•During harvest, fleshy berry tomato fruits (Solanum lycopersicum) were wounded at their stem scar. Within 3 d, this wound was rapidly sealed by a process covering the wound site with a membranous layer which effectively protects the tomato fruit from excessive water loss, nutrient elution and the entry of pathogens.
•Chemical analysis of the de novo synthesized stem scar tissue revealed the presence of aromatic and aliphatic components characteristic of the biopolyester suberin.
•Gene expression patterns associated with suberization were identified at the stem scar region. Changes in the relative abundance of different transcripts suggested a potential involvement of the plant hormone abscisic acid (ABA) in the wound-healing processes.
•The amount of ABA present in the stem scar tissue showed a significantly increased level during wound healing, whereas ABA-deficient mutants notabilis, flacca and sitiens were largely devoid of this rise in ABA levels. The mutant fruits showed a retarded and less efficient suberization response at the stem scar wound, whereas the rate and strength of this response were positively correlated with ABA content. These results clearly indicate in vivo the involvement of ABA in the suberization-based wound-healing processes at the stem scar tissue of tomato fruits.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Traditionally, tomato fruits are harvested individually from the vine. During the harvesting process, tomato fruits incur wounding caused by the removal of the receptacle at the fruit–pedicel junction. Rapid healing of the resulting stem scar wounds, including the development of a hydrophobic barrier to reduce water evaporation and/or transpiration, is crucial in protecting tomato fruits from microbial infection and dehydration during post-harvest storage. However, knowledge about the compositional nature, time course and regulation of wound-healing processes at the stem scar tissue of tomato fruits is very limited.
Suberin is a cell wall-associated biopolymer which, in many cases, acts as an interface separating living plant cells from their environment (Pollard et al., 2008; Ranathunge et al., 2011). Suberin consists of a polyaliphatic domain, typically in close association with a polyaromatic domain which is derived from ferulic acid (Bernards, 2002). Suberin is a complex biopolyester based on long-chain α,ω-alkandioic acids and ω-hydroxyalkanoic acids in a glycerol-bridged network (Graça & Santos, 2007). How these monomeric units are assembled at a macromolecular level remains hypothetical (Bernards, 2002; Franke & Schreiber, 2007). The exact qualitative and quantitative compositions of suberin monomers vary in different species.
Wound-induced suberization of tomato fruit surfaces has been described by Dean & Kolattukudy (1976). However, this process is complex and poorly understood. A rapid reduction in water permeance at the wound surface is essential to keep cells viable in order to allow for the subsequent formation of a suberized barrier. So far, it is not known how wounding induces the suberization of cells. Several studies on potato tubers have suggested that the abscisic acid (ABA) stress signal response may be involved in the regulation of wound healing (Soliday et al., 1978; Lulai et al., 2008). Plants that are challenged by drought stress recruit ABA as an endogenous signal to initiate adaptive responses (Bray, 1997; Zhu, 2002). On wounding, ABA levels of tomato leaves are preferentially increased near the wound site (Birkenmeier & Ryan, 1998). As a result, ABA has been suggested to accumulate on wounding because of desiccation. Nevertheless, the role of ABA in the wound-healing process of tomato fruits remains to be elucidated. The blocking of ABA biosynthesis through genetic mutants could provide a reliable means of determining the role of ABA in wound healing. Three ABA-deficient mutants have been induced in tomato lines by X-irradiation (Stubbe, 1957, 1958, 1959). Plants homozygous for the mutant alleles are known as notabilis, flacca and sitiens. All have characteristically low endogenous ABA concentrations and, consequently, still allow the formation of a small amount of ABA (Tal & Nevo, 1973).
The astomatous fruits of tomato have been proven to be suitable for the study of the function of the cuticular membrane with respect to the barrier properties (Vogg et al., 2004; Leide et al., 2007, 2011). Therefore, tomato fruits can be regarded as a model system to functionally analyze surface modifications, such as wound healing. The purpose of this study was to characterize the nature, time course, regulation and functional efficiency of the wound-healing processes at the stem scar tissue of tomato fruits. In this regard, gene expression patterns in the stem scar region and the relevance of ABA in the regulation of wound healing were investigated.
