Journal of Plant Physiology 176 (2015) 129137

Contents lists available at ScienceDirect

Journal of Plant Physiology

journal homepage: www.elsevier.com/locate/jplph

Physiology

Lower cell wall pectin solubilisation and galactose loss during early
fruit development in apple (Malus x domestica) cultivar Scifresh are
associated with slower softening rate
Jovyn K.T. Ng a,b, , Roswitha Schrder b , David A. Brummell c , Paul W. Sutherland b ,
Ian C. Hallett b , Bronwen G. Smith a , Laurence D. Melton a , Jason W. Johnston d
a

Food Science, School of Chemical Sciences, The University of Auckland, Private Bag 92019, Auckland, New Zealand
The New Zealand Institute for Plant & Food Research Limited, Mount Albert Research Centre, Private Bag 92169, Auckland 1142, New Zealand
c
The New Zealand Institute for Plant & Food Research Limited, Food Industry Science Centre, Private Bag 11600, Palmerston North 4442, New Zealand
d
The New Zealand Institute for Plant & Food Research Limited, Private Bag 1401, Havelock North 4157, New Zealand
b

a r t i c l e

i n f o

Article history:
Received 31 October 2014
Received in revised form
14 December 2014
Accepted 15 December 2014
Available online 5 January 2015
Keywords:
Cell wall
Fruit development
Fruit softening
Arabinan
Galactan

a b s t r a c t
Substantial differences in softening behaviour can exist between fruit even within the same species. Apple
cultivars Royal Gala and Scifresh soften at different rates despite having a similar genetic background
and producing similar amounts of ethylene during ripening. An examination of cell wall metabolism
from the fruitlet to the ripe stages showed that in both cultivars pectin solubilisation increased during
cell expansion, declined at the mature stage and then increased again during ripening. This process was
much less pronounced in the slower softening Scifresh than in Royal Gala at every developmental
stage examined, consistent with less cell separation and softening in this cultivar. Both cultivars also
exhibited a progressive loss of pectic galactan and arabinan side chains during development. The cell wall
content of arabinose residues was similar in both cultivars, but the galactose residue content in Scifresh
remained higher than that of Royal Gala at every developmental stage. The higher content of cell wall
galactose residue in Scifresh cell walls correlated with a lower -galactosidase activity and more intense
immunolabelling of RG-I galactan side chains in both microscopy sections and glycan microarrays. A high
cell wall galactan content has been associated with reduced cell wall porosity, which may restrict access
of cell wall-modifying enzymes and thus maintain better structural integrity later in development. The
data suggest that the composition and structure of the cell wall at very early development stages may
inuence subsequent cell wall loosening, and may even predispose the walls ensuing properties.
2014 Elsevier GmbH. All rights reserved.

Introduction
After pollination, apple fruit have a phase of intensive cell division lasting 35 weeks, followed by a phase of cell expansion and

Abbreviations:
AFase, -arabinofuranosidase; Ara, arabinose; BGal, galactosidase; CWM, cell wall material; Gal, galactose; Glc, glucose; HG,
homogalacturonan; RG-I, rhamnogalacturonan I; Rha, rhamnose; UA, uronic acid;
WS, water solubles; Xyl, xylose.
Corresponding author at: Present address: The New Zealand Institute for Plant &
Food Research Limited, Food Industry Science Centre, Private Bag 11600, Palmerston
North 4442, New Zealand. Tel.: +64 63556151.
E-mail addresses: jovyn.ng@plantandfood.co.nz (J.K.T. Ng),
rosie.schroeder@plantandfood.co.nz (R. Schrder),
david.brummell@plantandfood.co.nz (D.A. Brummell),
paul.sutherland@plantandfood.co.nz (P.W. Sutherland),
ian.hallett@plantandfood.co.nz (I.C. Hallett), bsmith.auckland@gmail.com
(B.G. Smith), l.melton@auckland.ac.nz (L.D. Melton),
jason.johnston@plantandfood.co.nz (J.W. Johnston).
http://dx.doi.org/10.1016/j.jplph.2014.12.012
0176-1617/ 2014 Elsevier GmbH. All rights reserved.

maturation, where the fruit reach physiological maturity and the

competence to ripen (Lakso et al., 1995). As the cells expand, and
the fruit grows, size is gained mainly through water uptake and
increase in parenchyma cell volume, together with an increase in
intercellular air spaces (Volz et al., 2003). Throughout cell expansion and maturation, the apple parenchyma tissue loses rmness,
in part due to changes in overall fruit anatomy and types of tissue,
as well as changes to the shape and size of cells and how they are
packed and adhere to each other (Harker et al., 1997). As the fruit
ripens, there are major changes in the chemistry of the primary cell
wall, the membranes, and in cell turgor. The primary cell wall is
considered to consist of a framework of cellulose microbrils held
together by a complex matrix of hemicelluloses (predominantly
xyloglucan and xylan) and pectins. During ripening-related softening, hemicellulose modications are often subtle, but pectins are
solubilised and extensively altered. The two major pectins are the
largely unbranched homogalacturonan (HG), which becomes demethylesteried and depolymerised, and rhamnogalacturonan-I

