Available online at www.sciencedirect.

com

Postharvest Biology and Technology 47 (2008) 315–324

Relationship between texture and pectin composition

of two apple cultivars during storage
Ludivine Billy a,b , Emira Mehinagic a,∗ , Gaëlle Royer a , Catherine M.G.C. Renard c ,
Gaëlle Arvisenet b , Carole Prost b , Frédérique Jourjon a
a
Ecole Supérieure d’Agriculture, Groupe de Recherche en Agro-industrie sur les Produits et les Procédés (GRAPPE),
55 rue Rabelais, B.P. 30 748, 49007 Angers Cedex 01, France
b ENITIAA, UMR CNRS GEPEA 6144, Laboratoire de Biochimie, rue de la Géraudière, B.P. 82 225, 44322 Nantes Cedex 03, France
c UMR A 408 INRA, Université d’Avignon, SQPOV, Sécurité et Qualité des Produits d’Origine Végétale,

Domaine St. Paul, 84 914 Avignon Cedex 09, France

Received 3 April 2007; accepted 13 July 2007

Abstract
The texture of two apple cultivars was characterised by sensory and instrumental methods for five different storage periods. The aim of this
study was to explain the changes in apple texture during storage by different physical (penetrometry, compression) and chemical measurements
(extraction and analysis of pectin composition). The emphasis was on determining the most relevant biochemical markers in relation to different
sensory properties of apple texture.
Contrary to ‘Fuji’, ‘Golden Delicious’ fruit softened easily during storage, became mealy and had higher neutral sugar concentrations in their
alcohol-insoluble residues (AIR) and more galacturonic acid in the water-soluble pectin extracts (WSP). The most relevant biochemical marker
linked to texture change was the galacturonic acid content in the water-soluble pectin extracts. High and positive correlation coefficients were
observed between sensory mealiness (R = 0.84) and galacturonic acid content in the WSP while, sensory crunchiness and instrumentally measured
firmness were negatively correlated with this component. The total neutral sugar content in the alcohol-insoluble residues and in the water-soluble
pectin fractions also changed with apple texture properties.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Apple; Malus domestica Borkh; Texture; Cell wall; Mechanical properties; Sensory analysis

1. Introduction fruit. A better understanding of the relationship between fruit

Apple texture is one of the most important quality proper-
ties that influence consumer acceptability of these fruit (Stow,
1995). Thus, postharvest softening of apples during storage and
market distribution is a serious commercial problem resulting
in quality losses for growers and distributors. Therefore, our
research aims to explain the development of apple softening
during storage using different physical (penetrometry, compres-
sion) and chemical measurements (extraction and analysis of
pectin composition).
Knowledge of the mechanism of fruit tissue softening, cell-
wall structure and cell-wall breakdown is very important for
understanding and improving the texture and quality of apple
∗ Corresponding author. Tel.: +33 2 41 23 55 55; fax: +33 2 41 23 55 65.
E-mail address: e.mehinagic@groupe-esa.com (E. Mehinagic).
texture and biochemical composition could lead to improve-
ments in quality control and process design in the food industry
and the marketplace. Many studies have already addressed either
the changes in apple texture during storage (Abbott et al., 1984;
Grotte et al., 2001; Johnston et al., 2001; Mehinagic et al., 2004;
Varela et al., 2007) or changes in pectin composition during
softening (Yoshioka et al., 1994; Massiot et al., 1996; Nara et
al., 2001), but few studies tried to connect these two phenom-
ena. Moreover, for the majority of these, sensory perception of
changes in fruit texture was never carried out. Ben and Gaweda
(1985) showed high correlation coefficients between the content
of protopectin (non-cellulosic neutral sugar residues) and the
firmness of ‘Jonathan’ apples in one storage season only. A study
performed on ‘Golden Delicious’ apples attempted to determine
the possible contribution of changes in cell-wall composition to
fruit softening under various storage conditions (Siddiqui et al.,
1996). More recently, Pena and Carpita (2004) reported a study