Materials and Methods
Fruits of tomato (Solanum lycopersicum L.) cv MicroTom were investigated. MicroTom wild-type and lecer6 mutant fruits with a deficiency in a fatty acid β-ketoacyl-coenzyme A synthase were used. The identification and characterization of this insertional mutant have been reported previously (Vogg et al., 2004; Leide et al., 2007). The MicroTom plants were cultivated in a growth chamber with c. 75% relative humidity, a 14-h photoperiod at 450 μmol m−2 s−1 and a day : night temperature regime of 22°C : 18°C. Plants were watered daily and fertilized with 1‰ Hakaphos Blau nutrient solution (w/v; Compo, Münster, Germany) once a week. In addition, fruits from cv Rheinlands Ruhm (RR), cv Ailsa Craig (AC) and the ABA-deficient mutants flacca (RR background), sitiens (RR background) and notabilis (AC background), cultivated under glasshouse conditions, were analyzed. Seeds of the RR and AC lines were obtained from the Tomato Genetics Resource Center (TGRC; University of California, Davis, CA, USA). Mature green fruits were harvested between 17 and 23 d after flowering, when fruit growth was completed, and immediately utilized or stored in darkness at 25°C until further processing.
Enzymatic isolation of fruit surface membrane layers
Immediately after harvest, or within a 6-d period of storage, the stem scar tissues of tomato fruits were enzymatically isolated according to Schönherr & Riederer (1986). For this purpose, cylindrical sections of the stem scar region, with a diameter of 4 mm for MicroTom and 8 mm for RR and AC, were excised from the tomato fruits with a cork borer and subjected to enzymatic isolation. The individual weights of the air-dried surface membrane layers were determined.
Tomato fruit cuticular membranes were injured by producing small incisions (c. 10 mm in length) with a scalpel. After 2 d of storage, the wounded tomato fruits were harvested and the wounded cuticular membranes were excised and isolated enzymatically.
Fluorescence microscopic analysis
Cylindrical sections of the stem scar region of tomato fruits were embedded in Tissue-Tek medium (Sakura Finetek, Zoeterwoude, The Netherlands) and cooled to − 22°C. Frozen longitudinal sections (30 μm) were obtained using a Leica CM 1900 cryostat (Leica Microsystems, Nussloch, Germany). Sections were investigated by UV fluorescence microscopy using a Leica DMR HC microscope system (excitation filter BP 450–490, suppression filter LP 515). Images were recorded using a Leica DC 50 digital camera combined with Leica IM 1000 software (version 1.2).
Toluidine blue infiltration
To visualize the barrier efficiency of the stem scar tissue, toluidine blue staining was adapted from Tanaka et al. (2004). Tomato fruits freshly harvested or fruits that had been stored for 4 d were submerged in 0.025% toluidine blue (v/v; Sigma) for 10 min at 4°C. The infiltration of toluidine blue solution into the tomato fruits through the stem scar region was inspected visually. Photographs were taken with a Leica DMZ 16 binocular microscope (Leica Microsystems).
Pathogen infection assay
Fusarium solani DSM 1164 was propagated on 3.9% potato dextrose agar (w/v; Sigma-Aldrich) in darkness at 25°C. Small agar plugs (5 × 5 mm2), containing actively growing mycelium of F. solani, were applied directly onto the stem scar area of freshly harvested or 4 d-stored tomato fruits, which underwent wound-healing processes. The inoculated tomato fruits were incubated for 7 d in a dark moist chamber at 25°C. Finally, tomato fruits were cut into halves and fungal penetration through the stem scar tissue was assessed by checking for a brownish discoloration of the tomato fruit tissue beneath the stem scar wound as an unequivocal indicator of fungal infection. Photographs were taken with a Leica DMZ 16 binocular microscope (Leica Microsystems).
Characterization of permeance for water
The permeance for water was determined for harvested tomato fruits with respect to the time of storage. The amount of water transpired was measured using a balance with a precision of 0.1 mg (Sartorius AC210S, Göttingen, Germany). The water flow rate (F, in g s−1) of individual tomato fruits was determined from the slope of a linear regression line fitted through the gravimetric data. The coefficient of determination (r2) averaged 0.999. The flux (J, in g m−2 s−1) was calculated by dividing F by the total fruit surface area. Between measurements, the tomato fruits were stored at 25°C over dry silica gel. Under these conditions, the external water vapor concentration was essentially zero. The vapor phase-based driving force (Δc) for transpiration was therefore 23.07 g m−3. To calculate the water permeance (P, in m s−1) based on the water vapor concentration, J was divided by Δc.
Characterization of ion leakage
The electrical conductivity of the incubation solution was determined for freshly harvested tomato fruits or fruits that had been stored for 4 d. Tomato fruits were washed with deionized water and, subsequently, incubated in a volume of 5 ml of deionized water at 25°C. The conductance of the ions released into the water was measured using an electrical conductivity meter (Twin Cond B-173; Horiba, Kyoto, Japan).
