A

ccispiarcc

io©

K

1

abwawit

0d

Postharvest Biology and Technology 43 (2007) 335–342

Analysis of the main secondary metabolites produced in tomato(Lycopersicon esculentum, Mill.) epicarp tissue during fruit

ripening using fluorescence techniques

A. Lai a,∗, E. Santangelo b, G.P. Soressi c, R. Fantoni a

a ENEA, Divisione Fisica Applicata, Centro Ricerche Frascati, Via Enrico Fermi 45, 00044 Frascati, Roma, Italyb Consorzio Agrital Ricerche, Viale dell’Industria 24, 00057 Maccarese, Roma, Italy

c Department of Agrobiology and Agrochemistry, Tuscia University, Via S.C. de Lellis s.n.c., 01100 Viterbo, Italy

Received 14 July 2005; accepted 15 September 2006

bstract

Tomato (Lycopersicon esculentum L.) fruit are an important source of antioxidant (mainly pigment) compounds, as well as lycopene, �-arotene, ascorbic acid and polyphenols. Differentiation of the final product in the market requires an accurate evaluation of these value-addingompounds. Because of this, we have undertaken a comparison of the spectral characterisation of the tomato fruit surface pigments from themmature to over-ripe stage, using spectroscopy techniques based on visible fluorescence emission upon excitation in the same or ultravioletpectral regions. The aim was to verify the spectral band for optimal conditions for fruit harvesting using non-destructive techniques. Theattern of pigment composition changed markedly during ripening and showed progressive disappearance of chlorophyll with a concomitantncrease in carotenoids until the fully ripe stage. The main fluorescence spectral features belonging to anthocyanins, flavonoids, carotenoidsnd chlorophyll a after excitation of skin tomato pigments at different laser wavelengths was identified. In comparing, the fluorescence spectralatios at the excitation wavelength λexc = 266 nm, significant differences were obtained for the spectral ratios of chlorophyll/flavonoids andarotenoids/chlorophyll. Positive correlation coefficients were found for the carotenoids/flavonoids (0.780) ratios and negative ones for thearotenoids/chlorophyll ratios (−0.513).

Analysis of fluorescence resulted in determination of the most useful laser radiation for remote non-invasive measurements with laser-nduced fluoresence (LIF): for the ripening stage, λexc = 266 nm was the optimal laser wavelength, since the induced fluorescence spectrabtained appeared to differ with the physiological stage of the fruit.

2007 Elsevier B.V. All rights reserved.

induced

t(eRkr

eywords: Lycopersicon esculentum; Fruit ripening; Skin pigments; Laser-

. Introduction

Tomato (Lycopersicon esculentum L.) ripening involvesnumber of physiological processes that include the visi-

le breakdown of chlorophyll and build-up of carotenoids,ith massive accumulation of antioxidant components such

s lycopene and �-carotene (Laval-Martin et al., 1975)

ithin the plastids. Besides carotenoids, the contents of other

mportant antioxidant compounds, such as ascorbic acid, �-ocopherol and phenolics, vary during ripening, thus varying

∗ Corresponding author. Tel.: +39 06 94005207; fax: +39 06 94005312.E-mail address: lai@frascati.enea.it.it (A. Lai).

ipiieu

925-5214/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.postharvbio.2006.09.016

fluorescence

he nutritional value and the antioxidant activity of the fruitGiovanelli et al., 1999; Martınez-Valverde et al., 2002; Raffot al., 2002; Cano et al., 2003; Slimestad and Verheul, 2005).ipening conditions affect both the antioxidant accumulationinetics and the final contents, which are higher in postharvestipened than in vine-ripened fruit (Giovanelli et al., 1999).

