards at 450 nmThree separate sets of calibration curves were built each set was injected three times in the same day(intraday assay) and three times in three different days (interday assay) The accuracy was reported asthe discrepancies between the calibration curves performed intraday and interday and the results wereexpressed as relative standard deviation RSD () A recovery test was performed spiking two sampleswith two known amounts of carotenoids (50 and 100 microg mLminus1 final concentration) and taking intoaccount the overestimation due to the target analytes already present in the samples The concentrationof the target carotenoids was expressed as microg gminus1 dw

29 Statistics

The ShapirondashWilk and KolmororovndashSmirnov procedures were performed to verify that thedata had a normal distribution and the Levene OrsquoBrien and Bartlet tests were conducted to verifythe homogeneity of variances Then all morphometric nutritional and functional quality datawere subjected to analysis of variance (two-way ANOVA) using IBM SPSS 20 software package(wwwibmcomsoftwareanalyticsspss) The means were separated by Tukeyrsquos honestly significantdifference (HSD) test (significance level 005) Butterhead lettuce cultivar main effects were comparedby t-Test

3 Results and Discussion

31 Growth Response Fresh Yield Dry Matter and Radiation Use Efficiency

Inter and intra-specific genetic variability is among the most important preharvest factors whichinfluence lettucersquos phenotypic and biochemical traits Lettuce presents within the same speciesa variety of colors sizes textures and shapes [45ndash47] In our study two cultivars of green and redSalanova were evaluated from a productive and nutritional point of view in response to different Feconcentrations in the nutrient solution

For leaf number per plant marketable fresh yield and leaf dry matter percentage no significantinteraction between cultivar (C) and Fe nutrient solution concentration (I) was observed whereas leafarea dry biomass and radiation use efficiency (RUE) were significantly affected by the interaction ofthese two factors (Table 1) Irrespective of the Fe concentration in the nutrient solution the red Salanovahad higher marketable yield and percentage dry matter than those recorded in green Salanova plantsby 93 and 70 respectively (Table 1) Analogous genotypic variation in marketable fresh yieldand leaf dry matter content has been previously demonstrated over seven iceberg cultivars (lsquoEquinosrsquolsquoIce Castlersquo lsquoMetaliarsquo lsquoNum 189rsquo lsquoSilvinasrsquo lsquoOmbrinasrsquo and lsquoVanguardiarsquo [36])

Agronomy 2019 9 290 4 of 21

calculated Finally radiation use efficiency (RUE) was expressed as the shoot dry biomass divided bycumulative daily intercepted photosynthetically active radiation (PAR)

23 Collection of Samples for Mineral and Nutritional Quality Analyses

Part of the dried leaf tissue of green and red Salanova plants was used for macro-mineral andFe analyses For the identification and quantification of total ascorbic acid lipophilic antioxidantactivity (LAA) phenolic acids and carotenoid compounds by spectophotometry and HPLC-DADfresh samples of three plants per experimental unit were instantly frozen in liquid nitrogen and storedat minus80 C before lyophilizing them in a Christ Alpha 1-4 (Osterode Germany) freeze drier

24 Mineral Analysis by Ion Chromatography and ICP-OES

For mineral analysis 250 mg of dried green and red butterhead lettuce leaves were ground at05 mm in a Wiley Mill and then suspended in 50 mL of ultrapure water (Milli-Q Merck MilliporeDarmstadt Germany) and shaken in water bath (ShakeTemp SW22 Julabo Seelbach Germany) at80 C for 10 min The solution was centrifuged at 6000 rpm for 10 min (R-10 M Remi ElektrotechnikLimited India) then filtered through a 045 microm nylon syringe filter (Phenomenex Torrance CAUSA) and analyzed by ion chromatography (ICS-3000 Dionex Sunnyvale CA USA) coupled to aconductivity detector An IonPac CG12A (4 times 250 mm Dionex Corporation) guard column and IonPacCS12A (4 times 250 mm Dionex Corporation) analytical column were used for the K Ca and Mg analysiswhile for nitrate and P determination an IonPac AG11-HC guard (4 times 50 mm) column and IonPacAS11-HC analytical column (4 times 250 mm) were adopted as detailed in Rouphael et al [41] Nitratewas expressed as mg kgminus1 fresh weight (fw) on the basis of each samplersquos original dry weight (dw)while P K Ca and Mg were expressed as g kgminus1 dw

In addition to macro-minerals analysis the Fe content was also measured in green and red Salanovaleaf tissue Each sample was subjected to a first phase of acid digestion performed using a commercialhigh-pressure laboratory microwave oven (Mars plus CEM Italy) operating at an energy output of1800 W Approximately 300 mg of each dry sample was inserted directly into a microwave-closed vesselTwo mL of 30 (mm) H2O2 05 mL of 37 HCl and 75 mL of HNO3 69 solution were added toeach vessel The heating program was performed in one step Temperature was ramped linearly from25 to 180 C in 37 min then held at 180 C for 15 min After the digestion procedure and subsequentcooling samples were transferred into a Teflon beaker and total volume was made up to 25 mL withMilli-Q water The digest solution was then filtered (DISMIC 25HP PTFE syringe filter pore size045 microm Toyo Roshi Kaisha Ltd Japan) and stored in a screw cap plastic tube (Nalgene New YorkNY USA) Blanks were prepared in each lot of samples The reagents of super-pure grade used forthe microwave-assisted digestions were Hydrochloric acid (36 HCl) nitric acid (69 HNO3) andhydrogen peroxide (30 H2O2) (Merck Darmstadt Germany) High-purity water (18 MΩ cmminus1)from a Milli-Q water purification system (Millipore Bedford MA USA) was used for the dilutionof the standards for preparing samples throughout the chemical process and for final rinsing of theacid-cleaned vessels glasses and plastic utensils For this work tomato leaves (SRM 1573a) were usedas external certified reference material The calculated concentration for Fe was very close (within ca2) to the expected one 368 microg gminus1 certified value versus 37175 microg gminus1 determined value

Fe quantification was performed using an inductively coupled plasma optical emissionspectrometer (ICP-OES) with an axially viewed configuration (8000 DV PerkinElmer Shelton CTUSA) equipped with an ultrasonic nebulizer To assess Fe concentration calibration standards wereprepared treated equally to samples before dilution For detection we have chosen the frequency withthe lowest interferences high analytical signal and background ratio line at 2599 nm

Agronomy 2019 9 290 5 of 21

25 Total Ascorbic Acid Analysis

The total ascorbic acid defined as dehydroascorbic (DHA) and ascorbic acid (AA) was determinedby UVndashVis spectrophotometry (Hach DR 2000 Hach Co Loveland CO USA) as described byKampfenkel et al [42] Briefly 400 mg of sample fresh plant tissues were extracted with trichloroaceticacid (TCA) 6 200 microL of the extract was solubilized with 22-dipyridyl (14454 Sigma-Aldrich StLouis MO USA) The assay is based on the reduction of Fe3+ to Fe2+ by total ascorbic acid andthe spectrophotometric detection of Fe2+ complexes with 22-dipyridyl DHA is reduced to ASA bypre-incubation of the sample with dithiothreitol (DTT) The absorbance of the solution was measuredat 525 nm and data were expressed as mg AA per 100 g fw

26 Lipophilic Antioxidant Activity Analysis

The lipophilic antioxidant activity (LAA) was extracted from freeze-dried butterhead lettuceleaves (200 mg) with methanol and the antioxidant activity of this extract was measured with the22prime-azinobis 3-ethylbenzothiazoline-6-sulfonic acid ABTS method [43] Similarly to the total ascorbicacid LAA was determined by UVndashVis spectrophotometry The absorbance of the solutions wasmeasured at 734 nm LAA fraction was expressed as mmol Trolox (6-hydroxy-2578-tetramethylchroman-2-carboxylic acid) per 100 g dw

27 Phenolic Acids and Anthocyanins Identification and Quantification

400 mg of lyophilized samples was solubilized in a solution of methanolwaterformic acid(50455 vvv 12 mL) as described by Llorach et al [28] to determine phenolic acids as hydroxycinnamicderivatives The suspensions were sonicated for 30 min and then centrifuged (2500 g for 30 min at4 C) After a second centrifugation of supernatants at 21100 g for 15 min at 4 C samples were filteredthrough 022 microm cellulose filters (Phenomenex) A reversed phase C18 column (Prodigy 250 times 46 mm5 microm Phenomenex Torrance CA) equipped with a C18 security guard (40 times 30 mm Phenomenex)was used for the separation of hydroxycinnamic derivatives and anthocyanins 20microL of each extract wasinjected and the following mobile phases was used (A) water formic acid (955 vv) and (B) methanolthrough the following gradient of solvent B (t in [min][B]) (05) (2540) (3240) The flow ratewas 1 mL minminus1 LC column was installed onto a binary system (LC-10AD Shimadzu Kyoto Japan)equipped with a DAD (SPD-M10A Shimadzu Kyoto Japan) and a Series 200 autosampler (PerkinElmer Waltham MA) Chlorogenic and chicoric acids at 330 nm were used for the calibration curves ofhydroxycinnamic derivatives Identification of caffeoyl-meso-tartaric acid and caffeoyl-tartaric acidwas performed by LC-MSMS experiments

The chromatographic profiles of reference curves and samples were recorded in multiple reactionmonitoring mode (MRM) by using an API 3000 triple quadrupole (ABSciex Carlsbad CA) Negativeelectrospray ionization was used for detection and source parameters were selected as follows sprayvoltage minus42 kV capillary temperature 400 C dwell time 100 ms nebulizer gas and cad gas were set to10 and 12 respectively (arbitrary units) Target compounds [M-H] were analyzed using mass transitionsgiven in parentheses Chicoric acid (mz 473rarr311 293) chlorogenic acid (mz 353rarr191) caffeoyltartaric acid (mz 311rarr179 149 retention time 158 min) caffeoyl-meso-tartaric acid (mz 311rarr179 149retention time 178 min) The concentration of phenolic acids was reported as mg 100 gminus1 of dw

Anthocyanins were also measured within the same LC-DAD chromatographic runs at 520 nmand the concentration calculated by using cyanidin as reference standard to calculate the concentrationThe results were reported as microg of cyanidin equivalent per g of samples

28 Carotenoids Identification and Quantification

One gram of lyophilized samples was used to determine carotenoids content following the methodof Vallverduacute-Queralt et al [44] with slight modifications Samples were solubilized in ethanolhexane(43 vv 25 mL) with 1 BHT vortexed at 22 C for 30 s and sonicated for 5 min in the dark Then

Agronomy 2019 9 290 6 of 21

the solution was centrifuged (2500 g 4 C 10 min) and filtered through 045 microm nylon syringe filters(Phenomenex Torrance CA USA) The extracts were dried in N and the dried extracts were dissolvedin 1 BHT in chloroform 20 microL of each sample was injected onto a C18 column (Prodigy 250 times 46mm 5 microm Phenomenex Torrance C A USA) with a C18 security guard (40 times 30 mm Phenomenex)Two mobile phases were used (A) acetonitrile hexane methanol and dichloromethane (4222vvvv) and (B) acetonitrile Carotenoids were eluted at 08 mL minminus1 through the following gradientof solvent B (t in [min][B]) (070) (2060) (3030) (402) Carotenoids were quantified by a binaryLC-10AD system connected to a DAD (SPD-M10A Shimadzu Kyoto Japan) equipped with a Series200 auto-sampler (Perkin Elmer Waltham MA USA) Violaxanthin neoxanthin β-cryptoxanthinlutein and β-carotene were used as reference standards Identification of the peaks was achieved bycomparison of UV-vis spectra and retention times of eluted compounds with pure standards at 450 nmThree separate sets of calibration curves were built each set was injected three times in the same day(intraday assay) and three times in three different days (interday assay) The accuracy was reported asthe discrepancies between the calibration curves performed intraday and interday and the results wereexpressed as relative standard deviation RSD () A recovery test was performed spiking two sampleswith two known amounts of carotenoids (50 and 100 microg mLminus1 final concentration) and taking intoaccount the overestimation due to the target analytes already present in the samples The concentrationof the target carotenoids was expressed as microg gminus1 dw

29 Statistics

The ShapirondashWilk and KolmororovndashSmirnov procedures were performed to verify that thedata had a normal distribution and the Levene OrsquoBrien and Bartlet tests were conducted to verifythe homogeneity of variances Then all morphometric nutritional and functional quality datawere subjected to analysis of variance (two-way ANOVA) using IBM SPSS 20 software package(wwwibmcomsoftwareanalyticsspss) The means were separated by Tukeyrsquos honestly significantdifference (HSD) test (significance level 005) Butterhead lettuce cultivar main effects were comparedby t-Test

3 Results and Discussion

31 Growth Response Fresh Yield Dry Matter and Radiation Use Efficiency

Inter and intra-specific genetic variability is among the most important preharvest factors whichinfluence lettucersquos phenotypic and biochemical traits Lettuce presents within the same speciesa variety of colors sizes textures and shapes [45ndash47] In our study two cultivars of green and redSalanova were evaluated from a productive and nutritional point of view in response to different Feconcentrations in the nutrient solution