130

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

(RG-I), which becomes depleted of its large galactan and arabinan

side chains (Brummell, 2006). Ripening-related softening is a complex multi-component process and the extent to which cell wall
changes occur varies greatly between species. In fruit that soften to
a melting texture during ripening (e.g. kiwifruit, tomato, or European pear) some of these changes are extensive compared with
fruit that soften partially and remain relatively rm and crisp (e.g.
apple, watermelon, or Asian pear).
Substantial differences exist between apple cultivars in both
softening behaviour (Oraguzie et al., 2007) and cell wall composition (Galvez-Lopez et al., 2011), but research to understand
softening processes and relate them to cultivar differences have
mainly focussed on cell wall changes during ripening. Changes to
the hemicellulosic polysaccharides appear to be minor in apple,
since Siddiqui et al. (1996) and Percy et al. (1997) reported minimal
depolymerisation of hemicelluloses and no depolymerisation of
xyloglucan, although Ito et al. (2004) found small changes in the ne
structure of water-soluble and cell-wall bound xyloglucan during
fruit development and ripening. In contrast, there are substantial
changes in the pectic polymers, particularly a decrease in the galactose (Gal) content of the wall (Gross and Sams, 1984; Fischer et al.,
1994), combined with an increase in pectin solubilisation (Redgwell
et al., 1997a) but with limited pectin depolymerisation (Yoshioka
et al., 1992; Fischer et al., 1994; Redgwell et al., 1997a). The loss
of RG-I galactan side chains occurs in most fruit species (Redgwell
et al., 1997b), although some studies report the loss of arabinans
as playing a more important role in apple softening, particularly in
mealiness (Tong et al., 1999; Nobile et al., 2011) and oversoftening
and Carpita, 2004). Pena
and Carpita (2004) found that the
(Pena
selective loss of highly branched arabinans from RG-I preceded the
loss of rmness in the four apple cultivars analysed. A large increase
in water-soluble polyuronide occurred in Golden Delicious but not
in Fuji (Billy et al., 2008) and arabinose (Ara) was lost from cell
walls of stored and ripened Braeburn and Macoun but not Cox
or Honeycrisp (Redgwell et al., 1997b; Tong et al., 1999), indicating
that these may be cultivar-specic events.
The loss of Gal and/or Ara residues from the wall also seems to
play a role during apple fruit growth and maturation, well before
ripening-related softening starts. The rate of loss of Gal was highest during the cell expansion phase of Red Delicious and then
and Carpita, 2004). Gal loss also
decreased up to maturation (Pena
occurred at a high rate during the cell expansion phase in Golden
Delicious and then slowed after harvest (Fischer et al., 1994).
Gal loss continued during postharvest softening in most cultivars
(Fischer et al., 1994; Redgwell et al., 1997b; Gwanpua et al., 2014). In
Mondial, both -galactosidase (BGal) and -arabinofuranosidase
(AFase) activities were low throughout growth and maturation,
but then increased when fruit softened during ripening (Goulao
et al., 2007). BGal activity also increased during fruit development
in Granny Smith apples (Ross et al., 1994).
With different softening rates, yet similar genetic background,
overlapping growing season and similar ethylene production
during ripening, the rapidly-softening Royal Gala and slowlysoftening Scifresh apple cultivars are ideal models for comparison.
Ng et al. (2013) showed that cell wall structures related to tissue microstructure and cell adhesion developed early during fruit
growth in these cultivars, and that these led to genotypic differences in softening rates well before the induction of the ripening
process. It thus seems likely that the way the wall is composed and
assembled from early development may play an important role in
affecting subsequent fruit rmness changes, in addition to the better studied ripening-related disassembly events. The aim of this
study was to relate pectin solubilisation and modication during
fruit growth to genotypic differences in ripening-related softening. By comparing the polysaccharides extracted from cell walls of
fruitlet, expanding, mature, and ripe fruit with the activities of BGal

and AFase and galactan and arabinan epitope abundance, we monitored the temporal and spatial distribution of the polysaccharides
and their changes during development in the cell wall.
Materials and methods
Physiological assessments of fruit and sampling
Apple fruit (Malus x domestica Borkh.) Scifresh and Royal Gala
were sampled from the Plant & Food Research orchard, Havelock
North, New Zealand and assessed as described in Ng et al. (2013).
Briey, twenty apples of each cultivar were collected at 40 days
after full bloom (DAFB) (fruitlet; end of cell division phase and
beginning of cell expansion), 70 DAFB (expanding fruit; cell expansion phase), and 120 (Royal Gala) or 140 (Scifresh) (mature fruit).
Mature fruits were ripened in the cold (0.5 C) for 20 weeks (ripe
fruit). Cortex tissue from 20 fruits at each of these stages was
pooled, diced, and frozen in liquid nitrogen and stored at 80 C.
This tissue was used for three replications of cell wall extractions.
Cell wall preparation and extraction
Cell wall isolation and extractions were performed as described
in Ng et al. (2013) to obtain the water-soluble fraction (WS)
and cell wall material (CWM). The CWM was further fractionated
by sequential treatment for 6 h in 0.05 M CDTA (trans-1,2diaminocyclohexane-N,N,N ,N -tetraacetic acid) containing potassium acetate buffer pH 6.5, followed by 0.05 M Na2 CO3 containing
20 mM NaBH4 for 17 h, 1 M KOH containing 20 mM NaBH4 for
17 h, and 4 M KOH containing 20 mM NaBH4 for 48 h (Melton
and Smith, 2005). After each extraction, the insoluble pellets
were homogenised, centrifuged, ltered, and rinsed twice with
water. Supernatants were combined accordingly to give the CDTA-,
Na2 CO3 -, 1 M KOH-, and 4 M KOH-soluble extracts. The remaining
insoluble pellet was termed cell wall residue. Alkaline fractions
were neutralised with glacial acetic acid and dialysed (SigmaAldrich; dialysis membrane molecular weight cut off 12 kDa)
exhaustively against water at 4 C to remove salts, concentrated
under rotary evaporation, freeze-dried, and weighed to determine
yields.
Protein extractions and enzyme assays
Ground frozen apple tissue (0.25 g) was extracted in 0.5 mL
0.2 M sodium acetate, 10 mM dithiothreitol, 50 mM NaCl, protease inhibitor (Roche Diagnostics) (pH 4.7) with 25 mg of
polyvinylpolypyrrolidone (Sigma-Aldrich, USA). The mixture was
vortexed, centrifuged at 10,000 g for 10 min and the pellet resuspended in 0.25 mL of the extraction buffer. After centrifugation,
the two supernatants were combined to give the buffer-soluble
protein extract. The pellet was then extracted with 0.3 M MES [2(N-morpholino)ethanesulfonic acid], 10 mM dithiothreitol, 0.6 M
NaCl, and protease inhibitor (pH 6.0) using the same procedure
as described above, except that the pellets were incubated on ice
for 1 h to aid solubilisation of protein from the cell wall. Supernatants were collected to give the salt-soluble (presumably cell
wall-bound) extract.
-galactosidase (BGal) and -arabinofuranosidase (AFase)
activities were assayed using p-nitrophenyl--D-galactopyranoside (Sigma-Aldrich) and p-nitrophenyl--l-arabinofuranoside
(Sigma-Aldrich) as substrates, respectively. Protein extracts (20 L)
were incubated with the substrate in 20 L of 1 M sodium acetate
buffer (pH 4.6) at 28 C for 1 h for BGal (Ross et al., 1994); and
in 20 L of 0.1 M sodium citrate buffer (pH 5.0) at 37 C for 3 h
for AFase (Yoshioka et al., 1995). Reactions were terminated by
adding 80 L of 20 mM Na2 CO3 and absorbance was determined

