Metabolic profiling of ‘Conference’ pears under low oxygen stress

Postharvest Biology and Technology 51 (2009) 123–130

Contents lists available at ScienceDirect

Postharvest Biology and Technology journal homepage: www.elsevier.com/locate/postharvbio

Metabolic profiling of ‘Conference’ pears under low oxygen stress Romina Pedreschi a,∗ , Christine Franck a , Jeroen Lammertyn a , Alexander Erban b , Joachim Kopka b , Maarten Hertog a , Bert Verlinden a , Bart Nicolaï a a b

Flanders Centre of Postharvest Technology/BIOSYST-MeBioS, Catholic University of Leuven, Willem de Croylaan 42, 3001 Leuven, Belgium Max Planck Institute for Molecular Plant Physiology, Am Mühlenberg 1, 14476 Golm bei Potsdam, Germany

a r t i c l e

i n f o

Article history: Received 13 April 2008 Accepted 28 May 2008 Keywords: Core breakdown GC–MS Metabolic profiling Metabolomics Oxidative stress Pyrus communis L

a b s t r a c t Pears (Pyrus communis L. cv. Conference) may develop core breakdown when stored under low oxygen or elevated carbon dioxide conditions. This physiological disorder is characterized by the development of brown spots due to oxidation of phenolic compounds, and eventually, cavities in the center of the fruit. Based on metabolic profiling of brown and sound tissue using GC-EI-TOF-MS, the hypothesis that this disorder is due to an imbalance between oxidative and reductive processes at the cellular level was investigated. Brown tissue was clearly characterized by a distinctive pattern in changes which included a decrease of malic acid and an increase in fumaric acid and gamma aminobutyric acid (GABA), which indicated a reduced metabolic activity at the level of the Krebs cycle and a putative block of the GABA shunt pathway. Increased gluconic acid concentration might be related to ascorbic acid degradation due to insufficient reducing equivalents or to an impaired pentose phosphate pathway. For the first time, GABA and gluconic acid have been shown to be metabolic markers for core breakdown. The concentrations of other compounds which are believed to be related to hypoxic stress response such as trehalose and putrescine were also considerably higher in brown tissue than in sound tissue. The concentration of some sugars which are typically found in xyloglucans also increased during brown development, possibly indicating cell wall breakdown due to enzymatic processes or chemical reactions of hydroxyl radicals. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Pyrus communis L. cv. Conference is one of the most important pear cultivars in Europe and it represents more than 75% of the total Belgian pear production. To extend their storage life, pears are cooled down to −1 ◦ C for 21 d after harvest before storing them under controlled atmosphere (CA) conditions. A reduced O2 concentration (2.5–3 kPa) and a slightly elevated CO2 concentration (0.7–1 kPa) in combination with low temperature (−0.5 to −1 ◦ C) are applied to prolong the storage life of pears, by reducing the respiration and other metabolic reactions and by preventing microbial growth. However, if the oxygen partial pressure is too low or the carbon dioxide partial pressure too high, the metabolism may change from aerobic to anaerobic and this may cause accumulation of volatiles and associated physiological disorders such as core breakdown.

Abbreviations: CA, controlled atmosphere; GC-EI-TOF-MS, gas chromatography/electrospray ionization-time of flight-mass spectrometry; GABA, gamma aminobutyric acid; GC–MS, gas chromatography–mass spectrometry; MSTFA, Nmethyl-N-(trimethylsilyl)-trifluoroacetamide. ∗ Corresponding author. Tel.: +32 16 320592; fax: +32 16 322955. E-mail address: [email protected] (R. Pedreschi). 0925-5214/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.postharvbio.2008.05.019

Core breakdown is characterized by the development of brown spots and the formation of internal cavities during controlled atmosphere storage without external symptoms. Additional factors that favor the incidence of core breakdown are pre-harvest factors such as harvest date, orchard characteristics, seasonal variation which includes fruit set, position of the fruit on the tree, weather conditions, fruit size and ascorbic acid content (Franck et al., 2007). The incidence of core breakdown may cause loss of commercial value of the batch of pears. The main factor that initiates the chain of events eventually resulting in core browning is believed to be the storage atmosphere composition. A too low O2 partial pressure in combination with a too high CO2 partial pressure in the storage atmosphere may lead to local hypoxic conditions in the center of the pear due to the diffusion resistance of the pear cortex tissue (Lammertyn et al., 2003; Ho et al., 2006). This results in oxidative stress and changes in the normal cellular metabolism of the respiratory towards the energetically far less efficient fermentation pathways. As a consequence, too little energy is available for repair of membrane damage caused by reactive oxygen species (Larrigaudière et al., 2001; Veltman et al., 2003). When membrane damage occurs, the normal cellular compartmentalization is lost and phenolic substrates may be enzymatically oxidized to o-quinones and, eventually, brown colored polymers are responsible for the actual browning symptoms.

