Plant Physiology and Biochemistry 130 (2018) 127–138

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry

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

Research article

Genotypic and environmental effects on the level of ascorbic acid, phenolic T

compounds and related gene expression during pineapple fruit development
and ripening
Mathieu Léchaudela,∗, Marie Darnauderyb, Thierry Joëtc, Patrick Fournierd, Jacques Joase
a
CIRAD, UMR QUALISUD, F-97130, Capesterre-Belle-Eau, Guadeloupe, France
b
CIRAD, UPR HORTSYS, F-97455, Saint-Pierre, Réunion, France
c
IRD, UMR DIADE, BP 64501, F-34394, Montpellier, France
d
CIRAD, UPR GECO, F-97455, Saint-Pierre, Réunion, France
e
CIRAD, UMR QUALISUD, F-34398, Montpellier, France

A R T I C LE I N FO A B S T R A C T

Keywords: Pineapple (Ananas comosus (L.) Merr.) is a non-climacteric tropical fruit whose ripening could be accompanied
Antioxidant by oxidative processes and the concurrent activation of enzymatic and non-enzymatic reactive oxygen species
Cultivar (ROS) scavenging systems. To better understand the variability of these processes among climatic environments
Environment or genotypes in pineapple, the temporal expression dynamics for genes encoding oxidative and antioxidative
Gene expression
stress enzymes were analyzed by real-time RT-PCR during fruit development and ripening, among three culti-
Pineapple
Reactive oxygen species
vars: Queen Victoria, Flhoran 41 and MD-2 hybrid, and in two climatic areas. Pineapple development and
Non-climacteric ripening ripening involved changes in the levels of transcripts encoding for polyphenol oxidase and transcripts involved
in the first steps of the phenylpropanoid pathway and in the balance of ROS, especially those encoding for
ascorbate peroxydase and metallothioneins, regardless of the cultivar. Our results confirm the same dynamic in
gene expression from the two environmental crop areas, however climatic conditions influenced the level of the
expression of the major transcripts studied that were linked to these oxidative and antioxidant metabolisms.
MT3a and MT3b transcripts were not influenced by genetic factor. The genetic effect was not significant on the
various transcripts linked to the first steps of the phenylpropanoid pathway and to phenol oxidation, except 4CL
ones. In ripe pineapple, highly significant relationships were found between the contents in antioxidant meta-
bolites, i.e., ascorbic acid and total phenolic compounds, and the transcript levels of genes involved in the
enzymatic ROS-scavenging system and in the biosynthesis or regeneration of ROS-scavenging compounds, like
phenylpropanoids, ascorbic acid, metallothioneins.

1. Introduction
Pineapple Ananas comosus (L.) Merr. is a multiple fruit consisting of
coalesced berries. It is currently the third most important tropical fruit
after banana and mango in terms of worldwide production. Pineapple is
cultivated mainly for fresh or canned fruit and juice. Pineapple flesh is
rich in biologically-active substances such as vitamins and phenolic
compounds (Gil et al., 2006; Montero-Calderón et al., 2010), which are
important for the human diet due to their antioxidant and anti-in-
flammatory properties (Poiroux-Gonord et al., 2010). Physicochemical
characteristics involved in pineapple quality, including antioxidants,
evolve during the 3–5 months of fruit development. The optimal quality
is reached at a full maturity stage that corresponds to the extent of the
yellow or red-orange skin color area, depending on the cultivar, and to
a high sweetness-to-acidity balance and aroma sensation (Brat et al.,
2004; Wei et al., 2011).
Fleshy fruits are generally classified into two physiological groups,
climacteric and non-climacteric, according to their respiratory activity
and associated ethylene biosynthesis profiles during ripening.
Pineapple is an example of a non-climacteric fruit and can present
oxidative processes during its ripening (Moyle et al., 2005), as observed
for other non-climacteric fruits such as strawberry (Aharoni et al.,

Abbreviations: βtub, beta tubulin; 60sRP, 60s RP; MT3a, MT3 Metallothionein; MT3b, MT3 Metallothionein; PAL, phenylalanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-
coumarate-CoA ligase; PPO1, Polyphenol oxidase; PPO2, Polyphenol oxidase; CAT, Catalase; Fe-SOD, Fe Superoxide dismutase; [Cu/Zn]-SOD, [Cu/Zn] Superoxide dismutase; APx1,
Ascorbate peroxidase; APx3, Ascorbate peroxidase; GR, Glutathione Reductase; MDHAR, Monodehydroascorbate reductase; QV, Queen Victoria; RL41, Flhoran 41; MD2, MD-2 hybrid
∗
Corresponding author.
E-mail address: mathieu.lechaudel@cirad.fr (M. Léchaudel).