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

131

Table 1
Yields of cell wall material (CWM, as mg g1 FW) and water-soluble polyuronide (mg UA g1 CWM) from cell walls of Royal Gala (RG) and Scifresh (SF) fruitlet, expanding
and mature fruit at the indicated days after full bloom (DAFB), and ripe fruit cold-stored for 20 weeks. Results are means of three extraction replicates on fruit harvested
from one season SD.
Extract

CWM
Water-soluble polyuronide

Fruitlet (40 DAFB)

Expanding (70 DAFB)

Mature (120/140 DAFB)

Stored, ripe (20 weeks)

RG

SF

RG

SF

RG

SF

RG

SF

64.3 5.3
13.0 2.4

51.2 8.3
6.4 1.4

19.9 2.2
50.6 11.8

20.4 4.3
22.1 7.9

16.1 3.6
43.4 13.4

17.1 3.5
12.1 8.8

7.1 2.9
63.3 15.5

9.4 2.3
40.8 12.3

at 415 nm. For background reactions, the termination solution was

added before addition of the enzyme extract. These background values were subtracted from values obtained using the active enzyme
extract, and enzyme activities expressed as mol p-nitrophenol
released per unit time per mg protein, using p-nitrophenol (SigmaAldrich) in the corresponding buffer as standard.
Protein gel blot analysis
Protein from ground frozen apple tissue was extracted, separated by SDS-PAGE and electroblotted to a membrane as described
in Ng et al. (2013). Proteins were reacted with polyclonal antibodies raised against BGal from kiwifruit (courtesy of SL Johnston),
at a dilution of 1:1000 (v/v) in TBS buffer containing 5% non-fat
milk powder. Membranes were incubated with a goat anti-rabbit
secondary antibody conjugated to alkaline phosphatase (SigmaAldrich) and visualised using 1-StepTM NBT/BCIP staining (Thermo
Scientic, USA).

for 48 h in 5% (w/v) skimmed milk powder in PBS at 4 C with

gentle agitation. Arrays were probed with monoclonal antibodies LM5, LM6, and LM13 (PlantProbes, Leeds, UK) at a dilution
of 1:200 in 5% skimmed milk powder in PBS, and incubated for
2 h at room temperature. The membranes were then washed
three times with PBS for 10 min. The secondary antibody (rabbit
anti-rat IgG-alkaline phosphatase, Sigma) was added at a concentration of 1:1000 in 5% skimmed milk powder in PBS and was
incubated for 2 h at room temperature. After extensive washing
in PBS, the arrays were developed using NBT/BCIP in an alkaline buffer [100 mM NaCl, 5 mM MgCl2 , 100 mM diethanolamine,
pH 9.5]. Arrays were scanned using a high-resolution (1200 dpi)
desktop scanner (Epson Perfection 4490 Photo) and the images
saved as negative, 16-bit TIFF les. The images were analysed
using Xplore Image Processing Software (LabNEXT), and mean pixel
value for spot integral intensity quantied and presented as a
heatmap.
Analytical methods and statistical analyses

Size exclusion chromatography

WS and Na2 CO3 -soluble fractions (2.5 mg freeze-dried powder)
were re-dissolved and separated on a column of Sepharose CL-2B
(GE Healthcare, USA) as described in Ng et al. (2013). The 1 M KOH
and 4 M KOH-soluble fractions (5.0 mg freeze-dried powder) were
dissolved in 0.5 mL water and eluted through a column of Sephacryl
S-300 (GE Healthcare, 1.5 75 cm) using 0.05 M sodium acetate
(pH 6.0), 0.125 M NaCl, 0.05% (v/v) chlorobutanol buffer with an
average ow rate of 5.5 mL h1 . Column fractions were assayed for
uronic acid (UA) content (Blumenkrantz and Asboe-Hansen, 1973).
Columns were calibrated with linear dextran size markers T10
(10 kDa), T40 (40 kDa), T500 (500 kDa) and Blue Dextran (2 MDa)
(GE Healthcare).
Immunolabelling
Fixation, sectioning, immunolabelling, and visualisation were
carried out as described in Ng et al. (2013), using monoclonal antibodies LM5 and LM6 (PlantProbes, Leeds, UK) at dilutions of 1:200
and 1:20, respectively, in 0.1% (w/v) bovine serum albumin in phosphate buffered saline (PBS), and secondary antibody goat anti-rat
IgG AlexaFluor 488 (Molecular Probes, Oregon, USA) at a 1:600 dilution in PBS. Two sets of images are presented: dual-labelling with
calcouor (blue) and antibody labelling manipulated to a hue angle
of pink for visual clarity, and (in Supplementary Data) antibody
immunouorescence labelling as green.
Glycan arrays
The extraction of cell wall glycans, printing, and probing of
microarrays were carried out as described in Moller et al. (2012).
Samples were printed in triplicate and in three dilutions (0-, 5, and 25-fold) onto a nitrocellulose membrane (Sigma-Aldrich),
using two capillary pins on a contact microarray robot (Xact II
Microarrayer, LabNEXT, USA). After printing, arrays were blocked