Effective monitoring of crop production needs continuousnformation to be collected on the metabolic state and thehysical and chemical cell environment of the fruit. Current

maging techniques allow monitoring of changes in the phys-ological state of plants to be performed on-line, in situ andven non-invasively. Non-destructive technologies for eval-ating the quality of a food might be helpful for improving

3 gy and Technology 43 (2007) 335–342

ptcmobso

isaaStltmofotr

adTtwwwtb

2

2

mdastTbi1

ywc

(t

Table 1Colour classification of tomatoes (Standards for Grade of Fresh Tomatoes(7 CFR 51), US Department of Agriculture)

F1 (green): The surface of the tomato is completely green. The shade ofgreen may vary from light to dark

F2 (breaker): There is a definite break in the colour from green to tannishyellow with pink or red skin not covering more than 10% of the surface

F3 (turning): More than 10% but no more than 30% of the surface, in thegreen aggregate, shows a definite change in colour from green to tannishyellow, pink, red, or a combination of these colours

F4 (pink): More than 30%, but no more than 60% of the surface, in the

F

F

2

oiplet(Taeatg

2

asrscmd

(ethe optics and the sample holder, and a third component foremitted fluorescence analysis. The two monochromators arecontrolled by an interface with the motors installed in thecalculator. The apparatus allows both the emission spectrum

Table 2Characteristics of the PTI Quantamaster spectrofluorimeter system

Component Type Characteristics

Xenon lamp 75 W Emission band 240–1000 nm

36 A. Lai et al. / Postharvest Biolo

ostharvest conditioning and handling with practical resultshat will be reflected in the quality of the fruit delivered toonsumers. Polder et al. (2004) demonstrated that on-lineeasurements of the spatial distribution of the concentration

f lycopene, �-carotene, lutein, chlorophyll a and chlorophyllis possible by using spectral imaging. This allows on-line

orting of tomato fruit using criteria such as the concentrationf the compounds and the uniformity of distribution.

From this point of view, the application of fluorescence,n particular chlorophyll fluorescence, techniques are beingtudied in a number of species with respect to ripening, stor-ge and postharvest physiology (DeEll et al., 1996; Toivonennd DeEll, 1998; Prange et al., 2002; Bron et al., 2004;chofield et al., 2005). The use of a non-destructive optical

echnique, such as laser-induced fluorescence (LIF) to fol-ow the pigment changes during ripening, could be a usefulool to provide information on changes in fruit pigments. The

ain advantages are rapid measurements on a large numberf samples, which thereafter remain intact and may be usedor further analysis. As well, it allows both the setting upf parameters for the optimisation of harvest date and selec-ion of genotypes and ripening stages of particular genotypes,icher in healthy compounds.

The aim of this work, therefore, was to use tomatos a model plant to test fluorescence techniques for non-estructive analysis of pigment changes during fruit ripening.he changes occurring in the main surface pigments of

he tomato fruit at different stages of ripening (from greenith a predominance of chlorophylls, to fully red fruitith carotenoids predominating and lycopene in particular)ere evaluated. The analysis included secondary compounds

hat strongly influence ripening and provide nutritionalenefits.

. Materials and methods

.1. Plants and fruit samples

Plants of tomato (Lycopersicon esculentum L. cv. Mar-ande) were grown until maturity in the open air in pots

uring summertime with temperatures between 20 and 30 ◦Cnd 30–70% R.H. Tomato fruit were harvested at six differenttages of ripeness, from immature to completely mature, inerms of their external colour (from light green to dark red).he different ripening stages of each fruit were characterisedy progressive numbers (from F1 = green to F6 = red) accord-ng to the Standards for Grade of Fresh Tomatoes (USDA,991) (Table 1).

The contents of pigments in over-ripe fruit were also anal-sed. In order to induce a over-ripening, fruit at the F6 stageere harvested and stored for 30 days at 28 ◦C. The arbitrary

ode, F7, was assigned to these samples.The total number of fruit used in the experiment was 21

three fruit per ripening stage). Each fruit was measured threeimes.

MEEP

aggregate, shows pinkish red or red colour5 (light red): More than 60% but no more than 90%, in the aggregate,shows pinkish red or red colour

6 (red): More than 90% of the surface, in the aggregate, shows red colour

.2. Pigment extraction

In order to determine the fluorescence bands of pigmentsf the outer surface of the fruit, the pigments were extractedn absolute methanol. The outer surface of the fruit for thisurpose comprised the skin including the first layer of under-ying pericarp. In order to make detachment of the sampleasy, the fruit were frozen for 2 h at −20 ◦C. A section 1 mmhick (three sections for each fruit) corresponding to the skinplus a little layer of pericarp), was then taken using a scalpel.he samples were soaked in 10 ml of methanol for 2 minnd then homogenised for 10 min. In order to evaluate, theffect of filtration of the sample on fluorescence emission, anmount equal to 5 ml of homogenised sample obtained fromomatoes at stage F1 was filtered through a double layer ofauze and the remaining 5 ml analysed without filtration.