https://doi.org/10.1016/j.plaphy.2018.06.041
Received 18 March 2018; Received in revised form 4 June 2018; Accepted 28 June 2018
Available online 30 June 2018
0981-9428/ © 2018 Elsevier Masson SAS. All rights reserved.
M. Léchaudel et al.
2002) and grape (Pilati et al., 2007). In contrast to the well-studied
climacteric fruits such as tomato, apple and banana, the process of
development and ripening of non-climacteric fruits is less well known
(Cherian et al., 2014). This oxidative processes require an increase in
reactive oxygen species (ROS), which are highly reactive molecules that
may cause oxidative damage to biological macromolecules, including
proteins, DNA and lipids (Apel and Hirt, 2004). Various cellular enzy-
matic and non-enzymatic mechanisms have the capacity to scavenge
ROS to modulate their steady-state concentrations, avoiding their po-
tential toxicity while allowing them to function as signal molecules
(Mittler et al., 2004). A few studies have reported that the ripening of
non-climacteric fruit is accompanied by, in particular, the accumulation
of transcripts of genes involved in the enzyme-mediated scavenging
system and in the antioxidant compound metabolism and regeneration
(Aharoni and O'Connell, 2002; Pilati et al., 2007). Further studies are
required to analyze genotypic and environmental effects on expression
profiling of these ripening-related genes, as described in this current
report.
The enzymatic ROS scavenging mechanisms in plants include su-
peroxide dismutase (SOD), catalase (CAT), the ascorbate peroxidase,
and the ascorbate-glutathione cycle (Apel and Hirt, 2004). Indeed,
hydrogen peroxide can be converted into water, involving the oxidation
and the regeneration of ascorbate, glutathione and NADPH by enzymes
such as ascorbate peroxydase (APX), gluthatione reductase (GR) and
monodehydroascorbate reductase (MDHAR), as described by Apel and
Hirt (2004). The levels of the transcripts of genes involved in the en-
zymatic detoxification of ROS have been shown to be modulated during
non-climacteric ripening, especially those of APX and SOD in grape and
pineapple (Koia et al., 2012; Pilati et al., 2007), and those of CAT in
citrus (Peroni et al., 2007), which were up-regulated.
In addition to enzymatic scavenging pathways, fruit cells have a
network of low-molecular-mass antioxidant molecules such as ascor-
bate or phenolic compounds, which are also involved in non-enzymatic
protection against oxidative stress and damage caused by ROS. Ascorbic
acid is one of the most frequently studied and powerful antioxidants
(Smirnoff, 2000). Phenolic compounds possess ideal structural chem-
istry for free radical and hydrogen peroxide scavenging activities, and
for altering peroxidation kinetics (Takahama and Oniki, 1997). These
antioxidants can work together to limit oxidative stress (Takahama and
Oniki, 1997). H2O2 can also be reduced by peroxidases using phenolics
as primary electron donors, making it possible for ascorbic acid to re-
duce phenoxyl radicals generated by this oxidation. The levels of these
antioxidants have been reported to vary during fruit ripening. An in-
creased contribution of the reduced form of ascorbate to the total as-
corbate pool was observed in berries at the latter stages of ripening
(Melino et al., 2009). At these stages, the expression of MDHAR and
dehydroascorbate reductase (DHAR) genes encoding ascorbic acid re-
cycling enzymes was up-regulated. Studies on grape berries and
strawberries, two non-climacteric fruit models that accumulate phe-
nolic compounds, reported that fruit ripening was associated with the
synthesis of metabolites from the phenylpropanoid pathway, assisted by
the up-regulation of phenylalanine ammonia lyase (PAL), cinnamic acid
4-hydroxylase (C4H) and 4-coumarate:CoA ligase (4CL) genes, involved
in the first three committed steps of the pathway (Almeida et al., 2007;
Sweetman et al., 2012).
The phenolic composition of fruits may be modified by oxidative
reactions involved in the antioxidant activity of the phenols and oxi-
dative browning. In fact, phenolic compounds are substrates for enzy-
matic browning reactions that are initially induced by enzymatic oxi-
dation of phenolic compounds by polyphenol oxidases (PPO) and that
form colored quinones. These reactions significantly diminish consumer
acceptance, storage life and the value of the fruit products, especially in
the case of pineapple, which is known to develop browning symptoms
(Raimbault et al., 2010; Zhou et al., 2003a). However, high levels of
PPO activity reduce the free-oxygen level available for ROS production
and may be one of elements that contribute to the lowering of the
128
Plant Physiology and Biochemistry 130 (2018) 127–138
cellular ROS level during stress conditions (Ortega-García and Peragón,
2009). Thus, the induction of PPO expression has been linked to fruit
tolerance to stress conditions. Indeed, the two PPO genes identified in
pineapple have been shown to be highly up-regulated in response to
chilling and wounding (Stewart et al., 2001).
Another group of low-molecular-weight compounds are me-
tallothioneins, small cysteine-rich proteins that provide thiols for metal
chelation. Metallothioneins are not only involved in maintaining
homeostasis of essential metals and metal detoxification, but are also
involved in a range of physiological processes, including ROS scaven-
ging (Mir et al., 2004). Various studies have shown that the expression
of metallothionein genes increased during fruit development and was
highest in mature fruit (Moyle et al., 2005; Pandit et al., 2010). These
authors suggested that metallothioneins may be involved in scavenging
ROS generated during fruit ripening.
The molecular mechanisms involved in non-climacteric fruit ri-
pening have been studied in grape (Pilati et al., 2007), pea (Matamoros
et al., 2010) and strawberry (Aharoni et al., 2002) to gain a broader
knowledge about the variability of fruit quality and maturity. Recent
studies on pineapple analyzed changes in gene expression during ri-
pening of Smooth Cayenne cv (Koia et al., 2012) and involved in the
cold response (Chen et al., 2016). Little is known about the effects of
the climatic environments and of the genotypes on the expression of
genes related to the oxidative balance since non-climacteric fruits may
contain a transcriptional program responsive to oxidative stress in-
duced during ripening (Aharoni and O'Connell, 2002). A better un-
derstanding would contribute to adjusting growing practices according
to the cultivar and the climatic area in order to produce high-quality
fruit in a changing environment. The temporal dynamics of the ex-
pression of genes coding oxidative and antioxidative stress enzymes
were therefore analyzed in pineapple fruit. For this purpose, the levels
of transcripts of several ROS scavenging enzymes and of antioxidant
molecule synthesis and recycling were identified during the develop-
ment and ripening of pineapple. To evaluate the seasonal and cultivar-
specific variations, three cultivars, Queen Victoria, Flhoran 41 and MD-
2 hybrid, for which no molecular markers approach exists, and two
climatic areas were studied at different development stages. In ripe
fruits, the relationships between antioxidant compound content, i.e.,
ascorbic acid and total phenolic compounds, and transcript levels re-
lated to oxidative stress and antioxidant system are discussed.
2. Materials and methods
2.1. Experimental sites and fruit material
Pineapple fruits (Ananas Comosus Merr L.) from three cultivars,
Queen Victoria (QV) clone ‘RE43’, Flhoran 41 (RL41) clone ‘RL41’, and
MD-2 hybrid (MD2) clone ‘DLNE’, were grown in two different climatic
areas in Reunion Island. The first field was located in the southwest of
the island at CIRAD's Bassin Plat Research Station, St Pierre, 150 m
above sea level (55°29′20.64″E, 21°19′21.62″S), corresponding to a
drier climatic area than the eastern one, with a cumulative rainfall of
380 mm and an average temperature of 24.1 °C during the period of
fruit growth studied. The second field was located at 290 m above sea
level in the east of the island, St Benoit (55°42′12.86″E, 21°05′53.85″S),
corresponding to a very wet area with a cumulative rainfall of 1680 mm
and an average temperature of 21.7 °C during the period of fruit growth
studied. Each field was identically managed following the locally re-
commended cultural practices (Fournier, 2011).
Fruits from QV of the East field were not studied due to the risk of
the development of fruitlet core rot disease in this climatic area, due to
Fusarium ananatum that is known to involve a response in the accu-
mulation of phenolic acids (Barral et al., 2017).
Fruits were harvested at five development stages from 30 to 140
days after flowering. Flowering was defined as 50% of inflorescences
with at least one open corolla. The different developmental stages were
M. Léchaudel et al.
selected on the basis of the thermal time sum (Léchaudel et al., 2010) as
of flowering for the three first stages since the peel color of pineapple is
green at these stages, and on the basis of the peel color of pineapples for
the two last maturity stages since changes in peel color occur according
to fruit ripening at these stages. Fruits were harvested at 520 (H1), 750
(H2) and 1080 (H3) degree days, with a base temperature of 10 °C
(Léchaudel et al., 2010). For the last two harvest stages, fruits were
harvested at a turning stage (H4) corresponding to the beginning of
changes in peel color, i.e., yellow for the QV and MD2 cultivars and red
for the RL41 cultivar, and at a ripe stage (H5) corresponding to a totally
colored fruit. Four to eight fruits were selected per condition, i.e., for
each maturity stage, each pineapple cultivar, and each climatic area.
Fruits were weighed with and without their crown and the number of
fruitlets per spiral was counted. After every harvest, peel tissues of each
fruit were excised and pulp tissues were sampled from lower to upper
parts of fruit and randomly separated in two flesh sub-samples. The first
one was immediately frozen in liquid nitrogen. Frozen pulp tissues were
mixed and then stored at −80 °C to study the expression profiling of the
various genes studied and for phenolic compounds analysis. The second
flesh sample was used to prepare pulp juice for chemical analysis to
characterize each development stage, i.e., titratable acidity, and total
soluble solids content of the flesh. This second sample of flesh was
mixed using a Grindomix blender (Retsch, Haan, Germany) to prepare a
volume of pineapple juice needed to measure ascorbic acid content,
titratable acidity, and total soluble solids content of the flesh. Total
soluble solids content was measured using a refractometer (ATC-1E,
Atago, Tokyo, Japan). Titratable acidity was determined by titrating
10 mL of fruit juice with a 0.1 mol L−1 NaOH solution, using phe-
nolphthalein as a color indicator.
2.2. Determination of ascorbic acid and total phenolic compound content
The amount of ascorbic acid in the juice sample (in mg 100 mL−1)
of each fruit was determined according to the titration method based on
the reduction of the blue dye, 2, 6-dichlorophenolindophenol, by as-
corbic acid (AOAC, 1998). The endpoint of the titration is indicated by
the appearance of the pink acid form of the dye. Five mL of each juice
sample were treated with 5 mL of metaphosphoric acid (3%), and then
titrated against a freshly standardized 2, 6-dichlorophenolindophenol
solution. A standard curve was obtained with 10 mL of standard as-
corbic acid solution at different concentrations.
The analysis of total phenolic compounds was conducted with an
ultrasound-assisted technique, using the sample of mixed pulp stored at
−80 °C after its lyophilization. Total phenolic compounds were ex-
tracted by adding 20 mL of 80% ethanol to 1 g of powdered sample. To
improve compound extraction, the mixture was briefly vortexed and
incubated in a sonicating bath (Transsonic TI-H, Elma Hans
Schmidbauer GmbH & Co., Singen, Germany) at 35 °C for 10 min, in a
continuous mode at a fixed frequency of 35 kHz, then filtered and
evaporated. The dry extract was solubilized with 100% methanol,
transferred and adjusted in a 10-mL graduated flask, and then stored at
−80 °C for 1 h for the precipitation of sugars and residues of fiber and
proteins (extract A). One mL of the supernatant was added with 100 mg
of polyvinylpolypyrrolidone for adsorption of phenolic compounds, and
then homogenized and incubated in the dark for 12 h at 4 °C on an
orbital shaker. The homogenate was centrifuged at 1000 g for 5 min at
4 °C and the supernatant was recovered (extract B). One hundred μL of
extracts A and B were added to 2.5 mL of Folin-Ciocalteau reagent,
briefly vortexed and incubated for 2 min at 20 °C. Two mL of an aqu-
eous solution containing 7.5 g Na2CO3 per 100 mL were added to the
two mixtures containing either extract A or B, briefly vortexed, and
incubated for 15 min at 50 °C. The reaction was stopped by placing the
reaction tubes in an ice bath. The absorbance of each mixture was read
at 750 nm with a Thermospectronic Helios spectrophotometer (Thermo
Fisher Scientific, Inc., Waltham, MA, USA) to determine the total con-
centration of phenols by difference. The concentrations of the phenols
129
Plant Physiology and Biochemistry 130 (2018) 127–138
were calculated using a standard curve of gallic acid and expressed as
milligrams of gallic acid per 100 g of fresh weight.
2.3. RNA extraction and retro-transcription
First, the stored samples of frozen pulp of each fruit were lyophi-
lized before being ground with a mortar in liquid nitrogen. For each
sample corresponding to one fruit per condition, 20 mg of lyophilized
powder were prepared in a 1.5-mL tube. RNA extraction was carried out
with the RNeasy® Mini Kit (Qiagen, Valencia, CA, USA) according to the
manufacturer's instructions (RNeasy Plant Mini Kit protocol) with β-
mercaptoethanol (1/5) added to the buffer before homogenization with
an Ultra-Turrax T10 (IKA, Staufen, Germany). RNA samples were
treated with RQ1 RNase free DNase I (Promega, Madison, WI, USA). To
clean and concentrate RNA samples, RNeasy MiniElute Cleanup col-
umns (Qiagen, Valencia, CA, USA) were then used. Yield and purity
were determined by spectrophotometry with a Nanodrop 8000 (Thermo
Fisher Scientific, Inc., Waltham, MA, USA), with a ratio greater than 1.9
between the absorbance at 260 nm and at 280 nm.
Two independent reverse transcription reactions were performed
for each sample using the ImPro-II reverse transcription system
(Promega, Madison, WI, USA) as recommended by the manufacturer's
protocol, using a total of 1 μg of total RNA as a template in a 20-μL
reaction mixture. Identical reactions without the reverse transcriptase
were performed to check for the absence of genomic DNA (no-RT
controls). The cDNA samples were then diluted (1:200).
2.4. Primer design
The nucleotide sequences used to design primers were derived from
the GenBank Nucleotide Sequence Database (NCBI). Primers were de-
signed using Primer3 software (http://frodo.wi.mit.edu/cgi-bin/
primer3/primer3_www.cgi) with the following criteria: to amplify
PCR products from 90 to 150 bp, to obtain a primer size of between 20
and 22 bp, a melting temperature (Tm) of between 61 and 63 °C, and a
GC content comprised between 40 and 60%. All primer pairs (Table 1)
were tested for their efficiency to amplify the specific target cDNA from
different fruit development stages. The specificity of the PCR products
generated for each set of primers was tested in several ways: (1) by
checking the ‘dissociation curve’, the melt gradient in which fluores-
cence decreases at a single discrete temperature, indicating separation
of the two strands of a single cDNA species; (2) by assessing correct
band size and absence of other non-specific products by agarose gel
electrophoresis of PCR products (Fig. S1); and (3) by purification of
PCR products with the MineElute PCR Purification Kit (Qiagen, Va-
lencia, CA, USA) and further sequencing of all amplicons using specific
primers.
2.5. Real-time RT-PCR assays
Real-Time PCR was performed in an optical 96-well plate with an
ABI Prism 7000 sequence detection system (Applied Biosystems, Foster
City, CA, USA), using SYBR Green to monitor dsDNA synthesis.
Reactions contained 5 μl cDNA, 6.5 μl SYBR Green Master Mix reagent
and a final concentration of 0.2 μM (each) for forward and reverse
primers. The cycling conditions for amplification included 2 min at
50 °C for incubation, 10 min at 95 °C for the initial polymerase activa-
tion step, followed by 45 cycles at 95 °C for 15 s and at 60 °C for 1 min.
Data were analyzed using SDS 2.1 software (Applied Biosystems,
Thermo Fisher Scientific, Inc., Waltham, MA, USA) to determine cycle
threshold (Ct) values. The PCR efficiency (E) was estimated for each
gene with LinReg software (Heart Failure Research Center, Amsterdam,
The Netherlands), which uses absolute fluorescence data captured
during the exponential phase of amplification of each reaction using the
equation (1 + E) = 10(−1/slope) (Ramakers et al., 2003). Most primer
pairs had efficiencies comprised between 1.8 and 2 (Table 1). Efficiency
M. Léchaudel et al. Plant Physiology and Biochemistry 130 (2018) 127–138