0925-5214/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.postharvbio.2007.07.011
316 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324
of the chronology of the biochemical events associated with loss
of firm texture and cell separation, and Lo Scalzo et al. (2005)
described the results of research on pectin composition just after
reddening and after cold storage of reddened ‘Annurca’ apple
fruit in different atmospheres.
Loss of firm texture and cell adhesion are processes asso-
ciated with ripening that affect fruit quality and postharvest
storage. The cell walls of fruit have received considerable atten-
tion, as changes occurring in their structure and composition are
clearly determining in fruit firmness (Knee, 1973a,b; Bartley,
1976; De Vries et al., 1981; Renard et al., 1990b, 1991a,b,c,d;
Renard and Thibault, 1993; Massiot et al., 1994). Indeed, it is
well known that separation and degradation of water-soluble
pectic substances, erosion of the middle lamella and disintegra-
tion of the primary cell wall lead to softening of fruit flesh. Pectic
substances are the most abundant class of macromolecules
within the cell-wall matrix and are the main components of the
middle lamellae. They have a role in the adhesion between cells,
and in regulation of intercellular adhesion. Consequently, pectic
substances have been assigned an important role in changes in
texture of fruit tissues (Yoshioka et al., 1992; Grant Reid, 1997)
and notably, changes in pectin composition have been impli-
cated in decreased adhesion between cells (Ilker and Szczesniak,
1990). During fruit softening, pectins typically undergo solubil-
isation and depolymerisation, which are thought to contribute
to wall loosening and disintegration. In apple, there is usu-
ally an increase in water-soluble pectin (WSP), and a decrease
in galactose and arabinose residues (Knee, 1973b), with little
depolymerisation occurring in any pectin fraction during ripen-
ing (Yoshioka et al., 1992; Fischer and Amado, 1994; Fischer
et al., 1994). However, the association of biochemical changes
with changes in fruit texture is still unclear; for example, it is
still not known if the processes of pectin solubilisation and loss
of galactose are causal, coincidental, or a consequence of apple
softening (Redgwell et al., 1997).
We aimed to study the relationship between texture and pectin
composition of stored apples. Two cultivars with different tex-
tural characteristics were studied in order to investigate whether
the differences in texture between different apple cultivars were
due to differences in their pectin composition. ‘Fuji’ is a crunchy
and juicy apple that stores well and ‘Golden Delicious’ apples
become mealy during storage (Mehinagic et al., 2004). The
apples were stored for up to 7 months at 2 ◦ C. Their texture
was assessed using two deformation tests (penetrometry and
compression) and sensory analysis. At the same time, cell-wall
content and pectin extractability of these fruit were characterised
in order to try to relate texture modifications and sensory per-
ceptions to different chemical changes occurring in apple cell
walls.
2. Materials and methods
2.1. Fruit
Two different apple cultivars (Malus domestica Borkh),
‘Fuji’ (FU) and ‘Golden Delicious’ (GO), were studied. One
hundred and forty fruit of each cultivar harvested in 2004
at commercial maturity were provided by a fruit-bearing test
center (“La Morinière”, Saint Epain, France). Their ripening
stage was verified by starch regression measurements. These
fruit were selected on the basis of uniformity and absence of
damage or blemishes. They were all stored in the same cool
room at 2 ◦ C and under ambient atmospheric pressure and
humidity for five different periods (1, 2, 3, 4 and 7 months).
The choice of these storage periods was based on a previ-
ous study which showed that they resulted in a large range
of apple mechanical properties (Mehinagic et al., 2004). For
each storage period, 28 fruit per cultivar were brought up to
room temperature 24 h before being analysed. Half of each
fruit (n = 28) was analysed by sensory panel and the second
half by penetrometry (n = 14) or compression (n = 14). Sensory
and mechanical methods were carried out simultaneously and
immediately after the apples were cut. For penetrometry and
compression tests, apples were manually and carefully peeled by
the same operator using the same vegetable knife. Fruit analysed
by penetrometry were then sampled (taking care to remove the
crushed part before sampling) and freeze-dried prior to chemical
analysis.
2.2. Sensory evaluation
The sensory panel included 14 permanent trained panellists
from the staff of the Ecole Supérieure d’Agriculture (ESA,
Angers, France). These panellists, selected in 1999, are trained
every week to evaluate apple texture according to the recom-
mendations of AFNOR (1995) and of Fortin and Desplancke
(1998). Their performances are evaluated once a year to ensure
reliability. The sensory attributes studied on apple flesh were
crunchiness, juiciness, mealiness, chewiness and fondant. These
sensory attributes are defined and described in Table 1. During
a sensory session, each panellist peeled and analysed two dif-
ferent half fruit from each cultivar. The washed unpeeled half
apples, coded by a random three-digit code, were randomly pre-
sented to the panellists, at room temperature and under red light
illumination to avoid apple cultivar identification. A 10 cm line
scale was used for evaluation. The left end of the scale corre-
sponded to the lowest intensity (value 0) and the right end to
the highest (value 10). Each panellist used mineral water as a
rinse between sample analyses. Five sessions, corresponding to
five storage periods, were organised. In this way, each panellist
analysed 2 cultivars × 2 half apples during each session. Thus,
Table 1
Texture sensory descriptors used for apples
Texture attribute Definition (Mehinagic et al., 2003)
Crunchiness Force required for the first bite and the noise resulting
from this bite
Juiciness Amount of liquid released on mastication just before
swallowing
Mealiness Dry and crumbly texture
Chewiness Time and number of chewing movements needed to
grind the sample prior to swallowing
Fondant Force required to crush a piece of apple between the
tongue and palate
 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324
Fig. 1. The penetrometric curve obtained on peeled half apples, using the MTS
(Synergie 200H) traction machine (cylindrical probe with a 4 mm diameter con-
vex tip; penetration speed of 50 mm min−1 ; depth of 10 mm) (FLC: flesh limit
compression force; S: slope of the force-deformation curve; D: deformation asso-
ciated with flesh limit compression force; WFLC : work associated with FLC; W7 :
work required to attain a flesh deformation of 7 mm).
for each cultivar and each storage period, 28 fruit samples were
analysed.
2.3. Penetrometry
According to Mehinagic et al. (2004), a cylindrical probe with
a 4 mm diameter convex tip was used to perforate peeled half
apples in an MTS (Synergie 2000H) traction machine. Peeled
half apples were placed with the flat side facing down on the
plate and punched on their rounded side in an equatorial position.
The equipment used detected the sample automatically; pene-
tration speed was set at 50 mm min−1 , and the test was stopped
after penetration to 10 mm. This distance was short enough to
avoid any influence of the core in the measurement. Fifty points
were recorded during a penetrometric test. Penetrometry curves
were analysed and five parameters were studied (Fig. 1): flesh
limit compression force (FLC), deformation associated with
flesh limit compression force (D), slope of the force-deformation
curve (S), work associated with FLC (WFLC ) and work required
to attain a flesh deformation of 7 mm (W7 ) (Duprat et al., 2000).
Definitions of these parameters are given in Table 2. For each
cultivar and each storage period, 14 fruit samples were analysed.
2.4. Compression
According to Mehinagic et al. (2004), two parallel plates
(50 mm of diameter and 7 mm of thickness) were used to com-
press peeled half apples. These were placed with the flat side
facing down on the plate, with the equatorial section of their
rounded side targeted in the same MTS (Synergie 2000H) trac-
tion machine. The equipment detected the diameter and height
of the sample automatically: the height of peeled half apples was
Table 2
Definitions of penetrometric parameters
Parameters Calculation
FLC (N) Maximal force before the drop in force at the moment of penetration
D (mm) Deformation associated with FLC
S (N mm−1 ) Gradient on the curve (between 0 and FLC)
WFLC (N mm) Area under the curve between 0 and D
W7 (N mm) Area under the curve between 0 and 7 mm
317
Fig. 2. Double compression force/time curve obtained on peeled half
apples, using an MTS (Synergie 200H) traction machine (2 parallel plates;
50 mm min−1 ; 7 mm) (1: the first compression; 2: the second compression; H:
maximal force; WH : surface area under the compression curve; S: slope of the
compression).
compressed by 20% and the loading rate of the crosshead was
50 mm min−1 . Two successive compressions were carried out on
each sample. Compression curves were analysed and six param-
eters were studied (Fig. 2): maximal force associated with the
first compression, usually described as hardness (H1 ), maximal
force associated with the second compression (H2 ), surface area
under the first compression curve (WH1 ), surface area under the
second compression curve (WH2 ), slope of the first compression
(S1 ) and slope of the second compression (S2 ). For each cultivar
and each storage period, 14 fruit samples were analysed.
2.5. Apple powder preparation
Peeled apple pieces were freeze-dried, then frozen with liquid
N2 and immediately homogenised in a commercial blender for
1 min. The apple powders obtained were prepared in triplicate
for each batch (2 cultivars × 5 storage durations) and stored in
a desiccator under vacuum prior to chemical analysis.
2.6. Alcohol-insoluble residue (AIR) preparation
Cell walls were prepared as alcohol-insoluble residues for
each sample triplicate. According to Renard (2005), the apple
powder (7 g) was blended with 50 mL of 70% ethanol in an
empty 75 mL Sep-pack prep column (InterchimTM ) equipped
with a sinter, porosity 20 ␮m. The mixture was stirred at room
temperature for 20 min then filtered under vacuum. The residue
was re-extracted with 70 mL of 70% ethanol eight times, as
described above. At the end of these washings, the filtrates were
sugar-free (absence of sugars was tested by the phenol-sulphuric
Definitions (Duprat et al., 2000; Grotte et al., 2001)
Force representing the limit of flesh elasticity
Gradient measuring penetrometric firmness
Work required for rupture of the flesh
Work required to attain a flesh deformation of 7 mm
318 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324