.3. Spectrofluorescence analysis

About 5 ml of methanol-extracted samples were put intocuvette (1 cm path length) and the fluorescence measured

pectrofluorimetrically. In order to associate the peak fluo-escence with the chemical compounds present in the fruitkin, standards of anthocyanins, flavonoids, carotenoids andhlorophyll a (all Sigma UK) were dissolved in absoluteethanol (1 mg/l) and measured under the same sample con-

itions.Table 2 shows the main features of the spectrofluorimeter

PTI Quantamaster). This is a system comprising a device forxcitation at variable wavelength, a compartment containing

onochromator Czerny–Turner 0.2 m, f/4xcitation 300 g/mm Bandpass = 5 nmmission 300/500 g/mm Bandpass = 5 nmhotomoltiplicator R928P Hamamatsu

A. Lai et al. / Postharvest Biology and Technology 43 (2007) 335–342 337

F

esbUscm

2

mmftIe

lvmtaeldme

3

3

n

Fig. 2. (a) Fluorescence emission at different λ of excitation of quercetinstandard (flavonol with three aromatic rings and three OH groups) dissolvedin methanol (1 mg/ml) (a.u.: arbitrary unity); (b) fluorescence emission atdwi

stsp(qgr((prititfl

ig. 1. Schematic illustration of the LIF (laser-induced fluorescence) system.

xcited at a fixed excitation wavelength (λexc) and the excitedpectra measured at a fixed emission wavelength (λem) toe acquired. Fluorescence bidimensional maps in the visibleV are created by performing repeated scanning of emis-

ion spectra, varying the excitation wavelength. The operatorarries out the measurement protocol automatically throughacro functions that use the instrument software.

.4. LIF system

In order to perform non-invasive fluorescence measure-ents, two fruit at extreme ripening conditions were used:ature green (F1) and advanced stage of maturation (F7). The

ruit, after cleaning with paper, were put on a flat surface andhe external pigments in situ analysed using the LIF (Lasernduced Fluorescence) system (Fig. 1). The measurement onach fruit was repeated three times.

The available excitation wavelengths, determined by theaser used, were 532 nm (visible) and 355 or 266 nm (ultra-iolet). The laser source was pulsed and the fluorescenceeasurement performed by acquiring the spectrum on a mul-

ichannel analyser into a temporal window (300 ns) openingt the end of the laser (pulse). This operated at a 10 Hz rep-tition rate and the measurement was mediated with 100aser impulses. The measurements were carried out withoutamaging the surface: following the excitation by monochro-atic low intensity non-focused laser radiation, fluorescence

mitted at the surface was registered.

. Results and discussion

.1. Pigment fluorescence spectra

The fluorescence spectrum of methanol as the solvent didot interfere with pigment fluorescence spectra emission and

iwa(

ifferent λ of excitation of pelargonidine chloride standard (anthocyaninith three aromatic rings, four OH groups and a chetonic group) dissolved

n methanol (1 mg/ml) (log scale).

howed the typical Raman peaks of OH for each excita-ion wavelength (λexc) (data not shown). The fluorescencepectra of the standards assigned the groups of fluorescentigments to chlorophyll, flavonoids (Fig. 2a) anthocyaninsFig. 2b), and carotenoids. For the flavonoid standard,uercetin (a flavonol with three aromatic rings and three OHroups) was used. Three distinct emission bands of compa-able intensity were identified: 300 nm (λexc = 266 nm), bluebetween 450 and 500 nm, λexc = 266 and 355 nm) and yellow570–580 nm, λexc = 480 nm) (Fig. 2a). As an anthocyanin,elargonidine chloride (an anthocyanin with three aromaticings, four OH groups and a chetonic group) was used, andn this case, the UV band (λexc = 266 nm, emission 380 nm),he blue band (λexc = 355–400 nm, emission 450 nm) andntense red bands (showing emission at 600 nm excited overhe whole visible range) were recognised (Fig. 2b). Theuorescence emission band spectra of chlorophyll a were reg-

stered at about 680 nm, as expected (Chappelle et al., 1984),hile the �-carotene spectrum appeared as an expanded

nd roughly structured band, ranging from blue to green420–580 nm).