For leaf number per plant marketable fresh yield and leaf dry matter percentage no significantinteraction between cultivar (C) and Fe nutrient solution concentration (I) was observed whereas leafarea dry biomass and radiation use efficiency (RUE) were significantly affected by the interaction ofthese two factors (Table 1) Irrespective of the Fe concentration in the nutrient solution the red Salanovahad higher marketable yield and percentage dry matter than those recorded in green Salanova plantsby 93 and 70 respectively (Table 1) Analogous genotypic variation in marketable fresh yieldand leaf dry matter content has been previously demonstrated over seven iceberg cultivars (lsquoEquinosrsquolsquoIce Castlersquo lsquoMetaliarsquo lsquoNum 189rsquo lsquoSilvinasrsquo lsquoOmbrinasrsquo and lsquoVanguardiarsquo [36])

Agronomy 2019 9 290 5 of 21

25 Total Ascorbic Acid Analysis

The total ascorbic acid defined as dehydroascorbic (DHA) and ascorbic acid (AA) was determinedby UVndashVis spectrophotometry (Hach DR 2000 Hach Co Loveland CO USA) as described byKampfenkel et al [42] Briefly 400 mg of sample fresh plant tissues were extracted with trichloroaceticacid (TCA) 6 200 microL of the extract was solubilized with 22-dipyridyl (14454 Sigma-Aldrich StLouis MO USA) The assay is based on the reduction of Fe3+ to Fe2+ by total ascorbic acid andthe spectrophotometric detection of Fe2+ complexes with 22-dipyridyl DHA is reduced to ASA bypre-incubation of the sample with dithiothreitol (DTT) The absorbance of the solution was measuredat 525 nm and data were expressed as mg AA per 100 g fw

26 Lipophilic Antioxidant Activity Analysis

The lipophilic antioxidant activity (LAA) was extracted from freeze-dried butterhead lettuceleaves (200 mg) with methanol and the antioxidant activity of this extract was measured with the22prime-azinobis 3-ethylbenzothiazoline-6-sulfonic acid ABTS method [43] Similarly to the total ascorbicacid LAA was determined by UVndashVis spectrophotometry The absorbance of the solutions wasmeasured at 734 nm LAA fraction was expressed as mmol Trolox (6-hydroxy-2578-tetramethylchroman-2-carboxylic acid) per 100 g dw

27 Phenolic Acids and Anthocyanins Identification and Quantification

400 mg of lyophilized samples was solubilized in a solution of methanolwaterformic acid(50455 vvv 12 mL) as described by Llorach et al [28] to determine phenolic acids as hydroxycinnamicderivatives The suspensions were sonicated for 30 min and then centrifuged (2500 g for 30 min at4 C) After a second centrifugation of supernatants at 21100 g for 15 min at 4 C samples were filteredthrough 022 microm cellulose filters (Phenomenex) A reversed phase C18 column (Prodigy 250 times 46 mm5 microm Phenomenex Torrance CA) equipped with a C18 security guard (40 times 30 mm Phenomenex)was used for the separation of hydroxycinnamic derivatives and anthocyanins 20microL of each extract wasinjected and the following mobile phases was used (A) water formic acid (955 vv) and (B) methanolthrough the following gradient of solvent B (t in [min][B]) (05) (2540) (3240) The flow ratewas 1 mL minminus1 LC column was installed onto a binary system (LC-10AD Shimadzu Kyoto Japan)equipped with a DAD (SPD-M10A Shimadzu Kyoto Japan) and a Series 200 autosampler (PerkinElmer Waltham MA) Chlorogenic and chicoric acids at 330 nm were used for the calibration curves ofhydroxycinnamic derivatives Identification of caffeoyl-meso-tartaric acid and caffeoyl-tartaric acidwas performed by LC-MSMS experiments

The chromatographic profiles of reference curves and samples were recorded in multiple reactionmonitoring mode (MRM) by using an API 3000 triple quadrupole (ABSciex Carlsbad CA) Negativeelectrospray ionization was used for detection and source parameters were selected as follows sprayvoltage minus42 kV capillary temperature 400 C dwell time 100 ms nebulizer gas and cad gas were set to10 and 12 respectively (arbitrary units) Target compounds [M-H] were analyzed using mass transitionsgiven in parentheses Chicoric acid (mz 473rarr311 293) chlorogenic acid (mz 353rarr191) caffeoyltartaric acid (mz 311rarr179 149 retention time 158 min) caffeoyl-meso-tartaric acid (mz 311rarr179 149retention time 178 min) The concentration of phenolic acids was reported as mg 100 gminus1 of dw

Anthocyanins were also measured within the same LC-DAD chromatographic runs at 520 nmand the concentration calculated by using cyanidin as reference standard to calculate the concentrationThe results were reported as microg of cyanidin equivalent per g of samples

28 Carotenoids Identification and Quantification

One gram of lyophilized samples was used to determine carotenoids content following the methodof Vallverduacute-Queralt et al [44] with slight modifications Samples were solubilized in ethanolhexane(43 vv 25 mL) with 1 BHT vortexed at 22 C for 30 s and sonicated for 5 min in the dark Then

Agronomy 2019 9 290 6 of 21

the solution was centrifuged (2500 g 4 C 10 min) and filtered through 045 microm nylon syringe filters(Phenomenex Torrance CA USA) The extracts were dried in N and the dried extracts were dissolvedin 1 BHT in chloroform 20 microL of each sample was injected onto a C18 column (Prodigy 250 times 46mm 5 microm Phenomenex Torrance C A USA) with a C18 security guard (40 times 30 mm Phenomenex)Two mobile phases were used (A) acetonitrile hexane methanol and dichloromethane (4222vvvv) and (B) acetonitrile Carotenoids were eluted at 08 mL minminus1 through the following gradientof solvent B (t in [min][B]) (070) (2060) (3030) (402) Carotenoids were quantified by a binaryLC-10AD system connected to a DAD (SPD-M10A Shimadzu Kyoto Japan) equipped with a Series200 auto-sampler (Perkin Elmer Waltham MA USA) Violaxanthin neoxanthin β-cryptoxanthinlutein and β-carotene were used as reference standards Identification of the peaks was achieved bycomparison of UV-vis spectra and retention times of eluted compounds with pure standards at 450 nmThree separate sets of calibration curves were built each set was injected three times in the same day(intraday assay) and three times in three different days (interday assay) The accuracy was reported asthe discrepancies between the calibration curves performed intraday and interday and the results wereexpressed as relative standard deviation RSD () A recovery test was performed spiking two sampleswith two known amounts of carotenoids (50 and 100 microg mLminus1 final concentration) and taking intoaccount the overestimation due to the target analytes already present in the samples The concentrationof the target carotenoids was expressed as microg gminus1 dw

29 Statistics

The ShapirondashWilk and KolmororovndashSmirnov procedures were performed to verify that thedata had a normal distribution and the Levene OrsquoBrien and Bartlet tests were conducted to verifythe homogeneity of variances Then all morphometric nutritional and functional quality datawere subjected to analysis of variance (two-way ANOVA) using IBM SPSS 20 software package(wwwibmcomsoftwareanalyticsspss) The means were separated by Tukeyrsquos honestly significantdifference (HSD) test (significance level 005) Butterhead lettuce cultivar main effects were comparedby t-Test

3 Results and Discussion

31 Growth Response Fresh Yield Dry Matter and Radiation Use Efficiency

Inter and intra-specific genetic variability is among the most important preharvest factors whichinfluence lettucersquos phenotypic and biochemical traits Lettuce presents within the same speciesa variety of colors sizes textures and shapes [45ndash47] In our study two cultivars of green and redSalanova were evaluated from a productive and nutritional point of view in response to different Feconcentrations in the nutrient solution

For leaf number per plant marketable fresh yield and leaf dry matter percentage no significantinteraction between cultivar (C) and Fe nutrient solution concentration (I) was observed whereas leafarea dry biomass and radiation use efficiency (RUE) were significantly affected by the interaction ofthese two factors (Table 1) Irrespective of the Fe concentration in the nutrient solution the red Salanovahad higher marketable yield and percentage dry matter than those recorded in green Salanova plantsby 93 and 70 respectively (Table 1) Analogous genotypic variation in marketable fresh yieldand leaf dry matter content has been previously demonstrated over seven iceberg cultivars (lsquoEquinosrsquolsquoIce Castlersquo lsquoMetaliarsquo lsquoNum 189rsquo lsquoSilvinasrsquo lsquoOmbrinasrsquo and lsquoVanguardiarsquo [36])

Agronomy 2019 9 290 6 of 21

the solution was centrifuged (2500 g 4 C 10 min) and filtered through 045 microm nylon syringe filters(Phenomenex Torrance CA USA) The extracts were dried in N and the dried extracts were dissolvedin 1 BHT in chloroform 20 microL of each sample was injected onto a C18 column (Prodigy 250 times 46mm 5 microm Phenomenex Torrance C A USA) with a C18 security guard (40 times 30 mm Phenomenex)Two mobile phases were used (A) acetonitrile hexane methanol and dichloromethane (4222vvvv) and (B) acetonitrile Carotenoids were eluted at 08 mL minminus1 through the following gradientof solvent B (t in [min][B]) (070) (2060) (3030) (402) Carotenoids were quantified by a binaryLC-10AD system connected to a DAD (SPD-M10A Shimadzu Kyoto Japan) equipped with a Series200 auto-sampler (Perkin Elmer Waltham MA USA) Violaxanthin neoxanthin β-cryptoxanthinlutein and β-carotene were used as reference standards Identification of the peaks was achieved bycomparison of UV-vis spectra and retention times of eluted compounds with pure standards at 450 nmThree separate sets of calibration curves were built each set was injected three times in the same day(intraday assay) and three times in three different days (interday assay) The accuracy was reported asthe discrepancies between the calibration curves performed intraday and interday and the results wereexpressed as relative standard deviation RSD () A recovery test was performed spiking two sampleswith two known amounts of carotenoids (50 and 100 microg mLminus1 final concentration) and taking intoaccount the overestimation due to the target analytes already present in the samples The concentrationof the target carotenoids was expressed as microg gminus1 dw

29 Statistics

The ShapirondashWilk and KolmororovndashSmirnov procedures were performed to verify that thedata had a normal distribution and the Levene OrsquoBrien and Bartlet tests were conducted to verifythe homogeneity of variances Then all morphometric nutritional and functional quality datawere subjected to analysis of variance (two-way ANOVA) using IBM SPSS 20 software package(wwwibmcomsoftwareanalyticsspss) The means were separated by Tukeyrsquos honestly significantdifference (HSD) test (significance level 005) Butterhead lettuce cultivar main effects were comparedby t-Test

3 Results and Discussion

31 Growth Response Fresh Yield Dry Matter and Radiation Use Efficiency

Inter and intra-specific genetic variability is among the most important preharvest factors whichinfluence lettucersquos phenotypic and biochemical traits Lettuce presents within the same speciesa variety of colors sizes textures and shapes [45ndash47] In our study two cultivars of green and redSalanova were evaluated from a productive and nutritional point of view in response to different Feconcentrations in the nutrient solution

For leaf number per plant marketable fresh yield and leaf dry matter percentage no significantinteraction between cultivar (C) and Fe nutrient solution concentration (I) was observed whereas leafarea dry biomass and radiation use efficiency (RUE) were significantly affected by the interaction ofthese two factors (Table 1) Irrespective of the Fe concentration in the nutrient solution the red Salanovahad higher marketable yield and percentage dry matter than those recorded in green Salanova plantsby 93 and 70 respectively (Table 1) Analogous genotypic variation in marketable fresh yieldand leaf dry matter content has been previously demonstrated over seven iceberg cultivars (lsquoEquinosrsquolsquoIce Castlersquo lsquoMetaliarsquo lsquoNum 189rsquo lsquoSilvinasrsquo lsquoOmbrinasrsquo and lsquoVanguardiarsquo [36])

Agronomy 2019 9 290 7 of 21

Table 1 Analysis of variance and mean comparisons for leaf area leaf number fresh yield shoot dry biomass leaf dry matter percentage and radiation use efficiency(RUE) for green and red Salanova butterhead lettuce grown under increasing Fe concentration in the nutrient solution

Source of VarianceLeaf Area Leaf Number Fresh Biomass Dry Biomass Dry Matter RUE

(cm2 Plantminus1) (no Plantminus1) (g Plantminus1) (g Plantminus1) () (g molminus1)

Cultivar (C)Green Salanova 1070 plusmn 64 4594 plusmn 311 6073 plusmn 722 321 plusmn 019 532 plusmn 038 014 plusmn 001Red Salanova 1211 plusmn 74 5446 plusmn 233 6637 plusmn 785 376 plusmn 030 569 plusmn 031 017 plusmn 001

t-value 0000 0000 0081 0000 0016 0000Iron (mM Fe) (I)

0015 1196 plusmn 125 a 5010 plusmn 513 7150 plusmn 470 a 372 plusmn 049 a 519 plusmn 039 c 017 plusmn 002 a05 1185 plusmn 63 a 4968 plusmn 426 6740 plusmn 267 a 359 plusmn 021 ab 532 plusmn 019 bc 016 plusmn 001 ab1 1091 plusmn 64 b 4866 plusmn 648 6185 plusmn 552 b 343 plusmn 033 b 555 plusmn 016 b 015 plusmn 001 b2 1090 plusmn 96 b 5236 plusmn 498 5344 plusmn 291 c 319 plusmn 025 c 596 plusmn 027 a 014 plusmn 001 c

ns C times I

Green Salanova times 0015 mM Fe 1089 plusmn 48 cde 4557 plusmn 086 6767 plusmn 135 329 plusmn 007 de 486 plusmn 018 015 plusmn 000 cdGreen Salanova times 05 mM Fe 1135 plusmn 33 bcd 4680 plusmn 345 6613 plusmn 090 341 plusmn 010 bcd 516 plusmn 008 015 plusmn 000 bcGreen Salanova times 1 mM Fe 1052 plusmn 72 de 4296 plusmn 234 5751 plusmn 427 315 plusmn 014 de 549 plusmn 022 014 plusmn 001 cdGreen Salanova times 2 mM Fe 1004 plusmn 18 e 4845 plusmn 332 5161 plusmn 302 297 plusmn 009 e 577 plusmn 016 013 plusmn 000 d