Table 1
Details of primers and annotation used for real-time RT-PCR. PCR efficiencies were estimated for each primer pair using LinRegPCR Software (Ramakers et al., 2003).
Gene ID GeneBank Putative function Primer Sequence (5′-3′) Forward and Product size PCR Best match Accession E value
Accession reverse (bp) Efficiency (blastx)

MT3a CO732286 Metallothionein TTCTGCCTCATTTATTGGCTCAT 99 2,00 ACB10219 (Elaeis 5E-22

CCGTTTCCCTTCTTCACACAC guineensis)
MT3b CO732227 Metallothionein CCTCGTTTCTCTCTCATCTTTCG 104 1,95 CAB52582 (Elaeis 1E-28
TTTCCATGCCGTAGCTGTTTC guineensis)
PPO1 AY149880 Polyphenol oxidase 1 CGACGACTACCACAGCAGAGA 105 1,97 AAO16863 4E-170
CTTCCTTCTCCTTCCCACTCC
PPO2 AY149882 Polyphenol oxidase 2 GCCTCCGAATACTAGCTCACC 125 1,94 AAK29783 1E-166
CGGTTTGGGACTCTTACAAGG
PAL AY098511 Phenylalanine ammonia-lyase CCGAACAATTCCATCTTCCAA 147 1,96 AAM28276 0
CGGATAAGACCTGCACTCCTG
APx1 DT338077 Ascorbate peroxidase CTGAGCGGAGAGAAGGAAGGT 125 1,99 AEZ00894 (Elaeis 8E-153
GCATAATCGGCAAAGAAAGCA guineensis)
APx3 CO731513 Ascorbate peroxidase GTGAAGGCAAAACATCCAAGG 97 1,99 XP 002444620 5E-162
CAAAGTCAATGGTCGGTCCTC (Sorghum bicolor)
CAT2 DT339258 Catalase 2 TGCCCTGCACAACAATCACTA 91 1,97 ADU56198 (Musa 5E-139
AAGAGGAGTTTGCCCAGAAGG acuminata)
4CL DT338685 4-Coumarate CoA ligase AGGGGCAATTCGTTCTCCTAA 116 2,02 EAY84707 (Oryza 7E-109
CCACATTGACATCCCACAGAA sativa)
C4H DT339530 Cinnamate 4-hydroxylase GGCGACTGGTCAAGAACTTTG 140 1,96 XP 006850963
CCAATTTTACGCCGAGATCG (Amborella trichopoda)
MDHAR CO732023 Monodehydroascorbate AAAGCAAGGGCTCAATCCAG 149 1,95 ADF43731 (Lilium 1E-172
reductase CTCCCCCACTTCCAACACATA longiflorum)
GR DT336648 Glutathione Reductase ATGGGCTGTGGGTGATGTTA 150 2,00 XP 007222141 (Prunus 3E-176
AGGGGAGGGATGGAGAAAAC persica)
Fe-SOD CO730785 [Fe] Superoxide dismutase CCCCAAGTCTCACACGAAAAA 150 1,96 ABF68752 2E-131
TGTTCGCCCCAGTGTAGTTCT (Gymnadenia conopsea)
[Cu/Zn]- CO731743 [Cu/Zn] Superoxide dismutase GGCACATTTCAACCCAAACAA 106 2,00 AAX07164 (Lilium cv.) 7E-100
SOD CTCAGCCACTCCTTCAGCATT
SUI1 CO731666 SUI 1 translation factor GCGTTCTCTTGCCCTAATCC 150 1,94 EEE67305 (Oryza 1E-70
GCAAACGGATCGAAGGTAGTT sativa)
β-act CO731334 Beta actin ATGGAAGCTGCGGGTATTCA 101 1,96 ADV91594 0
CCACCACTGAGCACGATGTT
60sRP CO732287 60s RP GTAACAGCAATCACGCCCAAG 95 2,00 AAG27431 (Elaeis 5E-154
AGTGAATCCCCACTTCCTGCT guineensis)
β-tub DT338989 Beta tubulin AGGCGGAGAGCAACATGAAC 129 1,97 XP 006369648 (Populus 2E-173
CTCCTCGCGGTCTTTACATTTC trichocarpa)

values were taken into account in all subsequent calculations.
2.6. Housekeeping genes selection and relative gene expression calculation
The selection of appropriate reference genes for this study was
performed using geNorm [51] among the 18 genes studied, which in-
clude the four commonly used housekeeping genes, i.e., SUI1, β-actin,
60sRP and β-tubulin. The level of expression of each gene X was nor-
malized to the geometric mean of the expression levels of the two se-
lected reference genes (β-actin and SUI1), according to the formula:
⎛Ct (X ) − ⎛ Ct (R1) + Ct (R2) ⎞⎞
X ⎜
2
⎟ matic area and each developmental stage by analysis of variance (one-
= E⎝ ⎝ ⎠⎠
2
R1 × R2
where E is the estimation of PCR efficiency, Ct is the threshold cycle
and R1, R2 are the 2 selected reference genes. For relative gene ex-
pression analysis, a fold change expression value was computed using
the stage of maximal expression for QV cultivar of the field in the
Southwest area as a calibrator.
2.7. Statistical analysis
Firstly, the statistical significance of variations in gene expression
according to the developmental stage of pineapple fruit and to the
genotypic and environmental effects was verified by analysis of var-
iance (three-way ANOVA, with climatic area (Southwest and East),
cultivar (MD2 and RL41), and harvest (H1, H2, H3, H4 and H5) as
factors) on normalized gene expression values of 16 genes (beta tu-
bulin, 60s RP (Ribosomal Protein), Metallothionein A, Metallothionein
B, Phenylalanine ammonia-lyase, cinnamate 4-hydroxylase, 4-
coumarate-CoA ligase, Polyphenol oxidase 1, Polyphenol oxidase 2,
Catalase, [Fe] Superoxide dismutase, [Cu/Zn] Superoxide dismutase,
Ascorbate peroxidase 1, Ascorbate peroxidase 3, Glutathione reductase,
Monodehydroascorbate reductase).
Secondly, the statistical significance of variations in gene expression
according to the climatic area and the developmental stage of pineapple
fruit was verified for the pineapple cultivars MD2 and RL41 by analysis
of variance (two-way ANOVA, with climatic area (Southwest and East)
and harvest (H1, H2, H3, H4 and H5) as factors) on the normalized
expression of each studied genes.
The genotypic effect on gene expression were verified for each cli-
way ANOVA, with cultivar (QV, MD2 and RL41 for the Southwest area,
and MD2 and RL41 for the East one) as factor). Multiple comparisons
between means of gene expression at the various developmental stages
and for the different cultivars or climatic areas were performed using
the Tukey test to evaluate whether or not these means were sig-
nificantly different.
Principal component analysis (PCA) was performed on the average
of fold change values of the genes studied related to ROS scavenging
and of antioxidant molecule synthesis and recycling and the contents in
ascorbic acid and total phenolic compounds, using the FactoMineR
package (Lê et al., 2008) of R software (Team, 2007) to obtain an
overview of correlation among gene expression profiles and quality
traits.
All statistical analyses were computed using R software (Team,
2007).