Neutral sugar contents in the cell wall fractions were

determined by gas chromatography (GC) after hydrolysis in triuoroacetic acid (TFA), followed by the conversion of monosaccharides
to alditol acetates (Albersheim et al., 1967). For the insoluble
residue, non-cellulosic polysaccharides were hydrolysed with TFA,
while cellulosic polysaccharides were hydrolysed in a two-stage
sulphuric acid procedure (Harris et al., 1988) prior to conversion
into alditol acetates and GC analysis. UAs were determined as
described by Ahmed and Labavitch (1977) and Blumenkrantz and
Asboe-Hansen (1973) with galacturonic acid as a standard.
Two-way (for cell wall composition and enzyme assay results)
ANOVA (analysis of variance) analyses were conducted using the
Microsoft Excel 2007 for Windows software. Means were compared
using Fishers least signicant difference (LSD) post-test at P 0.05
using the SPSS (Version 15.0) software (IBM, USA).
Results
Royal Gala and Scifresh differ in the extent of pectin
solubilisation
The amount of water-soluble pectin increased substantially
from the fruitlet to the expanding stage, declined at fruit maturity
then increased again during ripening (Table 1). At all developmental stages, the rapidly-softening Royal Gala had about twice
the yield of water-soluble polyuronides as the slowly-softening
Scifresh, indicating major differences in pectin solubilisation
before ripening-related softening began. Size exclusion chromatography showed that the molecular weight of these polyuronides
increased overall in both cultivars during development, but some
populations of molecules were larger in Royal Gala than in
Scifresh at the fruitlet and ripe stages (Fig. 1AD).
At most developmental stages, both cultivars had similar yields
of Na2 CO3 -soluble uronic acid (UA) (Table S1). At the fruitlet stage,
however, Scifresh had only about half of the amount of UA as

132

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

Fig. 1. Size distribution of polyuronides in the different cell wall extracts. Water soluble (WS) (AD), Na2 CO3 soluble (EH), 1 M KOH soluble (IL) and 4 M KOH soluble
(MP) extracts from fruitlet (A, E, I, M), expanding (B, F, J, N), mature (C, G, K, O) and ripe (D, H, L, P) fruit of Royal Gala (solid line) and Scifresh (dashed line). Size exclusion
chromatography was carried out on Sepharose CL-2B (AH) and Sephacryl S300 (IP), and column fractions assayed for UA content. Results are shown from one typical
experiment, repeated over three seasons. Arrows indicate elution points of linear dextran molecular weight markers (2000, 500, 40 and 10 kDa).

Scifresh has more Gal in the cell wall and more highly branched
pectin
Sugar composition analysis of whole, unfractionated cell walls
showed that Scifresh had a much greater content of Gal residues
than did walls from Royal Gala (Fig. 2A). At the fruitlet stage,
cell walls from Scifresh contained 50% more Gal than did walls
from Royal Gala, and this difference was approximately maintained through fruit development and ripening even as Gal content
declined in both cultivars. Between the fruitlet stage and the

Gal content
(g mg-1 cell wall)

ripe stage, both cultivars lost just over half of their cell wall Gal.
The content of Ara in the wall was slightly higher in Scifresh
than in Royal Gala at the fruitlet stage, but after this both
content and loss during development were similar in the two
cultivars (Fig. 2B).

Ara content
(g mg-1 cell wall)

Royal Gala but twice as much Ara and Gal, indicating a high abundance of highly branched RG-I pectin. This polyuronide also had a
higher molecular weight than that in Royal Gala (Fig. 1E). After the
fruitlet stage, the molecular weight distribution of Na2 CO3 -soluble
polyuronide was similar in the two cultivars (Fig. 1FH).
The 1 M KOH-soluble and 4 M KOH-soluble extracts contained
substantial amounts of progressively more rmly bound pectin
together with lesser amounts of hemicelluloses, as indicated by
the high contents of UA, Ara, Gal, and rhamnose (Rha) (typical of
HG and RG-I) and lower contents of xylose (Xyl) and glucose (Glc)
(indicative of xylan and xyloglucan) (Table S1). There were no substantial differences in sugar composition between the cultivars at
any of the developmental stages (Table S1), but analysis by size
exclusion chromatography showed some differences between the
cultivars in polyuronide size distribution (Fig. 1IP). Particularly, in
both extracts at the mature stage (Fig. 1 K and O) a proportion of
the polyuronide in Royal Gala had already become depolymerised
to lower molecular weight, whereas in Scifresh this did not occur
until later in ripening.

Fig. 2. Content of Gal (A) and Ara (B) in cell walls of Royal Gala and Scifresh apple
fruit during development. Results are means of three extractions SD.

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

133

Fig. 3. Degree of pectin side-chain branching in different cell wall extracts of Royal Gala and Scifresh fruit based on neutral sugar composition. Branching was calculated
from the molar ratio of Ara plus Gal to Rha in the water-soluble (A), Na2 CO3 -soluble (B), 1 M KOH-soluble (C), 4 M KOH-soluble (D) and insoluble residue (E).

The molar ratio of Ara plus Gal to Rha residues in pectin can
be used to reveal the average length of neutral sugar side chains
attached to the RG-I backbone (Fig. 3). A calculation of this in
four soluble fractions and the insoluble residue shows that the
most highly branched pectin was in the 4 M KOH-soluble fraction
(Fig. 3D). The main difference between the cultivars was in the 1 M
and 4 M KOH-soluble fractions at the fruitlet and expanding stages,
where Scifresh pectin was more highly branched. At the mature
and ripe stages there was little difference between the cultivars in
most fractions, but there was an increase in water-soluble highly
branched pectin at the mature stage, and this remained slightly
higher in Scifresh than in Royal Gala at the ripe stage (Fig. 3A).

Scifresh has lower activity of BGal

Both cultivars possessed BGal and AFase activities in the buffersoluble and in the salt-soluble fraction, indicating the presence of
different isoforms (Fig. 4). In the faster softening Royal Gala, BGal
activity was generally higher than in Scifresh, and was substantially more at the ripe stage (Fig. 4A). Protein gel blot analysis using
an antibody raised against BGal from ripe kiwifruit conrmed a
higher abundance of BGal protein in Royal Gala, especially at the
ripe stage (Fig. 4A inset). AFase activity was approximately 5-fold
lower than BGal activity, and except for the fruitlet stage was similar in Royal Gala and Scifresh (Fig. 4B).