Red Salanova times 0015 mM Fe 1302 plusmn 50 a 5462 plusmn 190 7534 plusmn 305 415 plusmn 022 a 551 plusmn 019 019 plusmn 001 aRed Salanova times 05 mM Fe 1236 plusmn 31 ab 5257 plusmn 292 6866 plusmn 349 376 plusmn 012 b 548 plusmn 010 017 plusmn 001 bRed Salanova times 1 mM Fe 1130 plusmn 25 bcd 5435 plusmn 148 6620 plusmn 116 372 plusmn 007 bc 562 plusmn 007 017 plusmn 000 bRed Salanova times 2 mM Fe 1177plusmn 23 bc 5628 plusmn 226 5526 plusmn 143 340 plusmn 011 cd 616 plusmn 020 015 plusmn 001 bc

ns ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test(P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 8 of 21

When averaged over both Salanova cultivars the percentage of yield reduction in comparison tothe control treatment (0015 mM Fe) was 57 135 and 253 at 05 10 and 20 mM Fe concentrationin the nutrient solution respectively whereas an opposite trend was observed for the leaf dry matterpercentage (Table 1) Furthermore the highest total leaf area was recorded in red Salanova treatedwith both 0015 and 05 mM Fe Although the highest shoot dry biomass was recorded in red Salanovatreated with 0015 mM Fe increasing the Fe concentration in the nutrient solution from 0015 to 20decreased the dry biomass with more detrimental effect recorded in red-pigmented butterhead lettuceSpecifically the percentage of dry biomass reduction in comparison to control plants (0015 mM Fe)ranged from 43 to 97 in green Salanova and from 104 to 181 in red Salanova at 10 and20 mM Fe concentration in the nutrient solution respectively (Table 1) The high Fe-tolerance ofgreen-pigmented butterhead lettuce at both Fe concentrations (10 and 20 mM) may be due to thelower accumulation of Fe in leaf tissue compared to red-pigmented Salanova (Figure 1) Similarlyto the effects on shoot dry biomass the RUE in red Salanova under control Fe treatment (0015 mM)exhibited the highest values (Table 1)

In the current study diamino-di- (ortho-hydroxy phenyl acetic) acid (o o-EDDHA) was usedas Fe chelate because the final amount of dissolved Fe released in solution is greater than otherforms of Fe chelates and it can be considered as a good supplement in nutrient solutions for soillesscultivation [48ndash50] Roosta et al [49] have reported increases in the number of leaves and total leafarea of Capsicum annuum L after the application of 10 microM of EDDHA to the nutrient solution Similarresults have been also observed by the same authors for four lettuce varieties grown with 20 microM Fe inan NFT hydroponic system [50]

The results on growth parameters yield and shoot dry biomass of this study are in agreementwith those reported by Filho et al [51] who cultivated Cichorium intybus in an NFT using increasedFe concentrations in the nutrient solution (09 27 83 and 25 mg Lminus1) where plant height leaf numberper plant as well as plant fresh and dry weight were reduced as Fe concentration increased The sameauthors were able to identify the optimal Fe range (27 to 83 mg Lminus1) while the 25 mg Lminus1 applicationrate had the most harmful effects on plant growth and productivity While in fact Fe is essential forplant growth it is also involved in the Fenton reaction that leads to the formation of reactive oxygenspecies (ROS) which in turn can lead to cell destruction because they react with polyunsaturated fattyacids proteins and nucleic acids [35253] In the De Dorlodot et al [52] study three concentrationsof Fe2+ (0 125 and 250 mg Lminus1) were used for greenhouse cultivation of rice plants in hydroponicsystem The intermediate dose of 125 mg Lminus1 produced the maximum fresh and dry weights whereasat the higher dose of 250 mg Lminus1 a significant reduction in both fresh and dry plant weights wasincurred as well as in water content A putative mechanism involved in reduced fresh and dry biomassaccumulation might be the excessive exposure to Fe (especially at 20 mM) which increases peroxidehydrogen generation causing the overproduction of ROS which irreversibly leads to membrane lipidoxidation impairs cellular structure and damages DNA and proteins [5455]

32 Nitrate Content Mineral Composition and Iron Biofortification

The nitrate content recorded among the eight treatments was within the maximum nitrate contentallowable for the commercialization of fresh lettuce (4000-5000 mg NO3

minus kgminus1 fw depending onharvest period andor growing conditions) according to Commission regulation (EU) No 12582011 [56]The nitrate content varied considerably across the eight treatments (C times I interaction) with the highestnitrate concentration found in green Salanova treated with 2 mM Fe (Table 2) In the current experimentincreasing the Fe concentration in the nutrient solution from 0015 to 20 mM increased the nitratecontent in red Salanova (by 88) but especially in green Salanova (by 273) plants (Table 2) Similarresults have been reported by Liu et al [57] who also suggested a positive correlation between Fesupplementation and nitrate content in hydroponically grown lettuce leaves The high nitrate uptakeand accumulation under high Fe availability has been attributed to a molecular mechanism involvedin the up-regulation of LATS gene (coding for a low-affinity NO3

minus transporter) observed in corn saladgrown in a floating raft system [58]

Agronomy 2019 9 290 9 of 21

It has been demonstrated that a number of dietary macro-minerals such as P K Ca and Mgare crucial components of the human diet due to their multifaceted nutraceutical properties such aslowering blood pressure and hypertension (K) promoting bone health and reducing osteoporosis(P Ca and Mg) [29] Our results on the mineral profile of green and red-pigmented butterhead lettucewere proximate to those reported by the National Nutrient Database for Standard References [59]and by several authors [296061] on green and red leaf lettuce including the butterhead typeK (48ndash72 mg gminus1 dw) P (4ndash6 mg gminus1 dw) Mg (14ndash28 mg gminus1 dw) and Ca (4ndash10 mg gminus1 dw)

In our study a significant interaction between the tested factors was observed in the case of Caand Mg whereas P and K were affected only by cultivar and Fe concentration in the nutrient solutionwith no C times I interaction (Table 2) The P and K concentrations in red Salanova were higher (P lt

0001) by 77 and 101 respectively than those observed in green Salanova whereas an oppositetrend was recorded for Ca Moreover Ca and Mg concentrations were the highest in red Salanovatreated with 0015 mM Fe whereas the lowest values of Ca were also observed in the red-pigmentedcultivar treated with 20 mM Fe (Table 2) In fact Fe can cause alteration of mineral composition statusdue to the competition between Mg and Fe ions for occupying the chlorophyll ring Moreover asshown by De Dorlodot et al [52] the contents of rice plants grown under soilless conditions in P Caand Mg were reduced by Fe application rates of 125 and especially 250 mg Lminus1 as compared to thecontrol Furthermore K Ca and Mg reductions had been observed in radish broccoli alfalfa andmung bean after the nutrient solution was enriched with Fe [3] The decreased leaf mineral statusespecially at 10 and 20 mM Fe may be attributed to several mechanisms such as i) root injury(ie formation of root coat) caused by excessive Fe stress impairing nutrient absorption [62] ii) theincreasing competition between Fe and other cations in particular Mg for absorption sites owing toion transportersrsquo lack of specificity [63] and iii) lipid peroxidation and oxidation of enzyme sulfhydrylgroups causing the irreversible inhibition of plasma membrane H+-ATPases [64]

The importance of Fe for human health is linked to the synthesis of hemoglobin and oxygentransport Plants contain Fe only in trace amounts hence particular attention is given to this mineralfrom a human diet perspective especially for vegans who place vegetables at the core of their dietMoreover plants partly contain Fe in non-heme (non-chelated) forms which are less bioavailable thanthe heme Fe found in animal-based foods [29] However Fe biofortification can be achieved in leafyvegetables including lettuce [20] though its effectiveness can vary between speciescultivars [365]Based on the review of Kim and co-workers [29] 100 g of fresh lettuce in particular butterhead that wasused in the current experiment can provide 2 to 15 (without biofortification) of the recommendeddaily Fe intake of 8ndash18 mg dayminus1 according to age gender and body weight indicating that Febiofortification of lettuce leaves can potentially improve the nutritional status of humans In ourstudy the Fe accumulation in green and red butterhead lettuce ranged from 406 to 755 mg kgminus1 dwand from 660 to 930 mg kgminus1 dw respectively (Figure 1) Moreover the red leaf lettuce cultivar(avg 771 mg kgminus1 dw) accumulated 455 more Fe than the green one (avg 529 mg kgminus1 dw)(Figure 1) which is in agreement with previous results on red and green-pigmented lettuce [293261]Inversely to macro-minerals the Fe content recorded in the current experiment differed from thevalues reported in butterhead lettuce by Kawashima and Soares [61] (100 microg gminus1 dw) and Baslamand co-workers [61] (758-112 mg kgminus1 dw) Such differences in Fe content reported in the scientificliterature could be associated to different farming practices environmental conditions as well asto the cultivars tested [34] Regardless of cultivar the addition of 10 mM and especially 20 mMFe in the nutrient solution lettuce leaves incurred a significant increase of Fe content by 205 and537 (avg 661 and 843 mg kgminus1 dw respectively) (Figure 1) demonstrating that the productionof Fe-enriched lettuce with absence of defects and decay using closed soilless cultivation is feasibleHowever elucidating the physiological and especially the molecular mechanisms facilitating Fe uptakein interaction with genotype pose the future challenge confronting the horticultural industry beforeachieving the production of leafy vegetables of superior functional quality

Agronomy 2019 9 290 10 of 21

Table 2 Analysis of variance and mean comparisons leaf mineral composition in green and red Salanova butterhead lettuce grown under increasing Fe concentrationin the nutrient solution

Source of VarianceNitrate P K Ca Mg

(mg kgminus1 FW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW)

Cultivar (C)Green Salanova 2277 plusmn 224 481 plusmn 028 6267 plusmn 239 647 plusmn 072 201 plusmn 016Red Salanova 2105 plusmn 93 517 plusmn 016 6901 plusmn 318 573 plusmn 105 206 plusmn 032

t-value 0023 0001 0000 0056 0615Iron (mM Fe) (I)

0015 1991 plusmn 41 c 484 plusmn 030 b 6732 plusmn 416 ab 708 plusmn 041 a 230 plusmn 018 a05 2228 plusmn 115 b 510 plusmn 029 ab 6857 plusmn 405 a 650 plusmn 063 b 214 plusmn 005 ab1 2192 plusmn 136 b 485 plusmn 028 b 6542 plusmn 365 b 569 plusmn 059 c 197 plusmn 016 b2 2351 plusmn 226 a 515 plusmn 019 a 6205 plusmn 265 c 512 plusmn 076 c 174 plusmn 013 c

C times I

Green Salanova times 0015 mM Fe 1999 plusmn 27 c 458 plusmn 005 6371 plusmn 188 701 plusmn 062 abc 217 plusmn 018 abGreen Salanova times 05 mM Fe 2267 plusmn 163 b 488 plusmn 017 6512 plusmn 172 705 plusmn 030 ab 209plusmn 001 bcGreen Salanova times 1 mM Fe 2295 plusmn 109 b 467 plusmn 032 6218 plusmn 054 603 plusmn 062 bcd 193 plusmn 006 bcdGreen Salanova times 2 mM Fe 2545 plusmn 110 a 509 plusmn 026 5967 plusmn 033 579 plusmn 026 d 184 plusmn 005 cd

Red Salanova times 0015 mM Fe 1983 plusmn 57 c 510 plusmn 016 7093 plusmn 082 716 plusmn 017 a 243 plusmn 005 aRed Salanova times 05 mM Fe 2190 plusmn 42 bc 533 plusmn 016 7202 plusmn 150 595 plusmn 007 cd 219 plusmn 002 abRed Salanova times 1 mM Fe 2089 plusmn 50 bc 503 plusmn 006 6866 plusmn 121 535 plusmn 038 d 200 plusmn 023 bcRed Salanova times 2 mM Fe 2157 plusmn 53 bc 521 plusmn 012 6443 plusmn 064 445 plusmn 020 e 163 plusmn 007 d

ns ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to TukeyrsquosHSD test (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 11 of 21

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The most important role played by ascorbic acid is its radical scavenging power derived from its oxidation to the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to its ability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reduces extracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid including ascorbic and dehydroascorbic acid varied considerably across treatments with the highest values (194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10 and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and 10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmented lettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627 (Figure 2) Our findings are in agreement with other previous works where the concentration and activity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3] For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Fe concentration in the nutrient solution increased [3]

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The mostimportant role played by ascorbic acid is its radical scavenging power derived from its oxidationto the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to itsability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reducesextracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid includingascorbic and dehydroascorbic acid varied considerably across treatments with the highest values(194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmentedlettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627(Figure 2) Our findings are in agreement with other previous works where the concentration andactivity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3]For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Feconcentration in the nutrient solution increased [3]