method of Dubois et al., 1956). The alcohol-insoluble residues
were dried by solvent exchange with 96% ethanol (three times),
and acetone (three times), and finally overnight in an oven at
40 ◦ C.
2.7. Water-soluble pectin extraction
From each AIR preparation, water-soluble pectin was
extracted by water. For each extraction, approximately 100 mg
of cell-wall preparation and 6 mL of water (Renard, 2005)
were incubated for 2 h at room temperature in an empty 8 mL
Sep-pack prep column (Interchim TM ) equipped with a sinter,
porosity 20 ␮m, under slow planetary agitation. After incuba-
tion, the extract and cell walls were separated by filtration under
vacuum. This extraction was repeated three times for each AIR
preparation; the three water extracts were pooled and frozen
before analysis.
2.8. Analytical methods
France). The analysis of variance (ANOVA) was carried out
independently for each cultivar and for each of the studied vari-
ables measured by sensory analysis, penetrometry, compression
and pectin analysis. Two-way ANOVA was performed to deter-
mine significant differences in sensory quality between apples.
The factors studied were the panellist effect and the storage
period. An interaction of these two factors was evaluated for
each cultivar and each sensory attribute to make sure that these
main effects were not confused. One-way ANOVA was carried
out on instrumental measurements (penetrometry, compression
and pectin analysis) to test the effects of storage duration on
each cultivar. For each analysis, a significant level of 5% was
used. To compare the effect of 5 storage periods on the dif-
ferent attributes of the two cultivars, Fischer’s least significant
difference (LSD) procedures were applied separately to each cul-
tivar. Principal component analysis (PCA) was carried out on the
averaged data. Pearson’s correlation coefficients between texture
parameters and biochemical data were calculated between the
average values.

Galacturonic acid content (GA) was determined by auto-
mated m-hydroxydiphenyl assays on an Alliance Instruments
(Méry sur Oise, France) autoanalyser after sulphuric acid
prehydrolysis and hydrolysis (26N sulphuric acid, 1 h, room
temperature followed by 2N, 3 h, 100 ◦ C) for the alcohol-
insoluble residues (Saeman et al., 1954), or saponification (0.1N
NaOH, 30 min, room temperature) for water-soluble pectin
extracts (Thibault, 1979). The results are expressed as anhy-
drogalacturonic acid. Total neutral sugar content (NS) was
determined by the orcinol method (Tollier and Robin, 1979) on
the same hydrolysates and saponified pectin solutions. Methanol
was determined according to Klavons and Bennett (1986). The
degree of methylation (DM) was calculated as the molar ratio of
methanol to galacturonic acid. The degree of acetylation (DA)
was determined by high performance liquid chromatography
after saponification according to the method of Voragen et al.
(1986).
2.9. Statistical analysis
Data acquisition and statistical treatment were performed
with Statgraphics Plus 5.1 software (Sigmaplus, Toulouse,
3. Results and discussion
3.1. Effect of storage duration on sensory attributes and
mechanical parameters of fruit texture
Variance analysis was carried out independently for each of
the five sensory descriptors studied. For all sensory descriptors,
no interactions were found between panellist effect and storage
period for ‘Fuji’ apples while two significant interactions were
found for ‘Golden Delicious’. These two interactions involv-
ing mealiness and fondant can be explained by heterogeneity
within a fruit batch or differences in sensitivity between pan-
ellists. For ‘Golden Delicious’ apples, the storage effect was
significant for all descriptors except chewiness. Texture change
in ‘Fuji’ was more limited and the storage effect was significant
for three descriptors at the 5% level: mealiness, juiciness and
fondant. For both cultivars, the intensity of juiciness decreased
while the intensity of mealiness increased (Table 3). These
results confirmed those of Mehinagic et al. (2004). According
to Holt and Schoorl (1984), these changes are related to the
deterioration of the mechanical strength of apple tissue during
storage.