124

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

Finally, cell death results and the released cell sap diffuses towards the boundary of the fruit where the water is lost to the environment while cavities remain in the center of the pear (Franck et al., 2007). While the above hypothesis on the development of core breakdown is now generally accepted, many details remain to be elucidated. Instead of focusing on isolated phenomena such as gas transport (Lammertyn et al., 2003; Schouten et al., 2004; Ho et al., 2006, 2008), the respiratory activity (Lammertyn et al., 2001), lascorbic acid metabolism (Franck et al., 2003a,b; Veltman et al., 2003) or energy status and metabolism (Saquet et al., 2003), a more global biochemical profiling approach is required to better understand the metabolic events behind browning. An unbiased and sensitive analytical technique such as gas chromatography–mass spectrometry (GC–MS) which allows the simultaneous analysis of metabolites in a complex extract can be used to achieve this. This technique has been used in clinical medicine in order to diagnose human diseases (Jellum et al., 1988), aroma analysis (Buttery et al., 1988; Baldwin et al., 1991) and was introduced into plant research for metabolic profiling purposes in the late nineties in, for example, apricots (Katona et al., 1999) and potato tubers (Roessner et al., 2000). GC–MS based metabolome analysis has profound applications in obtaining new insights in plant biochemistry. The prerequisite and thus key challenge of metabolite profiling is the rapid, reliable and unambiguous identification of hundreds of metabolites in highly complex biological preparations. In this paper, we present the results obtained with gas chromatography/electrospray ionizationtime of flight-mass spectrometry (GC-EI-TOF-MS) technology. TOF technology differs from scanning type mass spectrometers such as quadrupole by their fast acquisition rates with unchanged mass spectra from peak front to tail (provided that the peak contains only one component). The fast acquisition rates with unchanged mass spectra are beneficial with respect to automated findings by mass spectral comparisons and spectral deconvolution of coeluting components (Veriotti and Sacks, 2000, 2001). 2. Materials and methods 2.1. Plant material and sampling procedure Pyrus communis L. (cv. Conference) pears were harvested on September 20th 2004 (two weeks after the optimal commercial harvest time) in the orchard of the Centre of Fruit Culture in Ril-

laar (Belgium). The pears were immediately placed under browning inducing conditions (1 kPa O2 , 10 kPa CO2 , −1 ◦ C) without prior cooling in air to enforce browning development. Samples were taken in January and March 2005, after 4 and 6 months of storage. Five brown affected pears (Fig. 1, left side) and 5 sound pears (Fig. 1, right side) were sampled in January and March. Pear slices (1 cm thickness) were cut perpendicularly to the longitudinal axis at 5 cm from the calyx end. From each slice, samples were taken with a cork borer (0.5 cm Ø) at different locations as shown in Fig. 1. In the case of brown pears, the outer samples (numbers 1–4) consisted almost exclusively of apparently sound tissue. The samples were transferred to numbered Eppendorf tubes and immediately frozen in liquid nitrogen. The tissue was homogenized using a mortar and a pestle containing liquid nitrogen. A representative sample of approximately 40 mg (FW) of tissue was used for subsequent extraction. 2.2. Extraction and derivatization Extraction, liquid partition, concentration to dryness, and methoxyamination of carbonyl moieties followed by derivation of acidic protons with N-methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) prior to GC-EI-TOF-MS analysis was performed as described by Fiehn et al. (2000) and Roessner et al. (2000) with minor modifications (Wagner et al., 2003). Five hundred ␮L methanol (including ribitol at a final concentration of 9 ␮g mL−1 ) was added to the samples and incubated for 15 min at 70 ◦ C. Subsequently, 200 ␮L chloroform was added, followed by 5 min incubation at 37 ◦ C. Finally, 500 ␮L water was added and after thoroughly vortexing and subsequent phase separation by centrifugation, 10 ␮L of the polar phase was dried in vacuo. Highly lipophilic components were removed by liquid partition into chloroform and discarded from this study. The dried residue was re-dissolved in 20 ␮L of 20 mg mL−1 methoxyamine HCl (Sigma) in pyridine and incubated for 90 min at 30 ◦ C. The silylation step was carried out by adding 40 ␮L MSTFA (Machery-Nagel, Düren, Germany) to the extract followed by 30 min incubation at 37 ◦ C. Five ␮L of a mixture of retention time standards was added to the final sample volume (n-dodecane (RI 1200), npentadecane (RI 1500), n-nonadecane (RI 1900), n-docosane (RI 2200), n-octacosane (RI 2800), n-dotriacontane (RI 3200), and nhexatriacontane (RI 3600), dissolved in pyridine at 0.22 mg mL−1 final concentration).

Fig. 1. Cross-section of pears stored under browning inducing conditions with core breakdown (left side), referred as brown (5–8) and sound tissue (1–4) and without core breakdown (right side) referred to as sound tissue. Numbers indicate sampling location. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.)

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

2.3. GC-EI-TOF-MS analysis A GC 6890 (Agilent Technologies, Palo Alto, CA, USA) was operated under electronic pressure control and equipped with a split/splitless capillary inlet. All samples were analyzed twice: first, a 1 ␮L split injection (split ratio 1:30) mostly for abundant sugars, and second, a 1 ␮L splitless injection for less concentrated polar components. Injection was followed by a 2 min pulse at 110 psi and injection temperature set to 230 ◦ C. The capillary column used was a VF-5 ms column (Varian, Darmstadt, Germany) with the following dimensions: 30 m × 0.25 mm inner diameter and a 0.25 ␮m film. Helium was used as carrier gas with constant flow at 1 mL min−1 . The temperature program was 2 min at 80 ◦ C followed by a 15 min ramp to 350 ◦ C and final heating for 2 min at 350 ◦ C. The transfer line to the mass spectrometer was set to 250 ◦ C. The time-of-flight (TOF) mass spectrometer was a Pegasus II MS system (Leco, St. Joseph, MI, USA) with an electron impact ionization (EI) source set to 200 ◦ C. Mass spectra were monitored with an acquisition rate of 6 spectra s−1 in the mass range m/z = 70–600. Tuning and all other settings of the mass spectrometer were according to manufacturer’s recommendations. More details have been reported by Erban et al. (2007).