Phenolic compounds including mainly phenolic acids and flavonoids refer to an importantgroup of secondary metabolites having great antioxidant activity and beneficial effects against chronicdiseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis ofthe lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile

Agronomy 2019 9 290 12 of 21

(Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenicacid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrationsMoreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for thetarget and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaricacid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as totalphenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce whichis in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]Agronomy 2019 9 x FOR PEER REVIEW 2 of 23

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Phenolic compounds including mainly phenolic acids and flavonoids refer to an important group of secondary metabolites having great antioxidant activity and beneficial effects against chronic diseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis of the lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile (Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenic acid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrations Moreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for the target and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaric acid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as total phenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce which is in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]

Concerning the Fe nutrient solution management no significant effects on target and total phenolic acids were observed in green Salanova with the increasing Fe application rate Contrarily to the green-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mM induced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmented lettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higher accumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mM in comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the two main phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentration in the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors of Fe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlation between the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated on several horticultural species such as broccoli mung beans grape berries and radish [367]

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Concerning the Fe nutrient solution management no significant effects on target and total phenolicacids were observed in green Salanova with the increasing Fe application rate Contrarily to thegreen-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mMinduced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmentedlettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higheraccumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mMin comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the twomain phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentrationin the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors ofFe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlationbetween the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated onseveral horticultural species such as broccoli mung beans grape berries and radish [367]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 8 of 21

When averaged over both Salanova cultivars the percentage of yield reduction in comparison tothe control treatment (0015 mM Fe) was 57 135 and 253 at 05 10 and 20 mM Fe concentrationin the nutrient solution respectively whereas an opposite trend was observed for the leaf dry matterpercentage (Table 1) Furthermore the highest total leaf area was recorded in red Salanova treatedwith both 0015 and 05 mM Fe Although the highest shoot dry biomass was recorded in red Salanovatreated with 0015 mM Fe increasing the Fe concentration in the nutrient solution from 0015 to 20decreased the dry biomass with more detrimental effect recorded in red-pigmented butterhead lettuceSpecifically the percentage of dry biomass reduction in comparison to control plants (0015 mM Fe)ranged from 43 to 97 in green Salanova and from 104 to 181 in red Salanova at 10 and20 mM Fe concentration in the nutrient solution respectively (Table 1) The high Fe-tolerance ofgreen-pigmented butterhead lettuce at both Fe concentrations (10 and 20 mM) may be due to thelower accumulation of Fe in leaf tissue compared to red-pigmented Salanova (Figure 1) Similarlyto the effects on shoot dry biomass the RUE in red Salanova under control Fe treatment (0015 mM)exhibited the highest values (Table 1)

In the current study diamino-di- (ortho-hydroxy phenyl acetic) acid (o o-EDDHA) was usedas Fe chelate because the final amount of dissolved Fe released in solution is greater than otherforms of Fe chelates and it can be considered as a good supplement in nutrient solutions for soillesscultivation [48ndash50] Roosta et al [49] have reported increases in the number of leaves and total leafarea of Capsicum annuum L after the application of 10 microM of EDDHA to the nutrient solution Similarresults have been also observed by the same authors for four lettuce varieties grown with 20 microM Fe inan NFT hydroponic system [50]

The results on growth parameters yield and shoot dry biomass of this study are in agreementwith those reported by Filho et al [51] who cultivated Cichorium intybus in an NFT using increasedFe concentrations in the nutrient solution (09 27 83 and 25 mg Lminus1) where plant height leaf numberper plant as well as plant fresh and dry weight were reduced as Fe concentration increased The sameauthors were able to identify the optimal Fe range (27 to 83 mg Lminus1) while the 25 mg Lminus1 applicationrate had the most harmful effects on plant growth and productivity While in fact Fe is essential forplant growth it is also involved in the Fenton reaction that leads to the formation of reactive oxygenspecies (ROS) which in turn can lead to cell destruction because they react with polyunsaturated fattyacids proteins and nucleic acids [35253] In the De Dorlodot et al [52] study three concentrationsof Fe2+ (0 125 and 250 mg Lminus1) were used for greenhouse cultivation of rice plants in hydroponicsystem The intermediate dose of 125 mg Lminus1 produced the maximum fresh and dry weights whereasat the higher dose of 250 mg Lminus1 a significant reduction in both fresh and dry plant weights wasincurred as well as in water content A putative mechanism involved in reduced fresh and dry biomassaccumulation might be the excessive exposure to Fe (especially at 20 mM) which increases peroxidehydrogen generation causing the overproduction of ROS which irreversibly leads to membrane lipidoxidation impairs cellular structure and damages DNA and proteins [5455]

32 Nitrate Content Mineral Composition and Iron Biofortification

The nitrate content recorded among the eight treatments was within the maximum nitrate contentallowable for the commercialization of fresh lettuce (4000-5000 mg NO3

minus kgminus1 fw depending onharvest period andor growing conditions) according to Commission regulation (EU) No 12582011 [56]The nitrate content varied considerably across the eight treatments (C times I interaction) with the highestnitrate concentration found in green Salanova treated with 2 mM Fe (Table 2) In the current experimentincreasing the Fe concentration in the nutrient solution from 0015 to 20 mM increased the nitratecontent in red Salanova (by 88) but especially in green Salanova (by 273) plants (Table 2) Similarresults have been reported by Liu et al [57] who also suggested a positive correlation between Fesupplementation and nitrate content in hydroponically grown lettuce leaves The high nitrate uptakeand accumulation under high Fe availability has been attributed to a molecular mechanism involvedin the up-regulation of LATS gene (coding for a low-affinity NO3

minus transporter) observed in corn saladgrown in a floating raft system [58]

Agronomy 2019 9 290 9 of 21

It has been demonstrated that a number of dietary macro-minerals such as P K Ca and Mgare crucial components of the human diet due to their multifaceted nutraceutical properties such aslowering blood pressure and hypertension (K) promoting bone health and reducing osteoporosis(P Ca and Mg) [29] Our results on the mineral profile of green and red-pigmented butterhead lettucewere proximate to those reported by the National Nutrient Database for Standard References [59]and by several authors [296061] on green and red leaf lettuce including the butterhead typeK (48ndash72 mg gminus1 dw) P (4ndash6 mg gminus1 dw) Mg (14ndash28 mg gminus1 dw) and Ca (4ndash10 mg gminus1 dw)

In our study a significant interaction between the tested factors was observed in the case of Caand Mg whereas P and K were affected only by cultivar and Fe concentration in the nutrient solutionwith no C times I interaction (Table 2) The P and K concentrations in red Salanova were higher (P lt

0001) by 77 and 101 respectively than those observed in green Salanova whereas an oppositetrend was recorded for Ca Moreover Ca and Mg concentrations were the highest in red Salanovatreated with 0015 mM Fe whereas the lowest values of Ca were also observed in the red-pigmentedcultivar treated with 20 mM Fe (Table 2) In fact Fe can cause alteration of mineral composition statusdue to the competition between Mg and Fe ions for occupying the chlorophyll ring Moreover asshown by De Dorlodot et al [52] the contents of rice plants grown under soilless conditions in P Caand Mg were reduced by Fe application rates of 125 and especially 250 mg Lminus1 as compared to thecontrol Furthermore K Ca and Mg reductions had been observed in radish broccoli alfalfa andmung bean after the nutrient solution was enriched with Fe [3] The decreased leaf mineral statusespecially at 10 and 20 mM Fe may be attributed to several mechanisms such as i) root injury(ie formation of root coat) caused by excessive Fe stress impairing nutrient absorption [62] ii) theincreasing competition between Fe and other cations in particular Mg for absorption sites owing toion transportersrsquo lack of specificity [63] and iii) lipid peroxidation and oxidation of enzyme sulfhydrylgroups causing the irreversible inhibition of plasma membrane H+-ATPases [64]

The importance of Fe for human health is linked to the synthesis of hemoglobin and oxygentransport Plants contain Fe only in trace amounts hence particular attention is given to this mineralfrom a human diet perspective especially for vegans who place vegetables at the core of their dietMoreover plants partly contain Fe in non-heme (non-chelated) forms which are less bioavailable thanthe heme Fe found in animal-based foods [29] However Fe biofortification can be achieved in leafyvegetables including lettuce [20] though its effectiveness can vary between speciescultivars [365]Based on the review of Kim and co-workers [29] 100 g of fresh lettuce in particular butterhead that wasused in the current experiment can provide 2 to 15 (without biofortification) of the recommendeddaily Fe intake of 8ndash18 mg dayminus1 according to age gender and body weight indicating that Febiofortification of lettuce leaves can potentially improve the nutritional status of humans In ourstudy the Fe accumulation in green and red butterhead lettuce ranged from 406 to 755 mg kgminus1 dwand from 660 to 930 mg kgminus1 dw respectively (Figure 1) Moreover the red leaf lettuce cultivar(avg 771 mg kgminus1 dw) accumulated 455 more Fe than the green one (avg 529 mg kgminus1 dw)(Figure 1) which is in agreement with previous results on red and green-pigmented lettuce [293261]Inversely to macro-minerals the Fe content recorded in the current experiment differed from thevalues reported in butterhead lettuce by Kawashima and Soares [61] (100 microg gminus1 dw) and Baslamand co-workers [61] (758-112 mg kgminus1 dw) Such differences in Fe content reported in the scientificliterature could be associated to different farming practices environmental conditions as well asto the cultivars tested [34] Regardless of cultivar the addition of 10 mM and especially 20 mMFe in the nutrient solution lettuce leaves incurred a significant increase of Fe content by 205 and537 (avg 661 and 843 mg kgminus1 dw respectively) (Figure 1) demonstrating that the productionof Fe-enriched lettuce with absence of defects and decay using closed soilless cultivation is feasibleHowever elucidating the physiological and especially the molecular mechanisms facilitating Fe uptakein interaction with genotype pose the future challenge confronting the horticultural industry beforeachieving the production of leafy vegetables of superior functional quality

Agronomy 2019 9 290 10 of 21

Table 2 Analysis of variance and mean comparisons leaf mineral composition in green and red Salanova butterhead lettuce grown under increasing Fe concentrationin the nutrient solution

Source of VarianceNitrate P K Ca Mg

(mg kgminus1 FW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW)

Cultivar (C)Green Salanova 2277 plusmn 224 481 plusmn 028 6267 plusmn 239 647 plusmn 072 201 plusmn 016Red Salanova 2105 plusmn 93 517 plusmn 016 6901 plusmn 318 573 plusmn 105 206 plusmn 032

t-value 0023 0001 0000 0056 0615Iron (mM Fe) (I)

0015 1991 plusmn 41 c 484 plusmn 030 b 6732 plusmn 416 ab 708 plusmn 041 a 230 plusmn 018 a05 2228 plusmn 115 b 510 plusmn 029 ab 6857 plusmn 405 a 650 plusmn 063 b 214 plusmn 005 ab1 2192 plusmn 136 b 485 plusmn 028 b 6542 plusmn 365 b 569 plusmn 059 c 197 plusmn 016 b2 2351 plusmn 226 a 515 plusmn 019 a 6205 plusmn 265 c 512 plusmn 076 c 174 plusmn 013 c

C times I

Green Salanova times 0015 mM Fe 1999 plusmn 27 c 458 plusmn 005 6371 plusmn 188 701 plusmn 062 abc 217 plusmn 018 abGreen Salanova times 05 mM Fe 2267 plusmn 163 b 488 plusmn 017 6512 plusmn 172 705 plusmn 030 ab 209plusmn 001 bcGreen Salanova times 1 mM Fe 2295 plusmn 109 b 467 plusmn 032 6218 plusmn 054 603 plusmn 062 bcd 193 plusmn 006 bcdGreen Salanova times 2 mM Fe 2545 plusmn 110 a 509 plusmn 026 5967 plusmn 033 579 plusmn 026 d 184 plusmn 005 cd

Red Salanova times 0015 mM Fe 1983 plusmn 57 c 510 plusmn 016 7093 plusmn 082 716 plusmn 017 a 243 plusmn 005 aRed Salanova times 05 mM Fe 2190 plusmn 42 bc 533 plusmn 016 7202 plusmn 150 595 plusmn 007 cd 219 plusmn 002 abRed Salanova times 1 mM Fe 2089 plusmn 50 bc 503 plusmn 006 6866 plusmn 121 535 plusmn 038 d 200 plusmn 023 bcRed Salanova times 2 mM Fe 2157 plusmn 53 bc 521 plusmn 012 6443 plusmn 064 445 plusmn 020 e 163 plusmn 007 d

ns ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to TukeyrsquosHSD test (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 11 of 21

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The most important role played by ascorbic acid is its radical scavenging power derived from its oxidation to the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to its ability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reduces extracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid including ascorbic and dehydroascorbic acid varied considerably across treatments with the highest values (194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10 and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and 10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmented lettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627 (Figure 2) Our findings are in agreement with other previous works where the concentration and activity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3] For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Fe concentration in the nutrient solution increased [3]

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The mostimportant role played by ascorbic acid is its radical scavenging power derived from its oxidationto the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to itsability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reducesextracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid includingascorbic and dehydroascorbic acid varied considerably across treatments with the highest values(194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmentedlettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627(Figure 2) Our findings are in agreement with other previous works where the concentration andactivity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3]For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Feconcentration in the nutrient solution increased [3]

