T(

FFa

b

a

ARA

KATDPF

1

wacKars(aTcd

fiLMad

0d

Postharvest Biology and Technology 69 (2012) 54–63

Contents lists available at SciVerse ScienceDirect

Postharvest Biology and Technology

journa l h o me pa g e: www.elsev ier .com/ locate /postharvbio

exture dynamics during postharvest cold storage ripening in appleMalus × domestica Borkh.)

abrizio Costaa,∗, Luca Cappellina, Marco Fontanaria, Sara Longhia, Walter Guerrab, Pierluigi Magnagoa,lavia Gasperia, Franco Biasioli a

Research and Innovation Centre, Edmund Mach Foundation, Via Mach 1, 38010 San Michele all’Adige (TN) ItalyLaimburg Research Centre for Agriculture and Forestry, Via Laimburg 6, 39040 Ora (BZ) Italy

r t i c l e i n f o

rticle history:eceived 15 December 2011ccepted 4 March 2012

eywords:pple fruitextureynamics

a b s t r a c t

Texture is a principal quality factor and represents one of the main priorities in apple postharvest man-agement and breeding programs designed for the creation of new ideotypes defined by a better fruitquality and extended storability. The apple panorama is characterized by a great variability of textureperformance due to specific functional regulation and genetic control of the physiological machinerydevoted to the degradation of the polysaccharide architecture of the middle lamella/cell wall structure.In this work we present an investigation of texture dynamics in apple, in terms of variation of severaltexture components dissected over two months of postharvest storage. Apple texture was assessed at

ostharvest storageruit storage index

harvest and after storage, by acquiring both mechanical and acoustic profiles in a collection of 83 applecultivars. The general texture variability, illustrated over the reduced hyperspace defined by principalcomponents, revealed a different variety distribution between the two stages. Time evolution plots andthe novel storage index presented here highlighted that each single texture component behaves differ-ently, as seen by some cultivars (i.e. ‘Fuji’) having a more favourable acoustic response after postharvest.The dissected fruit texture dynamics assessed in a set of reference apple cultivars are discussed.

. Introduction

The “crispy” flesh of apple represents a major quality trait,hich drives consumer preference (Harker et al., 2003). Texture

ttributes are directly associated with more general fruit qualityoncepts, such as fruit freshness and eating enjoyment (Fillion andilcast, 2002; Konopacka and Plocharski, 2004). Texture, as wells the other principal quality factors (Bourne, 2002), has beenecognized as a multi-factorial trait, being composed of severalub-traits grouped in two main categories, such as mechanicalfirmness, hardness, stiffness and elasticity) and acoustic (crispnessnd crunchiness) components, as reported in Szczesniak (2002).

hese two categories are largely distinguished by their physi-al nature and recording procedures. Mechanical components areetermined by a force displacement measured in Newtons, while

Abbreviations: Yield F., yield force (initial force); Max F., maximum force; Final F.,nal force; F. Peak, number of force peaks; Area, area below the mechanical profile; F.in. Dist., force linear distance; Young’s M., Young’s modulus (elasticity modulus);ean F., mean force; Ac. Peak, number of acoustic peaks; Max Ac. P., maximum

coustic pressure; Mean Ac. P., mean acoustic pressure; Ac. Lin. Dist., acoustic linearistance; � F., force difference; F. R., force ratio.∗ Corresponding author. Tel.: +39 0461 615387; fax: +39 0461 650956.

E-mail address: fabrizio.costa@fmach.it (F. Costa).

925-5214/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.postharvbio.2012.03.003

© 2012 Elsevier B.V. All rights reserved.

the acoustic signature is dependent upon the sound emission occur-ring at the onset of sample fracturing and measured in decibels(Duizer, 2001; Kilcast, 2004). Dissection and comprehension ofthe texture sub-traits is of basic importance, especially for theacoustic components, due to the fact that crispness is the texturesub-phenotype most appreciated by consumers (Hampson et al.,2000).

Despite the importance of texture in the definition of fruit qual-ity, the analytical strategies developed to measure and dissect fruittexture have been only recently improved. To date, fruit texture hasbeen characterized, in most cases, by empirical methods (Harkeret al., 2003). The fruit puncture test (penetrometer) has been largelyused, but this is strictly related only to cortex firmness assess-ment. More sophisticated equipment providing a better texturedescription have been recently proposed, employing a wide vari-ety of techniques, such as mechanical measurements (Mehinagicet al., 2004; Harker et al., 2006), sound recording during chew-ing (De Belie et al., 2000; Roudaut et al., 2002; Ioannides et al.,2007) and the detection of vibration generated by sample fractur-ing (Taniwaki et al., 2006; Taniwaki and Sakurai, 2008). However,

all these methodologies have been generally used only on a limitednumber of cultivars.

Among them, the combined and simultaneous analysis of bothtexture components (mechanical and acoustic) has been suggested

ogy and Technology 69 (2012) 54–63 55

aeeekdattcwag

tqaueq2e

mwsirf

2

2

hRSrCira

toowwt

2

bEfiatsiftd

Table 1List of the 83 apple cultivars employed in this study, with their respective code usedin Figs. 2, 3 and 5.

Code Cultivar

1 Almagold2 Ananas Renette3 Ariane4 Ariwa5 Baujade6 Baumans Renette7 Bellida8 Berner Rosen9 Boskoop

10 Braeburn11 Brina12 Brixner Plattling13 Calvilla Bianca14 Caudle15 CIVG19816 Coop 3917 MC 3818 Cripps Pink19 Cripps Red20 Croncels21 Dalinette22 Dalitron23 Dalla Rosa24 Delblush25 Delcorf26 Delcoros27 Delearly28 Delorina29 Early Gold30 Edelböhmer31 Fiamma32 Florina33 Fuji34 Gala (Baigent)35 Gala (Gala Schnitzer)36 Galmac37 Gelber Edelapfel38 Gloster39 Gold Pink40 Coop 3841 Golden Delicious42 Golden Orange43 Granny Smith44 Gravensteiner45 Idared46 Jonagold47 Kronprinz Rudolph48 La Flamboyante49 Ligol50 Magrè51 Maigold52 Milwa53 Minnewashta54 Morgenduft55 Napoleone56 Nevson57 Nicogreen58 Nicoter59 Permain Dorato60 Pilot61 Rafzubin62 Red Delicious (Camspur)63 Red Elstar64 Red Field65 Renetta Canada66 Renetta Champagne67 Rosa di Caldaro68 Rosa Doppia69 Roter Stettiner70 Gala (Tenroy)