125

outer) and storage time (4 months storage or ‘January’ samples and 6 months storage or ‘March’ samples). Only metabolites for which the R2 of the model was larger than 0.5 and for which the model was significant (p < 0.05) were retained. Sample grouping was carried out by means of a Tukey multiple comparison test (p < 0.05). The analysis was carried out in SAS Version 9.1 (The SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Metabolic profile of pear fruit tissue Due to the large concentration differences between sugars and other metabolites, it was necessary to analyze every sample twice. The splitless injection gave the most complex profiles (Fig. 2A, B) but the sugar fraction was overloaded. Therefore, the samples were analyzed again with a split injection and from these ‘diluted’ chromatograms (Fig. 2C), information about the sugar fraction was obtained. We were able to identify 64 compounds (Table 1). The compounds displayed in italics in Table 1 were strongly suggested by the NIST library but are not yet verified by authentic standards. 3.2. Metabolic changes in sound and brown pears

2.4. Data processing Chromatograms were acquired with ChromaTOFTM software (LECO, St. Joseph, MI, USA). Initial processing, namely baseline substraction, smoothing, and export of the processed chromatograms into a NetCDF file interchange format were performed within the ChromaTOFTM software. Retention index (RI) calibration and mass spectral deconvolution were performed with the ChromaTOFTM software. The NIST98 mass spectral search program (National Institute of Standards and Technology, Gaithersburg, MD, USA) is fully integrated with the ChromaTOFTM deconvolution software and was chosen as software platform for mass spectral comparison using the NIST98 mass spectral library (http://www.nist.gov/srd/mslist.htm) and the mass spectral (MS) and retention time index collection of the Golm Metabolome Database (Kopka et al., 2005; Schauer et al., 2005). Mass spectral matching was manually supervised and matches accepted with thresholds of match > 650 (with maximum match equal to 1000) and retention index deviation < 1.0%. Peak height representing arbitrary mass spectral ion currents of each mass fragment was normalized using the amount of the sample fresh weight and ribitol for internal standardization of volume variations.

The PLS-DA multivariate statistical approach accounted for 18% and 12% of the total variance (X variance) and 27% and 18% (Y variance) being explained by the first two latent variables (LV1 and LV2), respectively (Fig. 3). The further the variable is from the origin the more influential the variable in explaining the differences among the different treatments. The biplot is shown in Fig. 3. From the scores it is clear that brown and sound tissue have different metabolic profiles. There are also differences in relation to storage time (4 months or January and 6 months or March). Spatial gradients of metabolites can be observed but are relatively small. The VIP procedure outcome revealed that malic acid, threitol and erythritol were the major compounds to explain the difference between samples from sound and brown pears. These metabolites were present in higher amounts in sound pears. Samples from brown pears in January were characterized by enhanced concentrations of fumaric acid, succinic acid, gamma aminobutyric acid (GABA), putresceine and isoleucine, while the samples from March were higher in certain sugars (trehalose, fucose, and C5 sugars) and gluconic acid.

2.5. Statistical analysis Two different statistical approaches were used. The multivariate statistical analysis consisted of partial least squares discriminant analysis (PLS-DA). The different tested metabolites were considered as X-variables and the categorical variables (disorder, position and storage time) were considered as Y-variables. The data were mean centred and variables were weighed by their standard deviation to give them equal variance. The Variable Importance Plot (VIP) was used as a formal tool (Karp et al., 2005; Pedreschi et al., 2007) to rank the different metabolites in order of importance based on the correlation loadings. For more details about PLS-DA and VIP procedure, the reader is referred to Norden et al. (2005). All calculations were carried out in Unscrambler Version 9.6 (CAMO, A/S, Trondheim, Norway). The univariate statistical analysis was based on a general linear model composed of three factors and two levels per factor to test significant changes (p < 0.05) in individual metabolites. The factors and levels consisted of: disorder (brown, sound), position (inner,

Fig. 2. GC-EI-TOF-MS total ion chromatogram of a pear fruit extract. (A and B) Splitless injection; (C) split injection for improved quantification of abundant metabolites, such as glucose, fructose, malic or citric acid and sucrose.

126

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

Table 1 Metabolites identified by GC-EI-TOF-MS as components of a methanol extract from pear fruit. Amino acids

Organic acids

Sugars

Fatty acids

Sugar alcohols

Others

Alanine Asparagine Aspartic acid GABA* Glutamic acid Glycine Isoleucine* Phenylalanine Proline Pyroglutamic acid* Serine Threonine Valine

Benzoic acid 3-cis-Caffeoylquinic acid 3-trans-Caffeoylquinic acid Citric acid Dehydroabietic acid Fumaric acid* Gluconic acid* Glutaric acid Glyceric acid Lactic acid Malic acid* Quinic acid* Shikimic acid Succinic acid* Threonic acid*

Fructose Fucose* Galactose Glucose Glucose-6-P Lyxose Mannose* Raffinose Ribose Sucrose* Trehalose* Threose* Xylose*

Eicosanoic acid Heptadecanoic acid Hexadecanoic acid 9-Z-Hexadecenoic acid Nonanoic acid Octadecanoic acid 9-Z-Octadecenoic acid

Diethyleneglycol Erythritol* Glycerol myo-inositol Maltitol Sorbitol Threitol* Xylitol

Ethanolamine Hydroxylamine N,N-Di-(2-hydroxyethyl)-methanamine Putrescine* Triethanolamine Squalene Uracil Urea

The compounds displayed in italics were suggested by mass spectral match to reference spectra of the NIST library, but were not verified by authentic reference compounds. (*) Significantly different metabolites (p < 0.05) for which the general linear model had an R2 > 0.5.

This latter compound appeared to be typical for brown samples and was only found in trace amounts in samples from sound pears. The univariate statistical approach confirmed the above mentioned metabolites (p < 0.05). The multivariate statistical approach used is a good visualization tool and accounts also for correlations, which is crucial given the fact that the different metabolites are interrelated because either they belong to the same pathway or the different pathways are interrelated. However, to account for absolute statistically significant changes, the univariate approach can provide further information. The following results are based on the 18 metabolites for which the total model was significant at a confidence level of 0.05 and R2 of the model was higher than 0.5 (Fig. 4).