Phenolic compounds including mainly phenolic acids and flavonoids refer to an importantgroup of secondary metabolites having great antioxidant activity and beneficial effects against chronicdiseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis ofthe lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile

Agronomy 2019 9 290 12 of 21

(Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenicacid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrationsMoreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for thetarget and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaricacid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as totalphenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce whichis in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]Agronomy 2019 9 x FOR PEER REVIEW 2 of 23

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Phenolic compounds including mainly phenolic acids and flavonoids refer to an important group of secondary metabolites having great antioxidant activity and beneficial effects against chronic diseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis of the lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile (Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenic acid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrations Moreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for the target and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaric acid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as total phenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce which is in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]

Concerning the Fe nutrient solution management no significant effects on target and total phenolic acids were observed in green Salanova with the increasing Fe application rate Contrarily to the green-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mM induced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmented lettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higher accumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mM in comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the two main phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentration in the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors of Fe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlation between the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated on several horticultural species such as broccoli mung beans grape berries and radish [367]

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Concerning the Fe nutrient solution management no significant effects on target and total phenolicacids were observed in green Salanova with the increasing Fe application rate Contrarily to thegreen-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mMinduced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmentedlettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higheraccumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mMin comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the twomain phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentrationin the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors ofFe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlationbetween the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated onseveral horticultural species such as broccoli mung beans grape berries and radish [367]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 9 of 21

It has been demonstrated that a number of dietary macro-minerals such as P K Ca and Mgare crucial components of the human diet due to their multifaceted nutraceutical properties such aslowering blood pressure and hypertension (K) promoting bone health and reducing osteoporosis(P Ca and Mg) [29] Our results on the mineral profile of green and red-pigmented butterhead lettucewere proximate to those reported by the National Nutrient Database for Standard References [59]and by several authors [296061] on green and red leaf lettuce including the butterhead typeK (48ndash72 mg gminus1 dw) P (4ndash6 mg gminus1 dw) Mg (14ndash28 mg gminus1 dw) and Ca (4ndash10 mg gminus1 dw)

In our study a significant interaction between the tested factors was observed in the case of Caand Mg whereas P and K were affected only by cultivar and Fe concentration in the nutrient solutionwith no C times I interaction (Table 2) The P and K concentrations in red Salanova were higher (P lt

0001) by 77 and 101 respectively than those observed in green Salanova whereas an oppositetrend was recorded for Ca Moreover Ca and Mg concentrations were the highest in red Salanovatreated with 0015 mM Fe whereas the lowest values of Ca were also observed in the red-pigmentedcultivar treated with 20 mM Fe (Table 2) In fact Fe can cause alteration of mineral composition statusdue to the competition between Mg and Fe ions for occupying the chlorophyll ring Moreover asshown by De Dorlodot et al [52] the contents of rice plants grown under soilless conditions in P Caand Mg were reduced by Fe application rates of 125 and especially 250 mg Lminus1 as compared to thecontrol Furthermore K Ca and Mg reductions had been observed in radish broccoli alfalfa andmung bean after the nutrient solution was enriched with Fe [3] The decreased leaf mineral statusespecially at 10 and 20 mM Fe may be attributed to several mechanisms such as i) root injury(ie formation of root coat) caused by excessive Fe stress impairing nutrient absorption [62] ii) theincreasing competition between Fe and other cations in particular Mg for absorption sites owing toion transportersrsquo lack of specificity [63] and iii) lipid peroxidation and oxidation of enzyme sulfhydrylgroups causing the irreversible inhibition of plasma membrane H+-ATPases [64]

The importance of Fe for human health is linked to the synthesis of hemoglobin and oxygentransport Plants contain Fe only in trace amounts hence particular attention is given to this mineralfrom a human diet perspective especially for vegans who place vegetables at the core of their dietMoreover plants partly contain Fe in non-heme (non-chelated) forms which are less bioavailable thanthe heme Fe found in animal-based foods [29] However Fe biofortification can be achieved in leafyvegetables including lettuce [20] though its effectiveness can vary between speciescultivars [365]Based on the review of Kim and co-workers [29] 100 g of fresh lettuce in particular butterhead that wasused in the current experiment can provide 2 to 15 (without biofortification) of the recommendeddaily Fe intake of 8ndash18 mg dayminus1 according to age gender and body weight indicating that Febiofortification of lettuce leaves can potentially improve the nutritional status of humans In ourstudy the Fe accumulation in green and red butterhead lettuce ranged from 406 to 755 mg kgminus1 dwand from 660 to 930 mg kgminus1 dw respectively (Figure 1) Moreover the red leaf lettuce cultivar(avg 771 mg kgminus1 dw) accumulated 455 more Fe than the green one (avg 529 mg kgminus1 dw)(Figure 1) which is in agreement with previous results on red and green-pigmented lettuce [293261]Inversely to macro-minerals the Fe content recorded in the current experiment differed from thevalues reported in butterhead lettuce by Kawashima and Soares [61] (100 microg gminus1 dw) and Baslamand co-workers [61] (758-112 mg kgminus1 dw) Such differences in Fe content reported in the scientificliterature could be associated to different farming practices environmental conditions as well asto the cultivars tested [34] Regardless of cultivar the addition of 10 mM and especially 20 mMFe in the nutrient solution lettuce leaves incurred a significant increase of Fe content by 205 and537 (avg 661 and 843 mg kgminus1 dw respectively) (Figure 1) demonstrating that the productionof Fe-enriched lettuce with absence of defects and decay using closed soilless cultivation is feasibleHowever elucidating the physiological and especially the molecular mechanisms facilitating Fe uptakein interaction with genotype pose the future challenge confronting the horticultural industry beforeachieving the production of leafy vegetables of superior functional quality

Agronomy 2019 9 290 10 of 21

Table 2 Analysis of variance and mean comparisons leaf mineral composition in green and red Salanova butterhead lettuce grown under increasing Fe concentrationin the nutrient solution

Source of VarianceNitrate P K Ca Mg

(mg kgminus1 FW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW)

Cultivar (C)Green Salanova 2277 plusmn 224 481 plusmn 028 6267 plusmn 239 647 plusmn 072 201 plusmn 016Red Salanova 2105 plusmn 93 517 plusmn 016 6901 plusmn 318 573 plusmn 105 206 plusmn 032

t-value 0023 0001 0000 0056 0615Iron (mM Fe) (I)

0015 1991 plusmn 41 c 484 plusmn 030 b 6732 plusmn 416 ab 708 plusmn 041 a 230 plusmn 018 a05 2228 plusmn 115 b 510 plusmn 029 ab 6857 plusmn 405 a 650 plusmn 063 b 214 plusmn 005 ab1 2192 plusmn 136 b 485 plusmn 028 b 6542 plusmn 365 b 569 plusmn 059 c 197 plusmn 016 b2 2351 plusmn 226 a 515 plusmn 019 a 6205 plusmn 265 c 512 plusmn 076 c 174 plusmn 013 c

C times I

Green Salanova times 0015 mM Fe 1999 plusmn 27 c 458 plusmn 005 6371 plusmn 188 701 plusmn 062 abc 217 plusmn 018 abGreen Salanova times 05 mM Fe 2267 plusmn 163 b 488 plusmn 017 6512 plusmn 172 705 plusmn 030 ab 209plusmn 001 bcGreen Salanova times 1 mM Fe 2295 plusmn 109 b 467 plusmn 032 6218 plusmn 054 603 plusmn 062 bcd 193 plusmn 006 bcdGreen Salanova times 2 mM Fe 2545 plusmn 110 a 509 plusmn 026 5967 plusmn 033 579 plusmn 026 d 184 plusmn 005 cd

Red Salanova times 0015 mM Fe 1983 plusmn 57 c 510 plusmn 016 7093 plusmn 082 716 plusmn 017 a 243 plusmn 005 aRed Salanova times 05 mM Fe 2190 plusmn 42 bc 533 plusmn 016 7202 plusmn 150 595 plusmn 007 cd 219 plusmn 002 abRed Salanova times 1 mM Fe 2089 plusmn 50 bc 503 plusmn 006 6866 plusmn 121 535 plusmn 038 d 200 plusmn 023 bcRed Salanova times 2 mM Fe 2157 plusmn 53 bc 521 plusmn 012 6443 plusmn 064 445 plusmn 020 e 163 plusmn 007 d

ns ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to TukeyrsquosHSD test (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 11 of 21

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The most important role played by ascorbic acid is its radical scavenging power derived from its oxidation to the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to its ability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reduces extracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid including ascorbic and dehydroascorbic acid varied considerably across treatments with the highest values (194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10 and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and 10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmented lettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627 (Figure 2) Our findings are in agreement with other previous works where the concentration and activity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3] For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Fe concentration in the nutrient solution increased [3]

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The mostimportant role played by ascorbic acid is its radical scavenging power derived from its oxidationto the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to itsability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reducesextracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid includingascorbic and dehydroascorbic acid varied considerably across treatments with the highest values(194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmentedlettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627(Figure 2) Our findings are in agreement with other previous works where the concentration andactivity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3]For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Feconcentration in the nutrient solution increased [3]

Phenolic compounds including mainly phenolic acids and flavonoids refer to an importantgroup of secondary metabolites having great antioxidant activity and beneficial effects against chronicdiseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis ofthe lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile

Agronomy 2019 9 290 12 of 21

(Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenicacid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrationsMoreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for thetarget and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaricacid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as totalphenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce whichis in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]Agronomy 2019 9 x FOR PEER REVIEW 2 of 23

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Phenolic compounds including mainly phenolic acids and flavonoids refer to an important group of secondary metabolites having great antioxidant activity and beneficial effects against chronic diseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis of the lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile (Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenic acid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrations Moreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for the target and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaric acid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as total phenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce which is in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]

Concerning the Fe nutrient solution management no significant effects on target and total phenolic acids were observed in green Salanova with the increasing Fe application rate Contrarily to the green-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mM induced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmented lettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higher accumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mM in comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the two main phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentration in the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors of Fe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlation between the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated on several horticultural species such as broccoli mung beans grape berries and radish [367]

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Concerning the Fe nutrient solution management no significant effects on target and total phenolicacids were observed in green Salanova with the increasing Fe application rate Contrarily to thegreen-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mMinduced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmentedlettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higheraccumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mMin comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the twomain phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentrationin the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors ofFe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlationbetween the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated onseveral horticultural species such as broccoli mung beans grape berries and radish [367]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 10 of 21

Table 2 Analysis of variance and mean comparisons leaf mineral composition in green and red Salanova butterhead lettuce grown under increasing Fe concentrationin the nutrient solution

Source of VarianceNitrate P K Ca Mg

(mg kgminus1 FW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW) (g kgminus1 DW)

Cultivar (C)Green Salanova 2277 plusmn 224 481 plusmn 028 6267 plusmn 239 647 plusmn 072 201 plusmn 016Red Salanova 2105 plusmn 93 517 plusmn 016 6901 plusmn 318 573 plusmn 105 206 plusmn 032

t-value 0023 0001 0000 0056 0615Iron (mM Fe) (I)

0015 1991 plusmn 41 c 484 plusmn 030 b 6732 plusmn 416 ab 708 plusmn 041 a 230 plusmn 018 a05 2228 plusmn 115 b 510 plusmn 029 ab 6857 plusmn 405 a 650 plusmn 063 b 214 plusmn 005 ab1 2192 plusmn 136 b 485 plusmn 028 b 6542 plusmn 365 b 569 plusmn 059 c 197 plusmn 016 b2 2351 plusmn 226 a 515 plusmn 019 a 6205 plusmn 265 c 512 plusmn 076 c 174 plusmn 013 c

C times I

Green Salanova times 0015 mM Fe 1999 plusmn 27 c 458 plusmn 005 6371 plusmn 188 701 plusmn 062 abc 217 plusmn 018 abGreen Salanova times 05 mM Fe 2267 plusmn 163 b 488 plusmn 017 6512 plusmn 172 705 plusmn 030 ab 209plusmn 001 bcGreen Salanova times 1 mM Fe 2295 plusmn 109 b 467 plusmn 032 6218 plusmn 054 603 plusmn 062 bcd 193 plusmn 006 bcdGreen Salanova times 2 mM Fe 2545 plusmn 110 a 509 plusmn 026 5967 plusmn 033 579 plusmn 026 d 184 plusmn 005 cd

Red Salanova times 0015 mM Fe 1983 plusmn 57 c 510 plusmn 016 7093 plusmn 082 716 plusmn 017 a 243 plusmn 005 aRed Salanova times 05 mM Fe 2190 plusmn 42 bc 533 plusmn 016 7202 plusmn 150 595 plusmn 007 cd 219 plusmn 002 abRed Salanova times 1 mM Fe 2089 plusmn 50 bc 503 plusmn 006 6866 plusmn 121 535 plusmn 038 d 200 plusmn 023 bcRed Salanova times 2 mM Fe 2157 plusmn 53 bc 521 plusmn 012 6443 plusmn 064 445 plusmn 020 e 163 plusmn 007 d

ns ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to TukeyrsquosHSD test (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 11 of 21