F. Costa et al. / Postharvest Biol

s the most efficient strategy, enabling a complete and morexhaustive texture characterization (Varela et al., 2006; Zdunekt al., 2010). In apple the employment of a novel texture analyzer,quipped with an acoustic envelop device, has led, to the best of ournowledge, to the largest apple texture investigation performed toate (Costa et al., 2011). This latter study used both mechanical andcoustic signatures to describe texture variability of 86 apple cul-ivars after two months of cold storage. As additional validation ofhis proof of concept, the same methodology was exploited in theharacterization of two full-sib populations (Ballabio et al., 2011),hich allowed one of the largest QTL mapping investigations for

pple fruit texture (Longhi et al., 2011), directed to targetting theenomic regions involved in texture control.

New parameters defined to dissect texture variability, iden-ified by Costa et al. (2011), can thus be considered as noveluality indexes to be used for the selection of novel individu-ls derived from breeding programs. However, quality traits aresually assessed only once, and generally at harvest, yet apples gen-rally undergo long periods of postharvest storage, during whichuality traits evolve in a cultivar dependent fashion (Johnston et al.,001; Konopacka and Plocharski, 2004; Varela et al., 2006; Billyt al., 2008).

The aim of this work was to investigate changes, during twoonths of postharvest cold storage, in several texture parametersith a collection of 83 apple cultivars, employing the phenomic

trategy previously presented in Costa et al. (2011). The storagendex discussed here highlights a defined dynamics for each textu-al component, suggesting a unique texture postharvest physiologyor each apple cultivar.

. Materials and methods

.1. Plant materials

Texture dynamics was assessed on 83 apple cultivars (Table 1)arvested in 2010 in two experimental orchards, belonging to theesearch and Innovation Centre of FEM (Foundation Edmund Mach,an Michele, Italy) and the Laimburg Research Centre (Ora, Italy),espectively, as documented in the proof of concept published byosta et al. (2011). Fruit from each cultivar were harvested accord-

ng to the parameters that are normally used to monitor apple fruitipening and chosen to define the best picking time, such as skinnd seed colour, fruit firmness and starch conversion index.

The harvested fruit from each cultivar were then divided intowo sub-groups. The first was assessed within a day of harvest aftervernight storage at room temperature (about 20 ◦C), while the sec-nd was stored in normal atmospheres for two months, at 2 ◦C andith a relative humidity of approximately 95%. The stored applesere then kept overnight at room temperature (around 20 ◦C) prior

o instrumental analysis.

.2. Texture analysis

Texture was assessed using a TA-XTplus texture analyzer (Sta-le MicroSystem Ltd., Godalming, UK), equipped with an Acousticnvelop Device (AED), which simultaneously profiled a mechanicalorce displacement together with an acoustic response. The dynam-cs was investigated by assessing the texture behaviour at harvestnd after two months of cold storage, in order to underline theexture modifications occurring during postharvest ripening. Fruitample preparation and instrument settings are described in detail

n Costa et al. (2011). In brief, samples (composed by five biologicalruit samples and four technical replicates/samples, for a total ofwenty measurements per cultivar), consisted of a flesh disc 1.7 cmiameter and 1 cm thick.

71 Rubinola72 San Lugano73 Santana74 Saturn

56 F. Costa et al. / Postharvest Biology an

Table 1 (Continued)

Code Cultivar

75 Red Delicious (Evasni)76 Scifresh77 Scilate78 Shinano Gold79 Stayman Red80 Summerfree81 Tavola Bianca82 Weisser Rosmarin

c1a9aptfatm

tyai

TL

T

83 Weisser Wintertaffet

A force displacement profile was detected by a 5 kg loadingell (equipped with a 4 mm flat head probe) with a test speed of00 mm/min (and a post-test returning speed of 300 mm/min) and

trigger force of 5 g, compressing the disc until the deformation of0% (strain). Acoustic response was profiled by the AED device, with

frequency cut off set at 3.125 kHz. From the two profiles 14 mainarameters were derived (Table 2). The mechanical response washen characterized by yield force (or initial force), max force, finalorce, mean force, number of force peaks, area, force linear distance,nd Young’s modulus. The acoustic resolution was instead charac-erized by parameters such as number of acoustic peaks, max and

ean acoustic pressure and acoustic linear distance.Force direction parameters were represented by � force and

he force ratio, interpreted as the difference and ratio between theield (initial) and final force. These two parameters, related to thescending/descending direction of the mechanical profile, provide

nformation about flesh disc compression.

able 2ist of mechanical and acoustic parameters related to the texture profiling.

Mechanical parameters General description Unit

Yield force Force measured at theyield point

N

Max force Maximum force valuerecorded over the probe’stravel

N

Final force Force measured at the endof the probe’s travel

N

Mean force Mean force value over theentire mechanical profile

N

Force peak Number of counted forcepeaks

–

Area Area underlying themechanical profile

N%

Force linear distance Computation of the forcecurve length

–

Young’s modulus Elasticity modulus,computed as ratio betweenstress and strain

N%

Acoustic parameter General description Unit

Acoustic peak Number of the acousticpeaks calculated above thethreshold of 10 dB

–

Max acoustic pressure Highest acoustic peaksdetected on the soundpressure wave

dB

Mean acoustic pressure Mean value of the soundpressure recorded over theacoustic profile

dB

Acoustic linear distance Computed length of theacoustic profile

–

Force direction parameter General description Unit

� force Yield force–final force –Force ratio Yield force–final force –

he description of these parameters was adapted from Costa et al. (2011).

d Technology 69 (2012) 54–63

2.3. Data analysis

Texture parameters were acquired from the combined profilesby the software EXPONENT v4 (Stable MicroSystem Ltd., Godalm-ing, UK).

Texture variability for the 83 cultivars at both harvest and aftertwo months of cold storage, as well as the variable projection,was illustrated in a reduced hyperspace by Principal ComponentAnalysis computed with R package (R Development Core Team,R: A Language and Environment for Statistical Computing, Vienna,Austria, 2009). With the same software, the time evolution plotover the two months of cold storage of four representative texturalparameters (maximum force and Young’s modulus for the mechan-ical component, and number of acoustic peaks and mean acousticpressure for the acoustic signature) were also illustrated.