Immunolabelling of (1 4)--galactan with LM5 is more intense

in Scifresh
The monoclonal antibody LM5, specic for linear tetrasaccharides of Gal residues in (1 4)--D-galactan (McCartney et al.,
2000), labelled Scifresh more intensely than Royal Gala at
all developmental stages (Fig. 5, pink colour; Fig. S1). LM5
strongly labelled the outer regions of the cell walls but not
the middle lamella regions, especially at the fruitlet (Fig. 5A
and E) and expanding fruit stages (Fig. 5B and F). In fruitlet,
labelling appeared as thick bands, indicating abundant galactan
(Fig. 5A and E). In mature and ripe fruit, the labelling pattern became more irregular, and changed to a granular and
uneven distribution as cell wall Gal content declined. In both
cultivars, less labelling seemed to be present in cell junctions,
and more in cell wall areas of adjacent cells (Fig. 5 C, D, G
and H). The images also portray a characteristic of LM5 with
absence of labelling at plasmodesmatal elds (e.g. pl arrow in
Fig. 5A) (McCartney et al., 2000). Co-staining of sections with
calcouor (Fig. 5; blue colour) showed cellulose staining around
the inside of the cell walls towards the cell lumen in both cultivars.

Immunodetection of arabinan with monoclonal antibody LM6

specic for a linear pentasaccharide in (1 5)--l-arabinans
(Willats et al., 1998) was similar in both cultivars throughout development (Fig. S2, pink colour; Fig. S3). Labelling with
LM6 occurred on the outer side of cell walls, with a thick and
intense labelling pattern in fruitlet that became less intense
and homogeneous as fruit matured and ripened. Labelling in
ripe fruit was very uneven with patches of strong labelling in
some areas of the same cell, and very weak labelling in other
parts.
Glycan array proling
Detection of polysaccharide epitopes on thin sections can be
affected by the masking of one polysaccharide by another (Marcus
et al., 2008). Glycan array proling is a method for detection
of polysaccharide epitopes in solubilised extracts spotted onto
membranes, and can detect epitopes that might be partially or
completely obscured in muro. To conrm that galactan epitopes
were present in different abundances in developing Royal Gala and
Scifresh cell walls, glycan array proling was carried out on solubilised extracts using LM5 (Fig. 6). The heatmap shows that the LM5
galactan epitope was present at a higher abundance in Scifresh relative to Royal Gala at every developmental stage. The epitope was
most abundant at the expanding fruit stage, consistent with a large
proportion of cell wall Gal residues being present as (1 4)--Dgalactan (Fig. 2). The LM5 epitope had relatively higher abundances
in both Scifresh KOH extracts compared to Royal Gala, showing
that this galactan was rmly bound into the wall, and abundance
progressively declined as the fruit matured and ripened.
A comparison of LM6 arabinan epitope abundance revealed that
this epitope was also mainly present in the two KOH extracts,
and was more similar in abundance between the cultivars than
was the LM5 epitope. Interestingly, the abundance of the epitope
recognised by LM13 (a long linear stretch of (1 5)--l-arabinan,
Verhertbruggen et al., 2009) increased drastically in mature and
ripe fruit, indicating that arabinans were becoming less branched.
Discussion
Scifresh and Royal Gala fruit show quite different softening
behaviour during development (Ng et al., 2013). The mean rmness values measured by a texture analyzer puncture test showed
that Royal Gala softened from an initial value of 81.8 N in mature
fruit to 51.4 N in ripe fruit, whereas Scifresh showed much less
softening over the same period, declining from 89.9 N to 83.2 N (Ng
et al., 2013). Firmness loss was less and occurred at a much slower
rate in Scifresh. This was most accentuated in the cell expansion
phase, where Scifresh had a pronounced lag phase with minimal

134

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

Fig. 4. Specic activity of BGal (A) and AFase (B) during development of Royal Gala (RG) and Scifresh (SF) fruit. Bars represent total activity composed of buffer-soluble
(white bars) and salt-soluble (black bars) forms. Values are means SE of three protein extractions with duplicate activity assays. Inset in panel (A) shows a protein gel blot
using polyclonal antibodies raised against BGal from kiwifruit. Lanes 1, 2, 3, Royal Gala fruitlet, mature and ripe fruit; lanes 4, 5, 6, Scifresh fruitlet, mature and ripe fruit;
lane 7, BioRad Precision Plus Dual Protein Standard. An arrow indicates the presence of BGal protein at approx. 59 kDa.

change in rmness and higher accumulation of dry weight, and during ripening, where Scifresh barely lost rmness. In comparison,
Royal Gala lost rmness at a high rate throughout the cell division
phase and this continued at a reduced rate during ripening. Work
with candidate genes, QTLs and gene silencing has shown that the
ripening-related polygalacturonase gene PG1 provides an important component for apple softening in the ripening phase (Wakasa
et al., 2006; Atkinson et al., 2012; Longhi et al., 2013). However,
fruit also soften prior to the ripening phase, and during ripening
cultivars that do not possess immunodetectable PG1 protein such
as Scifresh (Ng et al., 2013) or transgenic Royal Gala suppressed
in PG1 expression (Atkinson et al., 2012) still soften although to a
reduced extent, implying that other factors are involved in softening. Two such factors could be pectin solubilisation and the
degradation of RG-I galactan and/or arabinan side-chains, which
occur in most fruit during softening but begin early in development
(Brummell, 2006).
During ripening-related softening, certain pectin populations
leave the conglomerate of polysaccharides in the cell wall and
become water-soluble (Redgwell et al., 1992, 1997a). This process
had already started at the fruitlet stage (Table 1), when the faster
softening Royal Gala had about twice the amount of water-soluble
pectin than Scifresh, a trend that continued until fruit were ripe.
Similarly, a comparison of the rapidly softening Golden Delicious
with the non-softening Fuji showed that an increase in watersoluble polyuronide during ripening was correlated with softening
(Billy et al., 2008). In Scifresh and Royal Gala, the water-soluble
pectin had a high UA content and a low Rha content in both cultivars