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The most important role played by ascorbic acid is its radical scavenging power derived from its oxidation to the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to its ability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reduces extracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid including ascorbic and dehydroascorbic acid varied considerably across treatments with the highest values (194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10 and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and 10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmented lettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627 (Figure 2) Our findings are in agreement with other previous works where the concentration and activity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3] For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Fe concentration in the nutrient solution increased [3]

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The mostimportant role played by ascorbic acid is its radical scavenging power derived from its oxidationto the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to itsability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reducesextracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid includingascorbic and dehydroascorbic acid varied considerably across treatments with the highest values(194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmentedlettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627(Figure 2) Our findings are in agreement with other previous works where the concentration andactivity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3]For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Feconcentration in the nutrient solution increased [3]

Phenolic compounds including mainly phenolic acids and flavonoids refer to an importantgroup of secondary metabolites having great antioxidant activity and beneficial effects against chronicdiseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis ofthe lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile

Agronomy 2019 9 290 12 of 21

(Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenicacid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrationsMoreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for thetarget and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaricacid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as totalphenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce whichis in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]Agronomy 2019 9 x FOR PEER REVIEW 2 of 23

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Phenolic compounds including mainly phenolic acids and flavonoids refer to an important group of secondary metabolites having great antioxidant activity and beneficial effects against chronic diseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis of the lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile (Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenic acid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrations Moreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for the target and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaric acid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as total phenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce which is in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]

Concerning the Fe nutrient solution management no significant effects on target and total phenolic acids were observed in green Salanova with the increasing Fe application rate Contrarily to the green-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mM induced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmented lettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higher accumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mM in comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the two main phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentration in the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors of Fe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlation between the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated on several horticultural species such as broccoli mung beans grape berries and radish [367]

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Concerning the Fe nutrient solution management no significant effects on target and total phenolicacids were observed in green Salanova with the increasing Fe application rate Contrarily to thegreen-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mMinduced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmentedlettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higheraccumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mMin comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the twomain phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentrationin the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors ofFe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlationbetween the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated onseveral horticultural species such as broccoli mung beans grape berries and radish [367]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 11 of 21

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The most important role played by ascorbic acid is its radical scavenging power derived from its oxidation to the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to its ability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reduces extracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid including ascorbic and dehydroascorbic acid varied considerably across treatments with the highest values (194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10 and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and 10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmented lettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627 (Figure 2) Our findings are in agreement with other previous works where the concentration and activity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3] For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Fe concentration in the nutrient solution increased [3]

Figure 1 Effects of cultivar and iron concentration in the nutrient solution on Fe accumulation inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

33 Hydrophilic Antioxidants Ascorbic Acid and Phenolics Profile

Many leafy vegetables including lettuce are regarded as a good sources of vitamin C The mostimportant role played by ascorbic acid is its radical scavenging power derived from its oxidationto the dehydroascorbate form [66] Ascorbate plays an important role in Fe uptake due to itsability to reduce Fe3+ It has been shown that the presence of ascorbate in the apoplast reducesextracellular Fe3+ facilitating Fe-uptake [66] In the current study the total ascorbic acid includingascorbic and dehydroascorbic acid varied considerably across treatments with the highest values(194 mg 100 gminus1 fw) found in red Salanova treated with 05 mM Fe and to a lesser extent under 10and 20 mM Fe (avg 134 mg 100 gminus1 fw (Figure 2) On the other hand the application of Fe at 05 and10 mM in the nutrient solution did not improve significantly the ascorbic acid in green-pigmentedlettuce but at 20 mM Fe the concentration of this important antioxidant molecule increased by 627(Figure 2) Our findings are in agreement with other previous works where the concentration andactivity of enzymatic and non-enzymatic antioxidant systems increases with the Fe content of plants [3]For instance ascorbic acid rose by 150 and 80 in alfalfa and broccoli respectively when the Feconcentration in the nutrient solution increased [3]

Phenolic compounds including mainly phenolic acids and flavonoids refer to an importantgroup of secondary metabolites having great antioxidant activity and beneficial effects against chronicdiseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis ofthe lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile

Agronomy 2019 9 290 12 of 21

(Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenicacid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrationsMoreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for thetarget and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaricacid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as totalphenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce whichis in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]Agronomy 2019 9 x FOR PEER REVIEW 2 of 23

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Phenolic compounds including mainly phenolic acids and flavonoids refer to an important group of secondary metabolites having great antioxidant activity and beneficial effects against chronic diseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis of the lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile (Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenic acid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrations Moreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for the target and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaric acid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as total phenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce which is in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]

Concerning the Fe nutrient solution management no significant effects on target and total phenolic acids were observed in green Salanova with the increasing Fe application rate Contrarily to the green-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mM induced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmented lettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higher accumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mM in comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the two main phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentration in the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors of Fe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlation between the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated on several horticultural species such as broccoli mung beans grape berries and radish [367]

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Concerning the Fe nutrient solution management no significant effects on target and total phenolicacids were observed in green Salanova with the increasing Fe application rate Contrarily to thegreen-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mMinduced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmentedlettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higheraccumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mMin comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the twomain phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentrationin the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors ofFe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlationbetween the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated onseveral horticultural species such as broccoli mung beans grape berries and radish [367]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 12 of 21

(Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenicacid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrationsMoreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for thetarget and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaricacid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as totalphenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce whichis in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]Agronomy 2019 9 x FOR PEER REVIEW 2 of 23

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Phenolic compounds including mainly phenolic acids and flavonoids refer to an important group of secondary metabolites having great antioxidant activity and beneficial effects against chronic diseases such as inflammation diabetes and some types of cancer [29] The HPLC-DAD analysis of the lettuce extracts provides a quantitativendashqualitative evaluation of the phenolic compounds profile (Table 3) Across treatments the most abundant compound was chicoric acid followed by chlorogenic acid while caffeoyl-meso-tartaric and caffeoyl-tartaric acid were detected in lower concentrations Moreover significant differences in the cultivarsrsquo response to Fe solution enrichment were found for the target and total phenolic acids as reflected by the C times I interaction (Table 3) Except for caffeoyl tartaric acid the greatest accumulation of chlorogenic chicoric caffeoyl meso tartaric acids as well as total phenolic acids were observed in red Salanova leaves in comparison to green-pigmented lettuce which is in agreement with the previous findings of Llorach et al [28] Colonna et al [34] and Kim et al [29]

Concerning the Fe nutrient solution management no significant effects on target and total phenolic acids were observed in green Salanova with the increasing Fe application rate Contrarily to the green-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mM induced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmented lettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higher accumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mM in comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the two main phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentration in the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors of Fe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlation between the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated on several horticultural species such as broccoli mung beans grape berries and radish [367]

Figure 2 Effects of cultivar and Fe concentration in the nutrient solution on total ascorbic acid inlettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005)The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Concerning the Fe nutrient solution management no significant effects on target and total phenolicacids were observed in green Salanova with the increasing Fe application rate Contrarily to thegreen-pigmented butterhead lettuce increasing the Fe concentration in the nutrient solution to 05 mMinduced a significant increase in the chlorogenic acid and total phenolics contents of red-pigmentedlettuce by 1101 and 291 respectively compared to the control treatment (0015 mM) and a higheraccumulation (+314) of caffeoyl meso tartaric acid was also observed in red Salanova at 10 mMin comparison to the control treatment (Table 3) Contrarily to caffeoyl meso tartaric acid the twomain phenolic acids (chlorogenic and chicoric) decreased in response to increasing Fe concentrationin the nutrient solution to 10 mM (Table 3) Since phenolic compounds are considered inhibitors ofFe bioavailability their increase could be linked to Fe-excess stress [3] The same type of correlationbetween the increased Fe (ie Fe-EDDHA) and phenolic concentrations has been demonstrated onseveral horticultural species such as broccoli mung beans grape berries and radish [367]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 13 of 21

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentationin different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova(Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in thenutrient solution elicited significant increase in the anthocyanin contents of plants compared to thosetreated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values(Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyaninconcentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to thecontrol treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was inline with the findings of Shi et al [69] who found that total anthocyanin concentration of CabernetSauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lowerthan that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe hasbeen associated to several mechanisms including the accumulation of sugar and the expression ofseveral structural genes in the flavonoid pathway Considering the significant influence of the geneticmaterial (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids(ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a costeffective tool to increase the phytochemicals content of hydroponically grown lettuce especially forred-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that maycounteract Fe bioavailability in human subjects

Agronomy 2019 9 x FOR PEER REVIEW 3 of 23

Anthocyanins which constitute a subgroup of flavonoids responsible for redndashpurple pigmentation in different Lactuca sativa L types [29] were expectedly detected only in red-pigmented Salanova (Figure 3) as previously demonstrated by Llorach et al [28] The application of 20 mM Fe in the nutrient solution elicited significant increase in the anthocyanin contents of plants compared to those treated with 0015 and 10 mM Fe whereas treatment with 05 mM exhibited intermediate values (Figure 3) Our results are in agreement with those of Mohammadi et al [68] where anthocyanin concentration in peppermint increased by 115 with 05g Lminus1 Fe2O3 treatment in comparison to the control treatment (0 g Lminus1 Fe2O3) The non-linear dose response recorded in our experiment was in line with the findings of Shi et al [69] who found that total anthocyanin concentration of Cabernet Sauvignon grapes under Fe deficient and excess treatments (0 23 92 and 184 microM Fe) were lower than that of the 42 microM Fe treated grapes The improvement of anthocyanin synthesis at 42 microM Fe has been associated to several mechanisms including the accumulation of sugar and the expression of several structural genes in the flavonoid pathway Considering the significant influence of the genetic material (red vs green pigmented cultivar) and Fe application rate on phenolic acids and flavonoids (ie anthocyanins) mild to high (05ndash20 mM) Fe application in nutrient solution can be used as a cost effective tool to increase the phytochemicals content of hydroponically grown lettuce especially for red-pigmented cultivars although such practice is likely to precipitate anti-nutritive effects that may counteract Fe bioavailability in human subjects

Iron (mmol L-1)0015 05 10 20

Ant

hocy

anin

s (μg

cyan

idin

eq g

-1 d

w)

0

5

10

15

20

25

30

c

b

c

a

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins content in red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Figure 3 Effects of cultivar and iron concentration in the nutrient solution on anthocyanins contentin red Salanova lettuce leaves Anthocyanins in green Salanova were not detected Different lettersindicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of threereplicates Vertical bars indicate plusmn standard deviation of means

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 14 of 21

Table 3 Analysis of variance and mean comparisons for target phenolic acids total phenolic acids and anthocyanins in green and red Salanova butterhead lettucegrown under increasing Fe concentration in the nutrient solution

Source of VarianceCaffeoyl Tartaric Acid Chlorogenic Acid Chicoric Acid Caffeoyl Meso Tartaric Acid

sumPhenolic Acids

(mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100 gminus1 DW) (mg 100gminus1 DW)

Cultivar (C)Green Salanova 1592 plusmn 66 730 plusmn 24 7262 plusmn 224 226 plusmn 14 981 plusmn 30Red Salanova 662 plusmn 32 5874 plusmn 294 11489 plusmn 527 2892 plusmn 78 2092 plusmn 75

t-value 0000 0000 0018 0000 0000Iron (mM Fe) (I)

0015 1081 plusmn 88 ab 2677 plusmn 188 b 11718 plusmn 313 a 1629 plusmn 131 a 1711 plusmn 55 a05 881 plusmn 17 b 4890 plusmn 475 a 1067 plusmn 539 a 1 1709 plusmn 165 a 1815 plusmn 114 a1 879 plusmn 45 b 1304 plusmn 91 c 4547 plusmn 118 b 1916 plusmn 201 a 865 plusmn 21 b2 1668 plusmn 85 a 4337 10566 plusmn 387 a 981 plusmn 94 b 1755 plusmn 74 a

C times I

Green Salanova times 0015 mM Fe 1828 plusmn 49 966 plusmn 10 e 8992 plusmn 122 bcd 439 plusmn 05 d 1223 plusmn 15 cGreen Salanova times 05 mM Fe 1000 plusmn 03 559 plusmn 05 e 6141 plusmn 252 cd 211 plusmn 06 d 791 plusmn 26 cGreen Salanova times 1 mM Fe 1265 plusmn 24 470 plusmn 08 e 5285 plusmn 95 d 127 plusmn 01 d 715 plusmn 13 cGreen Salanova times 2 mM Fe 2274 plusmn 81 925 plusmn 03 e 8632 plusmn 194 bcd 125plusmn 02 d 1196 plusmn 28 c

Red Salanova times 0015 mM Fe 333 plusmn 02 4388 plusmn 20 c 14444 plusmn 85 ab 2819 plusmn 10 b 2198 plusmn 10 bRed Salanova times 05 mM Fe 761 plusmn 18 9221 plusmn 35 a 15201 plusmn 215 a 3207 plusmn 19 ab 2839 plusmn 18 aRed Salanova times 1 mM Fe 493 plusmn 09 2137 plusmn 03 d 3810 plusmn 98 d 3705 plusmn 69 a 1015 plusmn 18 cRed Salanova times 2 mM Fe 1062 plusmn 26 7748 plusmn 98 b 12500 plusmn 473 abc 1837 plusmn 19 c 2315 plusmn 59 b

ns

ns Nonsignificant or significant at P le 005 001 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSD test (P = 005)Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 15 of 21