338 A. Lai et al. / Postharvest Biology and Technology 43 (2007) 335–342

F m fruit(

tfliTmwg

vddpestaw

tt(tpffltbcrmia

ig. 3. Fluorescence mapping of tomato pigments extract by methanol froF6); (d) over-ripe (F7) fruit.

In all the ripening stages of the fruit the fluorescence spec-ra of fluorescent pigments such as chlorophyll, anthocyanins,avonoids and carotenoids were detectable with different

ntensities related to the ripening stage of the fruit (Fig. 3).he spectra quality was better on the gauze-filtered speci-en, where the absorption effects and particulate scatteringere minimised. For the following assays, we operated withauze-filtered samples.

In Fig. 3, the results of the pigments present in the fruit har-ested over the stages F1, F4, F6, and F7, and their changesuring ripening are shown. Green tomatoes (F1 stage, Fig. 3a)o not contain carotenoids and the presence of chlorophyll isredominant. The presence of flavonoids is also detectableven if the fruit colour is green. At the green stage (F1

tage), the flavonoids are metabolised and incorporated intohe epicuticular waxes but not in the cutin matrix (Laguna etl., 1999) and are not visible on the fruit. It could be thatith extraction with methanol, fluorescence allows detec-

b

pd

skin at different ripening stages: (a) unripe (F1); (b) turning (F4); (c) ripe

ion of flavonoids that are in solution. During fruit ripeninghe flavonoids are progressively incorporated into the cutinLaguna et al., 1999) and the spectra progressively (F4) shiftowards carotenoids (Fig. 3b), but the presence of chloro-hyll was still detectable. This change continued until F6,ull ripening, when chlorophyll, but not anthocyanins andavonoids, completely disappeared and the carotenoid con-

ents strongly increased (Fig. 3c). In the over-ripe fruit (F7),oth anthocyanin and flavonoid fluorescence intensity wasomparable to that of the immature fruit (Fig. 3a and d). Thisesult suggests that the physiological condition of the fruit hadarkedly changed. The observed accumulation of flavonoids

n stored fruit at the F7 stage could be due to the increasedntioxidant activity as previously observed in broccoli flower

uds (Starzynska et al., 2003).

The excitation spectra allowed the different kind of chloro-hylls present in the fruit to be detected. By comparing theifferences in fluorescence intensity for chlorophyll a and

A. Lai et al. / Postharvest Biology and T

Ft

bgatw

3

mrsafi

ea

1psodλ

ecflt

TB

P

AFCC

iof0(fTteasaftbea(

rt(rmfes

3

tommbftzaa

se

ig. 4. Excitation spectra of chlorophyll a and b extracted by methanol fromomato fruit skin at different stages of ripening (a.u.: arbitrary unity).

during ripening (Fig. 4), it was possible to identify thereater height of absorption peaks observed for chlorophyll(435 nm) and b (470 nm) in the F2 and F3 stages. A varia-

ion in the ratio between the two types of pigments associatedith the ripening stages is evident (Fig. 4).

.2. Spectral ratios and correlation analysis

The fluorescence spectra, collected by measuring the pig-ents extracted from the fruit epicarp at each stage of

ipening, were analysed by integrating the fluorescence inten-ity of bands representative of each group of pigments; thenalysis was performed by using the data obtained from bothltered and non-filtered samples.

The excitation wavelengths used for the production of themission fluorescence and spectra were those available withNd:YAG laser (λexc = 266, 355 and 352 nm) (Table 3).

The chosen bandwidths were defined with a constant0 nm width, allowing non-invasive LIF measurements to beerformed on the surface of the entire fruit. Lacking internaltandards for intensity, data were successively elaborated bybtaining the spectral ratios between the intensities of theifferent couple of bands. The emission spectral ratios withexc = 266 nm refer to chlorophyll/flavonoids: R(em672/

m435), carotenoids/chlorophyll: R(em525/em672), antho-yanins/flavonoids: R(em335/em435) and carotenoids/avonoids: R(em525/em435). The results obtained for fil-

ered and non-filtered samples are given in Fig. 5. Differences

able 3ands of fluorescent pigments utilized for data spectrofluorimeter analysis

igment λexc (monochromatic) (nm) λemis (nm)

nthocyanins 266 335lavonoids 266 435arotenoids 355 525a

hlorophyll a 355 672a

a Emissions detected also in the case of the high excitation laser at 266 nm.