3. Results and discussion

3.1. Texture dynamics

This work employed the same technological approach describedby Costa et al. (2011) to assess texture dynamics over a period oftwo months of cold postharvest storage. This investigation aimed atillustrating the physical texture variation occurring during posthar-vest ripening on a set of 83 apple cultivars. To achieve this goal, thetwo texture components (mechanical and acoustic) were simulta-neously profiled, for each cultivar, at two specific times, at harvestand after two months of cold storage.

The comparison of the texture profiles highlighted three maintexture dynamics trends (Fig. 1). In the first case, represented bythe cultivars ‘Golden Delicious’ (selected as reference cultivar) and‘Magrè’, a general decrease in both mechanical and acoustic profileswas observed. In the second case, a stable physiological situationwas depicted, with a practically un-modified profile determinedbetween the two stages (‘Maigold’, Fig. 1). The third case, whichshed light on a particular comparison, showed that in a specific setof cultivars (‘Fuji’) the texture slightly improved over storage, in thesense that the texture profile magnitude (acoustic in particular) washigher after storage than after harvest.

Texture was then dissected as described in detail by Costaet al. (2011), and from the combined profiles (represented byboth mechanical and acoustic signatures) a set of parameters wasextracted and used as a novel analytical index to describe thisphenomenon (Table 2). The Pearson correlation computed com-paring all parameters (for both stages) revealed that these could begrouped into two main categories, mechanical and acoustic. Thisdistinct clustering revealed that the parameters were more corre-lated within each of the two groups, as illustrated by the correlationmatrix (Table 3), with the exception of the number of force peaks(confirming the results previously described by Costa et al., 2011),which underlined a higher association with the acoustic groupthan the mechanical one. The brittle fracturing progression of themechanical profile is in fact the causal event generating the acous-tic response, accompanied by a corresponding number of acousticpeaks (Vincent, 1998).

The two Pearson matrices illustrate similar correlation trendsamong the several texture parameters characterized at both stages,apart from two cases, represented by the force linear distance andYoung’s modulus, also known as elasticity modulus. At harvest, infact, these two parameters are independent, not being correlatedwith any of the other parameters, while after two months of stor-

age, the force linear distance was basically correlated with almostall the parameters (with the exceptions for final force, force peaksand max acoustic pressure), and the Young’s modulus was morecorrelated with the mechanical parameter set. This correlation

F. Costa et al. / Postharvest Biology and Technology 69 (2012) 54–63 57

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

Magrè

Fuji

Golden

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

Maigol d

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

Forc e (N ) Acoust ic (dB)

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

Magrè

Fuji

Golden

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

0 20 40 60 800

2

4

6

8

10

12

14

16

18

35

40

45

50

55

60

65

70

75

Maigol d

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

Forc e (N ) Acoust ic (dB)

Stra in (%)

)Bd(citsuocA)N(ecroF

Stra in (%)

Fig. 1. Texture dynamics over two months of cold storage. For each cultivar, left and right panels illustrate the texture profile at harvest and after two months of cold storage,respectively. Black line corresponds to the mechanical displacement profile and the grey line to the acoustic profile. X axis: 90% strain. Primary Y axis: force profile (N).Secondary Y axis: acoustic profile (dB).

58

F. Costa

et al.

/ Postharvest

Biology and

Technology 69

(2012) 54–63

Table 3Pearson correlation matrix at harvest (a), and after two months of postharvest cold storage (b). Significant correlations are in bold (r > 0.6, P-value ≤ 0.05).

(a) Harvest Yield F. Max F. Final F. F. Peak Area F. Lin. Dist. Young’s M. Mean F. Ac. Peak Max Ac. P. Mean Ac. P. Ac. Lin. Dist. � F. F.R.

Yield F. 1.000 0.746 0.536 −0.355 0.834 0.382 0.535 0.824 −0.356 −0.267 −0.440 −0.406 0.370 0.008Max F. 0.746 1.000 0.900 −0.406 0.915 0.315 0.386 0.917 −0.434 −0.320 −0.462 −0.481 −0.274 −0.018Final F. 0.536 0.900 1.000 −0.377 0.795 0.195 0.244 0.803 −0.402 −0.298 −0.411 −0.442 −0.586 −0.034F. Peak −0.355 −0.406 −0.377 1.000 −0.374 0.439 0.120 −0.393 0.657 0.614 0.683 0.692 0.074 0.064Area 0.834 0.915 0.795 −0.374 1.000 0.344 0.486 0.998 −0.423 −0.313 −0.471 −0.477 −0.075 −0.022F. Lin. Dist. 0.382 0.315 0.195 0.439 0.344 1.000 0.565 0.319 0.424 0.461 0.364 0.410 0.152 0.055Young’s M. 0.535 0.386 0.244 0.120 0.486 0.565 1.000 0.446 −0.041 0.105 −0.002 −0.029 0.245 0.034Mean F. 0.824 0.917 0.803 −0.393 0.998 0.319 0.446 1.000 −0.431 −0.327 −0.486 −0.488 −0.092 −0.025Ac. Peak −0.356 −0.434 −0.402 0.657 −0.423 0.424 −0.041 −0.431 1.000 0.765 0.770 0.973 0.101 0.031Max Ac.P. −0.267 −0.320 −0.298 0.614 −0.313 0.461 0.105 −0.327 0.765 1.000 0.700 0.781 0.071 0.054Mean Ac.P. −0.440 −0.462 −0.411 0.683 −0.471 0.364 −0.002 −0.486 0.770 0.700 1.000 0.860 0.030 0.072Ac. Lin. Dist. −0.406 −0.481 −0.442 0.692 −0.477 0.410 −0.029 −0.488 0.973 0.781 0.860 1.000 0.097 0.050� F. 0.370 −0.274 −0.586 0.074 −0.075 0.152 0.245 −0.092 0.101 0.071 0.030 0.097 1.000 0.045F. R. 0.008 −0.015 −0.034 0.064 −0.022 0.055 0.034 −0.025 0.031 0.054 0.072 0.050 0.045 1.000

(b) Postharvest Yield F. Max F. Final F. F. Peak Area F. Lin. Dist.Young’s M. Mean F. Ac. Peak Max Ac. P. Mean Ac. P. Ac. Lin. Dist. � F. F. R.