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by Ctimes I interaction In the current experiment the LAA rangedfrom 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM andred lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigmentswhich constitute an important quality trait in fresh vegetables since these antioxidant moleculesprevent the formation of free radicals in both plants and humans [70] A putative mechanism involvedin the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could beassociated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasingplants antioxidant capacity [65]

Agronomy 2019 9 x doi FOR PEER REVIEW wwwmdpicomjournalagronomy

34 Lipophilic Antioxidant Activity and Target Carotenoids Profile

The LAA was significantly affected by C times I interaction In the current experiment the LAA ranged from 26 to 71 mmol Trolox 100 gminus1 dw with the highest values recorded in green lettuce at 20 mM and red lettuce treated with 10 and 20 mM Fe (Figure 4) The LAA derives mainly from lipophilic pigments which constitute an important quality trait in fresh vegetables since these antioxidant molecules prevent the formation of free radicals in both plants and humans [70] A putative mechanism involved in the accumulation of LAA under moderate to high Fe concentrations (10 and 20 mM) could be associated to the absorption of excess Fe by ferritin preventing the formation of ROS and increasing plants antioxidant capacity [65]

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidant activity (LAA) in lettuce leaves Different letters indicate significant differences according to Tukeyrsquos HSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviation of means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits and vegetables and constitute important antioxidant components of lettuce [29] In the current study the carotenoid composition as a function of cultivars and Fe concentration in the nutrient solution are displayed in Table 4 where in red Salanova plants contained higher amounts of all the detected compounds compared to green Salanova verifying that the content in lipophilic pigments of butterhead lettuce is readily reflected in leaf pigmentation

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthin followed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthin no significant interaction between cultivar and Fe nutrient solution concentration was observed whereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interaction of these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the red Salanova had higher lutein and β-cryptoxanthin than those recorded in

Figure 4 Effects of cultivar and iron concentration in the nutrient solution lipophilic antioxidantactivity (LAA) in lettuce leaves Different letters indicate significant differences according to TukeyrsquosHSD test (P = 005) The values are means of three replicates Vertical bars indicate plusmn standard deviationof means

Carotenoids being lipid-soluble pigments contribute to the yellow-orange color of fruits andvegetables and constitute important antioxidant components of lettuce [29] In the current studythe carotenoid composition as a function of cultivars and Fe concentration in the nutrient solutionare displayed in Table 4 where in red Salanova plants contained higher amounts of all the detectedcompounds compared to green Salanova verifying that the content in lipophilic pigments of butterheadlettuce is readily reflected in leaf pigmentation

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 16 of 21

Table 4 Analysis of variance and mean comparisons for target carotenoids in green and red Salanova butterhead lettuce grown under increasing Fe concentration inthe nutrient solution

Source of VarianceViolaxanthin + Neoxanthin Lutein β-Cryptoxanthin β-Carotene

(microg Violaxanthin eq gminus1 DW) (microg gminus1 DW) (microg gminus1 DW) (microg gminus1 DW)

Cultivar (C)Green Salanova 6124 plusmn 52 2563 plusmn 39 4268 plusmn 61 2308 plusmn 27Red Salanova 12737 plusmn 170 6047 plusmn 65 10728 plusmn 87 4162 plusmn 48

t-value 0000 0000 0000 0000Iron (mM Fe) (I)

0015 8277 plusmn 267 c 3942 plusmn 188 b 7206 plusmn 332 bc 2907 plusmn 90 c05 9016 plusmn 383 bc 4219 plusmn 217 b 7015 plusmn 360 c 2987 plusmn 106 c1 9630 plusmn 361 b 4262 plusmn 161 b 7754 plusmn 352 a 3349 plusmn 84 b2 10800 plusmn 453 a 4798 plusmn 213 a 8017 plusmn 393 a 3695 plusmn 129 a

C times I

Green Salanova times 0015 mM Fe 5877 plusmn 18 d 2285 plusmn 29 4252 plusmn 42 2088 plusmn 3 eGreen Salanova times 05 mM Fe 5524 plusmn 9 d 2260 plusmn 26 3804 plusmn 92 2030 plusmn 1 eGreen Salanova times 1 mM Fe 6353 plusmn 42 d 2835 plusmn 37 4582 plusmn 68 2594 plusmn 14 dGreen Salanova times 2 mM Fe 6740 plusmn 13 d 2873 plusmn 23 4435 plusmn 26 2518 plusmn 6 d

Red Salanova times 0015 mM Fe 10677 plusmn 73 c 5598 plusmn 75 10159 plusmn 110 3726 plusmn 10 cRed Salanova times 05 mM Fe 12508 plusmn 36 b 6178 plusmn 42 10227 plusmn 74 3944 plusmn 20 bRed Salanova times 1 mM Fe 12906 plusmn 42 b 5688 plusmn 46 10926 plusmn 57 4105 plusmn 22 bRed Salanova times 2 mM Fe 14859 plusmn 132 a 6723 plusmn 43 11598 plusmn 16 4872 plusmn 17 a

ns ns

ns Nonsignificant or significant at P le 005 and 0001 respectively Different letters within each column indicate significant differences according to Tukeyrsquos HSDtest (P = 005) Cultivars main effects were compared by Studentrsquos t-test All data are expressed as mean plusmn standard deviation n = 3

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 17 of 21

The most abundant carotenoid compounds were violaxanthin + neoxanthin and β-cryptoxanthinfollowed by lutein while β-carotene was detected in lower levels (Table 4) For lutein and β-cryptoxanthinno significant interaction between cultivar and Fe nutrient solution concentration was observedwhereas violaxanthin + neoxanthin and β-carotene were significantly affected by the interactionof these two factors (Table 4) Irrespective of the Fe concentration in the nutrient solution the redSalanova had higher lutein and β-cryptoxanthin than those recorded in green Salanova plants by1359 and 1513 respectively (Table 4) Moreover when averaged over both lettuce cultivars thehighest lutein and β-cryptoxanthin concentrations were recorded at 20 mM and at both 10 and20 mM Fe respectively (Table 4) Similarly to the Fe effects observed on target phenolic acids limitedand nonsignificant variation on violaxanthin + neoxanthin and β-carotene concentrations in responseto increasing Fe concentration in the nutrient solution was observed in green Salanova On the otherhand our results also demonstrated that the addition of 20 mM Fe to the nutrient solution elicitedsignificant increase (392 and 308) of violaxanthin + neoxanthin and β-carotene compared to thecontrol treatment in red Salanova whereas treatments with 05 and 10 mM exhibited intermediatevalues (Table 4) An explanation to the premium quality of red Salanova in terms of carotenoids couldbe associated with the activation of molecular and physiological mechanisms necessary for adaptationto suboptimal conditions (excess of Fe) such as the biosynthesis and accumulation of secondarymetabolites for instance carotenoids [20] These antioxidant compounds are known to contributeto ROS scavenging and cellular water homeostasis [71] and more interestingly these secondarymetabolites are also important owing to their health-promoting effects

4 Conclusions

Nowadays consumers scientists nutritionists and vegetable growers are looking for functionalfoods with beneficial effects on human health and longevity Our findings highlighted thatbiofortification of butterhead lettuce with an essential micronutrient such as Fe could be facilitatedby closed soilless cultivation due to the constant exposure of root apparatus to the fortified nutrientsolution Our results indicate that fresh yield shoot dry biomass mineral composition (P and K) aswell as hydrophilic (ascorbic acid chlorogenic chicoric caffeoyl meso tartaric acids total phenolicsand anthocyanins) and lipophilic (violaxanthin + neoxanthin lutein β-cryptoxanthin and β-carotene)antioxidant molecules were strongly affected by genotype with higher nutritional and functionalquality traits recorded in the red-pigmented cultivar Since product quality enhancement is sometimesassociated with reduction of crop productivity a compromise is needed to find the sectio divinaOur findings demonstrated the possibility of producing lettuce Fe-enriched by 21 while incurringan acceptable reduction in fresh marketable yield (13) by supplying 10 mM Fe The addition of Fein the nutrient solution differentially modulated the functional quality in green and red pigmentedlettuce The application of 05 and 10 mM Fe in the nutrient solution improved target phenolicacids (chlorogenic and chicoric acids at 05 mM and caffeoyl tartaric acid at 10 mM Fe) as well astotal phenolics (at 05 mM Fe) and the application of 20 mM Fe enhanced the carotenoids profile(violaxanthin + neoxanthin and β-carotene) of red Salanova whereas no significant improvements infunctional quality traits were observed in green Salanova Overall Fe biofortification facilitated byclosed soilless cultivation could be used as an effective and sustainable tool for producing functionalfood especially in urban farms triggered by the rising food demand and the malnutrition owing tounbalanced diets

Author Contributions YR had the original idea set up the experimental protocol coordinated the research andwas significantly involved in writing CE-N carried out the whole experiment and was actively involved inanalysis and writing MG and AP were involved in analysis and writing SRS was involved in ICP analysisand data interpretation MCK and SDP contributed in improving the final version of the manuscript

Funding This research received no external funding

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 18 of 21

Acknowledgments We would like to acknowledge Societagrave Agricola Punzi Eboli (Salerno) for providing the plantmaterial (green and red Salanova) The authors are grateful for to Annamaria Palladino Mirella SorrentinoAntonio De Francesco for their technical assistance in the Fitotron Plant Growth Chamber experiment as well asto Sabrina De Pascale Paola Vitaglione and Antonio Dario Troise for providing the access to HPLC facilitiesand analysis

Conflicts of Interest The authors declare no conflict of interest

References

1 Murgia I Arosio P Tarantino D Soave C Biofortification for combating lsquohidden hungerrsquo for iron TrendsPlant Sci 2012 17 [CrossRef] [PubMed]

2 White PJ Broadley MR Biofortification of crops with seven mineral elements often lacking in humandietsndashiron zinc copper calcium magnesium selenium and iodine New Phytol 2009 182 49ndash84 [CrossRef][PubMed]

3 Przybysz A Wrochna M Małecka-Przybysz M Gawronska H Gawronski SW Vegetable sproutsenriched with iron Effects on yield ROS generation and antioxidative system Sci Hortic 2016 203 110ndash117[CrossRef]

4 Sperotto RA Ricachenevsky FK de Abreu Waldow V Palma Fett J Iron biofortification in rice Itrsquos along way to the top Plant Sci 2012 190 24ndash39 [CrossRef] [PubMed]

5 Abbaspour N Hurrell R Kelishadi R Review on iron and its importance for human health J Res MedSci 2014 19 164ndash174 [PubMed]

6 FAOWHO Human Vitamin and Mineral Requirements Report of a joint FAOWHO expert consultationBangkok Thailand FAO Rome Italy 2001

7 Gowthami VSS Ananda N Dry matter production yield and yield components of groundnut (Arachishypogaea L) genotypes as influenced by zinc and iron through ferti-ortification Indian J Agric Res 2017 51339ndash344 [CrossRef]

8 Gillooly M Bothwell TH Torrance JD MacPhail AP Derman DP Bezwoda WR Mills WCharlton RW The effects of organic acids phytates and polyphenols on the absorption of iron fromvegetables Br J Nutr 1983 49 331ndash342 [CrossRef] [PubMed]

9 Siegenberg D Baynes RD Bothwell TH Macfarlane BJ Lamparelli RD Car NG MacPhail PSchmidt U Tal A Mayet F Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenolsand phytates on nonheme-iron absorption Am J Clin Nutr 1991 53 537ndash541 [CrossRef]

10 Finkelstein JL Haas JD Mehta S Iron-biofortified staple food crops for improving iron status A reviewof the current evidence Curr Opin Biotechnol 2017 44 138ndash145 [CrossRef]

11 La Frano MR de Moura FF Boy E Loumlnnerdal B Burri BJ Bioavailability of iron zinc and provitaminA carotenoids in biofortified staple crops Nutr Rev 2014 72 289ndash307 [CrossRef]

12 Pereira EG Oliva MA Rosado-Souza L Mendes GC Colares DS Stopato CH Almeida AM Ironexcess affects rice photosynthesis through stomatal and non-stomatal limitations Plant Sci 2013 201ndash20281ndash92 [CrossRef] [PubMed]

13 De Souza Pinto S de Souza AE io Oliva MA Gusmatildeo Pereira EG Oxidative damage and photosyntheticimpairment in tropical rice cultivars upon exposure to excess iron Sci Agric 2016 73 217ndash226 [CrossRef]

14 Frei M Narh Tetteh R Razafindrazaka AL Fuh MA Wu LB Becker M Responses of rice to chronicand acute iron toxicity Genotypic differences and biofortification aspects Plant Soil 2016 408 149ndash161[CrossRef]

15 Adamski MJ Peters JA Danieloski R Bacarin MA Excess iron-induced changes in the photosyntheticcharacteristics of sweet potato J Plant Physiol 2011 168 2056ndash2062 [CrossRef] [PubMed]

16 Agarwal S Sairam RK Meena RC Tyagi A Srivastava GC Effect of Excess and Deficient Levels ofIron and Copper on Oxidative Stress and Antioxidant Enzymes Activity in Wheat J Plant Sci 2006 1 86ndash97[CrossRef]

17 Li X Ma H Jia P Wang J Jia L Zhang T Yang Y Chen H Wei X Responses of seedling growth andantioxidant activity to excess iron and copper in Triticum aestivum L Ecotoxicol Environ Saf 2012 86 47ndash53[CrossRef] [PubMed]

18 Hemalatha K Venkatesan S Impact of iron toxicity on certain enzymes and biochemical parameters of teaAsian J Biochem 2011 6 384ndash394 [CrossRef]