130
M. Léchaudel et al. Plant Physiology and Biochemistry 130 (2018) 127–138

Table 2
Characteristics of fruits (fruit fresh mass, in g; fruit fresh mass without crown, in g; fruitlet number; flesh titratable acidity, in meq 100 mL−1; flesh total soluble solid
content, in ° brix; flesh dry matter content, in %; flesh content in total phenolic compounds, in mg 100 g−1 fresh mass, and in ascorbic acid, in mg 100 mL−1)
harvested from the three cultivars in the two climatic areas at five developmental stages. Values are means ± standard errors of three to five fruits for each climatic
area, and each developmental stage at harvest and each cultivar.
Climatic area Cultivar Harvest Fruit fresh mass Fruit without crown Fruitlet Titratable Total Soluble Flesh dry matter Total phenolics Ascorbic acid
number acidity Solids content

Southwest MD2 H1 269.2 ± 21.0 204.0 ± 9.8 6±0

H2 643.2 ± 72.0 491.1 ± 77.4 8±1 2.8 ± 0.2 5.1 ± 0.2 6.6 ± 0.1 24.8 ± 0.7
H3 1267.5 ± 83.6 857.0 ± 62.0 11 ± 1 8.5 ± 0.6 8.5 ± 1.4 6.5 ± 0.2 40.9 ± 3.5
H4 1243.8 ± 142.5 955.8 ± 118.2 11 ± 1 12.4 ± 1.5 12.1 ± 0.6 13.8 ± 0.9 40.3 ± 3.5 44.7 ± 3.0
H5 1303.2 ± 119.4 1071.8 ± 83.8 10 ± 0 17.3 ± 1.3 11.6 ± 0.4 12.4 ± 0.8 58.5 ± 0.7 53.7 ± 5.1
QV H1 322.3 ± 55.9 277.7 ± 53.5 12 ± 1
H2 453.4 ± 10.1 383.8 ± 13.3 15 ± 1 3.9 ± 0.6 6.5 ± 0.2 8.0 ± 0.2 17.0 ± 0.7
H3 732.5 ± 28.6 556.3 ± 33.3 14 ± 1 8.0 ± 0.2 11.5 ± 0.3 12.3 ± 0.6 18.3 ± 1.7
H4 772.7 ± 30.0 628.8 ± 29.0 14 ± 1 15.6 ± 1.3 14.9 ± 0.7 16.2 ± 0.5 20.6 ± 1.4 23.3 ± 2.0
H5 771.2 ± 62.2 679.0 ± 59.6 11 ± 1 14.8 ± 0.8 15.6 ± 0.8 16.5 ± 0.7 23.5 ± 0.8 25.0 ± 2.4
RL41 H1 277.2 ± 8.6 240.7 ± 10.5 9±0
H2 498.4 ± 46.3 411.5 ± 48.6 8±1 3.7 ± 0.2 6.0 ± 0.0 7.5 ± 0.1 18.7 ± 0.9
H3 723.8 ± 35.5 517.8 ± 24.5 9±1 8.6 ± 0.6 8.1 ± 0.7 9.7 ± 0.7 16.6 ± 2.7
H4 1086.2 ± 49.7 849.2 ± 61.3 9±1 16.1 ± 1.0 12.2 ± 0.3 13.2 ± 0.4 16.6 ± 1.2 21.6 ± 1.1
H5 1202.6 ± 60.6 1106.6 ± 59.3 12 ± 1 17.9 ± 0.8 10.8 ± 1.2 13.2 ± 1.4 25.7 ± 2.3 20.5 ± 2.3
East MD2 H1 430.1 ± 24.1 314.2 ± 30.1 6±1
H2 678.0 ± 149.0 510.5 ± 123.2 10 ± 2 3.7 ± 0.4 5.3 ± 0.1 3.9 ± 0.1 21.6 ± 3.2
H3 722.0 ± 16.9 598.5 ± 12.8 10 ± 2 5.1 ± 0.9 7.2 ± 0.9 12.5 ± 2.4 48.3 ± 2.8
H4 1388.2 ± 53.1 1072.1 ± 47.9 9±1 13.2 ± 1.0 12.5 ± 0.6 13.7 ± 0.7 43.3 ± 1.0 65.6 ± 5.9
H5 1039.0 ± 142.8 815.1 ± 110.9 8±1 16.7 ± 1.0 13.8 ± 0.5 15.1 ± 0.6 31.2 ± 3.2 61.7 ± 3.2
RL41 H1 439.3 ± 22.0 355.7 ± 27.52 6±0
H2 793.8 ± 42.2 593.8 ± 30.19 10 ± 0 4.1 ± 0.9 6.0 ± 0.0 4.5 ± 0.5 17.3 ± 0.4
H3 1019.0 ± 49.3 787.5 ± 74.37 10 ± 0 6.9 ± 0.2 7.0 ± 0.2 8.2 ± 0.5 13.7 ± 1.4
H4 1410.5 ± 79.2 1152.3 ± 91.85 10 ± 1 12.7 ± 0.6 11.1 ± 0.5 11.7 ± 0.6 21.6 ± 1.8 28.8 ± 2.1
H5 1463.9 ± 152.1 1218.2 ± 145.9 10 ± 0 20.5 ± 3.9 13.0 ± 0.5 13.5 ± 0.6 22.4 ± 3.3 22.8 ± 0.9

3. Results
Fruits characteristics, both the size and the various quality traits,
varied according to cultivar and fruit growth conditions (Table 2).
Fruits from QV were smaller than from the two other cultivars related
to smaller fruitlet, as the number of fruits in QV was higher than those
in RL41 and MD2. The titratable acidity and the total soluble solid
content of the flesh increased with the developmental stage, regardless
of the cultivar and the climatic area. The contents in total phenolic
compounds tended to increase with the developmental stage, with
Table 3
Effects of climatic area of crop production, cultivar, and harvest stage on nor-
malized gene expression values of 15 genes (βtub, beta tubulin; 60sRP, 60s RP;
MT3a: MT3 Metallothionein; MT3b: MT3, Metallothionein; PAL, phenylalanine
ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate-CoA ligase;
PPO1, Polyphenol oxidase; PPO2, Polyphenol oxidase; CAT, Catalase; Fe-SOD,
Fe Superoxide dismutase; [Cu/Zn]-SOD, [Cu/Zn] Superoxide dismutase; APx1,
Ascorbate peroxidase; APx3, Ascorbate peroxidase; GR, Glutathione Reductase;
MDHAR, monodehydroascorbate reductase).
Transcripts Climatic area effect Cv. effect Harvest effect

higher values measured in the flesh of MD2 then in the one of QV and β-tub *** n.s. ***
RL41. In ripe fruits, i.e., H4 and H5 stages, the flesh content in ascorbic 60sRP *** *** n.s.
acid was also higher in MD2 than in the two other cultivars (Table 2). MT3a * n.s. ***
The results of the three-way ANOVAs, with climatic area, cultivar, MT3b *** n.s. ***
PAL ** * ***
and developmental stage as factors, indicated that the 16 monitored
C4H *** n.s. ***
genes showed significant changes in expression during pineapple de- 4CL n.s. ** ***
velopment, including the ripening process (P < 0.05, Table 3), except PPO 1 *** n.s. ***
60sRP transcript. Environmental and genetic effects on the 16 mon- PPO 2 *** n.s. ***
Fe-SOD *** *** ***
itored genes were not always significant.
[Cu/Zn]-SOD *** *** ***
Genes encoded proteins that are involved in the structure of the CAT2 *** *** ***
cytoskeleton, i.e., β-tubulin, showed variations of expression with fruit APx1 *** n.s. ***
development (Fig. 1). The expression of genes for 60sRP was transiently APx3 * ** ***
induced, as shown by a slight increase in fold changes between H1 and MDHAR *** *** ***
GR ** n.s. ***
H2 and a decrease of one between H3 and H5, only for MD2 in the
Southwest area. The pattern of β-tub was different since the highest ns, *, **, and *** correspond to the effect of climatic area of crop production,
expression was observed at the first stage, followed by a significant cultivar, or harvest stage: non-significant and significant at P < 0.05,
decline to low expression at the H4 and H5 stages, regardless of the P < 0.01, and P < 0.001, respectively.
cultivar.
Fold changes during fruit development of β-tub and 60sRP were 3.1. Phenolic compound synthesis and oxidation
significantly affected by the climatic area for both MD2 and RL41
(Table 3). For the Southwest climatic area, the levels of these transcripts The expression of PAL, C4H, 4CL, PPO1 and PPO2 significantly
were significantly influenced by the cultivar (Table 4), with the lowest varied during pineapple development, regardless of the cultivar (Fig. 2,
β-tub values for QV and RL41 and the highest 60sRP values for MD2 Table 3). The genotypic effect on the transcript levels of these genes was
(Fig. 1). not significant, except for PAL and 4CL, whereas this later was not in-
fluenced by the climatic area unlike the four other studied genes
(Table 3). These genes related to phenolic compound synthesis and

131
M. Léchaudel et al. Plant Physiology and Biochemistry 130 (2018) 127–138

Fig. 1. Expression dynamics of 7 genes (βtub, beta tubulin; 60sRP, 60s RP; MT3a: MT3 Metallothionein; MT3b: MT3, Metallothionein; CAT, Catalase; Fe-SOD, Fe
Superoxide dismutase; [Cu/Zn]-SOD, [Cu/Zn] Superoxide dismutase) during pineapple fruit development of three cultivars: QV (white bars), MD2 (gray bars) and
RL41 (black bars), in the two climatic areas (Southwest, SW, and East one), as revealed by relative quantification PCR. The maximum transcript level of the QV
cultivar was used as the calibrator for relative gene expression analysis. For each harvest stage and each cultivar, values are means of three to four fruits and bars
indicate standard errors.
ns, *, **, and *** correspond to the effect of cultivar for each harvest stage: non-significant and significant at P < 0.05, P < 0.01, and P < 0.001, respectively.
Different letters (in normal, bold, and italics for QV, MD2 and RL41, respectively) signify that the averages of gene expression levels were different between harvest
stages at P < 0.05 (according to Tukey's multiple comparison test), for each cultivar.