Table 3
Evolution of sensory quality of 2 apple cultivars during storage
Storage (months)/cultivar Crunchiness Juiciness Mealiness Chewiness Fondant

1/Fuji 7.6 6.9ab 1.8b 5.7 2.9ab

2/Fuji 7.9 6.2c 2.0ab 5.8 2.4b
3/Fuji 7.6 7.0a 1.7b 5.3 3.1a
4/Fuji 7.5 6.5bc 1.8b 5.3 3.5a
7/Fuji 7.5 6.2c 2.3a 5.3 3.5a
1/Golden 4.5ab 4.9a 3.9b 4.2 5.5c
2/Golden 4.6ab 4.3b 3.7b 4.1 5.4c
3/Golden 4.2bc 4.5ab 4.0b 3.8 6.6ab
4/Golden 3.7c 3.9b 4.8a 3.6 6.8a
7/Golden 4.9a 3.9b 4.8a 4.2 6.1bc

Different letters (a–c) in a column within a cultivar denote values that are significantly different at P < 0.05.
 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324 319

Fig. 3. Slope of the force-deformation curve (S), obtained by penetrometry (a) and maximal force associated with the second compression (H2 ) (b), of ‘Fuji’ (grey
columns) and ‘Golden Delicious’ apples (white columns) after different periods of storage (identical letters on columns for the same cultivar indicate that there is no
significant difference with P > 0.05).

It appeared that storage had a significant effect for all penetro-
metric parameters and for both cultivars. For ‘Golden Delicious’
apples, the effect of storage was also significant for all compres-
sion parameters while ‘Fuji’ apples seem more resistant. This
could be explained by the better texture conservation of ‘Fuji’
apples. Indeed, Jobling and McGlasson (1995) and Curry (1989)
showed that ‘Fuji’ apples maintain their firm texture during
storage.
As all penetrometric parameters were strongly correlated,
only the changes in the slope of the force-deformation curve
(S), corresponding to firmness, are represented in Fig. 3a. In
this figure, it can be seen that fruit firmness decreased sig-
nificantly between the second and third month of storage for
‘Fuji’ apples. For ‘Golden Delicious’ fruit, this parameter did
not change throughout the 7 months storage period. In Fig. 3b,
the maximal force (describing hardness) associated with the sec-
ond compression is shown for each storage step. The hardness
of ‘Fuji’ did not change at all, while ‘Golden Delicious’ apples
hardness decreased between the first and second month of stor-
age. Instrumental measurements and sensory analysis showed
a good concordance, with a general decrease in fruit firmness
during storage and better texture conservation of ‘Fuji’ apples.
Moreover, both methods showed that ‘Fuji’ apples were firmer
than ‘Golden Delicious’. These results were in accordance with
previous studies. They confirmed (Grotte et al., 2001; Mehinagic
et al., 2004) that ‘Fuji’ and ‘Golden Delicious’ apples present
mechanical and sensory textural differences. For the temporal
aspects, Mehinagic et al. (2004) found a significant change in
firmness between apples stored for 3 weeks and those stored for
20 or 30 weeks but they found no significant difference between
20 and 30 weeks of storage. Indeed, according to Johnston et
al. (2001), the decrease in firmness for many apple cultivars can
be characterised by a curve consisting of three distinct phases.
Firstly, fruit soften slowly, then more rapidly, and finally slowly
again.
3.2. Effect of storage duration on apple pectin extractability
For ‘Fuji’ apples, the storage effect was significant for five
biochemical parameters: the total neutral sugar content and the
degree of methylation in the AIR, the composition of the WSP
extracts (primarily galacturonic acid, less total neutral sugars)
and its degree of methylation. For ‘Golden Delicious’ apples,
three parameters changed significantly during storage: the galac-
turonic acid content and the total neutral sugar content in the AIR
and the degree of methylation in the WSP fractions.

Table 4
Composition of different apple pectin extracts during storage
Alcohol-insoluble residues composition Water-soluble pectin composition
Storage (months)/cultivar GA (mg/g) DM (%) DA (%) NS (mg/g) GA (mg/g) DM (%) NS (mg/g)

1/Fuji 292 75a 5 694a 8b 52b 29a

2/Fuji 270 73a 5 666ab 7b 70a 13b
3/Fuji 276 75a 5 629b 10b 75a 17b
4/Fuji 290 72ab 5 582c 23a 56b 16b
7/Fuji 286 69b 5 567c 23a 75a 15b
1/Golden 252b 75 6 716a 36 53c 18
2/Golden 276a 75 6 676b 35 62b 19
3/Golden 290a 74 5 676b 35 66b 16
4/Golden 275a 73 6 622c 29 81a 15
7/Golden 285a 74 5 668b 31 80a 16

Different letters in a column (a–c) within a cultivar denote values that are significantly different at P < 0.05. GA: Galacturonic acid; DM: degree of methylation; DA:
degree of acetylation; NS: neutral sugar.
320 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324