3.3. Metabolic differences due to position Differences in concentrations of metabolites for sound pears between the inner (5 –8 ) and outer part (1 –4 ) provide information about the radial distribution of metabolites. Seventeen out of the eighteen retained metabolites by the univariate approach were significantly different with respect to position and the interactions position × disorder and position × time, indicative of radial gradients of metabolites. These gradients were more pronounced in brown tissue. In January, only malic acid significantly differed between the inner and outer part in sound pears while in March also erythritol, threitol and xylose were significantly different. However, 10 metabolites from brown pears

Fig. 3. PLS-DA of samples representing fruit tissue of brown (closed symbols) and sound (opened symbols) pears from two different sampling times (j, January and m, March) and position (inner, outer). Circles: January inner; squares: January outer; triangles: March inner; diamonds: March outer. Sample scores and metabolite loadings (small open circles) are superimposed. The percentage explained variances are indicated on the axes. The analysis was based on the correlation matrix. Only the most important metabolites selected through the VIP procedure are named. Legends for the metabolites correspond to: eryt: erythritol, fuc: fucose, fum: fumaric acid, gaba: 4-amino butyric acid, gluc: glucose, gluconic: gluconic acid, glut: glutamic acid, isoleu: isoleucine, lyx: lyxose, mal: malic acid, mann: mannose, putr: putrescine, pyro: pyroglutamic acid, quin: quinic acid, succ: succinic acid, suc: sucrose, threitol: threitol, threonic: threonic acid, trehal: trehalose, treose: treose, xyl: xylose.

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

127

Fig. 4. Relative response ratios for the eighteen metabolites that were statistically significant by using the general linear model for which R2 > 0.5. The relative response ratio is obtained by dividing the peak area by the peak area of ribitol, the internal standard. Data are mean values ± S.E. (standard error).

varied in the radial direction (position) and might also be related to browning in January. These compounds were succinic acid, threose, threonic acid, xylose, fucose, putrescine, gluconic acid, trehalose, mannose and sucrose. In March, four metabolites from brown pears varied in the radial direction (isoleucine, GABA, pyroglutamic acid, and quinic acid). These results are confirmed by the PLS-DA analysis (Fig. 3) in which the effect of position was smaller than the factor disorder.

was observed: the xylose concentration was five times higher at 6 months of storage compared to 4 months of storage. In brown pears, as storage time increased, a decrease in fumaric acid, malic acid, erythritol and GABA was observed. However, threonic acid, fucose, quinic acid, gluconic acid, trehalose and mannose accumulated with storage time.

3.4. Metabolic differences due to disorder

4.1. Radial gradients of metabolites

Comparison between the inner part of brown tissue (5–8) and the inner part of sound tissue (5 –8 ) revealed significantly increased concentrations of fumaric acid, threose, GABA, pyroglutamic acid, fucose, gluconic acid, trehalose and sucrose in brown pears. Brown inner tissue was also characterized by reduced concentrations of malic acid, erythritol and threitol.

From Fig. 4 it is clear that metabolites are not uniformly distributed in pear fruit tissue and their concentration may vary considerably in the radial direction, particularly in the storage time of March. Similar observations have been made by Franck et al. (2003b) who mapped the l-ascorbic acid distribution in pear slices and found considerable radial gradients. For ascorbic acid synthesis at least a possible link to respiration pathways such as the tricarboxylic acid cycle has been reported (Nunes-Nesi et al., 2005). As several of the metabolites are involved in the respiratory pathways, their spatial distribution is likely a consequence of the gradients in O2 and CO2 in the tissue (Lammertyn et al., 2003). Geigenberger et al. (2000) also found a gradient of metabolites in potatoes reflecting the lower internal oxygen concentration towards the center of the

3.5. Effect of storage time The same trends were measured after 6 months of storage as in the samples taken after 4 months of storage (Fig. 4). There was a clear interaction time × disorder for some compounds. In all pears (brown and sound pears), a remarkable rise in xylose concentration

4. Discussion

128

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

tuber. For pears stored under browning inducing conditions (1 kPa O2 , 10 kPa CO2 ), Ho et al. (2008) predicted the formation of gas gradients, with concentrations of oxygen towards the center of the fruit low below the concentration needed by cytochrome c oxidase to maintain an aerobic metabolism. 4.2. Metabolites involved in browning The different metabolites involved in browning have been grouped according to their metabolic function. Thus, browning seems to be the result of: (1) A disturbed energy metabolism: The observed differences between brown and sound tissue point to a greatly altered respiration metabolism as a consequence of hypoxic stress. Malic acid appears to be a key metabolite to distinguish sound from brown tissue. The concentration of malic acid is clearly higher in sound than in brown tissue, that of fumaric and succinic acid vice versa. The hydration of fumaric acid to malic acid is catalyzed by fumarase; this enzyme has been shown to be down-regulated in Arabidopsis cell cultures submitted to oxidative stress (Sweetlove et al., 2002). Fumaric acid also has been shown to act as an activator of malate dehydrogenase (Grissom et al., 1983). Pedreschi et al. (2007) in tissues from ‘Conference’ pears with core breakdown found that the expression of fumarase was completely down-regulated in brown tissue which is consistent with the observed decrease of malic acid and increase of fumaric acid here. Further, the same authors found that malic enzyme, which catalyses the oxidative decarboxylation of malic acid to pyruvate, CO2 and NADPH, was up-regulated in brown tissue. Malic enzyme is involved in many functions in plants including fruit ripening, anabolic functions to provide NADPH and pyruvate for biosynthesis, catabolic functions to provide NADPH and pyruvate for energy production by respiration and the maintenance of intracellular pH (Edwards and Andreo, 1981). Malic enzyme is believed to be up regulated under stress conditions to provide energy for cellular repair processes and also substrates for fatty acid synthesis in order to repair membranes (Casati et al., 1999). Tecsi et al. (1996) investigated the activity of malic enzyme in cotyledons of marrow plants infected by cucumber mosaic virus. They found an increased activity of malic enzyme in the affected part compared to the apparently healthy part. Pedreschi et al. (2007) also found that the expression of malate dehydrogenase was down regulated in brown tissue; this is possibly an additional attempt of the plant cell to steer malic acid through the pathway catalyzed by malic enzyme. Both routes may lead to an increased fumaric acid concentration under reduced respiratory activity. Note that fumaric acid is a by-product of the urea cycle, which serves to eliminate excess nitrogen due to proteolysis and consecutive amino acid degradation. Also, a partial reversal of the Krebs cycle has been reported in fruit under anoxic conditions (Davies, 1980; Vanlerberghe et al., 1990), and exposure of suspension-cultured pear fruit cells to hypoxia resulted in an increased phosphoenolpyruvate carboxykinase activity (Nanos et al., 1994). High CO2 partial pressures during hypoxia might facilitate the conversion of oxalacetate from phosphoenolpyruvate which via the reversal of the Krebs cycle might also lead to accumulation of fumaric acid. While these metabolic routes might also be associated with the increase of fumaric acid, more research is required to validate this hypothesis. Future experiments should focus on the role of carbon dioxide and oxygen on the pear cell metabolism via metabolic profiling and metabolic flux analysis.