throughout development (Table S1), indicating that it consisted

mainly of HG. Water-soluble pectin is not attached to the cell wall,
and so does not contribute to wall strength. HG is also the main
component of the middle lamellae that hold cells together, and
an increase in cell separation is an important factor in apple fruit
and Carpita, 2004). There was howsoftening during ripening (Pena
ever, no correlation between the amount of water-soluble pectin
(Table 1) and actual rmness of fruit (Ng et al., 2013), as Royal Gala
had more water-soluble pectin than Scifresh even at the fruitlet
and expanding fruit stage, when the cortex tissue was still rmer.
Instead, it seemed to indicate the potential rate of apple fruit softening might be already established at the fruitlet stage (Nelmes and
Preston, 1968). The mechanism of pectin becoming water-soluble
is still unclear, but since these pectin populations become larger in
average molecular weight during growth and ripening (Fig. 1AD),
it seems unlikely to occur simply through enzymatic degradation of
the pectic backbone by polygalacturonase. However, some involvement of polygalacturonase in this process seems evident, as Royal
Gala apple fruit down-regulated in MdPG1 polygalacturonase have
reduced amounts of water-soluble pectin (Atkinson et al., 2012).
A high Gal content has been associated with cell elongation in
several tissues, followed by a decline at maturation as cell walls
became less extensible (Willats et al., 1999; Stolle-Smits et al.,
1999; McCartney et al., 2003; Leboeuf et al., 2004). Expanding apple
fruit showed a similar correlation, with the most highly branched
pectin and highest Gal content in fruitlets and in rapidly expanding fruit. Cell wall Gal content and RG-I branching were higher
in Scifresh than in Royal Gala during these early developmental

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

135

Fig. 5. Immunouorescence labelling of (1 4)--D-galactan in Royal Gala (AD) and Scifresh (EH) apple cortex tissue. The (1 4)--D-galactan epitope was detected
using monoclonal antibody LM5 (pink) and compared with cellulose stain calcouor (blue). Bar in panel A = 10 m for all micrographs. pl, plasmodesmata piteld; ml, middle
lamella.

stages (Figs. 2 and 3). Altered cell wall Gal content can also affect
the physical properties of the cell wall and tissue. In pea cotyledons,
pectic galactan appeared as a thin layer at the inner surface of the
wall late in development, and was correlated with a doubling of the
rmness of the cotyledons (McCartney et al., 2000). The increase in

unbranched arabinan (LM13 epitope, Fig. 6) at the mature and ripe

and Carpita (2004),
stages is consistent with the ndings of Pena
who reported that the loss of branched arabinans accompanied
softening in stored apple fruit. The high mobility of arabinan in
ripe Scifresh and Royal Gala apple fruit, possibly resulting from

136

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

Fig. 6. Glycan array proling of cell wall extracts from Royal Gala (RG) and Scifresh (SF) apple fruit during development. The relative abundances of three glycan epitopes:
linear tetrasaccharide in (1 4)--D-galactan, linear pentasaccharide in (1 5)--l-arabinan and long linear (1 5)--l-arabinan, as detected by monoclonal antibodies
(mAb) LM5, LM6 and LM13, respectively, were analysed in the water, Na2 CO3 , 1 M KOH and 4 M KOH-cell wall extracts. The colour intensity is proportional to the relative
mean spot intensity value (scale bar) relative to the highest signal intensity for each antibody, which was set to 100%.

reduced branching, was also observed in a solid-state 13 C nuclear

magnetic resonance study (Ng et al., 2014). Highly mobile arabinans
and galactans may represent those molecules that are attached at
only one end and extend into and ll pores in the wall, whereas less
mobile ones may provide important structural links in the wall,
through covalent bonding to xyloglucan (Popper and Fry, 2005)
and direct attachment to cellulose (Zykwinska et al., 2005, 2007),
where the degree of pectin-cellulose association is dependent on
the relative amount of xyloglucan present (Dick-Prez et al., 2011;
Zykwinska et al., 2008).
In both Scifresh and Royal Gala, BGal activity was high in
fruitlets, decreased to a minimum in mature fruit and increased
again during ripening (Fig. 4). AFase activity showed the same
trend, although the ripening-related increase was small. Wei et al.
(2010) found that both BGal and AFase activity increased during ripening-related softening in Golden Delicious and Fuji, but
where AFase activity was similar between the two cultivars, BGal
activity increased rapidly at the early ripening stage in the faster
softening Golden Delicious. Relative to Royal Gala, in the rm cultivar Scifresh lower Gal loss correlated with lower BGal activity, a
higher cell wall Gal content and more intense galactan labelling of
the cell walls throughout development.
No direct correlation between cell wall Gal content and
fruit softening has been found in other studies (Redgwell and
Harker, 1995; Redgwell et al., 1997b), but Srensen et al. (2000)
showed that a reduction of galactan content in potato tuber cell
walls due to overexpression of a fungal endo-1,4--galactanase
resulted in increased enzymatic solubility of pectin, presumably
through increased wall porosity. In tomato, down-regulation of the
early-expressed BGal isoform TBG4 resulted in rmer fruit, but suppression of isoforms occurring later in softening had no inuence
on fruit rmness, perhaps because greater cell wall porosity had
already been established by then (Smith et al., 2002). Therefore, it
seems possible that BGal action and a reduction in Gal content from
early in development may be an indirect but essential prerequisite