Yield F. 1.000 0.883 0.697 0.229 0.899 0.722 0.737 0.891 0.214 0.294 0.312 0.247 0.220 −0.053Max F. 0.883 1.000 0.872 0.274 0.960 0.729 0.663 0.959 0.220 0.296 0.363 0.269 −0.155 −0.196Final F. 0.697 0.872 1.000 0.370 0.868 0.598 0.505 0.873 0.245 0.253 0.410 0.318 −0.546 −0.497F. Peak 0.229 0.274 0.370 1.000 0.335 0.596 0.366 0.323 0.712 0.524 0.789 0.782 −0.236 −0.393Area 0.899 0.960 0.868 0.335 1.000 0.739 0.702 0.999 0.266 0.317 0.400 0.314 −0.130 −0.263F. Lin. Dist. 0.722 0.729 0.598 0.596 0.739 1.000 0.708 0.724 0.619 0.576 0.673 0.646 0.031 −0.113Young’s M. 0.737 0.663 0.505 0.366 0.702 0.708 1.000 0.681 0.310 0.354 0.417 0.357 0.173 −0.045Mean F. 0.891 0.959 0.873 0.323 0.999 0.724 0.681 1.000 0.255 0.307 0.389 0.302 −0.146 −0.271Ac. Peak 0.214 0.220 0.245 0.712 0.266 0.619 0.310 0.255 1.000 0.724 0.870 0.972 −0.082 −0.205Max Ac.P. 0.294 0.296 0.253 0.524 0.317 0.576 0.354 0.307 0.724 1.000 0.712 0.726 −0.001 −0.079Mean Ac.P. 0.312 0.363 0.410 0.789 0.400 0.673 0.417 0.389 0.870 0.712 1.000 0.931 −0.193 −0.317Ac. Lin. Dist. 0.247 0.269 0.318 0.782 0.314 0.646 0.357 0.302 0.972 0.726 0.931 1.000 −0.144 −0.273� F. 0.220 −0.155 −0.546 −0.236 −0.130 0.031 0.173 −0.146 −0.082 −0.01 −0.193 −0.144 1.000 0.614F. R. −0.053 −0.196 −0.497 −0.393 −0.263 −0.113 −0.045 −0.271 −0.205 −0.079 −0.317 −0.273 0.614 1.000

ogy an

cmpt

atatlmanud

oiFtd

mr36aiTTt

bs5‘dow

fpFcthtmc5nr

Cdld

mttelet2

F. Costa et al. / Postharvest Biol

hange, in particular for the fruit flesh elasticity, highlighted evenore the texture change during storage, indicating that distinct

hysiological events occur specifically for some of the dissectedexture components.

Taking into consideration the results of the correlation matrixnd the parameter grouping by the PCA presented in the next sec-ions, four parameters (two belonging to the mechanical profilend two to the acoustic signature) were selected to represent theexture dynamics over two months of cold storage, by a time evo-ution plot (Figs. 2 and 3). Cultivars plotted near the bisector line

aintained a stable texture during the two months of cold stor-ge, while those located further below the line presented importantegative textural changes. Cultivars plotted above the bisector (leftpper side of the plot) are characterized by positive texture changesuring storage.

Generally, the four profiles showed a wide variability in termsf trait dynamics. A more complete overview of the trait variabilitys provided by the boxplots reported in the Supplementary Dataig. S1 where, besides the dynamics of the selected parameters,he parameter variation between cultivars and within cultivars isepicted.

For the maximum force sub-trait (Fig. 2) registered over theechanical displacement profile three main cultivars showed

emarkable texture retention, such as 76 ‘Schifresh’, 60 ‘Pilot’ and3 ‘Fuji’. For this parameter, 64 ‘Red Field’, 37 ‘Gelber Edelapfel’,5 ‘Renetta Canada’, 23 ‘Dalla Rosa’, 13 ‘Calvilla Bianca’, 74 ‘Saturn’nd 27 ‘Delearly’ showed instead the most dramatic loss of max-mum force resolution assessed after two months of cold storage.hese data are in agreement with the results presented by Kühn andhybo (2001) which showed that ‘Saturn’ did not retain sufficientexture during storage.

For Young’s modulus, despite the overall variability, a high num-er of cultivars showed an un-modified behaviour between the twotages. In fact, 76 ‘Schifresh’, 33 ‘Fuji’, 7 ‘Bellida’, 75 ‘Red Delicious’,7 ‘Nicogreen’, 26 ‘Delcoros’, 51 ‘Maigold’, 69 ‘Roter Stettiner’, 21

Dalinette’ and 49 ‘Ligol’ showed an un-modified cortex elasticityuring the experimental period. From the data analysis performedn the mechanical profile it was evident that old apple varietiesere highly affected by a severe decrease in texture properties.

Within the texture component variation, it is worth noting thator the acoustic parameters, considered here for the time evolutionlot, an enhanced performance was observed for some cultivars.or the number of acoustic peaks (Fig. 3), for instance, besides theultivars with a stable profile, 33 ‘Fuji’, 77 ‘Scilate’, 69 ‘Roter Stet-iner’, 28 ‘Delorina’, 57 ‘Nicogreen’, 35 ‘Gala (Schnitzer)’ showedigher performance after two months of cold storage with respecto the harvest stage. A similar trend was also observed for the

ean acoustic pressure, where for 60 ‘Pilot’, 33 ‘Fuji’, 62 ‘Red Deli-ious (Camspur)’, 75 ‘Red Delicious (Evasni)’, 69 ‘Roter Stettiner’,7 ‘Nicogreen’, 56 ‘Nevson’, 26 ‘Delcoros’, 21 ‘Dalinette’, 47 ‘Kro-printz Rudolf’, 77 ‘Scilate’ and 22 ‘Dalitron’ the level of acousticesponse was higher after two months of postharvest storage.

For both the acoustic parameters considered here, 65 ‘Renettaanada’, 46 ‘Jonagold’, 31 ‘Fiamma’ and 50 ‘Magrè’ showed the mostramatic change during storage, loosing most of the acoustic reso-

ution. These varieties can be thus classified as mealy cultivars, asescribed elsewhere (Smedt et al., 1998).