Brown tissue is characterized by a higher sucrose concentration possibly as a consequence of the partial impairment of glycolysis and tricarboxylic acid cycle in brown pears. (2) Alteration in concentrations of metabolites dependent on energy metabolism pathways: The accumulation of gluconic acid could be the result of the degradation of ascorbic acid. 5-Keto-d gluconic acid is an intermediate in the degradation of ascorbic acid towards oxalic and l-tartaric acids (Debolt et al., 2007). This coincides with the reported lower concentration of ascorbic acid in brown tissue compared to the sound areas (Franck et al., 2003b) in a previous study. If the oxidized form of ascorbate (dehydroascorbate) is not reduced to ascorbate by dehydroascorbate reductase, which needs glutathione in its reduced form, rapid catabolism of ascorbic acid yielding gluconic acid as an intermediate, can occur. The gluconic acid accumulation in brown tissue seems to be an indication of lack of reducing equivalents. The considerable increase of the gluconic acid concentration in brown tissue could also be an indication that the pentose phosphate pathway is activated instead. However, accumulation of gluconic acid in brown tissue seems to indicate that this pathway is blocked in the oxidative part at the 6-phosphogluconate level and thus gluconic acid accumulates and NADPH production is limited. NADPH is important for biosynthetic processes including fatty acid synthesis, and to maintain the redox potential to protect against oxidative stresses (Kruger and von Schaewen, 2003). In addition the NADPH provided by the pentose phosphate pathway participates in the reduction of nitrate to nitrite for further downstream production of polyamines and GABA, eventually leading to succinate to enter the Krebs cycle. NAPDH is also required under stress conditions that cause overproduction of reactive oxygen species (Keyhani et al., 2006). These need to be scavenged by glutathione, which uses NADPH to return to its reduced form (Garnczarska, 2005). This finding is in accordance with our results. The apparently sound tissue from brown pears did not show as high an accumulation of GABA as brown tissue possibly indicating the pentose phosphate is still active to provide NADPH for other reactions that feed substrates to the Krebs cycle. This is supported by Pedreschi et al. (2007) who found an up-regulation of glutathione-S-transferase in sound tissue from sound pears but total disappearance in brown tissue. This is likely due to the presence of an adequate oxygen concentration above the minimum needed by cytochrome c oxidase to carry on respiratory processes in sound tissue. As discussed above, the GABA concentration was also considerably higher in brown than in sound pears (Fig. 4) indicating that possibly GABA in combination with malate and fumarate are important indicators for browning. GABA is considered to be a regulator of cytosolic pH, a reserve of C and/or N, and a signaling molecule in case of exposure to a biotic stress (Kinnersley and Turano, 2000). GABA is synthesized through decarboxylation of glutamic acid which is produced from ␣-ketoglutaric acid by transamination reactions. Anoxia and other stress conditions lead to cytosolic increases in Ca2+ , which stimulate glutamic acid decarboxylase activity (Ferreira de Sousa and Sodek, 2002). A decrease of glutamic acid was indeed found in brown tissue (data not shown). In addition, GABA:pyruvic acid transaminase is known to be inhibited in anaerobic conditions (Streeter and Thompson, 1972). Both phenomena cause an increase in GABA concentration. It is reasonable to expect that GABA may therefore accumulate in anaerobic conditions. Besides, a partial blocking of the TCA cycle was found between fumarate and malate resulting in fumaric acid accumulation, which could feed back towards GABA accumulation. The mea-