Table 4 oxidation were strongly induced during ripening, as seen by the sig-
Effects of cultivar and harvest stage for each climatic area of crop production on nificant highest fold change values either at H4 or at H5, stages cor-
normalized gene expression values of 15 genes (βtub, beta tubulin; 60sRP, 60s responding to visual changes in the external fruit color, indicating that
RP; MT3a: MT3 Metallothionein; MT3b: MT3, Metallothionein; PAL, phenyla- ripening had begun. The effect of the developmental stage on the levels
lanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate-CoA of these transcripts was confirmed for the East area, especially for MD2
ligase; PPO1, Polyphenol oxidase; PPO2, Polyphenol oxidase; CAT, Catalase;
(Fig. 2). Significantly higher levels of 4CL and PPO2 transcripts were
Fe-SOD, Fe Superoxide dismutase; [Cu/Zn]-SOD, [Cu/Zn] Superoxide dis-
measured as of the second stage of development in the East area. This
mutase; APx1, Ascorbate peroxidase; APx3, Ascorbate peroxidase; GR,
Glutathione Reductase; MDHAR, monodehydroascorbate reductase).
earlier induction of gene expression was observed as of the H2 stage,
both for the MD2 and RL41 cultivars. For the MD2 cultivar (Fig. 2), the
Transcripts Southwest climatic area East climatic area level of these three transcripts continued to increase until fruit ripening,
Cv. effect Harvest effect Cv. effect Harvest effect
whereas for RL41 (Fig. 2), stagnation and a decrease after the H3 de-
velopment stage were observed for the levels of 4CL and PPO2 tran-
β-tub *** *** n.s. *** scripts, respectively. In contrast, in the Southwest area, the expression
60sRP *** n.s. n.s. n.s. of PPO1 was not significantly influenced by the cultivar for any of the
MT3a n.s. *** n.s. ***
MT3b *** *** n.s. ***
five development stages (Table 4, Fig. 2). An increase in the fold change
PAL ** *** n.s. *** values of PPO1 was observed with ripening. A strong climatic area ef-
C4H * *** n.s. *** fect on the pattern of PPO1 expression was observed as well, with an
4CL * *** * *** induction of expression from the H2 development stage in fruits from
PPO 1 n.s. ** n.s. ***
the second area. The last transcript studied linked to phenolic synthesis,
PPO 2 * *** n.s. ***
Fe-SOD *** * *** i.e., C4H, had a pattern of expression during fruit development that
[Cu/Zn]-SOD *** *** ** *** depended on the cultivar (Fig. 2). No effect of developmental stage was
CAT2 n.s. *** ** *** observed for QV pineapple, whereas a significant induction of C4H
APx1 *** *** n.s. *** expression occurred during ripening of MD2 and RL41 pineapples
APx3 n.s. *** * ***
MDHAR n.s. *** *** ***
(Fig. 2). Moreover, as for the other genes linked to the phenylpropanoid
GR n.s. *** n.s. *** pathway, a strong climatic area effect was noted, with an earlier in-
crease of C4H mRNA levels as of the H2 development stage, both for the
ns, *, **, and *** correspond to the effect of cultivar or harvest stage for each RL41 and MD2 cultivars.
climatic area: non-significant and significant at P < 0.05, P < 0.01, and
P < 0.001, respectively.

132
M. Léchaudel et al. Plant Physiology and Biochemistry 130 (2018) 127–138

Fig. 2. Expression dynamics of 5 genes related to the phenolic pathway PAL, phenylalanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate-CoA
ligase; PPO1, Polyphenol oxidase; PPO2, Polyphenol oxidase) during pineapple fruit development of three cultivars: QV (white bars), MD2 (gray bars) and RL41
(black bars), in the two climatic areas (South West, SW, and East one), as revealed by relative quantification PCR. The maximum transcript level of the QV cultivar
was used as the calibrator for relative gene expression analysis. For each harvest stage, values are means of three to four fruits and bars indicate standard errors.
ns, *, **, and *** correspond to the effect of cultivar for each harvest stage: non-significant and significant at P < 0.05, P < 0.01, and P < 0.001, respectively.
Different letters (in normal, bold, and italics for QV, MD2 and RL41, respectively) signify that the averages of gene expression levels were different between harvest
stages at P < 0.05 (according to Tukey's multiple comparison test), for each cultivar.

3.2. Reactive oxygen species and enzymatic and non-enzymatic defense
system
Transcripts of genes encoding metalloenzymes, such as me-
tallothionein (MT3) and superoxide dismutase (SOD), showed sig-
nificant variations during pineapple development (Fig. 1, Table 3),
whereas no genotypic effect was noted for MT3. MT3a and MT3b
transcripts were strongly induced during ripening, at harvest stages H4
and H5, regardless of the pineapple cultivar. This significant effect of
harvest stage was observed in the two climatic areas as well (Table 4).
In the Southwest area, the significant effect of cultivar on the levels of
MT3a and MT3b transcripts were due to high levels in QV in the first
development stages for MT3a and to low levels in the same cultivar for
MT3b (Fig. 1). The levels of Fe-SOD and [Cu/Zn]-SOD transcripts were
affected by the development stage, regardless of the cultivar (Table 3).
However, the level of variation of gene expression was low, with a
slight decrease observed during the beginning of pineapple develop-
ment, followed by a slight increase during ripening, especially for QV
(Fig. 1). Significant differences between cultivars and climatic areas
were also noted for Fe-SOD and [Cu/Zn]-SOD transcripts (Tables 3 and
4).
The expression of genes linked to hydrogen peroxide catabolism,
such as CAT2, APx1, and APx3, was significantly influenced by the
environment and by the genotype, except APx1 (Table 3). The changes
in the expression of these three genes did not have the same pattern
during fruit development and ripening (Fig. 3). CAT2 genes had the
same levels between H1, H2 and H3 and then repressed at H4 and H5,
regardless of the cultivar in the South-West area. The pattern was quite
different in the East area, mainly for RL41, with higher levels and a
transient induction at H2 followed by a decrease in gene levels until H5
(Fig. 3). APx1 genes tended to be progressively repressed during fruit
development, except for QV in which it is induced in ripe fruit (Fig. 3).
In the second area for RL41 and MD2 (Fig. 3) this trend was less clear,
with a fluctuation in gene expression. The levels of APx3 transcripts
were induced during ripening, regardless of the cultivar in the various
areas (Fig. 3). The levels were higher in the MD2 fruit than in the RL41
ones of the East area at H4 and H5 (Fig. 3).
Ascorbic acid recycling, analyzed according to the patterns of GR
and MDHAR transcript variations during pineapple development in-
dicated a significant induction of the expression of these genes during
fruit ripening, as showed by the significant harvest effect in the two
climatic areas (Tables 3 and 4) and the significant higher gene levels at
H4 and H5 stages (Fig. 3). No genotypic effect was reported for levels of
GR transcripts (Table 2). However, in the Southwest area, at H5 stage
only significant higher levels of GR transcripts were observed in QV
than in MD2 followed by RL41 (Fig. 3), even if globally the cultivar
effect being not significant (Table 3). For MDHAR ones climatic area
and cultivar effects were noted, especially due to differences in ripe
fruit between cultivars at H5 and higher levels in fruit of MD2 from the
East (Fig. 3, Table 4).
3.3. Correlations between gene expression and contents in antioxidant
compounds in ripe pineapple
Principal component analysis (PCA) was used to identify the most
significant variables in the dataset. About 90% of the variability ob-
served could be explained by the first four components (Fig. 4). The
variables with higher scores on the first axis, PC1 (Fig. 4A) include
transcripts related to phenolic compound synthesis and oxidation such
as PAL, C4H, PPO1 and PPO2, with a contribution of 11%, 8%, 10%
and 11%, respectively, to oxidative stress such as MT3a, MT3b and
APx3, with a contribution of 6%, 12% and 10%, respectively, as well as
in ascorbic acid content and its recycling, i.e., MDHAR, with a con-
tribution of 12% and 8%, respectively. Correlations between the ori-
ginal variables and the first principal component indicated that tran-
scripts related to phenolic compound synthesis and oxidation such as
PPO2 (R = 0.88), PPO1 (R = 0.84), PAL (R = 0.89) and C4H
(R = 0.76), and to antioxidant response to oxidative stress represented

133
M. Léchaudel et al. Plant Physiology and Biochemistry 130 (2018) 127–138

Fig. 3. Expression dynamics of 4 genes APx1, Ascorbate peroxidase; APx3, Ascorbate peroxidase; GR, Glutathione Reductase; MDHAR, monodehydroascorbate
reductase) during pineapple fruit development of three cultivars: QV (white bars), MD2 (gray bars) and RL41 (black bars), in the two climatic areas (South West, SW,
and East one), as revealed by relative quantification PCR. The maximum transcript level of the QV cultivar was used as the calibrator for relative gene expression
analysis. For each harvest stage, values are means of three to four fruits and bars indicate standard errors.
ns, *, **, and *** correspond to the effect of cultivar for each harvest stage: non-significant and significant at P < 0.05, P < 0.01, and P < 0.001, respectively.
Different letters (in normal, bold, and italics for QV, MD2 and RL41, respectively) signify that the averages of gene expression levels were different between harvest
stages at P < 0.05 (according to Tukey's multiple comparison test), for each cultivar.

by MT3b (R = 0.92), APx3 (R = 0.84) and MT3a (R = 0.67), as well as
ascorbic acid content (R = 0.93) and its recycling, i.e., MDHAR
(R = 0.76), are positively correlated with PC1. The second axis, PC2
(Fig. 4B) is mainly built by the genes linked to the synthesis and
breakdown of hydrogen peroxide such as Fe-SOD, [Cu/Zn]-SOD, CAT,
and to phenolic synthesis, i.e., 4CL, and transcripts of GR, which con-
tribute to 12%, 15%, 15%, 11% and 18%, respectively. On PC2, the
main variable positively correlated with this axis was transcripts of GR
(R = 0.83), [Cu/Zn]-SOD (R = 0.76), and 4CL (R = 0.64), whereas
those of CAT (R = −0.75) and Fe-SOD (R = −0.68) were negatively
correlated with it. The main transcripts contributed to the third axis,
PC3 (Fig. 4D) were the transcripts related to phenolic compound
synthesis, such as 4CL (14%), C4H (7%) and total phenolic compound
content (21%), and to hydrogen peroxide and ascorbate balance, such
as [Cu/Zn]-SOD (9%), Fe-SOD (9%) and APx1 (16%) transcripts.
Variable related to transcripts of APx1 (R = 0.60) was positively cor-
related with the PC3 whereas this of total phenolic compound content
(R = −0.68) was negatively correlated with it. The variables with the
highest scores on the fourth axis, PC4 (Fig. 4B), are variables linked to
redox balance such as MDHAR, MT3a, Fe-SOD and APx1, and to phe-
nolic synthesis, i.e., 4CL, which contributed 16%, 19%, 14%, 27%, and
11% to this component, respectively. No variable was correlated with
the PC4 axis (with a P-value < 0.05). PC1 therefore opposed fruit with
a high expression in genes involved both in the phenolic pathway and
in antioxidant ROS scavenging. The second axis opposed fruit having
both enzymatic detoxification of ROS and non-enzymatic antioxidant
ones. Fig. 4C and D show that samples from the three cultivars are
distributed differently on axes 1 and 2, i.e., MD2 is characterized in the
area of PCA with high contents in ascorbic acid and total phenolic
compounds and high levels of PAL, C4H, PPO1, PPO2, MT3a, and MT3b
transcripts, whereas QV and RL41 are characterized by high levels of
CAT, [Cu-Zn]-SOD, GR and APx1.