The values obtained (Table 4) were close to those in the
literature. Renard et al. (1990a) and Renard (2005) found
that the galacturonic acid content of alcohol-insoluble residues
of ‘Golden Delicious’ apples is generally between 250 and
277 mg/g AIR. In the same way, the values of galacturonic acid
content of water-soluble pectin extracts agreed with those of
Fischer et al. (1994). While the total neutral sugar contents in
the WSP fractions were in accordance with those in the litera-
ture (Massiot et al., 1996), the total neutral sugar content in the
AIR of each cultivar seemed to be higher (Fischer and Amado,
1994). This can be explained by the different methods used:
these authors analysed the individual neutral sugar by gas chro-
matography and we analysed it by the orcinol method, which is
known to over-estimate the results, the calibration being done
with a hexose. The values for the degree of methylation of the
pectic substances of AIR (69–75%) were totally in accordance
with those reported by De Vries et al. (1981) (80%) and Renard
et al. (1990b) (72%), just as for the values for the degree of
acetylation (<10%) (Voragen et al., 1986). In the same way, the
degree of methylation of the pectic substances of WSP agreed
with previous studies (Renard et al., 1990a; Fischer et al., 1994;
Massiot et al., 1996).
In the AIR (Table 4), except for a significant increase between
the first and second month of storage for ‘Golden Delicious’
apples, no trend of an increase or a decrease in the galactur-
onic acid content during fruit storage was noticeable for both
cultivars. This observation was also valid for the degree of
methylation of apples even if a significant decrease in this one
was observed at the end of ‘Fuji’ apple storage. Moreover, no
changes were observed for the degree of acetylation. All these
results were in agreement with those in literature (Fischer and
Amado, 1994). However, the total neutral sugar content signif-
icantly decreased during apple storage. For both cultivars, this
decrease was more marked between the third and fourth month
of storage and between the first and second month of storage
for ‘Golden Delicious’. These results were in accordance with
Fischer and Amado (1994). According to Yoshioka et al. (1994),
this decrease is more specifically due to a decline in galactose
and arabinose contents.
In the WSP fractions (Table 4), for ‘Fuji’ apples, the content
of galacturonic acid increased significantly between the third and
fourth month of storage (Yoshioka et al., 1994; Massiot et al.,
1996; Lo Scalzo et al., 2005). This phenomenon, correspond-
ing to pectin solubilisation, might result from the enzymatic
cleavage of linkages between pectin and other cell-wall compo-
nents rather than from the degradation of pectin chains. Contrary
to the literature (Siddiqui et al., 1996; Nara et al., 2001) there
was no significant change in ‘Golden Delicious’. The degree of
methylation increased along with storage period. This increase
was particularly significant between the first and second month
of storage for both cultivars, and between the third and fourth
month of storage for ‘Golden Delicious’ apples. These results
confirm previous observations (Massiot et al., 1996) concerning
the biosynthesis of highly methylated pectin during apple stor-
age. The content of total neutral sugars decreased during storage
for ‘Fuji’ apples (Massiot et al., 1996; Nara et al., 2001). This
change was particularly significant between the first and second
month of storage. As previously shown by Massiot et al. (1996),
this decrease more particularly involves arabinose and galactose
contents. Yoshioka et al. (1994) suggested that these two types
of polyuronide may be lost from the side-chain polysaccharides
of the polyuronide molecules and may be involved in the solu-
bilisation of polyuronides. Contrary to ‘Fuji’ apples, this change
was not significant for ‘Golden Delicious’, as shown by Nara et
al. (2001).
To summarise, the total neutral sugar content in the AIR, the
composition of the WSP extracts (galacturonic acid and total
neutral sugars) and its degree of methylation showed the most
obvious changes during apple storage. Moreover, we showed
that the total neutral sugar content in the AIR and its degree of
acetylation, and the composition of the WSP fractions (galac-
turonic acid, total neutral sugars) of ‘Golden Delicious’ apples
were significantly different from those of ‘Fuji’. Consequently,
these five biochemical parameters could explain the mechanical
and sensory textural differences observed previously between
these two cultivars and storage periods.
3.3. Relationships between apple mechanical properties,
texture sensory attributes and biochemical composition of
cell-wall pectins
A PCA analysis was carried out on sensory, penetrometry,
compression and biochemical parameters. The first dimension
(PC1) described 66.5% of the variation amongst samples, and
the second (PC2) 15% (Fig. 4). Sensory parameters juiciness,
crunchiness and chewiness were highly related to the mechanical
parameters (firmness, hardness, slope of the force-deformation
curve) and negatively linked to mealiness and fondant. These two
Fig. 4. PCA plot of the two cultivars (‘Fuji’ and ‘Golden Delicious’) analysed
at five storage periods by five sensory attributes, four parameters measured by
penetrometry, six parameters measured by compression and seven biochemical
parameters in the plane defined by the first two principal components (GOX :
‘Golden Delicious’ apples stored for X months; FUX : ‘Fuji’ apples stored for X
months).
 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324 321

Table 5
Pearson’s correlation coefficients between instrumental attributes measured by mechanical parameters, sensory descriptors for texture, and biochemical parameters
for both cultivars (coefficients inferior to 0.6 were omitted)
Biochemical parameters Mechanical parameters Texture sensory descriptors

S H1 H2 S1 S2 Crunchiness Juiciness Mealiness Chewiness Fondant

GA WSP −0.94 −0.85 −0.85 −0.88 −0.78 −0.89 −0.86 0.84 −0.90 0.89
DA AIR −0.68 −0.64 −0.63 −0.63 −0.80 −0.76 0.81 −0.75 0.73

GA: Galacturonic acid; WSP: water-soluble pectin; DA: degree of acetylation; AIR: alcohol-insoluble residue.

Fig. 5. Changes in the galacturonic acid content in the water-soluble pectin extracts (—, GA WSP) and the maximal force associated with the second compression
(- -, H2 ) (a) or (- -) fondant (b) of ‘Fuji’ apples during storage.

attributes were related to the content of galacturonic acid in the
WSP extracts. Highly negative Pearson’s correlation coefficients
were found between the content of galacturonic acid in the WSP
extracts, mechanical parameters and fruit juiciness, crunchiness
and chewiness (Table 5). Indeed, the content of galacturonic
acid in water-soluble pectin extracts of ‘Fuji’ apples increased
at the same time as the hardness (H2 ) decreased or as the fondant
increased (Fig. 5). In the same way, correct Pearson’s correlation
coefficients were obtained between the degrees of acetylation in
the AIR and fruit texture parameters.
Peasron’s correlation coefficients between mechanical and
sensory parameters of texture and the total neutral sugar content
in the AIR and in the WSP extracts were not high enough to be
significant. However, for both cultivars, the total neutral sugar
content in the AIR seemed to follow the same profile as the
firmness (Fig. 6a), confirming Ben and Gaweda’s (1985) results.