The fact that the acoustic response can be higher after storageight depend on a specific physiological event associated with

he cell wall and the cortex intercellular space. It is well knownhat the mechanical changes occurring in the fruit cortex are gov-rned by a complex enzymatic disassembly of the cell wall/middle

amella polysaccharide architecture. These degradation processesventually result in a weakening of the chemical binding, leadingo a reduction in textural performance (Brummell and Harpster,001; Bourne, 2002; Rose et al., 2003). In addition, crispness, that is

d Technology 69 (2012) 54–63 59

the acoustic response upon mechanical compression, also dependson the turgor pressure and the air space within the cortex cell(Brummell, 2006; Saladie et al., 2007; Thomas et al., 2008). Releaseof the internal air pressure during cell wall breaking stimulates theparticle equilibrium, inducing crispness (Duizer, 2001). In apple,it has been reported that the volume occupied by air increasesduring fruit growth (Drazeta et al., 2004), accompanied by a pro-portional decline in fruit density. This process continues also duringpostharvest storage (Perring and Pearson, 1988; Harker and Hallett,1992).

Thus, during storage, apples loose density and increase involume, and the intercellular air fraction differs among several cul-tivars (Drazeta et al., 2004), supporting the considerable texturalvariation observed in this study over the apple cultivar collectionassessed. The air volume enhancement in some cultivars duringstorage, together with retention of a high level of turgor pressureand middle lamella integrity, might be the reason for a high acousticresponse after two months of cold storage.

In breeding and QTL mapping, fruit quality traits are generallyevaluated at harvest, but on the basis of our findings it would beof fundamental importance for a proper evaluation of storage abil-ity to be carried out, to identify the changes of these traits duringpostharvest. Fruit texture is the feature that changes most duringapple storage, thus an indication of the textural performance foreach apple cultivar would be of great value for breeders in orderto identify the best parents of new individuals with a favourableretention of quality features over storage.

To evaluate the storage potential, regarded as the change of eachdissected texture parameter that is employed here, a storage index(SI) was computed as SI = log2(Ti2M/TiH), where TiH is the valueof the ‘i’ texture parameter measured at harvest, and Ti2M is thevalue of the same parameter measured after 2 months of cold stor-age. Positive SI values indicate a texture sub-trait enhancement,while negative values point to a loss of textural performance dur-ing storage. An SI equal to zero means stable maintenance of thetextural trait under investigation. The SI for each parameter acrossthe 83 cultivars is illustrated in Fig. 4 and Supplementary Data Fig.S2.

An SI plot provides information on the dynamics of the tex-ture dissected into single parameters and consequent cultivarstorability. From the general picture of the texture dynamics rep-resented by the SI it is interesting to note that varieties suchas ‘Dalinette’, ‘Delcoros’, ‘Delorina’, ‘Gala (Schnitzer)’, ‘KronprintzRudolph’, ‘Maigold’, ‘Nevson’, ‘Nicogreen’, ‘Pilot’, ‘Scilate’ and ‘RoterStettiner’ coupled a decreased value in the mechanical parameterwith an increased acoustic response during postharvest ripening(Fig. 4). It is worth noting that ‘Fuji’, the cultivar well known forgreat storability and general fruit quality (Wakasa et al., 2006; Weiet al., 2010), has a positive SI for almost all the dissected texturetraits (Fig. 4), indicating that after two months of cold storage noneof the textural parameters decreased, but, on the contrary, they hada greater performance.

3.2. Principal component analysis of the texture variabilityassessed at harvest and after two months of cold storage

The distribution of the varieties in a reduced hyperspace definedby the first two principal components (Fig. 5a and b), computedover the texture performance at harvest and postharvest, showedclear differences between the two stages considered. The differentvariety distribution on the PCA plot confirmed the texture dynamics

occurring during postharvest ripening, as initially observed by thetime evolution plots. The physiological changes during this phaseheavily affect the general trait variability, initially supported by thedifferent PC loads.