uba(wmtflg2cw

echnology 43 (2007) 335–342 339

n pigment contents related to fruit ripening were foundnly for the spectral ratio chlorophyll/flavonoids (decreasingrom 2.5 and 3 for F1 filtered and non-filtered samples to.092 for both filtered and non-filtered F5, F6 samples)Fig. 5a) and carotenoids/chlorophyll (increasing from 0.14or F1 to 8.05 and 6.9 for F5 and F6, respectively) (Fig. 5b).he differences between the highest and lowest values of

he other spectral ratios were not relevant (more or less 0.3)xcept for the F7 sample with the carotenoids/chlorophyll,nthocyanins/flavonoids and carotenoids/flavonoids ratiosignificantly higher than other ripening stages (Fig. 5b–d)ccording to antioxidant compound accumulation due toruit senescing during storage. For the samples F2–F4,he spectral ratios of anthocyanins/flavonoids showed aroad difference between filtered and non-filtered; theffect is probably due to fluorescence emitted by the largermounts of cellulose present into the non-filtered onesFig. 5c).

A correlation analysis was performed between spectralatios of the principal fluorescent pigments. The correla-ion coefficients were positive for the carotenoids/flavonoids0.780) ratios and negative for the carotenoids/chlorophyllatios (−0.513), matching chloroplast–chromoplast transfor-ation during fruit ripening. A mild positive correlation was

ound for the anthocyanins/flavonoids ratios (r = 0.651), asxpected because anthocyanins and flavonoids are synthe-ized and accumulate within the same cells.

.3. LIF measurements

The results from the analysis of fluorescence by the spec-rofluorimeter, performed on the pigments present on theuter surface of tomato fruit, allowed for a choice to beade of the laser radiation most useful for remote measure-ents with LIF. The evaluation was carried out by operating

oth with unripe (F1) and over-ripe (F7) fruit. For the F1ruit the following areas were considered: the top and bot-om green surface, a more green area and a damaged necroticone (Fig. 6). For F7 fruit, healthy top and bottom surfacesnd the wilted zone of fruit, the result of over-ripening, werenalysed (Fig. 7).

Exciting into the UV at 266 nm, the induced fluorescencepectra obtained by LIF appeared to be different in differ-nt physiological stages of the fruit (Figs. 6 and 7). Thenripe fruit had an area of maximum fluorescence emissionetween 330 and 450 nm, corresponding to the anthocyaninsnd flavonoids (Fig. 6). Another smaller band around 530 nmcarotenoids) was also observed (Fig. 6). These pigmentsere already detected in F1 fruit pigments extracted inethanol (Fig. 3a). This result confirms that the LIF spec-

roscopy technique is a useful tool to detect, non-invasively,avonoid, anthocyanins and carotenoids pigments in unripe

reen tomato as reported for apple and carrots (Wulf et al.,005). In the over-ripe F7 fruit (Fig. 7), the maximum fluores-ence was observed between 450 and 550 nm (carotenoids)hile the flavonoids fluorescence was not as evident as in

340 A. Lai et al. / Postharvest Biology and Technology 43 (2007) 335–342

F tomatoF

tttm

sdtnd

Ffs

aa

eins

ig. 5. Spectral ratios of the different pigments extracted by methanol from7) ((�) filtered sample; (�) non-filtered sample).

he methanol extracted F7 fruit (Fig. 3d). At this stage,he flavonoids are in a deeper layer of epicarp tissue whilehe radiation penetration of UV excitation is only a few

icrons.From these results, it appears that the fluorescence inten-

ity varies for each fruit stage analysed. Since the bandwidthoes not change, the variation observed is connected to

he different concentrations of pigments. In contrast, in theecrotic area of F1 fruit the band shape was completelyifferent, reaching the maximum fluorescence emission at

ig. 6. LIF spectra of different outer zones in the unripe F1 skin of tomatoruit (λex = 266 nm not focused). In the panel the expanded red zone of thepectrum (wavelength above 600 nm) is shown.

po6pi

Ff

fruit skin, according to different ripening stages (F1–F6 and over-ripe fruit

round 430 nm (flavonoids), while the emission band ofnthocyanins completely disappeared (Fig. 6).