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

sured GABA accumulation in brown tissue, therefore, gives biochemical evidence for the presence of anoxic zones in pears. Accumulation of GABA and other amino acids under anaerobiosis has been related to an increased activity of the glutamine synthetase/glutamine synthase cycle, for example in rice roots (Reggiani et al., 2000). The increased level of putrescine, a polyamine, in brown tissue is probably an indication of oxidative stress. Ye et al. (1997) found that oxidative stress increased the level of putrescine, and arginine and ornithine decarboxylase in Conyza bonariensis and wheat. Changes in transcripts related to putrescine have also been found in Arabidopsis cells submitted to oxidative stress (Baxter et al., 2007). The high putrescine content encountered in brown tissue can be associated to the biosynthesis of putrescine via the arginine decarboxylase pathway. The synthesis of putrescine involves the release of fumarate as a by product (Slocum et al., 1984). Notice that high fumaric acid content was found in brown tissue. (3) Collapsed antioxidant system: Accumulation of threonic acid in brown tissue might be related to ascorbic acid breakdown. It has been previously reported for ‘Conference’ pears submitted to stress that brown tissue contains a significantly lower amount of ascorbic acid (Franck et al., 2003a,b). Pedreschi et al. (2007) showed total down-regulation of ascorbate peroxidase, glutathione S transferase and monodehydroascorbate reductase in brown tissue indicative of total impairment of the ascorbateglutathione cycle. The increased concentration of the disaccharide trehalose in brown tissue provides further evidence of stress. The role of trehalose in plants is not completely elucidated, but it has been found to accumulate in a wide variety of organisms that withstand drought, salt, heat, or freeze stress (Elbein et al., 2003; Avonce et al., 2004). Trehalose-6-P or trehalose-6-P synthase have been shown to have an important role in controlling sugar metabolism in yeast through the regulation of the glycolysis, for instance, at the level of hexokinase activity (Hohmann et al., 1993). Trehalose also acts as a stabilizer and protectant of membranes by oxygen radicals in yeast. We have shown that trehalose is apparently associated with the hypoxic stress metabolism in core breakdown. An enhanced tolerance for abiotic stresses has been reported in tomato to be related to ˜ trehalose biosynthesis (Cortina and Culianez-Marcia, 2005). The decrease of polyols such threitol and erythritol in brown tissue is another indication of stress. Polyols have been reported to actively respond to abiotic stresses (Noiraud et al., 2001). (4) Collapsed cell wall architecture: Fucose has been found in xyloglucan, complex pectins as well as in some glycans from glycoproteins (Darley et al., 2001); mannose, xylose and lyxose have also been identified in xyloglucan which is a primary cell wall hemicellulose (Miller and Fry, 2001). Their increased concentration in brown tissue might indicate breakdown of the cell wall complex, either by enzymatic action of xyloglucan endotransglycosylase or by hydroxyl radicals. Whether the latter is a significant mechanism of cell wall breakdown is, however, still unclear (Miller and Fry, 2001). The remarkable five-fold increase of the concentration of xylose at 6 months of storage compared to 4 months seems to confirm the hypothesis of cell wall breakdown as these processes continue after cell death unlike other metabolic processes. 5. Conclusion The above results clearly indicate an impaired respiratory metabolism. Major changes were observed at the level of the Krebs cycle, with a decrease in malic acid, and an increase in succinic and

129

fumaric acid concentration. These changes corresponded well to changes at the proteome level (Pedreschi et al., 2007). The accumulation of gluconic acid in brown tissue indicated also impairment of the pentose phosphate pathway most probably with an insufficient production of NADPH for membrane repair processes and to maintain the cellular redox state. The accumulation of gluconic and threonic acids might also be the result of ascorbic acid catabolism due to lack of reducing equivalents. The imbalance between GABA and putrescine compared to a reduced glutamate concentration in brown tissue may be considered as an indicator of hypoxic stresses. The concentration of other compounds, which are believed to be related to an oxidative stress response such as trehalose and putrescine was also considerably higher in brown tissue than in sound tissue. Finally, the concentration of some sugars, which are typically found in xyloglucans also increased during the core browning process, possibly indicating primary cell wall breakdown due to enzymatic processes or by action of hydroxyl radicals. The results indicate that core breakdown development is a consequence of an imbalance between oxidative and reductive processes caused by too low oxygen or too high carbon dioxide conditions which lead to a deficiency of reducing equivalents for defensive mechanisms, cell damage repair processes and biosynthesis reactions. A correlation analysis in situ with oxygen and carbon dioxide concentrations and respiratory activity will help to elucidate pool changes and fluxes through the pentose phosphate and glycolysis pathways and thus may increase our understanding of this browning disorder. Quantitative techniques such as metabolic flux analysis are required to further elucidate the biochemical chain of events which lead to core breakdown. Acknowledgements This research was financially supported by the Research Council of the K.U. Leuven (project IDO 00/008; OT 04/31) and the Institute for the Promotion of Innovation by Science and Technology in Flanders (IWT). R. Pedreschi extends the acknowledgment to the International Relations Office of the K.U. Leuven (IRO scholarship). References Avonce, N., Leyman, B., Mascorro-Gallardo, J.O., Van Dijck, P., Thevelein, J.M., Iturriaga, G., 2004. The arabidopsis trehalose-6-P synthase AtTPS1 gene is a regulator of glucose, abscisic acid and stress signalling. Plant Physiol. 136, 3649–3659. Baldwin, E., Nisperos-Carriedo, M., Bakerm, R., Scott, J., 1991. Quantitative analysis of flavor parameters in six Florida tomato cultivars (Lycopersicon esculentum Mill.). J. Agric. Food Chem. 39, 1135–1140. Baxter, C., Redestig, H., Schauer, N., Repsilber, D., Patil, K., Nielsen, J., Liu, J., Fernie, A., Sweetlove, L., 2007. The metabolic response of heterotrophic Arabidopsis cells to oxidative stress. Plant Physiol. 143, 312–325. Buttery, R.G., Teranishi, R., Ling, L.C., Flath, R.A., Stern, D.J., 1988. Quantitative studies on origins of fresh tomato aroma volatiles. J. Agric. Food Chem. 36, 1247–1250. Casati, P., Drinkovich, M.F., Edwards, G.E., Andreo, C.S., 1999. Malate metabolism by NADP-malic enzyme in plant defense. Photosyn. Res. 61, 99–105. ˜ Cortina, C., Culianez-Marcia, F., 2005. Tomato abiotic stress enhanced tolerance by trehalose biosynthesis. Plant Sci. 169, 75–82. Darley, C.P., Forrester, A.M., Mc-Queen-Mason, S.J., 2001. The molecular basis of plant cell wall extension. Plant Mol. Biol. 47, 179–195. Davies, D.D., 1980. In: Stumpf, R.K., Conn, E.E. (Eds.), Anaerobic Metabolism and the Production of Organic Acids. The Biochemistry of Plants: A Comprehensive Treatise. Academic Press, New York, pp. 581–611. Debolt, S., Melino, V., Ford, C., 2007. Ascorbate as a biosynthetic precursor in plants. Ann. Bot. 99, 3–8. Edwards, G.E., Andreo, C.S., 1981. NADP-malic enzyme from plants. Photosyn. Res. 31, 1845–1857. Elbein, A.D., Pan, Y.T., Pastusza, I., Carroll, D., 2003. New insights on trehalose: a multifunctional molecule. Glycobiology 13, 17R–27R. Erban, A., Schauer, N., Fernie, A.R., Kopka, J., 2007. Non-supervised construction and application of mass spectral and retention time index libraries from timeof-flight GC-MS metabolite profiles. In: Weckwerth, W. (Ed.), Metabolomics: Methods and Protocols. Humana Press, Totowa, pp. 19–38. Ferreira de Sousa, C.A., Sodek, L., 2002. The metabolic response of plants to oxygen deficiency. Braz. J. Plant Physiol. 14, 83–94.