Fig. 6. Changes in the total neutral sugar content in the alcohol-insoluble residue (—, NS AIR) (a) and in the water-soluble pectin extracts (—, NS WSP) (b) and the
slope of the force-deformation curve (- -, S), obtained by penetrometry, of ‘Fuji’ () and ‘Golden Delicious’ (䊉) apples during storage.
322 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324
Fig. 7. Changes in the total neutral sugar content in the alcohol-insoluble residue (—, NS AIR) and the (- -) fondant of ‘Fuji’ apples (a) or the (- -) juiciness of
‘Golden Delicious’ apples (b) during storage.
Moreover, this biochemical parameter changed in the same way
as juiciness of ‘Golden Delicious’ apples and in the opposite way
to fondant of ‘Fuji’ (Fig. 7). As Yoshioka et al. (1994) showed
during softening of ‘Red gold’ apples, the firmness of ‘Fuji’
and ‘Golden Delicious’ also changed according to the content
of total neutral sugars in their WSP (Fig. 6b). Pena and Carpita
(2004) observed that the major biochemical event associated
with storage in four M. domestica cultivars is a marked decrease
in arabinose content, which always precedes the loss of fruit
firmness measured by both breaking strength and compression
resistance.
Tomato is the model system of choice for studying textural
changes during ripening of fleshy fruit. Recently, Van Dijk et
al. (2006) showed that polygalacturonase seems a more suit-
able candidate than ␤-galactosidase to explain the firmness
decrease. However, extrapolation of findings in tomato to other
fruit species may not be true. Redgwell et al. (1997) showed
temporal differences in fruit softening between eight species.
‘Braeburn’ and ‘Cox’s Orange Pippin’ apples softened to a
much lesser degree than ‘Billington’ plum. Moreover, cell-wall-
degrading enzymes responsible for softening are not the same
from one fruit species to another. Ripening avocado, tomato and
peach fruit possess relatively high levels of polygalacturonase
activity while it has been reported to be absent in other species
including strawberry, apple and melon (Brummell and Harpster,
2001). Consequently, fruit softening may have to be treated on
a case-by-case basis. Moreover, Brummell and Harpster (2001)
highlighted that no single cell-wall-modifying enzyme can be
identified as being necessary and sufficient for textural changes.
Fruit softening is a multi-gene trait with each enzyme activ-
ity having its own role to play in textural changes. Goulao
et al. (2007) identified different cell-wall-degrading enzymes
during growth and ripening of ‘Mondial Gala’ apples. Dur-
ing the ripening of Capsicum annuum fruit, Priya Sethu et al.
(1996) found that firmness of bell pepper decreases concomi-
tantly with the increase in polygalacturonase activity and with
the decrease in pectin methyl esterase. According to Prabha
and Bhagyalakshmi (1998), amylase, pectinases, xylanase and
laminarinase in banana pulp may play a dominant role in the
loosening of cell structure and tissue softening during ripening.
However, even though several enzyme activities seem to oper-
ate during the entire ripening period of papaya fruit, Manrique
and Lajolo (2004) observed that the period of greater firmness
variation in papaya fruit did not match the periods in which
the loss of galactose and the solubilisation of hemicellulose
were greater. However, a decreasing value of the neutral sug-
ars/acidic sugars ratio was observed by these authors during
solubilisation, giving a key enzymatic role to a polygalactur-
onase in papaya softening. Ali et al. (2004) suggested that
pectin methyl esterase and ␤-galactosidase activity seem rele-
vant and might contribute significantly to the observed variations
in softening rates amongst tropical fruit. These previous studies
revealed that fruit softening during storage is a complex process;
the action of enzymes and proteins modifies cell-wall structure
and consequently, fruit texture implying fruit softening. How-
ever, without transgenic experiments, it is difficult to determine
which cell-wall-modifying enzymes are responsible of fruit soft-
ening. According to Owino et al. (2005), the individual roles of
these enzymes are still being addressed in part through use of
gene silencing techniques (antisense and co-suppression). But
another technique is the use of “chimeric” genes to simultane-
ously silence several target enzymes in order to manipulate entire
biochemical pathways. Proteomic strategies will also be useful
and an additional approach in identifying genes regulating fruit
texture is the use tomato pleiotropic mutations.
4. Conclusion
Apple texture was studied by sensory, physical and chem-
ical means in order to research the most relevant biochemical
markers in relation to changes in different sensory properties
of apple texture. Two cultivars with different textural charac-
teristics were studied for five different storage periods. Fruit
softening confirmed previous studies as well as the observed
change in pectin composition. Indeed, major changes concern
soluble pectin composition. This study showed that the most
 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324
relevant biochemical marker which seemed linked to texture
change was the galacturonic acid content analysed in the water-
soluble pectin extracts. Indeed, high and positive correlation
coefficients were observed between sensory mealiness and this