Analysing the insert of Fig. 6, where the fluorescencemission in the red region (chlorophyll) is reported, it ismportant to stress that, notwithstanding that chlorophyll isot excited in a resonant manner (λex = 266 nm), an emis-ion band between 670 and 730 nm, distinctive of the greenigment, was recognisable. Such an emission was detectedn both inner and outer healthy tissue surfaces (peak around

80 nm), while for the greener and the necrotic areas, thiseak was not present at all. The last observation may benterpreted as representing a delay in chlorophyll biosynthe-

ig. 7. LIF spectra of different outer zones in the over-ripe F7 skin of tomatoruit (λex = 266 nm not focused).

gy and T

sc

4

npfoSistleocor

ffiapb

•

•

•

•

ccfsictcsspt

tdre

tict

A

Ot

R

B

C

C

C

D

G

L

L

M

P

P

R

S

A. Lai et al. / Postharvest Biolo

is and degradation, respectively, for the two physiologicalonditions considered.

. Conclusion

Spectroscopy techniques, including fluorescence tech-iques, are already widely used as diagnostic tools forlant stress, and imaging techniques have been proposedor presymptomatic non-destructive monitoring of changesf the physiological state of plants (Chaerle and Van dertraeten, 2000). These methods may also be powerful tools

n order to detect physiological changes occurring in plants,uch as fruit ripening (Wulf et al., 2005), and in particular,hey allow a study of changes in important secondary metabo-ites during fruit ripening (Polder et al., 2004; Baranskit al., 2005). Therefore, the first step has been the studyf flavonoids, anthocyanins, chlorophyll and carotenoidshanges during ripening, using fluorescence techniques inrder to characterize the single wavelength and its spectralatio.

The results allowed us to assign the fluorescence bandsor each pigment extracted from the outer surface of tomatoruit; specifically, the main fluorescence bands of pigmentsnvolved in ripening (flavonoids, anthocyanins, chlorophyll,nd carotenoids) have been identified (Figs. 2 and 3). Suchigments have been tagged with the following fluorescenceands:

flavonoids, three emission bands: 300, 450, and 500 nm(λex = 266 nm);anthocyanins: the emission band at 330, 380 and 450 nm(λex = 266, 355, and 400 nm), and one more at 600 nm(λex = 355, 400 and 450 nm);chlorophyll, a characteristic band at 680 nm(λex = 355 nm);�-carotene, one wide band spanned from 420 to 580 nm(λex = 355 nm).

Furthermore, the results show that the spectral ratio ofarotenoids/chlorophyll: R(em525/em672) is the best indi-ator of the stage of fruit ripening correlated with enhancedruit quality characteristics (Fig. 5b). In fact, this ratiohows that at the F5 stage, the carotenoids increase (approx-mately 10 times more than at the F4 stage) and thehlorophyll disappears (Fig. 5b). The increase in the spec-ral ratio of anthocyanins/flavonoids: R(em335/em435) andarotenoids/flavonoids: R(em525/em435) observed at the F7tage (Fig. 5b–d) suggests increased antioxidant activity intressed fruit (Fig. 3d). These spectral ratios could be alsoroposed as an index to evaluate oxidative stress in toma-oes.

The results obtained with the LIF technique demonstrate

hat it is possible to monitor fruit surface pigments andamage non-invasively. In this case, the assignment of fluo-escence bands has allowed for the choice of the most usefulxcitation laser radiation (λ = 266 and 355 nm). In perspec-

S

echnology 43 (2007) 335–342 341

ive, the use of the LIF technique combined with fluorescencemaging, and tomato mutants altered in the skin and/or pulpolour, might shed new light on possible connections betweenhe different pathways.

cknowledgement

This work was supported by the National Project TECN-DIS (quality control of agrofood by means of non-intrusive

echniques).

eferences

Baranski, R., Baranska, M., Schulz, H., 2005. Changes in carotenoid contentand distribution in living plant tissue can be observed and mapped in situusing NIR-FT-Raman spectroscopy. Planta 222, 448–457.

ron, I.U., Ribeiro, R.V., Azzolini, M., Jacomino, A.P., Machado, E.C.,2004. Chlorophyll fluorescence as a tool to evaluate the ripen-ing of “Golden” papaya fruit. Postharvest Biol. Technol. 33, 163–173.