130

R. Pedreschi et al. / Postharvest Biology and Technology 51 (2009) 123–130

Fiehn, O., Kopka, J., Trethewey, R.N., Willmitzer, L., 2000. Identification of uncommon plant metabolites based on calculation of elemental compositions using gas chromatography and quadrupole mass spectrometry. Anal. Chem. 72, 3575–3580. Franck, C., Baetens, M., Lammertyn, J., Verboven, P., Nicolaï, B.M., 2003a. Ascorbic acid content during fruit development and postharvest storage of Conference pears. J. Agric. Food Chem. 51, 4757–4763. Franck, et al., Franck, C., Baetens, M., Lammertyn, J., Scheerlinck, N., Nicolaï, B.M., 2003b. Ascorbic acid mapping to study core breakdown. Postharvest Biol. Technol. 30, 133–142. Franck, C., Lammertyn, J., Ho, Q.T., Verboven, P., Verlinden, B., Nicolaï, B.M., 2007. Browning disorders in pear fruit. Postharvest Biol. Technol. 43, 1–13. Garnczarska, M., 2005. Response of the ascorbate-glutathione cycle to reaeration following hypoxia in lupine roots. Plant Physiol. Biochem. 43, 583– 590. Geigenberger, P., Fernie, A.R., Gibon, Y., Crist, M., Stitt, M., 2000. Metabolic activity decreases as an adaptative response to low internal oxygen in growing potato tubers. Biol. Chem. 381, 723–740. Grissom, C.B., Canellas, P.F., Wedding, R.T., 1983. Allosteric regulation of the NAD malic enzyme from cauliflower: activation by fumarate and coenzyme A. Arch. Biochem. Biophys. 220, 133–144. Ho, Q.T., Verlinden, B.E., Verboven, P., Vandewalle, S., Nicolaï, B.M., 2006. A permeation-diffusion-reaction model of gas transport in cellular tissue of plant materials. J. Exp. Bot. 57, 4215–4224. Ho, Q., Verboven, P., Verlinden, B., Lammertyn, J., Vandewalle, S., Nicolaï, B., 2008. A continuum model for metabolic gas exchange in pear fruit. PLoS Comput. Biol. 4, 1–13. Hohmann, S., Neves, M.J., de Koning, W., Alijo, R., Ramos, J., Thevelein, J.M., 1993. The growth and signalling defects of the ggs1 (fdp1/byp1) deletion mutant on glucose are suppressed by a deletion of the gene encoding hexokinase PII. Curr. Genet. 23, 281–289. Jellum, E., Kvittingen, E.A., Stokke, O., 1988. Mass spectrometry in diagnosis of metabolic disorders. Biomed. Environ. Mass Spectrom. 16, 57–62. Karp, N., Graffin, J., Lilley, K., 2005. Application of partial least squares discriminant analysis to two-dimensional difference gel studies in expression proteomics. Proteomics 5, 81–90. Katona, Zs.F., Sass, P., Molnár-Perl, I., 1999. Simultaneous determination of sugars, sugar alcohols, acids and amino acids in apricots by gas chromatography-mass spectrometry. J. Chromatogr. B 847, 91–102. Keyhani, E., Ghamsari, L., Keyhani, J., Hadizadeh, M., 2006. Antioxidant enzymes during hypoxia–anoxia signalling events in Crocus sativus L. Ann. N. Y Acad. Sci. 1091, 65–75. Kinnersley, A.M., Turano, F.J., 2000. Gamma-aminobutyric acid (GABA) and plant responses to stress. Crit. Rev. Plant. Sci. 19, 479–509. Kopka, J., Schauer, N., Krueger, S., Birkemeyer, C., Usadel, B., Bergmuller, E., Dormann, P., Weckwerth, W., Gibon, Y., Stitt, M., Willmitzer, L., Fernie, A.R., Steinhauser, D., 2005. GMD@CSB. DB: the golm metabolome database. Bioinformatics 21, 1635–1638. Kruger, N.J., von Schaewen, A., 2003. The oxidative pentose phosphate pathway: structure and organisation. Curr. Opin. Plant. Biol. 6, 236–246. Lammertyn, J., Franck, C., Verlinden, B.E., Nicolaï, B.M., 2001. Comparative study of the O2 , CO2 and temperature effect on respiration between Conference pear cell protoplasts in suspension and intact pears. J. Exp. Bot. 362, 1769– 1777. Lammertyn, J., Scheerlinck, N., Jancsók, P., Verlinden, B.E., Nicolaï, B.M., 2003. A respiration-diffusion model for Conference pears I: model development and validation. Postharvest Biol. Technol. 30, 29–42. Larrigaudiere, C., Pinto, E., Lentheric, I., Vendrell, M., 2001. Involvement of oxidative processes in the development of core browning in controlled atmosphere pears. J. Hort. Sci. Biotechnol. 76, 157–162. Miller, J.G., Fry, S.C., 2001. Characteristics of xyloglucan after attack by hydroxyl radicals. Carbohydr. Res. 332, 389–403.