that affects cell wall porosity and increases the accessibility of cell
wall-modifying enzymes to substrates during later developmental
stages.
Scifresh and Royal Gala, two apple cultivars with similar
genetic background, growth behaviour and ethylene production
during ripening, have a different microstructure and softening
behaviour that manifest themselves at the fruitlet stage (Ng et al.,
2013). In this study, we found that even from the fruitlet stage
Scifresh showed less water-soluble pectin and more pectic galactan side chains, especially in cell wall areas of adjacent cells rather
than cell junctions. A reduced pectin solubilisation suggests that
Scifresh cell walls can better retain their integrity, which together
with the lower BGal activity and resultant increased abundance
of galactan side chains in Scifresh could result in lower cell wall
porosity and restrict access of cell wall-modifying enzymes during
fruit development, thereby maintaining wall strength and consequently a slower softening rate. A high cell wall galactan content
and low pectin solubilisation do not necessarily directly correlate
with fruit rmness per se, but appear to be either a prerequisite
or sets up the wall for changes that result in less fruit softening
later in ripening. As cell wall differences were evident between the
fruitlet and expanding stages, future studies aimed at investigating
the softening behaviour of apples should therefore target this rapid
growth phase for genes and enzymes involved not only in the modication but also in the synthesis of cell wall polysaccharides. More
cultivars need to be studied to verify that the differences between
the faster softening Royal Gala and slower softening Scifresh are
applicable to a wider range of genotypes with different softening characteristics, and furthermore, whether these differences are
generic features of fruit growth and ripening-related softening.
Acknowledgements
This research was funded by The New Zealand Ministry of Business, Innovation and Employment (CO6X0705). The authors thank

J.K.T. Ng et al. / Journal of Plant Physiology 176 (2015) 129137

Sarah Johnston for the gift of the kiwifruit BGal antibody, Isabel
Moller for help with setting up the microarray robot, and Ross
Atkinson for critically reading the manuscript.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.jplph.
2014.12.012.
References
Ahmed AER, Labavitch JM. A simplied method for accurate determination of cell
wall uronide content. J Food Biochem 1977;1:3615.
Albersheim P, Nevins DJ, English PD, Karr A. A method for the analysis of sugars in
plant cell-wall polysaccharides by gas-liquid chromatography. Carbohydr Res
1967;5:3405.
Atkinson RG, Sutherland PW, Johnston SL, Gunaseelan K, Hallett IC, Mitra D, et al.
Down-regulation of POLYGALACTURONASE1 alters rmness, tensile strength and
water loss in apple (Malus x domestica) fruit. BMC Plant Biol 2012;12:129.
Billy L, Mehinagic E, Royer G, Renard CMGC, Arvisenet G, Prost C, et al. Relationship
between texture and pectin composition of two apple cultivars during storage.
Postharvest Biol Technol 2008;47:31524.
Blumenkrantz N, Asboe-Hansen G. New method for quantitative determination of
uronic acids. Anal Biochem 1973;54:4849.
Brummell DA. Cell wall disassembly in ripening fruit. Funct Plant Biol
2006;33:10319.
Dick-Prez M, Zhang Y, Hayes J, Salazar A, Zabotina OA, Hong M. Structure and
interactions of plant cell-wall polysaccharides by two- and three-dimensional
magic-angle-spinning solid-state NMR. Biochemistry 2011;50:9891000.
Fischer M, Arrigoni E, Amado R. Changes in the pectic substances of apples during
development and postharvest ripening. Part 2: analysis of the pectic fractions.
Carbohydr Polym 1994;25:16775.
Galvez-Lopez D, Laurens F, Quemener B, Lahaye M. Variability of cell wall polysaccharides composition and hemicellulose enzymatic prole in an apple progeny.
Int J Biol Macromol 2011;49:11049.
Goulao LF, Santos J, deSousa I, Oliveira CM. Patterns of enzymatic activity of cell
wall-modifying enzymes during growth and ripening of apples. Postharvest Biol
Technol 2007;43:30718.
Gross KC, Sams CE. Changes in cell wall neutral sugar composition during fruit
ripening: a species survey. Phytochemistry 1984;23:245761.
Gwanpua SG, Van Buggenhout S, Verlinden BE, Christiaens S, Shpigelman A, Vicent
V, et al. Pectin modications and the role of pectin-degrading enzymes during
postharvest softening of Jonagold apples. Food Chem 2014;158:28391.
Harker FR, Redgwell RJ, Hallett IC, Murray SH, Carter G. Texture of fresh fruit. Hortic
Rev 1997;20:121224.
Harris PJ, Blakeney AB, Henry RJ, Stone BA. Gas chromatographic determination
of the monosaccharides of plant cell wall preparations. J Assoc Off Anal Chem
1988;71:2725.
Ito S, Mitsuishi Y, Okuno T, Kato Y. Changes in the structure of xyloglucan of apple
fruit during development. J Japan Soc Hortic Sci 2004;73:516.
Lakso AN, Corelli-Grappadelli L, Barnard J, Gofnet MC. An expolinear model of the
growth pattern of the apple fruit. J Hortic Sci 1995;70:38994.
Leboeuf E, Thoiron S, Lahaye M. Physico-chemical characteristics of cell walls from
Arabidopsis thaliana microcalli showing different adhesion strengths. J Exp Bot
2004;55:208797.
Longhi S, Hamblin MT, Trainotti L, Peace CP, Velasco R, Costa F. A candidate gene
based approach validates Md-PG1 as the main responsible for a QTL impacting
fruit texture in apple (Malus x domestica Borkh). BMC Plant Biol 2013;13:37.
Marcus SE, Verhertbruggen Y, Herve C, Ordaz-Ortiz JJ, Farkas V, Pedersen HL, et al.
Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant
cell walls. BMC Plant Biol 2008;8:60.
McCartney L, Ormerod AP, Gidley MJ, Knox JP. Temporal and spatial regulation of
pectic (1 4)--D-galactan in cell walls of developing pea cotyledons: implications for mechanical properties. Plant J 2000;22:10513.
McCartney L, Steele-King CG, Jordan E, Knox JP. Cell wall pectic (1 4)--D-galactan
marks the acceleration of cell elongation in Arabidopsis seedling root meristem.
Plant J 2003;33:44754.
Melton LD, Smith BG. Isolation of plant cell walls and fractionation of cell wall
polysaccharides. In: Wrolstad RE, editor. Handbook of food analytical chemistry:
water, proteins, enzymes, lipids and carbohydrates. New Jersey, USA: John Wiley
& Sons Hoboken; 2005. p. 697719.
Moller IE, Pettolino FA, Hart C, Lampugnani ER, Willats WGT, Bacic A. Glycan proling of plant cell wall polymers using microarrays. J Visual Exp 2012;70:
e4238.