60 F. Costa et al. / Postharvest Biology and Technology 69 (2012) 54–63

5 10 15 20 25

510

15

20

25

Max F

orc

e a

fter

2 m

onth

sMax For ce

1

2

3

4

5

6

7 8

9

10

11

12

13

14

15

16

17

18

19

20

21 22

23

24

25

26

27

28

29

30 3132

3334

3536

37

38

39

40

41

42

43

444546

47

48

49

50

51

5253 54

55

56

57

58

59

6061

62

63

64

65

66

67

68

69

70

71

7273

74

75

76

77

78

79

80

8182

83

0.5 1.0 1.5 2.0 2.5

0.5

1.0

1.5

2.0

2.5

1

2

34

5

67

8

9

10

1112

13

14

15

16

17

18

19

20

21 22

23

24

25

26

27

28

29

30

31

32

3334

35

36

3738

39

40

41

42

43

444546

47

4849

50

51

52

53

54

55

56

57

58

59

60

61

62

6364

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

8283

0.5 1.0 1.5 2.0 2.5

0.5

1.0

1.5

2.0

2.5

1

2

34

5

67

8

9

10

1112

13

14

15

16

17

18

19

20

21 22

23

24

25

26

27

28

29

30

31

32

3334

35

36

3738

39

40

41

42

43

444546

47

4849

50

51

52

53

54

55

56

57

58

59

60

61

62

6364

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

8283

5 10 15 20 25

510

15

20

25

Max F

orc

e a

fter

2 m

onth

sMax For ce

1

2

3

4

5

6

7 8

9

10

11

12

13

14

15

16

17

18

19

20

21 22

23

24

25

26

27

28

29

30 3132

3334

3536

37

38

39

40

41

42

43

444546

47

48

49

50

51

5253 54

55

56

57

58

59

6061

62

63

64

65

66

67

68

69

70

71

7273

74

75

76

77

78

79

80

8182

83

0.5 1.0 1.5 2.0 2.5

0.5

1.0

1.5

2.0

2.5

1

2

34

5

67

8

9

10

1112

13

14

15

16

17

18

19

20

21 22

23

24

25

26

27

28

29

30

31

32

3334

35

36

3738

39

40

41

42

43

444546

47

4849

50

51

52

53

54

55

56

57

58

59

60

61

62

6364

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

8283

0.5 1.0 1.5 2.0 2.5

0.5

1.0

1.5

2.0

2.5

1

2

34

5

67

8

9

10

1112

13

14

15

16

17

18

19

20

21 22

23

24

25

26

27

28

29

30

31

32

3334

35

36

3738

39

40

41

42

43

444546

47

4849

50

51

52

53

54

55

56

57

58

59

60

61

62

6364

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

8283

Young’s Modulus

Yo

un

g’s

Mo

du

lus a

fte

r 2

mo

nth

s

and Y

aPPwshp

prctvmAba

F

Max Force at harvestMax Force at harvest

Fig. 2. Time evolution plot of two mechanical parameters: maximum force

At harvest, the first two variables (PC1: 42% and PC2: 33%)ccounted together for almost the same variance explained by theC1 computed after two months of cold storage (PC1: 70% andC2: 12%). Besides this, from the eigenvalue plot (not shown) itas observed that while after two months the first two PCs were

ufficient to capture almost the entire texture variability (83%), atarvest a third component (14%) was required to reach a similarercentage of total variability expressed.

It is also interesting to compare the variable projections on therincipal components between the two stages. Both the PCA cor-esponding to harvest (Fig. 5a) and postharvest (Fig. 5b) display aonsistent grouping of the variable projections (in agreement withhe findings of Costa et al. (2011) for the two months posthar-est storage). Indeed, in both cases the variables were clustered inechanical, acoustic and derived force direction parameter groups.

t harvest (Fig. 5a) this latter group was not well characterized,eing projected mainly on the third PC which is not plotted, whilefter two months of cold storage it is more clearly clustered and

0 50 10 0 150

050

10

015

0

Number of Acoustic Peaks at harvest

Nu

mb

er

of A

co

ustic P

ea

ks a

fte

r 2

mo

nth

s

Number of Acoustic Peaks

1

2

3

45

6

78

9

10

11

12

13

14

15

16

1718

19

20

21

22

23

24

25

26

27

28

29

30 3132

33

34

3536

37

38

39

40

41

42

43

44

45

4647

48

49

50

51

52

53

54

55

56

5758

59

60

61

62

63

6465

66

67 68

69

7071

72

73

74

7576

77

78

79

80

81

82

83

0 50 10 0 150

050

10

015

0

Number of Acoustic Peaks at harvest

Nu

mb

er

of A

co

ustic P

ea

ks a

fte

r 2

mo

nth

s

Number of Acoustic Peaks

1

2

3

45

6

78

9

10

11

12

13

14

15

16

1718

19

20

21

22

23

24

25

26

27

28

29

30 3132

33

34

3536

37

38

39

40

41

42

43

44

45

4647

48

49

50

51

52

53

54

55

56

5758

59

60

61

62

63

6465

66

67 68

69

7071

72

73

74

7576

77

78

79

80

81

82

83

ig. 3. Time evolution plot of two acoustic parameters: number of acoustic peaks and me

Young’s Modulus at harvest

oung’s modulus. Cultivars are indicated with the code presented in Table 1.

projected towards an opposite direction with respect to the acous-tic parameters. This difference is related to the association betweenthe force direction parameters and the flesh compression orienta-tion. � force and force ratio, describe the type of mechanical profileprogression (and its relative level of magnitude). A mechanical pro-file with a decreasing trend means that the initial force is higherthan the final, due to the continuous rupture of the cell wall lay-ers upon compression, where the maximum resistance is exertedby the superficial tension of the sample. In the opposite case, anincreasing profile, the cell layers are instead compressed by theprobe, offering thus a higher mechanical resistance.

At harvest, the force direction can be related to the fruit fleshelasticity, and the close projection of these two parameters withthe Young’s modulus (arrows included in the force directioncircle but closer to the mechanical parameter group in Fig. 5a)

supports this hypothesis. After two months these two parameterswere strongly affected by the loss of humidity and internal turgorpressure, which can be quite severe in some cultivars, determining

40 42 44 46 48 50 52

40

42

44

46

48

50

52

Mean Acoustic Pressure at harvest

Me

an

Aco

ustic P

ressu

re a

fte

r 2

mo

nth

s

Mean Acoustic Pressure

1

2

3

4

5

67

8

9

10

11

12

13

14

15

16

17

1819

20

2122

23

24

25

26

27

28

29

30

31

32

33

34

3536

37

3839

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

40 42 44 46 48 50 52

40

42

44

46

48

50

52

Mean Acoustic Pressure at harvest

Me

an

Aco

ustic P

ressu

re a

fte

r 2

mo

nth

s

Mean Acoustic Pressure

1

2

3

4

5

67

8

9

10

11

12

13

14

15

16

17

1819

20

2122

23

24

25

26

27

28

29

30

31

32

33

34

3536

37

3839

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

an acoustic pressure. Cultivars are indicated with the code presented in Table 1.

F. Costa et al. / Postharvest Biology and Technology 69 (2012) 54–63 61

Fig. 4. Storage Index (SI) for the texture dynamics of twelve apple cultivars. X axis: texture parameters. Y axis: SI value. For each cultivar the colour of the histogram barsare: black – yield force, dark grey – max force, light grey – final force, white – force peaks, horizontal striped – area, vertical striped – force linear distance, left to right downstriped – Young’s modulus, left to right up striped – mean force, black dots – acoustic peaks, horizontal dashed – max acoustic pressure, vertical dashed – mean acousticpressure, white dots – acoustic linear distance.