Agronomy 2019 9 290 19 of 21

19 Rouphael Y Kyriacou MC Petropoulos SA De Pascale S Colla G Improving vegetable quality incontrolled environments Sci Hortic 2018 234 275ndash289 [CrossRef]

20 Rouphael Y Kyriacou MC Enhancing quality of fresh vegetables through salinity eustress andbiofortification applications facilitated by soilless cultivation Front Plant Sci 2018 9 1254 [CrossRef]

21 Rouphael Y Colla G Growth yield fruit quality and nutrient uptake of hydroponically cultivated zucchinisquash as affected by irrigation systems and growing seasons Sci Hortic 2005 105 177ndash195 [CrossRef]

22 Rouphael Y Rea E Cardarelli M Bitterlich M Schwarz D Colla G Can Adverse Effects of Acidity andAluminum Toxicity Be Alleviated by Appropriate Rootstock Selection in Cucumber Front Plant Sci 2016 7[CrossRef] [PubMed]

23 Fallovo C Rouphael Y Rea E Battistelli A Colla G Nutrient solution concentration and growing seasonaffect yield and quality of Lactuca sativa L var acephala in floating raft culture J Sci Food Agric 2009 891682ndash1689 [CrossRef]

24 Rouphael Y Colla G Bernardo L Kane D Trevisan M Lucini L Zinc Excess Triggered PolyaminesAccumulation in Lettuce Root Metabolome As Compared to Osmotic Stress under High Salinity FrontPlant Sci 2016 [CrossRef] [PubMed]

25 Colla G Rouphael Y Cardarelli M Svecova E Rea E Lucini L Effects of saline stress on mineralcomposition phenolic acids and flavonoids in leaves of artichoke and cardoon genotypes grown in floatingsystem J Sci Food Agric 2013 93 1119ndash1127 [CrossRef] [PubMed]

26 Borgognone D Rouphael Y Cardarelli M Lucini L Colla G Changes in Biomass Mineral Compositionand Quality of Cardoon in Response to NO3-Cl- Ratio and Nitrate Deprivation from the Nutrient SolutionFront Plant Sci 2016 7 [CrossRef]

27 Fanasca S Colla G Maiani G Venneria E Rouphael Y Azzini E Saccardo F Changes in AntioxidantContent of Tomato Fruits in Response to Cultivar and Nutrient Solution Composition J Agric Food Chem2006 54 4319ndash4325 [CrossRef] [PubMed]

28 Llorach R Martiacutenez-Saacutenchez A Tomaacutes-Barbern FA Gil MI Ferreres F Characterisation of polyphenolsand antioxidant properties of five lettuce varieties and escarole Food Chem 2008 108 1028ndash1038 [CrossRef]

29 Kim MJ Moon Y Tou JC Mou B Waterland NL Nutritional value bioactive compounds and healthbeneficts of lettuce (Lactuca sativa L) J Food Comp Anal 2016 49 19ndash34 [CrossRef]

30 Kyracou MC El-Nakhel C Graziani G Pannico A Soteriou G Giordano M Ritieni A De Pascale SRouphael Y Functional quality in novel food sources Genotypic variation in the nutritive and phytochemicalcomposition of thirteen microgreens species Food Chem 2019 277 107ndash118 [CrossRef]

31 Nicolle C Carnat A Fraisse D Lamaison JL Rock E Michel H Amouroux P Remesy CCharacterisation and variation of antioxidant micronutrients in lettuce (Lactuca sativa folium) J Sci FoodAgric 2004 84 2061ndash2069 [CrossRef]

32 Mou B Genetic variation of Beta-carotene and Lutein Contents in Lettuce J Am Soc Hort Sci 2005 130870ndash876 [CrossRef]

33 Lebeda A Ryder EJ Grube R Dolezalov AI Kristkova E Lettuce (Asteraceae Lactuca spp) In GeneticRessources Chromosone Engineering and Crop Improvement Singh RK Ed Vegetable Crops CRC PressTailor and Francis Group Boca Raton FL USA 2007 Volume 3 pp 377ndash472

34 Colonna E Rouphael Y Barbieri G De Pascale S Nutritional quality of ten leafy vegetables harvested attwo light intensities Food Chem 2016 199 702ndash710 [CrossRef] [PubMed]

35 Rouphael Y Cardarelli M Bassal A Leonardi C Giuffrida F Colla G Vegetable quality as affected bygenetic agronomic and environmental factors J Food Agric Environ 2012 10 680ndash688

36 Rouphael Y Kyriacou MC Vitaglione P Giordano M Pannico A Colantuono A De Pascale SGenotypic variation in nutritional and antioxidant profile among iceberg lettuce cultivars Acta Sci PolHortorum Cultus 2017 16 37ndash45 [CrossRef]

37 Tyksinski W Komosa A After effect of iron chelates on the yielding and iron content in greenhouse lettuceActa Sci Pol Hortorum Cultus 2008 7 3ndash10

38 Kozik E Tyksintildeski W Bosiacki MA Comparison of the efficiency of organic and mineral iron compoundsin the greenhouse cultivation of lettuce J Elementol 2011 16 59ndash68 [CrossRef]

39 Inoue K Kondo S Adachi A Yokota H Production of iron enriched vegetables Effect of feeding timeon the rate of increase in foliar iron content and foliar injury J Hortic Sci Biotechnol 2000 75 209ndash213[CrossRef]

Agronomy 2019 9 290 20 of 21

40 Tomasi N Pinton R Dalla Costa L Cortella G Terzano R Mimmo T Scampicchio M Cesco S Newlsquosolutionsrsquo for floating cultivation system of ready-to-eat salad A review Trends Food Sci Technol 2015 46267ndash276 [CrossRef]

41 Rouphael Y Colla G Giordano M El-Nakhel C Kyriacou MC De Pascale S Foliar applications of alegume-derived protein hydrolysate elicit dose-dependent increases of growth leaf mineral compositionyield and fruit quality in two greenhouse tomato cultivars Sci Hortic 2017 226 353ndash360 [CrossRef]

42 Kampfenkel K Van Montagu M Inzeacute D Extraction and determination of ascorbate and 504dehydroascorbate from plant tissue Anal Biochem 1995 225 165ndash167 [CrossRef]

43 Pellegrini N Re R Yang M Rice-Evans C Screening of dietary carotenoids and carotenoid-rich fruitextracts for antioxidant activities applying 22prime-azinobis(3-ethylenebenzothiazoline-6- sulfonic acid radicalcation decolorization assay Meth Enzymol 1999 299 379ndash384 [CrossRef]

44 Vallverduacute-Queralt A Oms-Oliu G Odriozola-Serrano I Lamuela-Raventoacutes RM Martiacuten-Belloso OElez-Martiacutenez P Metabolite profiling of phenolic and carotenoid contents in tomatoes after moderate-intensitypulsed electric field treatments Food Chem 2013 136 199ndash205 [CrossRef] [PubMed]

45 Lebeda A Doležalovaacute I Astley D Representation of wild Lactuca spp (Asteraceae Lactuceae) in worldgenebank collections Genetic Res Crop Evol 2004 51 167ndash174 [CrossRef]

46 Lebeda A Doležalovaacute I Feraacutekovaacute V Astley D Geographical distribution of wild Lactuca spp (AsteraceaeLactuceae) Botanic Rev 2004 70 328ndash356 [CrossRef]

47 Lebeda A Kriacutestkovaacute E Kitner M Mieslerovaacute B Jemelkovaacute M Pink DAC Wild Lactuca species theirgenetic diversity resistance to diseases and pests and exploitation in lettuce breeding Eur J Plant Pathol2014 138 597ndash640 [CrossRef]

48 Urrestarazu M Alvaro JE Moreno S Remediation of Iron Chlorosis by the Addition of Fe-oo-EDDHA inthe Nutrient Solution Applied to Soilless Culture HortScience 2008 43 1434ndash1436 [CrossRef]

49 Roosta HR Pourebrahimi M Hamidpour M Effects of bicarbonate and different Fe sources on vegetativegrowth and physiological characteristics of bell pepper (Capsicum annuum L) plants in hydroponic systemJ Plant Nutr 2015 38 397ndash416 [CrossRef]

50 Roosta HR Jalali M Shahrbabaki SMAV Effect of Nano Fe-Chelate Fe-Eddha and FeSO4 on VegetativeGrowth Physiological Parameters and Some Nutrient Elements Concentrations of Four Varieties of Lettuce(Lactuca sativa L) In NFT System J Plant Nutr 2015 38 2176ndash2184 [CrossRef]

51 Filho ABC Cortez JWM de Sordi D Urrestarazu M Common Chicory Performance as Influenced byIron Concentration in the Nutrient Solution J Plant Nutr 2015 38 1489ndash1494 [CrossRef]

52 De Dorlodot S Lutts S Pierre Bertin P Effects of Ferrous Iron Toxicity on the Growth and MineralComposition of an Interspecific Rice J Plant Nutr 2005 28 1ndash20 [CrossRef]

53 Briat JF Dubos C Gaymard F Iron nutrition biomass production and plant product quality Trends PlantSci 2015 20 [CrossRef] [PubMed]

54 Nagajyoti PC Lee KD Sreekanth TVM Heavy metals Occurrence and toxicity for plants A reviewEnviron Chem Lett 2010 8 199ndash216 [CrossRef]

55 Kobayashi T Nishizawa NK Iron uptake translocation and regulation in higher plants Annu Rev PlantBiol 2012 63 131ndash152 [CrossRef] [PubMed]

56 Commission Regulation (EU) 12582011 Commission Regulation (EU) No 12582011 of 2 of December 2011amending Regulation (EC) No 18812006 as Regardsmaximum Level for Nitrates in Foodstuffs Official Journal ofthe European Union Brussels Belgum 2011

57 Liu W Yang Q Du L Cheng R Zhou W Nutrient supplementation increased growth and nitrateconcentration of lettuce cultivated hydroponically with biogas slurry Acta Agric Scand Section B Soil PlantSci 2011 61 391ndash394 [CrossRef]

58 Iacuzzo F Gottardi S Tomasi N Savoia E Tommasi R Cortella G Terzano R Pinton R DallaCosta L Cesco S Corn salad (Valerianella locusta (L) Laterr) growth in a water-saving floating system asaffected by iron and sulfate availability J Sci Food Agric 2011 91 344ndash354 [CrossRef]

59 USDA National Nutrient database for Standard Reference Release 28 USDA Washington DC USA 201560 Kawashima LM Soares LMV Mineral profile of raw and cooked leafy vegetables consumed in Southern

Brazil J Food Compos Anal 2003 16 605ndash611 [CrossRef]

Agronomy 2019 9 290 21 of 21

61 Baslam M Idoia G Nieves G The arbuscular mycorrhizal symbiosis can overcome reductions inyield andnutritional quality in greenhouse-lettuces cultivated atinappropriate growing seasons Sci Hortic 2013 164145ndash154 [CrossRef]

62 Snowden RED Wheeler D Chemical changes in selected wetland plant species with increasing Fe supplywith specific reference to root precipitates and Fe tolerance New Phytol 1995 131 503ndash520 [CrossRef]

63 Lombi EKL Tearall JR Horwath JR Zhao FJ Hawkesford MJ McGrath SP Influence of iron statuson cadmium and zinc uptake by different ecotypes of the hyperaccumulator Thlaspi caerulescens Plant Physiol2002 128 1359ndash1367 [CrossRef]

64 Souza-Santos P Ramos RS Ferreira ST Carvalho-Alves PC Iron-induced oxidative damage of cornroot plasma membrane H+-ATPase Biochem Biophys Acta 2001 1512 357ndash366 [CrossRef]

65 Li K Hu G Yu S Tang Q Liu J Effect of the iron biofortification on enzymes activities and antioxidantproperties in germinated brown rice J Food Meas Characteriz 2018 12 789ndash799 [CrossRef]

66 Smirnoff N Ascorbic acid metabolism and functions A comparison of plants and mammals Free RadicBiol Med 2018 122 116ndash129 [CrossRef] [PubMed]

67 Shi P Song C Chen H Duan B Zhang Z Meng J Foliar applications of iron promote flavonoidsaccumulation in grape berry of Vitis vinifera cv Merlot grown in the iron deficiency soil Food Chem 2018253 164ndash170 [CrossRef] [PubMed]

68 Mohammadi M Hoseini NM Chaichi MR Alipour H Dashtaki M Safikhani S Influence of nano-ironoxide and zinc sulfate on physiological characteristics of peppermint Comm Soil Sci Plant Anal 2018 492315ndash2326 [CrossRef]

69 Shi P Li B Chen H Song C Meng J Zi Z Zhang Z Iron supply affects anthocyanin content andrelated gene expression in berries of Vitis vinifera cv Cabernet Sauvignon Molecules 2017 22 283 [CrossRef][PubMed]

70 Khanam UKS Oba S Yanase E Murakami Y Phenolic acids flavonoids and total antioxidant capacityof selected leafy vegetables J Funct Foods 2012 4 979ndash987 [CrossRef]

71 Orsini F Maggio A Rouphael Y De Pascale S ldquoPhysiological quality of organically grown vegetablesSci Hortic 2016 208 131ndash139 [CrossRef]

copy 2019 by the authors Licensee MDPI Basel Switzerland This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (httpcreativecommonsorglicensesby40)