137

Nelmes BJ, Preston RD. Wall development in apple fruits: a study of the life history
of a parenchyma cell. J Exp Bot 1968;19:496518.
Ng JKT, Schrder R, Sutherland PW, Hallett IC, Hall MI, Prakash R, et al. Cell wall
structures leading to cultivar differences in softening rates develop early during
apple (Malus x domestica) fruit growth. BMC Plant Biol 2013;13:183.
Ng JKT, Zujovic ZD, Smith BG, Johnston JW, Schrder R, Melton LD. Solid-state 13 C
NMR study of the mobility of polysaccharides in the cell walls of two apple
cultivars of different rmness. Carbohydr Res 2014;386:16.
Nobile PM, Wattebled F, Quencini V, Girardi CL, Lormeau M, Laurens F. Identication
of a novel -l-arabinofuranosidase gene associated with mealiness in apple. J
Exp Bot 2011;62:430921.
Oraguzie NC, Volz RK, Whitworth CJ, Bassett HCM, Hall AJ, Gardiner SE. Inuence of
Md-ACS1 allelotype and harvest season within an apple germplasm collection on
fruit softening during cold air storage. Postharvest Biol Technol 2007;44:2129.
MJ, Carpita NC. Loss of highly branched arabinans and debranching of rhamnoPena
galacturonan I accompany the loss of rm texture and cell separation during
prolonged storage of apple. Plant Physiol 2004;135:130513.
Percy AE, Melton LD, Jameson PE. Xyloglucan and hemicelluloses in the cell wall
during apple fruit development and ripening. Plant Sci 1997;125:319.
Popper ZA, Fry SC. Widespread occurrence of a covalent linkage between xyloglucan
and acidic polysaccharides in suspension-cultured angiosperm cells. Ann Bot
2005;96:919.
Redgwell RJ, Harker R. Softening of kiwifruit discs effect of inhibition of galactose
loss from cell walls. Phytochemistry 1995;39:131923.
Redgwell RJ, Melton LD, Brasch DJ. Cell wall dissolution in ripening kiwifruit (Actinidia deliciosa). Solubilization of the pectic polymers. Plant Physiol 1992;98:
7181.
Redgwell RJ, MacRae EA, Hallett I, Fischer M, Perry J, Harker R. In vivo and in vitro
swelling of cell walls during fruit ripening. Planta 1997a;203:16273.
Redgwell RJ, Fisher M, Kendal E, MacRae EA. Galactose loss and fruit ripening: highmolecular weight arabinogalactans in the pectin polysaccharides of fruit cell
walls. Planta 1997b;203:17481.
Ross GS, Wegrzyn T, MacRae EA, Redgwell RJ. Apple -galactosidase. Activity against
cell wall polysaccharides and characterization of a related cDNA clone. Plant
Physiol 1994;106:5218.
Siddiqui S, Brackmann A, Streif J, Bangerth F. Controlled-atmosphere storage
of apples: cell wall composition and fruit softening. J Hortic Sci 1996;71:
61320.
Smith DL, Abbott JA, Gross KC. Down-regulation of tomato beta-galactosidase 4
results in decreased fruit softening. Plant Physiol 2002;129:175562.
Srensen SO, Pauly M, Bush M, Skjt M, McCann MC, Borkhardt B, et al. Pectin
engineering: modication of potato pectin by in vivo expression of an endo1,4--D-galactanase. Proc Natl Acad Sci 2000;97:763944.
Stolle-Smits T, Beekhuizen JG, Kok MTC, Pijnenburg M, Recourt K, Derksen J, et al.
Changes in cell wall polysaccharides of green bean pods during development.
Plant Physiol 1999;121:36372.
Tong C, Krueger D, Vickers Z, Bedford D, Luby J, El-Shiekh A, et al. Comparison of
softening-related changes during storage of Honeycrisp apples, its parents, and
Delicious. J Am Soc Hortic Sci 1999;124:40715.
Verhertbruggen Y, Marcus SE, Haeger A, Verhoef R, Schols HA, McCleary BV, et al.
Developmental complexity of arabinan polysaccharides and their processing in
plant cell walls. Plant J 2009;59:41325.
Volz RK, Harker FR, Lang S. Firmness decline in Gala apple during fruit development.
J Am Soc Hortic Sci 2003;128:797802.
Wakasa Y, Kudo H, Ishikawa R, Akada S, Senda M, Niizeki M, et al. Low expression of
an endopolygalacturonase gene in apple fruit with long-term storage potential.
Postharvest Biol Technol 2006;39:1938.
Wei J, Ma F, Shi S, Qi X, Zhu X, Yuan J. Changes and postharvest regulation of activity
and gene expression of enzymes related to cell wall degradation in ripening
apple fruit. Postharvest Biol Technol 2010;56:14754.
Willats WGT, Marcus SE, Knox JP. Generation of a monoclonal antibody specic to
(1 5)--l-arabinan. Carbohydr Res 1998;308:14952.
Willats WG, Steele-King C, Marcus SE, Knox JP. Side chains of pectic polysaccharides are regulated in relation to cell proliferation and cell differentiation. Plant
J 1999;20:61928.
Yoshioka H, Aoba K, Kashimura Y. Molecular weight and degree of methoxylation in
cell wall polyuronide during softening in pear and apple fruit. J Am Soc Hortic
Sci 1992;117:6006.
Yoshioka H, Kashimura Y, Kaneko K. -D-Galactosidase and -larabinofuranosidase activities during the softening of apples. J Japan Soc
Hortic Sci 1995;63:8718.
Zykwinska AW, Ralet MCJ, Garnier CD, Thibault JFJ. Evidence for in vitro binding of
pectin side chains to cellulose. Plant Physiol 2005;139:397407.
Zykwinska A, Thibault J-F, Ralet MC. Organisation of pectic arabinan and galactan
side chains in association with cellulose microbrils in primary cell walls and
related models envisaged. J Exp Bot 2007;58:1795802.
Zykwinska A, Thibault J-F, Ralet MC. Competitive binding of pectin and xyloglucan
with primary cell wall cellulose. Carbohydr Polym 2008;74:95761.