marker while sensory crunchiness and instrumentally measured
firmness were negatively correlated to this biochemical compo-
nent. There were divergences between the two cultivars studied.
The physical parameters of the texture of ‘Golden Delicious’
apples were more affected than those of ‘Fuji’ while the chem-
ical composition of cell-wall pectin evaluated in higher extent
for ‘Fuji’ than for ‘Golden Delicious’ apples. This means that
other factors such as genetic and enzymatic metabolisms should
be taken into account in further studies dealing with fruit soft-
ening mechanisms. It would be necessary to relate the observed
changes in cell-wall pectin to the microscopic aspects (size and
shape of the cells, thickness of the cell wall, etc.) of fruit tissues
as well as cell turgidity (water content, osmotic pressure, vol-
ume of the intercellular space, etc.) in order to explain changes
in fruit texture on a macroscopic scale.
Acknowledgements
This research was supported by the Conseil Régional des Pays
de la Loire. The authors are grateful to Marie-Jeanne Crépeau
and Matthieu Vicaire for their contribution to this study. They
would also like to thank Stéphanie Khaldi and Nadège Gibouin
for their technical assistance as well as Ronan Symoneaux and
the members of the trained sensory panel. We are grateful to the
fruit-bearing test center “La Morinière” for providing fruit and
storage facilities.
References
Abbott, J.A., Watada, A.E., Massie, D.R., 1984. Sensory and instrument mea-
surement of apple texture. J. Am. Soc. Hort. Sci. 109, 221–228.
AFNOR, 1995. Analyse sensorielle-Recherche et sélection de descripteurs
pour l’élaboration d’un profil sensorial, par approche multidimensionnelle.
Contrôle de la qualité des produits alimentaires—Analyse sensorielle, 5ème
éd. AFNOR, Paris, pp. 276–310.
Ali, Z.M., Chin, L.-H., Lazan, H., 2004. A comparative study on wall degrading
enzymes, pectin modifications and softening during ripening of selected
tropical fruit. Plant Sci. 167, 317–327.
Bartley, I.M., 1976. Changes in the glucans of ripening apples. Phytochemistry
15, 625–626.
Ben, J., Gaweda, M., 1985. Changes of pectic compounds in Jonathan apples
under various storage conditions. Acta Physiol. Plant 7, 45–54.
Brummell, D.A., Harpster, M.H., 2001. Cell wall metabolism in fruit softening
and quality and its manipulation in transgenic plants. Plant Mol. Biol. 47,
311–340.
Curry, E.A., 1989. Effect of harvest date and oxygen level on storability of
late season apple cultivars. In: Fellman, J.K. (Ed.), Proceedings of the Fifth
International CA Research Conferences. Wanatchee, Wasch, pp. 103–109.
De Vries, J.A., Voragen, A.G.J., Rombouts, F.M., Pilnik, W., 1981. Extraction
and purification of pectins from alcohol insoluble solids from ripe and unripe
apples. Carbohydr. Polym. 1, 117–127.
Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Col-
orimetric method for determination of sugars and related substances. Anal.
Chem. 28, 350–356.
Duprat, F., Grotte, M., Loonis, D., Pietri, E., 2000. Etude de la possibilité de
mesurer simultanément la fermeté de la chair et l’épiderme des pommes.
Sci. Alim. 20, 253–265.
323
Fischer, M., Amado, R., 1994. Changes in the pectic substances of apples
during development and postharvest ripening. Part 1: Analysis of the alcohol-
insoluble residue. Carbohydr. Polym. 25, 161–166.
Fischer, M., Arrigoni, E., Amado, R., 1994. Changes in the pectic substances of
apples during development and postharvest ripening. Part 2: Analysis of the
pectic fractions. Carbohydr. Polym. 25, 167–175.
Fortin, J., Desplancke, C., 1998. Guide d’entraı̂nement d’un jury de dégustation.
Edisem.
Grant Reid, J.S., 1997. Carbohydrate metabolism: structural carbohydrates. In:
Dey, P.M., Harborne, J.B. (Eds.), Plant Biochemistry, vol. 5. Academic Press,
USA, pp. 205–236.
Goulao, L.F., Santos, J., de Sousa, I., Oliveira, C.M., 2007. Patterns of enzy-
matic activity of cell wall-modifying enzymes during growth and ripening
of apples. Postharvest Biol. Technol. 43, 307–318.
Grotte, M., Duprat, F., Loonis, D., Piétri, E., 2001. Mechanical properties of the
skin and the flesh of apples. Int. J. Food Prop. 4, 149–161.
Holt, J., Schoorl, D., 1984. Mechanical properties and texture of stored apples.
J. Text. Stud. 15, 377–394.
Ilker, R., Szczesniak, A.S., 1990. Review paper. Structural and chemical bases
for texture of plant foodstuffs. J. Text. Stud. 21, 1–36.
Jobling, J.J., McGlasson, W.B., 1995. A comparison of ethylene production,
maturity and controlled atmosphere storage life of Gala, Fuji and Lady
Williams apples (Malus domestica Borkh). Postharvest Biol. Technol. 6,
209–218.
Johnston, J.W., Hewett, E.W., Hertog, M.L.A.T.M., 2001. Temperature induces
differential softening responses in apple cultivars. Postharvest Biol. Technol.
23, 186–197.
Klavons, J.A., Bennett, R.D., 1986. Determination of methanol using alcohol
oxidase and its application to methyl ester content of pectins. J. Agric. Food
Chem. 34, 597–599.
Knee, M., 1973a. Polysaccharides and glycoproteins of apple fruit cell walls.
Phytochemistry 12, 637–653.
Knee, M., 1973b. Polysaccharide changes in cell walls of ripening apples. Phy-
tochemistry 12, 1543–1549.
Lo Scalzo, R., Forni, E., Lupi, D., Giudetti, G., Testoni, A., 2005. Changes of
pectic composition of ‘Annurca’ apple fruit after storage. Food Chem. 93,
521–530.