Nanos, G.D., Romani, R.J., Kader, A.A., 1994. Respiratory metabolism of pear fruit and cultured cells exposed to hypoxic atmospheres: associated change in activities of key enzymes. J. Am. Soc. Hort. Sci. 119, 288–294. Noiraud, N., Maurousset, L., Lemoine, R., 2001. Transport of polyols in higher plants. Plant Physiol. Biochem. 39, 717–728. Norden, B., Broberg, P., Lindberg, C., Plymoth, A., 2005. Analysis and understanding of high dimensionality data by means of multivariate data analysis. Chem. Biodivers. 2, 1487–1494. Nunes-Nesi, A., Carrari, F., Lytovchenko, A., Smith, A., Loureiro, M., Ratcliffe, R., Sweetlove, L., Fernie, A., 2005. Enhanced photosynthetic performance and growth as a consequence of decreasing mitochondrial malate dehydrogenase activity in transgenic tomato plants. Plant Physiol. 137, 611–622. Pedreschi, R., Vanstreels, E., Carpentier, S., Hertog, M., Lammertyn, J., Robben, J., Noben, J.P., Swennen, R., Vanderleyden, J., Nicolaï, B.M., 2007. Proteomic analysis of core breakdown disorder in Conference pears (Pyrus communis L.). Proteomics 7, 2083–2099. Reggiani, R., Nebuloni, M., Mattana, M., Brambilla, I., 2000. Anaerobic accumulation of amino acids in rice roots: role of glutamine synthetase/glutamate synthase cycle. Amino Acids 18, 207–217. Roessner, U., Wagner, C., Kopka, J., Trethewey, R.N., Willmitzer, L., 2000. Simultaneous analysis of metabolites in potato tuber by gas chromatography–mass spectrometry. Plant J. 23, 131–142. Saquet, A., Streif, J., Bangerth, F., 2003. Energy metabolism and membrane lipid alterations in relation to brown heart development in ‘Conference’ pears during delayed controlled atmosphere storage. Postharvest Biol. Technol. 30, 123– 132. Schauer, N., Steinhauser, D., Strelkov, S., Schomburg, D., Allison, G., Moritz, T., Lundgren, K., Roessner-Tunali, U., Forbes, M.G., Willmitzer, L., Fernie, A.R., Kopka, J., 2005. GC-MS libraries for the rapid identification of metabolites in complex biological samples. FEBS Lett. 579, 1332–1337. Schouten, R.E., Veltman, R.H., De Wild, H.P.J., Koopen, T.J., Staal, M.G., Tijskens, L.M.M., 2004. Determination of O2 and CO2 permeance, internal respiration and fermentation for a batch of pears (cv. Conference). Postharvest Biol. Technol. 32, 289–298. Slocum, R.D., Kaur-Sawhney, R., Galston, A.W., 1984. The physiology and biochemistry of polyamides in plants. Arch. Biochem. Biophys. 35, 283–303. Streeter, J.G., Thompson, J.F., 1972. Anaerobic accumulation of ␥-aminobutyric acid and alanine in radish leaves (Raphanus sativus L.). Plant Physiol. 49, 572– 578. Sweetlove, L.J., Heazlewood, J.L., Herald, V., Holtzapffel, R., Day, D.A., Leaver, C.J., Millar, A.H., 2002. The impact of oxidative stress on Arabidopsis mitochondria. Plant J. 32, 891–904. Tecsi, L.I., Alison, M.S., Andrews, J.M., Leegood, R.C., 1996. A spatial analysis of physiological changes associated with infection of cotyledons of marrow plants with cucumber mosaic virus. Plant Physiol. 111, 975–985. Vanlerberghe, G.C., Feil, R., Turpin, D.H., 1990. Anaerobic metabolism in the N limited green alga Selenastrum minutum I. Regulation of carbon metabolism and succinate as a fermentation product. Plant Physiol. 94, 1116–1123. Veltman, R., Lenthéric, L., van der Plas, L., Peppelenbos, H., 2003. Internal browning in pear fruit (Pyrus communis L. cv. Conference) may be a result of a limited availability of energy and antioxidants. Postharvest Biol. Technol. 28, 295–302. Veriotti, T., Sacks, R., 2000. High-speed GC/MS of gasoline-range hydrocarbon compounds using a pressure-tunable column ensemble and time-of-flight detection. Anal Chem. 72, 3063–3069. Veriotti, T., Sacks, R., 2001. High-speed GC and GC/time-of-flight MS of lemon and lime oil samples. Anal Chem. 73, 4395–4402. Wagner, C., Sefkow, M., Kopka, J., 2003. Construction and application of a mass spectral and retention time index database generated from plant GC/EI-TOF-MS metabolite profiles. Phytochemistry 62, 887–900. Ye, H.H., Muller, J., Zhang, Gressel, J., 1997. Constitutively elevated levels of putrescine and putrescine-generating enzymes correlated with oxidant stress resistance in Conyza bonariensis and Wheat B. Plant Physiol. 115, 1443–1451.