ano, A., Acosta, M., Arnao, M.B., 2003. Hydrophilic and lipophilic antioxi-dant activity changes during on-vine ripening of tomatoes (Lycopersiconesculentum Mill.). Postharvest Biol. Technol. 28, 59–65.

haerle, L., Van der Straeten, D., 2000. Imaging techniques and the earlydetection of plant stress. Trends Plant Sci. 5, 495–501.

happelle, E.W., Wood, F.M., McMurtrey III, J.E., Newcomb, W.W., 1984.Laser-induced fluorescence of green palnts. 1. A technique for remotedetection of plant stress and species differentiation. Appl. Opt. 23,134–138.

eEll, J.R., Prange, R.K., Murr, D.P., 1996. Chlorophyll fluorescence ofdelicious apples as harvest a potential predictor of superficial scalddevelopment during storage. Postharvest Biol. Technol. 9, 1–6.

iovanelli, G., Lavelli, V., Peri, C., Nobili, S., 1999. Variation in antioxidantcomponents of tomato during vine and post-harvest ripening. J. Sci. FoodAgric. 79, 1583–1588.

aguna, L., Casado, C.G., Heredia, A., 1999. Flavonoids biosynthesis intomato fruit cuticles after in vivo incorporation of 3H-phenylalanineprecursor. Physiol. Plant. 105, 491–498.

aval-Martin, D., Quennemet, J., Moneger, R., 1975. Pigment evolution inLycopersicon esculentum fruits during growth and ripening. Phytochem-istry 14, 2357–2362.

artınez-Valverde, I., Periago, M.J., Provan, G., Chesson, A., 2002. Pheno-lic compounds, lycopene and antioxidant activity in commercial varietiesof tomato (Lycopersicon esculentum). J. Sci. Food Agric. 82, 323–330.

older, G., van der Heijden, G.W.A.M., van der Voet, H., Young, I.T., 2004.Measuring surface distribution of carotenes and chlorophyll in ripeningtomatoes using imaging spectrometry. Postharvest Biol. Technol. 34,117–129.

range, R.K., DeLong, J.M., Leyte, J.C., Harrison, P.A., 2002. Oxygenconcentration affects chlorophyll fluorescence in chlorophyll-containingfruit. Postharvest Biol. Technol. 24, 201–205.

affo, A., Leonardi, C., Fogliano, V., Ambrosino, P., Salucci, M., Gen-naro, L., Bugianesi, R., Giuffrida, F., Quaglia, G., 2002. Nutritionalvalue of cherry tomatoes (Lycopersicon esculentum cv. Naomi F1) har-vested at different ripening stages. J. Agric. Food Chem. 50, 6550–6556.

chofield, R.A., DeEll, J.R., Murr, D.P., Jenni, S., 2005. Determining

the storage potential of iceberg lettuce with chlorophyll fluorescence.Postharvest Biol. Technol. 38, 163–173.

limestad, R., Verheul, M.J., 2005. Content of chalconaringenin and chloro-genic acid in cherry tomatoes is strongly reduced during postharvestripening. J. Agric. Food Chem. 53, 7251–7256.

3 gy and T

S

T

U

42 A. Lai et al. / Postharvest Biolo

tarzynska, A., Leja, M., Mareczek, A., 2003. Physiological changes inthe antioxidant system of broccoli flower buds senescening during

short-term storage, related to temperature and packing. Plant Sci. 165,1387–1395.

oivonen, P.M.A., DeEll, J.R., 1998. Differences in chlorophyll fluores-cence and chlorophyll content of broccoli associated with maturity andsampling section. Postharvest Biol. Technol. 14, 61–64.

W

echnology 43 (2007) 335–342

SDA, 1991. United States Standards for Grades of Fresh Tomatoes. UnitedStates Department of Agriculture, Agricultural Marketing Service, Fruit

and Vegetable Division, Fresh Products Branch, pp. 1–14.

ulf, J.S., Geyer, M., Nicolaı, B., Zude, M., 2005. Non-destructive assess-ment of pigments in apple fruit and carrot by laser-induced fluorescencespectroscopy (LIFS) measured at different time-gate positions. ActaHortic. 682, 1387–1394.