Manrique, G.D., Lajolo, F.M., 2004. Cell-wall polysaccharide modifications
during postharvest ripening of papaya fruit (Carica papaya). Postharvest
Biol. Technol. 33, 11–26.
Massiot, P., Baron, A., Drilleau, J.F., 1994. Characterisation and enzymatic
hydrolysis of cell-wall polysaccharides from different tissue zones of apple.
Carbohydr. Polym. 25, 145–154.
Massiot, P., Baron, A., Drilleau, J.F., 1996. Effect of storage of apple on the
enzymatic hydrolysis of cell wall polysaccharides. Carbohydr. Polym. 29,
301–307.
Mehinagic, E., Royer, G., Bertrand, D., Symoneaux, R., Laurens, F., Jourjon, F.,
2003. Relationship between sensory analysis, penetrometry and visible-NIR
spectroscopy of apples belonging to different cultivars. Food Qual. Prefer.
14, 473–484.
Mehinagic, E., Royer, G., Symoneaux, R., Bertrand, D., Jourjon, F., 2004.
Prediction of the sensory quality of apples by physical measurements.
Postharvest Biol. Technol. 34, 257–269.
Nara, K., Kato, Y., Motomura, Y., 2001. Involvement of terminal-arabinose
and -galactose pectic compounds in mealiness of apple fruit during storage.
Postharvest Biol. Technol. 22, 141–150.
Owino, W.O., Ambuko, J.L., Mathooko, F.M., 2005. Molecular basis of cell wall
degradation during fruit ripening and senescence. Stewart Postharvest Rev.
3, 1–10.
Pena, M.J., Carpita, N.C., 2004. Loss of highly branched arabinans and
debranching of rhamnogalacturonan I accompany loss of firm texture and
cell separation during prolonged storage of apple. Plant Physiol. 135, 1305–
1313.
Prabha, T.N., Bhagyalakshmi, N., 1998. Carbohydrate metabolism in ripening
banana fruit. Phytochemistry 48, 915–919.
Priya Sethu, K.M., Prabha, T.N., Tharanathan, R.N., 1996. Postharvest bio-
chemical changes associated with the softening phenomenon in Capsicum
annuum fruit. Phytochemistry 42, 961–966.
324 L. Billy et al. / Postharvest Biology and Technology 47 (2008) 315–324
Redgwell, R.J., Fischer, M., Kendal, E., MacRae, E.A., 1997. Galactose loss
and fruit ripening, high-molecular-weight arabinogalactans in the pectic
polysaccharides of fruit cell walls. Planta 203, 174–181.
Renard, C.M.G.C., Thibault, J.F., Voragen, A.G.J., 1990a. Etude chimique et
enzymatique de la structure de la paroi de pomme. Cahier Scientifique et
Technique, 341–347.
Renard, C.M.G.C., Voragen, A.G.J., Thibault, J.F., Pilnik, W., 1990b. Studies
on apple protopectin. I: Extraction of insoluble pectin by chemical means.
Carbohydr. Polym. 12, 9–25.
Renard, C.M.G.C., Searle van Leeuwen, M.J.F., Voragen, A.G.J., Thibault, J.F.,
Pilnik, W., 1991a. Studies on apple protopectin. II: Apple cell wall degra-
dation by pure polysaccharides and their combinations. Carbohydr. Polym.
14, 295–314.
Renard, C.M.G.C., Schols, H.A., Voragen, A.G.J., Thibault, J.F., Pilnik, W.,
1991b. Studies on apple protopectin. III: Characterization of the mate-
rial extracted by pure polysaccharidases from apple cell walls. Carbohydr.
Polym. 15, 13–32.
Renard, C.M.G.C., Voragen, A.G.J., Thibault, J.F., Pilnik, W., 1991c. Studies on
apple protopectin. IV: Apple xyloglucans and influence of pectin extraction
treatments on their solubility. Carbohydr. Polym. 15, 387–403.
Renard, C.M.G.C., Voragen, A.G.J., Thibault, J.F., Pilnik, W., 1991d. Studies on
apple protopectin. V: Structural studies on enzymatically extracted pectins.
Carbohydr. Polym. 16, 137–154.
Renard, C.M.G.C., Thibault, J.F., 1993. Studies on apple protopectin. VI: Extrac-
tion of pectins from apple cell walls with rhamnogalacturonase. Carbohydr.
Polym. 22, 203–210.
Renard, C.M.G.C., 2005. Variability in cell wall preparations, quantification and
comparison of common methods. Carbohydr. Polym. 60, 512–522.
Saeman, J.F., Moore, W.E., Mitchell, R.L., Millett, M.A., 1954. Techniques for
the determination of pulp constituents by quantitative paper chromatography.
Tappi 37, 336–343.
Siddiqui, S., Brackmann, A., Streif, J., Bangerth, F., 1996. Controlled atmo-
sphere storage of apples, cell wall composition and fruit softening. J. Hort.
Sci. 71, 613–620.
Stow, J., 1995. Quality measurements of apples. Postharvest News Inf. 6, 32–33.
Thibault, J.F., 1979. Automatisation du dosage des substances pectiques par la
méthode au méta-hydroxydiphényl. Lebensm. -Wiss. Technol. 12, 247–251.
Tollier, M.T., Robin, J.P., 1979. Adaptation de la méthode à l’orcinol sulfurique
au dosage automatique des glucides neutres totaux, conditions d’application
aux extraits d’origine végétale. Ann. Technol. Agric. 28, 1–15.
Varela, P., Salavdor, A., Fiszman, S., 2007. Changes in apple tissue with storage
time: rheological, textural and microstructural analyses. J. Food Eng. 78,
622–662.
Voragen, A.G.J., Schols, H.A., Pilnik, W., 1986. Determination of the degree
of methylation and acetylation of pectins by h.p.l.c. Food Hydrocolloid 1,
65–70.
Van Dijk, C., Boeriu, C., Stolle-Smits, T., Tijskens, L.M.M., 2006. The firm-
ness of stored tomatoes (cv Tradiro). 2: Kinetic and Near Infrared models
to describe pectin degrading enzymes and firmness loss. J. Food Eng. 77,
585–593.
Yoshioka, H., Aoba, K., Kashimura, Y., 1992. Molecular weight and degree of
methoxylation in cell wall polyuronide during softening in pear and apple
fruit. J. Am. Soc. Hort. Sci. 117, 600–606.
Yoshioka, H., Kashimura, Y., Kaneko, K., 1994. Solubilization and distribution
of neutral sugar residues derived from polyuronides during the softening in
apple fruit. J. Jpn. Soc. Hort. Sci. 63, 173–182.