134
M. Léchaudel et al. Plant Physiology and Biochemistry 130 (2018) 127–138

Fig. 4. Segregation of pineapples according to their expression profiles of genes and biochemical characteristics determined by principal component analysis (PCA).
Vectors (A and B) represent the loadings of expression profiles of genes and biochemical traits. Symbols (C and D) represent pineapple fruits of cv. MD2, RL41 and
QV, harvested at two maturity stages (H4, and H5) and under two climatic area conditions (Southwest and East). Abbreviations: (MT3a: MT3 Metallothionein; MT3b:
MT3, Metallothionein; PAL, phenylalanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate-CoA ligase; PPO1, Polyphenol oxidase; PPO2,
Polyphenol oxidase; CAT, Catalase; [Cu/Zn]-SOD, [Cu/Zn] Superoxide dismutase; Fe-SOD, Fe Superoxide dismutase; APx1, Ascorbate peroxidase; APx3, Ascorbate
peroxidase; GR, Glutathione Reductase; MDHAR, monodehydroascorbate reductase; TPP and AsA, contents in total phenolic compounds and ascorbate, respectively.

4. Discussion
4.1. Effect of maturity stage and environment on the expression of genes
involved in phenolic compound synthesis and oxidation according to the
pineapple cultivar
Genes involved in the first steps of the phenylpropanoid pathway
were up-regulated during the ripening of pineapple, regardless of the
cultivar, as shown by the significantly high levels of PAL, 4CL and, to a
lesser extent, C4H, at stages H4 and H5. These stages corresponded to
harvests when external fruit color has begun to change and a fully ripe
maturity has been reached (Brat et al., 2004; Wardy et al., 2009). An
increase in mRNA levels of PAL during fruit ripening has been shown in
other non-climacteric fruits like grape (Ali et al., 2011) and strawberry
(Pombo et al., 2011). PAL and 4CL are generally considered as key
enzymes for the biosynthesis of the different classes of phenylpropanoid
products because a high correlation has been reported between the
transcript levels of PAL and 4CL and the abundance of phenolic com-
pounds in grape varieties (Gatto et al., 2008). The last transcript studied
in the phenylpropanoid pathway, C4H, was not strongly induced during
ripening in QV compared to the two other cultivars. The expression of
C4H was significantly strongly induced in MD2 ripening, which may be
an indication of the particular expression of genes of this cultivar,
which is known to be one of the pineapple cultivars that accumulates
the highest contents in total phenolic compounds (Lu et al., 2014).
The sequence of the two pineapple PPO genes determined for the
three cultivars studied is taken from sequences described by Stewart
et al. (2001) on Smooth Cayenne cultivar. The authors indicated that
the expression of PINPPO1 and PINPPO2 was low in developing fruit,
but was strongly up-regulated in response to chilling and wounding. In
this study, results showed that PPO2 was up-regulated during pineapple
ripening. This increase in gene expression could involve a de novo
synthesis of the PPO enzyme in response to ripening of pineapple fruit,
explaining the significantly higher PPO activity reported in half-ripe
fruit compared to that of fruit where the yellow coloration just ap-
peared in the basal portion (Soares et al., 2005). Indeed it was observed
in other fruits such as olive (Ortega-García et al., 2008) that both the
PPO activity and the amount of PPO protein significantly increased
during fruit maturation. These changes in mRNA levels suggests that
PPO may be involved in the regulation of the oxidative processes in-
volved during pineapple ripening since PPO oxidation generates hy-
drogen peroxide (Robards et al., 1999). This reaction occurs after cell
walls are disrupted since phenolic compounds and polyphenoloxidases
are not predominantly located in the same compartment (Robards
et al., 1999). In pineapple, the increase in the activity of poly-
galacturonases, involved in pectin degradation, was reported at the
beginning of the fruit maturation, concomitantly with the peel chlor-
ophyll breakdown, and until the end of this process (Soler, 1991). The
other cDNA of PPO, PPO1, was also affected by fruit ripening in the two
areas, and seemed to be inducible more by environmental factors (Zhou
et al., 2003b) than genetic one, as no cultivar effect was reported. The
same pattern of gene expression was observed for other genes involved
in the metabolism of phenolic compounds, such as PAL, C4H and 4CL.
The developmental stage effect on the expression of PAL, 4CL and PPO2
was significant in the East area, and the up-regulation was not mea-
sured at the same time. The increase in transcript levels, especially
observed in RL41, was actually induced earlier in this area where
temperatures were significantly lower, around 3 °C less than the
average of daily mean temperatures recorded during the fruit devel-
opment phase in the other climatic area. This induction of expression of