-0.4 -0.2 0.0 0.2 0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

PC1

12

3

4

5

6

7

8

9

10

11

1213

14

15

16

17

1819

20

21

22

23

24

25

26

27

28

29

30

31

32

3334

35

36

37

38

39

40

41

42

43

4445

4647

48

49

50

51

5253

54

55

5657

58

59

6061

62

63

64

65

66

67

68

69

7071

72

73

74

75

76

77

78

79

80

81

82

83

-10 0 10

-10

-50

51

0

-0.4 -0.2 0.0 0.2 0.4

-0.2

-0.1

0.0

0.1

0.2

0.3

PC1

1

23

45

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22 2324

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

4647

48

49

5051

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

6768

69

70

71

72

73

7475

76

77

78

79

80

8182

83

-15 -10 -5 0 5 10 15

-50

51

0

a b

Force d irect ionMechani cal paramet ers

Acousti c paramet ers

Acousti c paramet ers

Mechani cal paramet ers

Force d irect ion

-0.4 -0.2 0.0 0.2 0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

PC1

PC

2

12

3

4

5

6

7

8

9

10

11

1213

14

15

16

17

1819

20

21

22

23

24

25

26

27

28

29

30

31

32

3334

35

36

37

38

39

40

41

42

43

4445

4647

48

49

50

51

5253

54

55

5657

58

59

6061

62

63

64

65

66

67

68

69

7071

72

73

74

75

76

77

78

79

80

81

82

83

-10 0 10

-10

-50

51

0

-0.4 -0.2 0.0 0.2 0.4

-0.2

-0.1

0.0

0.1

0.2

0.3

PC1

PC

2

1

23

45

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22 2324

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

4647

48

49

5051

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

6768

69

70

71

72

73

7475

76

77

78

79

80

8182

83

-15 -10 -5 0 5 10 15

-50

51

0

a b

Force d irect ionMechani cal paramet ers

Acousti c paramet ers

Acousti c paramet ers

Mechani cal paramet ers

Force d irect ion

Fig. 5. PCA plots illustrating the texture variability at harvest (a) and after two months of cold storage (b). For both plots the variable projections related to mechanical,acoustic and force direction parameters are shown.

6 ogy an

anoTamtpatdi

rictg

pmiacowpcta

4

tnatfiTtaa

am

cvatspcs

fctivrln

2 F. Costa et al. / Postharvest Biol

mealy texture. Mealiness is the contrasting phenotype to crisp-ess, and because of this the force direction is plotted with anpposite orientation with respect to the acoustic parameters.he fact that the mechanical and acoustic groups of variablesre well separated in both stages highlights the reliability of thisethodology in dissecting fruit texture complexity. In addition,

his result also confirmed that the mechanical and acoustic com-onents are two distinct phenomena, and both must be considerednd analyzed for a comprehensive characterization of apple fruitexture. The consistent clustering of the several parameters in twoistinct categories based on their nature (mechanical and acoustic)

s also consistent with the correlation matrix (Table 3).In the Supplementary Data Fig. S3 a PCA on the storage indexes is

eported to provide more details on the relative changes in mechan-cal and acoustic parameters between cultivars over two months ofold storage. The relative position of each cultivar is directly relatedo the results already presented in Figs. 2 and 3, and again, therouping of the mechanical and acoustic parameters is evident.

The different variability observed between harvest and theostharvest stages, possibly controlled by a distinct physiologicalechanism, was also observed by Costa et al. (2010) in a QTL exper-

ment based on a bi-parental population. Apple fruit are harvestedt a late physiological ripening stage oriented towards fruit senes-ence, where most of the transcription machinery has been turnedff. However, some important physiological pathways, such as cellall metabolism, may still remain active through the climacterichase. This late physiological process, together with the geneticonstitution of the elements involved in these pathways, can con-rol the great variation of fruit texture performance observed inpples.

. Conclusion

In this work, new methodology (TA-XTplus AED) developedo simultaneously profile combined mechanical and acoustic sig-atures was employed to investigate the texture dynamics over

two months postharvest storage of 83 apple cultivars. Fruitexture is the physiological result of a series of important modi-cations occurring in the cell wall/middle lamella during ripening.he coordinated action of the several degrading enzymes definehe different components on which texture is composed, andre grouped in two main categories, such as mechanical andcoustic.

The structural polysaccharide remodelling process, which startst the early stages of fruit development, continues throughout fruitaturation and ripening, towards the postharvest stage.For each cultivar the changes in parameters investigated indi-

ated a distinct dynamics, and it is worth noting that in somearieties the acoustic parameters tended to increase during stor-ge, confirming the importance of a detailed dissection of theextural properties. The precise description of fruit texture at twopecific times of fruit ripening (harvest and after two months ofostharvest) allowed investigation of the changes in each textureomponent, depicting a dynamics over the two months of coldtorage.

The storage index presented here can be considered as a use-ul indication of the storage potential of a representative set ofultivars. Postharvest management, in particular, can benefit fromhis index in order to adopt the best storage practice, by takingnto consideration the specific texture performance of each culti-

ar. Moreover, breeding programs can also take advantage of theseesults for selection of the most favourable parents to improve theong storability trait in new apple accessions, one of the most eco-omically relevant aspects for apple fruit quality.

d Technology 69 (2012) 54–63

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.postharvbio.2012.03.003.

References

Ballabio, D., Consonni, V., Costa, F., 2011. Relationship between apple texture andrheological parameters by means of multivariate analysis. Chem. Intell. Lab.Syst., doi:10.1016/j.chemolab.2011.11.002.

Billy, L., Mehinagic, E., Royer, G., Renard, C.M.G.C., Arvisenet, G., Prost, C., Jourjon,F., 2008. Relationship between texture and pectin composition of two applecultivars during storage. Postharvest Biol. Technol. 47, 315–324.

Bourne, M., 2002. Food Texture and Viscosity: Concept and measurement, seconded. San Diego, Academic Press.

Brummell, D.A., 2006. Cell wall disassembly in ripening fruit. Funct. Plant Biol. 33,103–119.

Brummell, D.A., Harpster, M.H., 2001. Cell wall metabolism in fruit softening andquality and its manipulation in transgenic plants. Plant Mol. Biol. 47, 311–340.

Costa, F., Cappellin, L., Longhi, S., Guerra, W., Magnano, P., Porro, D., Soukoulis,C., Salvi, S., Velasco, R., Biasioli, F., Gasperi, F., 2011. Assessment of apple(Malus × domestica Borkh.) fruit texture by a combined acoustic-mechanicalprofiling strategy. Postharvest Biol. Technol. 6, 21–28.

Costa, F., Peace, C.P., Stella, S., Serra, S., Musacchi, S., Bazzani, M., Sansavini, S., Vande Weg, W.E., 2010. QTL dynamics for fruit firmness and softening aroundan ethylene-dependent polygalacturonase gene in apple (Malus × domesticaBorkh.). J. Exp. Bot. 61, 3029–3039.