135
M. Léchaudel et al.
genes involved in the phenylpropanoid pathway and in polyphenol
oxydase could be related to the cooler temperatures recorded in this
area. Indeed, a similar increase in PAL mRNA levels was observed in
response to low-temperature storage in grape (Sanchez-Ballesta et al.,
2007). The activation of phenylpropanoid gene expression was ob-
served in the first stage of storage, which could be related to the per-
ception by the fruit of a change in temperature. In this study, lower
temperatures were recorded in the second area as of the flowering
stage. In pineapple, studies on PAL activity showed that an increase was
induced by cold temperatures (Zhou et al., 2003a). Expression of the
two pineapple PPO cDNAs, PINPPO1 and PINPPO2, was strongly up-
regulated in response to chilling (Zhou et al., 2003b). Thus, the similar
dynamic of PPO and PAL, 4CL expression changes with pineapple de-
velopment, in the various cultivars and according to environmental
conditions, and the correlations between these variables and the first
principal component, suggested that different steps of phenylpropanoid
metabolism and PPO are coordinately regulated. The coordinate reg-
ulation of PPO and phenylpropanoid biosynthesis underscores the im-
portance of PPO in fruit response to stress (Newman et al., 2011).
4.2. Effect of maturity stage and environment on the expression of ROS-
scavenging genes among different pineapple cultivars
Pineapple fruit development and ripening were associated with the
up-regulation of genes involved in ROS metabolism and in the enzy-
matic antioxidant system of defense to ROS. These results were in ac-
cordance with a recent study on pineapple gene expression profiles
during fruit ripening, identifying a cluster of six pineapple genes
thought to be involved in antioxidant activity, oxygen and ROS meta-
bolism (Koia et al., 2012). Ripening of non-climacteric fruit such as
grape is considered to be an oxidative process (Pilati et al., 2007). It was
revealed in strawberry (Aharoni et al., 2002; Aharoni and O'Connell,
2002), another non-climacteric fruit, that active gene expression was
induced to cope with oxidative stress conditions during ripening. The
accumulation of 10 cDNAs encoding putative stress response proteins
upon ripening was reported in grape berry (Davies and Robinson,
2000), suggested that ripening itself may impose stress conditions on
the fruit. One of these sequences encoded a metallothionein-like protein
of type 3 (MT3), which is associated with fruit ripening and that has
various functions, including roles in activated oxygen detoxification
and control of cellular redox potential. An increase in MT3a and MT3b
was also revealed in this study, regardless of the cultivar and the en-
vironmental conditions during pineapple development. The superoxide
dismutase (Fe-SOD) and catalase (CAT2) transcripts that codify for two
important detoxifying enzymes were not identified as being up-regu-
lated during ripening of pineapple fruit, regardless of the cultivar. CAT2
transcript had a typical pattern compared to other transcripts as it was
transiently induced during fruit development and then repressed in ripe
fruit, regardless of the cultivar and the climatic area. Transcripts of SOD
and CAT were not significantly modulated in studies on berry ripening
either (Pilati et al., 2007). As in Smooth Cayenne (Koia et al., 2012), the
increase in the expression of the [Cu/Zn] SOD gene during fruit ri-
pening was reported, but not always significant and after a decrease
between H1 and H3. Higher levels of CAT2 transcripts at H2 stage were
observed in fruits from the second area, which can be related to a fruit
response to the lower temperatures in this area than in the first one.
Indeed, CAT2 mRNA levels were enhanced in fruit during storage at
cold temperature (Figueroa-Yáñez et al., 2012).
The two transcripts of APX showed different trends, one reflecting a
decrease in expression with the progression of ripening only in a cli-
matic area, and the other an increase in expression with fruit devel-
opment and ripening, regardless of the cultivar and climatic area stu-
died. The pattern of APx3, with the highest expression at the ripe stage,
was in accordance with the one observed in Smooth Cayenne (Koia
et al., 2012). APX and CAT are generally considered as the main en-
zymes responsible for H2O2 removal in the chloroplast, peroxisomes
136
Plant Physiology and Biochemistry 130 (2018) 127–138
and mitochondria (Mittler et al., 2004). The oxidized form of ascorbate
produced after the reduction of H2O2 to water by APX is regenerated
through the ascorbate recycling pathway, which plays an important
role in the response and adaptation to the stress (Stevens et al., 2008)
and also contributes to the regulation of ROS levels. Both GR and
MDHAR transcripts were induced during the ripening of pineapple,
regardless of the cultivar and the climatic area studied. A significant up-
regulation of MDHAR transcripts was reported in post-veraison berries
(Melino et al., 2009) as well, and it has been suggested that this reg-
ulation leads to an increase in the contribution of the reduced form of
AsA to the total AsA pool of berries during the later stages of ripening.
Although the regulation of the detoxifying enzymes is very complex
and occurs at different levels (transcriptional-, post-transcriptional
regulation, subcellular compartmentalization, etc.), as reported in
several studies (Matamoros et al., 2010; Mittler et al., 2004; Peroni
et al., 2007), most of the gene expression followed in this study ap-
peared to be unchanged from H1 to H2, and to be strongly modulated
from H4 to H5. These changes in gene expression pointed out the oc-
currence of an up-regulation of transcripts in response to oxidative
processes involved during pineapple ripening, as in the case of other
non-climacteric fruits such as strawberry (Aharoni et al., 2002) and
grape berry (Pilati et al., 2007). However, to confirm that pineapple
ripening is accompanied by changes in redox balance it would be in-
teresting to analyze changes in reactive oxygen species, as reported on
other pineapple cultivar (Liu et al., 2017) or reactive nitrogen species
during the pineapple development and ripening, as proposed on pepper
(Corpas et al., 2018).
4.3. Correlations between gene expression and quality traits in ripe
pineapple
Results of PCA analysis indicated that the data variability could be
explained by four main components. The first one opposed fruit having
both high levels of transcripts involved in polyphenol synthesis and
oxidation of compounds involved in system defense to cope with oxi-
dative stress and in antioxidant ROS scavenging, i.e., ascorbate, MT3a
and MT3b. H2O2 balance may play a biological role, activating the
response in the antioxidant defense system as pineapple fruit ripens. A
microarray-based approach on pineapple has suggested that redox ac-
tivity is a prominent biological mechanism during pineapple fruit ri-
pening (Koia et al., 2012). Our results tented to support this hypothesis,
pointing out the up-regulation of the expression of genes related to the
scavenging of reactive oxygen species that accompanies ripening, in-
cluding phenylpropanoid, ascorbate and metallothionein metabolisms
and the enzymatic antioxidant system, as it was reported on other non-
climacteric fruit (Alós et al., 2013; Pilati et al., 2007; Racchi, 2013).
The other axes were mainly linked to transcripts related to ascorbate-
glutathion recycling and to enzymatic antioxidant system defense,
suggesting the involvement of these metabolic pathways in ripe pine-
apple. The role of CAT and APX enzymes in pineapple ripening has been
indicated in a study that reported higher activity of these enzymes in
ripe fruit after storage compared to pineapple fruit at harvest (Zhou
et al., 2003a). The capacity of AsA recycling was suggested to be one of
the mechanisms necessary to maintain the pool of this antioxidant in
the later stages of ripening of several fruit species such as tomato
(Mellidou et al., 2012) and pea (Matamoros et al., 2010). However, the
expression enhancement of the genes related to this response depends
on the cultivar. The two first axes separated cultivars RL41 and MD2
due to differences in non-enzymatic antioxidant defense, i.e., poly-
phenol and ascorbic acid compounds, metallothionein transcripts, and
in enzymatic antioxidant defense, i.e., APx1 and APx3. MD2 was the
cultivar characterized by the highest contents in these antioxidant
compounds (Lu et al., 2014; Raimbault et al., 2010). PCA analysis did
not indicate any correlation between transcripts of the phenylpropa-
noid pathway providing coumaroyl-CoA and the total phenolic com-
pound content. However, the role of PAL in the regulation of
M. Léchaudel et al.
polyphenolic compound synthesis is complex (Gill and Tuteja, 2010),
and a high content in phenolic compounds without higher levels in PAL
transcripts has already been reported in another ripe non-climacteric
fruit (Concha et al., 2013).
5. Conclusion
The temporal dynamics of the expression of genes coding oxidative
and antioxidative stress analyzed during pineapple fruit development
has enabled us to carry out a comprehensive description of different
transcriptional events related to redox balance that occurred according
to several cultivars and climatic conditions. Pineapple development and
ripening involve changes in the levels of the transcripts linked to the
first steps of the phenylpropanoid pathway such as PAL, 4CL, to phenol
oxidation, i.e., PPO2, and to the ROS balance involving the enzymatic
and non-enzymatic antioxidant system of defense to ROS, especially
genes of APX1, APX3, GR, MDHAR, MT3a and MT3b, regardless of the
cultivars, and of the environmental conditions of crop production.
These changes indicate that gene expression is induced to cope with
oxidative stress conditions that accompany the ripening of this non-
climacteric fruit. Moreover, the activation of phenylpropanoid meta-
bolism may play a role in the development of protective barriers in
stress-damaged cells. Our results confirm the influence of the environ-
mental conditions of crop production on the level of the expression of
these transcripts. However, MT3a and MT3b transcripts were not in-
fluenced by genetic and environmental factor. The genetic effect was
not noted on the various transcripts linked to the first steps of the
phenylpropanoid pathway and to phenol oxidation, except 4CL ones. In
ripe pineapple, relationships were obtained between the contents in
antioxidant metabolites and the expression of genes related to enzymes
involved in the scavenging of reactive oxygen species.
Author contributions
Mathieu Léchaudel, Thierry Joët, Patrick Fournier, and Jacques
Joas designed research; Jacques Joas, Mathieu Léchaudel, Marie
Darnaudery, and Thierry Joët performed research; Mathieu Léchaudel,
Marie Darnaudery, and Thierry Joët analyzed data; Mathieu Léchaudel,
Marie Darnaudery, Thierry Joët and Jacques Joas wrote the paper. All
authors approved the final version of the manuscript.
Acknowledgements
We would like to thank B. Abufera and G. Tullus (CIRAD, Banana,
Plantain and Pineapple Cropping Systems Research Unit, Reunion
Island) for their technical assistance during field experiments and J.
Minier (CIRAD, QualiSud Joint Research Unit, Reunion Island) for as-
sistance with the biochemical analyses. The authors gratefully ac-
knowledge G. Wagman for revising the manuscript and editing the