De Belie, N., De Smedt, V., De Baerdemaeker, J., 2000. Principal component analysisof chewing sounds to detect differences in apple crispness. Postharvest Biol.Technol. 18, 109–119.

Drazeta, L., Lang, A., Hall, A.J., Volz, R.V., Jameson, P.E., 2004. Air volume measure-ments of ‘Braeburn’ apple fruit. J. Exp. Bot. 399, 1061–1069.

Duizer, L., 2001. A review of acoustic research for studying the sensory perceptionof crisp crunchy and crackly textures. Trends Food Sci. Technol. 12, 17–24.

Fillion, L., Kilcast, D., 2002. Consumer perception of crispness and crunchiness infruits and vegetables. Food Qual. Prefer. 13, 23–29.

Hampson, C.R., Quamme, H.A., Hall, J.W., MacDonald, R.A., King, M.C., Cliff, M.A.,2000. Sensory evaluation as a selection tool in apple breeding. Euphytica 111,79–90.

Harker, F.R., Hallett, I.C., 1992. Physiological changes associated with developmentof mealiness of apple fruit cool storage. HortScience 27, 1291–1294.

Harker, F.R., Gunson, F.A., Jaeger, S.R., 2003. The case of fruit quality: an interpreta-tive review of consumer attitudes, and preferences for apples. Postharvest Biol.Technol. 28, 333–347.

Harker, F.R., White, A., Gunson, F.A., Hallett, I.C., De Silva, H.N., 2006. Instrumentalmeasurement of apple texture: a comparison of the single-edge notched testand the penetrometer. Postharvest Biol. Technol. 39, 185–192.

Ioannides, Y., Howarth, M.S., Raithatha, C., Defernez, M., Kemsley, E.K., Smith, A.C.,2007. Texture analysis of Red Delicious fruit: towards multiple measurementsof individual fruit. Food Qual. Prefer. 18, 825–833.

Johnston, J.W., Hewett, E.W., Banks, N.H., Harker, F.R., Hertog, M.L.A.T.M., 2001. Phys-ical changes in apple texture with fruit temperature: effects of cultivar and timein storage. Postharvest Biol. Technol. 23, 13–21.

Kilcast, D., 2004. Texture in food. Solid Foods, vol. 2. Cambridge, Woodhead Publish-ing Limited.

Konopacka, D., Plocharski, W.J., 2004. Effect of storage conditions on the relationshipbetween apple firmness and texture acceptability. Postharvest Biol. Technol. 32,205–211.

Kühn, B.F., Thybo, A.K., 2001. Sensory quality of scab-resistant apple cultivars.Postharvest Biol. Technol. 23, 41–50.

Longhi, S., Moretto, M., Viola, R., Velasco, R., Costa, F., 2011. ComprehensiveQTL mapping survey dissects the complex fruit texture physiology in apple(Malus × domestica Borkh.). J. Exp. Bot. 63 (3), 1107–1121.

Mehinagic, E., Royer, G., Symoneaux, R., Bertrand, D., Jourjon, F., 2004. Predictionof the sensory quality of apple by physical measurements. Postharvest Biol.Technol. 34, 257–269.

Perring, M.A., Pearson, K., 1988. Physical changes in stored apple. J. Sci. Food Agric.44, 193–200.

Rose, J.K.C., Catalá, C., Gonzalez-Carranza Zinnia, H., Roberts, J., 2003. Plant cell walldisassembly. In: The Plant Cell Wall. Annual Plant Reviews Series. BlackwellPublishing, pp. 264–324.

Roudaut, G., Dacremont, C., Pamies, B.V., Colas, B., Meste, M.L., 2002. Crispness: acritical review on sensory and material science approaches. Food Sci. Technol.13, 217–227.

Saladie, M., Matas, A.J., Isaacson, T., Jenks, M.A., Goodwin, S.M., Niklas, K.J., Ren, X.L.,Labavitch, J.M., Shackel, K.A., Fernie, A.R., Lytovchenko, A., O’Neill, M.A., Watkins,C.B., Rose, J.K.C., 2007. A re-evaluation of the key factors that influence tomato

fruit softening and integrity. Plant Physiol. 144, 1012–1028.

Smedt, V., De Pauwels, E., Baerdemaeker, J., De Nicolaï, B., 1998. Microscopic obser-vation of mealiness in apples: a quantitative approach. Postharvest Biol. Technol.14, 151–158.

Szczesniak, A.S., 2002. Texture is a sensory propriety. Food Qual. Prefer. 13, 215–225.

ogy an

T

T

T

V

Wei, J., Ma, F., Shi, S., Qi, X., Zhu, X., Yuan, J., 2010. Changes and postharvest regulationof activity and gene expression of enzymes related to cell wall degradation in

F. Costa et al. / Postharvest Biol

aniwaki, M., Hanada, T., Sakurai, N., 2006. Device for acoustic measure-ment of food texture using a piezoelectric sensor. Food Res. Int. 39,1099–1105.

aniwaki, M., Sakurai, N., 2008. Texture measurement of cabbage using an acousticalvibration method. Postharvest Biol. Technol. 50, 176–181.

homas, T.R., Shackel, K.A., Matthews, M.A., 2008. Mesocarp cell turgor in Vitis

vinifera L. berries throughout development and its relation to firmness growth,and the onset of ripening. Planta 228, 1067–1076.

arela, P., Chen, J., Fiszman, S., Povey, M.J.W., 2006. Crispness assessment of roastedalmonds by an integrated approach to texture description: texture, acoustic,sensory and structure. J. Chemometrics 20, 311–320.

d Technology 69 (2012) 54–63 63

Vincent, J.F.V., 1998. The quantification of crispness. J. Sci. Food Agric. 78, 162–168.Wakasa, Y., Kudo, H., Ishikawa, R., Akada, S., Senda, M., Niizeki, M., Harada, T., 2006.

Low expression of an endopolygalacturonase gene in apple fruit with long-termstorage potential. Postharvest Biol. Technol. 39, 193–198.

ripening apple fruit. Postharvest Biol. Techol. 56, 147–154.Zdunek, A., Konopacka, D., Jesionkowska, K., 2010. Crispness and crunchiness judg-

ment of apple based on contact acoustic emission. J. Text. Stud. 41, 75–91.