English.
Appendix ASupplementary data
Supplementary data related to this article can be found at http://dx.
doi.org/10.1016/j.plaphy.2018.06.041.
References
Aharoni, A., Keizer, L.C.P., Van Den Broeck, H.C., Blanco-Portales, R., Muñoz-Blanco, J.,
Bois, G., et al., 2002. Novel insight into vascular, stress, and auxin-dependent and
-independent gene expression programs in strawberry, a non-climacteric fruit. Plant
Physiol. 129, 1019–1031.
Aharoni, A., O'Connell, A., 2002. Gene expression analysis of strawberry achene and
receptacle maturation using DNA microarrays. J. Exp. Bot. 53, 2073–2087.
Ali, M., Howard, S., Chen, S., Wang, Y., Yu, O., Kovacs, L., et al., 2011. Berry skin de-
velopment in Norton grape: distinct patterns of transcriptional regulation and fla-
vonoid biosynthesis. BMC Plant Biol. 11, 7.
Almeida, J.R.M., D'Amico, E., Preuss, A., Carbone, F., de Vos, C.H.R., Deiml, B., et al.,
137
Plant Physiology and Biochemistry 130 (2018) 127–138
2007. Characterization of major enzymes and genes involved in flavonoid and
proanthocyanidin biosynthesis during fruit development in strawberry (Fragaria ×
ananassa). Arch. Biochem. Biophys. 465, 61–71.
Alós, E., Rodrigo, M.J., Zacarías, L., 2013. Transcriptomic analysis of genes involved in
the biosynthesis, recycling and degradation of L-ascorbic acid in pepper fruits
(Capsicum annuum L.). Plant Sci. 207, 2–11.
AOAC, 1998. Ascorbic acid in vitamin preparations and juices. In: Official Methods of
Analysis, sixteenth ed. AOAC International, Arlington, Virginia, pp. 16–18.
Apel, K., Hirt, H., 2004. REACTIVE OXYGEN SPECIES: metabolism, oxidative stress, and
signal transduction. Annu. Rev. Plant Biol. 55, 373–399.
Barral, B., Chillet, M., Minier, J., Léchaudel, M., Schorr-Galindo, S., 2017. Evaluating the
response to Fusarium ananatum inoculation and antifungal activity of phenolic acids
in pineapple. Fungal Biol. 121, 1045–1053.
Brat, P., Hoang, L.N.T., Soler, A., Reynes, M., Brillouet, J.-M., 2004. Physicochemical
characterization of a new pineapple hybrid (FLHORAN41 Cv.). J. Agric. Food Chem.
52, 6170–6177.
Chen, C., Zhang, Y., Xu, Z., Luan, A., Mao, Q., Feng, J., Xie, T., Gong, X., Wang, X., Chen,
H., Het, Y., 2016. Transcriptome profiling of the pineapple under low temperature to
facilitate its breeding for cold tolerance. PLoS One 11 (9), e0163315.
Cherian, S., Figueroa, C.R., Nair, H., 2014. ‘Movers and shakers’ in the regulation of fruit
ripening: a cross-dissection of climacteric versus non-climacteric fruit. J. Exp. Bot. 65,
4705–4722.
Concha, C.M., Figueroa, N.E., Poblete, L.A., Oñate, F.A., Schwab, W., Figueroa, C.R.,
2013. Methyl jasmonate treatment induces changes in fruit ripening by modifying the
expression of several ripening genes in Fragaria chiloensis fruit. Plant Physiol.
Biochem. 70, 433–444.
Corpas, F.J., Freschi, L., Rodríguez-Ruiz, M., Mioto, P.T., González-Gordo, S., Palma, J.M.,
2018. Nitro-oxidative metabolism during fruit ripening. J. Exp. Bot. 69 (14) erx453.
Davies, C., Robinson, S., 2000. Differential screening indicates a dramatic change in
mRNA profiles during grape berry ripening. Cloning and characterization of cDNAs
encoding putative cell wall and stress response proteins. Plant Physiol. 122, 803–812.
Figueroa-Yáñez, L., Cano-Sosa, J., Castaño, E., Arroyo-Herrera, A.-L., Caamal-Velazquez,
J.H., Sanchez-Teyer, F., et al., 2012. Phylogenetic relationships and expression in
response to low temperature of a catalase gene in banana (Musa acuminata cv. “Grand
Nain”) fruit. Plant Cell, Tissue and Organ Culture (PCTOC) 109, 429–438.
Fournier, P., 2011. La culture de l'ananas Victoria à la Réunion pour l'exportation. Recueil
de bonnes pratiques. CIRAD, Reunion Island, France.
Gatto, P., Vrhovsek, U., Muth, J., Segala, C., Romualdi, C., Fontana, P., et al., 2008.
Ripening and genotype control stilbene accumulation in healthy grapes. J. Agric.
Food Chem. 56, 11773–11785.
Gil, M.I., Aguayo, E., Kader, A.A., 2006. Quality changes and nutrient retention in fresh-
cut versus whole fruits during storage. J. Agric. Food Chem. 54, 4284–4296.
Gill, S.S., Tuteja, N., 2010. Reactive oxygen species and antioxidant machinery in abiotic
stress tolerance in crop plants. Plant Physiol. Biochem. 48, 909–930.
Koia, J.H., Moyle, R.L., Botella, J.R., 2012. Microarray analysis of gene expression profiles
in ripening pineapple fruits. BMC Plant Biol. 12, 240.
Lê, S., Josse, J., Husson, F., 2008. FactoMineR: an R package for multivariate analysis. J.
Stat. Software 25, 1–18.
Léchaudel, M., Damour, G., Fournier, P., Joas, J., Jahiel, M., 2010. How to predict the
harvest date of tropical fruit: from simple methods to complex models. Acta Hortic.
(Wagening.) 880, 175–182.
Liu, J., He, C., Shen, F., Zhang, K., Zhu, S., 2017. The crown plays an important role in
maintaining quality of harvested pineapple. Postharvest Biol. Technol. 124, 18–24.
Lu, X.-H., Sun, D.-Q., Wu, Q.-S., Liu, S.-H., Sun, G.-M., 2014. Physico-chemical properties,
antioxidant activity and mineral contents of pineapple genotypes grown in China.
Molecules 19, 8518–8532.
Matamoros, M.A., Loscos, J., Dietz, K.-J., Aparicio-Tejo, P.M., Becana, M., 2010. Function
of antioxidant enzymes and metabolites during maturation of pea fruits. J. Exp. Bot.
61, 87–97.
Melino, V., Soole, K., Ford, C., 2009. Ascorbate metabolism and the developmental de-
mand for tartaric and oxalic acids in ripening grape berries. BMC Plant Biol. 9, 145.
Mellidou, I., Keulemans, J., Kanellis, A., Davey, M., 2012. Regulation of fruit ascorbic acid
concentrations during ripening in high and low vitamin C tomato cultivars. BMC
Plant Biol. 12, 239.
Mir, G., Domènech, J., Huguet, G., Guo, W.-J., Goldsbrough, P., Atrian, S., et al., 2004. A
plant type 2 metallothionein (MT) from cork tissue responds to oxidative stress. J.
Exp. Bot. 55, 2483–2493.
Mittler, R., Vanderauwera, S., Gollery, M., Van Breusegem, F., 2004. Reactive oxygen
gene network of plants. Trends Plant Sci. 9, 490–498.
Montero-Calderón, M., Rojas-Graü, M., Martín-Belloso, O., 2010. Mechanical and che-
mical properties of Gold cultivar pineapple flesh (Ananas comosus). Eur. Food Res.
Technol. 230, 675–686.
Moyle, R., Fairbairn, D.J., Ripi, J., Crowe, M., Botella, J.R., 2005. Developing pineapple
fruit has a small transcriptome dominated by metallothionein. J. Exp. Bot. 56,
101–112.
Newman, S.M., Tantasawat, P., Steffens, J.C., 2011. Tomato polyphenol oxidase B Is
spatially and temporally regulated during development and in response to ethylene.
Molecules 16, 493–517.
Ortega-García, F., Blanco, S., Peinado, M.Á., Peragón, J., 2008. Polyphenol oxidase and its
relationship with oleuropein concentration in fruits and leaves of olive (Olea euro-
paea) cv. ‘Picual’ trees during fruit ripening. Tree Physiol. 28, 45–54.
Ortega-García, F., Peragón, J., 2009. The response of phenylalanine ammonia-lyase,
polyphenol oxidase and phenols to cold stress in the olive tree (Olea europaea L. cv.
Picual). J. Sci. Food Agric. 89, 1565–1573.
Pandit, S.S., Kulkarni, R.S., Giri, A.P., Köllner, T.G., Degenhardt, J., Gershenzon, J., et al.,
2010. Expression profiling of various genes during the fruit development and
M. Léchaudel et al.
ripening of mango. Plant Physiol. Biochemist. 48, 426–433.
Peroni, L.A., Ferreira, R.R., Figueira, A., Machado, M.A., Stach-Machado, D.R., 2007.
Expression profile of oxidative and antioxidative stress enzymes based on ESTs ap-
proach of citrus. Genet. Mol. Biol. 30, 872–880.
Pilati, S., Perazzolli, M., Malossini, A., Cestaro, A., Demattè, L., Fontana, P., et al., 2007.
Genome-wide transcriptional analysis of grapevine berry ripening reveals a set of
genes similarly modulated during three seasons and the occurrence of an oxidative
burst at vèraison. BMC Genom. 8, 428.
Poiroux-Gonord, F., Bidel, L.P.R., Fanciullino, A.L., Gautier, H., Lauri-Lopez, F., Urban, L.,
2010. Health benefits of vitamins and secondary metabolites of fruits and vegetables
and prospects to increase their concentrations by agronomic approaches. J. Agric.
Food Chem. 58, 12065–12082.
Pombo, M.A., Martínez, G.A., Civello, P.M., 2011. Cloning of FaPAL6 gene from straw-
berry fruit and characterization of its expression and enzymatic activity in two cul-
tivars with different anthocyanin accumulation. Plant Sci. 181, 111–118.
Racchi, M., 2013. Antioxidant defenses in plants with attention to prunus and citrus spp.
Antioxidants 2, 340–369.
Raimbault, A.-K., Marie-Alphonsine, P.-A., Horry, J.-P., Francois-Haugrin, M., Romuald,
K., Soler, A., 2010. Polyphenol oxidase and peroxidase expression in four pineapple
varieties (Ananas comosus L.) after a chilling injury. J. Agric. Food Chem. 59,
342–348.
Ramakers, C., Ruijter, J.M., Deprez, R.H., Moorman, A.F.M., 2003. Assumption-free
analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci.
Lett. 339, 62–66.
Robards, K., Prenzler, P.D., Tucker, G., Swatsitang, P., Glover, W., 1999. Phenolic com-
pounds and their role in oxidative processes in fruits. Food Chem. 66, 401–436.
Sanchez-Ballesta, M.T., Romero, I., Jiménez, J.B., Orea, J.M., González-Ureña, Á.,
Escribano, M.I., et al., 2007. Involvement of the phenylpropanoid pathway in the
response of table grapes to low temperature and high CO2 levels. Postharvest Biol.
Technol. 46, 29–35.
138
Plant Physiology and Biochemistry 130 (2018) 127–138
Smirnoff, N., 2000. Ascorbic acid: metabolism and functions of a multi-facetted molecule.
Curr. Opin. Plant Biol. 3, 229–235.
Soares, A.G., Trugo, L.C., Botrel, N., da Silva Souza, L.F., 2005. Reduction of internal
browning of pineapple fruit (Ananas comusus L.) by preharvest soil application of
potassium. Postharvest Biol. Technol. 35, 201–207.
Soler, A., 1991. Maturation et sénescence de l'ananas (Ananas comosus (L.) Merr) en Côte
d'Ivoire. Université des Sciences de Montpellier, Montpellier, pp. 262.
Stevens, R., Page, D., Gouble, B., Garchery, C., Zamir, D., Causse, M., 2008. Tomato fruit
ascorbic acid content is linked with monodehydroascorbate reductase activity and
tolerance to chilling stress. Plant. Cell Environ. 31, 1086–1096.
Stewart, R.J., Sawyer, B.J.B., Bucheli, C.S., Robinson, S.P., 2001. Polyphenol oxidase is
induced by chilling and wounding in pineapple. Funct. Plant Biol. 28, 181–191.
Sweetman, C., Wong, D.C., Ford, C.M., Drew, D.P., 2012. Transcriptome analysis at four
developmental stages of grape berry (Vitis vinifera cv. Shiraz) provides insights into
regulated and coordinated gene expression. BMC Genom. 13, 691.
Takahama, U., Oniki, T., 1997. A peroxidase/phenolics/ascorbate system can scavenge
hydrogen peroxide in plant cells. Physiol. Planta 101, 845–852.
Team, R.D.C., R: A language and environment for statistical computing, 2007. R
Foundation for Statistical Computing, Vienna, Austria.
Wardy, W., Saalia, F.K., Steiner-Asiedu, M., Budu, A.S., Sefa-Dedeh, S., 2009. A com-
parison of some physical, chemical and sensory attributes of three pineapple (Ananas
comosus) varieties grown in Ghana. Afr. J. Food Sci. 3, 94–99.
Wei, C.B., Liu, S.H., Liu, Y.G., Lv, L.L., Yang, W.X., Sun, G.M., 2011. Characteristic aroma
compounds from different pineapple parts. Molecules 16, 5104–5112.
Zhou, Y., Dahler, J.M., Underhill, S.J.R., Wills, R.B.H., 2003a. Enzymes associated with
blackheart development in pineapple fruit. Food Chem. 80, 565–572.
Zhou, Y., O'Hare, T.J., Jobin-Decor, M., Underhill, S.J.R., Wills, R.B.H., Graham, M.W.,
2003b. Transcriptional regulation of a pineapple polyphenol oxidase gene and its
relationship to blackheart. Plant Biotechnol. J. 1, 463–478.