Extraction and determination of ABA content
For the extraction of endogenous ABA, the stem scar tissues of tomato fruits were excised with a scalpel and immediately frozen in liquid nitrogen. The frozen explants were homogenized in liquid nitrogen and twice extracted with 80% methanol at − 20°C overnight. The methanolic extracts were collected, passed through a C18 Sep-Pak cartridge and evaporated to the water phase. The aqueous residue was acidified to pH 3.0 and partitioned three times against ethyl acetate. The organic fractions were combined and evaporated to dryness and taken up in Tris-buffered saline, pH 7.8 (Peuke et al., 1994). The purified samples were analyzed using an enzyme-linked immunosorbent assay according to Weiler et al. (1986). The immunoassay was performed using rabbit anti-mouse immunoglobulin, monoclonal anti-ABA antibody and ABA-conjugated alkaline phosphatase as a tracer. The alkaline phosphatase activity was colorimetrically assayed with p-nitrophenylphosphate as a substrate. The absorbance of p-nitrophenol was measured photometrically at 405 nm. The ABA concentration was calculated using a linear regression from ABA standards.
Depolymerization and analysis of suberin
The total soluble wax-like compounds were removed by immersing 1.2 mg of enzymatically isolated stem scar tissue and wounded cuticular tissue of tomato fruits in chloroform at room temperature twice for 2 min. The fruit surface membranes were then transesterified with 1 ml of methanolic boron trifluoride (c. 1.3 M boron trifluoride in methanol; Fluka, Steinheim, Germany) at 70°C overnight to release methyl esters of suberin monomers. Sodium chloride-saturated aqueous solution, chloroform and n-dotriacontane (Sigma-Aldrich) as internal standard were added to all reaction mixtures. From this two-phase system, the depolymerized transmethylated components were extracted three times with chloroform. The combined organic phases were dried over sodium sulfate (anhydrous; Applichem, Darmstadt, Germany) and filtered. The organic solvent was evaporated under a continuous flow of nitrogen. Before gas chromatographic analysis, suberin compounds were transformed into the corresponding trimethylsilyl derivatives using N,O-bis-trimethylsilyl-trifluoroacetamide (Macherey-Nagel, Düren, Germany) in pyridine (Merck, Darmstadt, Germany). The qualitative and quantitative compositions were analyzed as described in detail previously (Leide et al., 2007).
Immediately after harvest, or within a 4-d period of storage, the stem scar tissues of tomato fruits were excised with a scalpel and immediately frozen in liquid nitrogen. Total RNA from 100 mg of homogenized stem scar tissue was isolated using the RNeasy mini kit (Qiagen), complying with the manufacturer’s instructions, including optional washing steps and a deoxyribonuclease digestion in solution (RNase-free DNase Set; Qiagen). After deoxyribonuclease digestion in solution, total RNA was extracted with one volume of phenol/chloroform/isoamyl alcohol (25 : 24 : 1, v/v/v; Roth, Karlsruhe, Germany) to remove proteins. The aqueous phase was washed with one volume of chloroform/isoamyl alcohol (24 : 1, v/v; Roth).
The RNA-containing aqueous phase was mixed with 2.5 volumes of 100% ethanol and 0.1 volume of 3 M sodium acetate, pH 6.5, to precipitate the RNA. Subsequently, the RNA was pelleted by centrifugation. The RNA pellet was washed with 70% ice-cold ethanol, air dried at room temperature and, finally, dissolved in diethyl pyrocarbonate-treated deionized water.
For cDNA synthesis, the concentration and purity of the isolated total RNA were quantified by measuring the UV absorbance using a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The integrity of RNA was controlled by electrophoresis on a denaturing agarose gel.
Aliquots of 3.85 μg of total RNA were converted in the presence of 2.5 μM oligo(dT)20 primer into first-strand cDNA using the SuperScript III/RNaseOut reverse transcriptase (Invitrogen), according to the manufacturer’s instructions.
The quantitative, fluorescence-based, real-time reverse-transcription PCR was performed using the KAPA SYBR fast qPCR master mix universal (Peqlab, Erlangen, Germany), complying with the manufacturer’s instructions, and 19.5 ng cDNA template per reaction. Gene-specific primer pairs were designed by Primer3 input version 0.4.0 software (http://frodo.wi.mit.edu/primer3/), controlled by in silico PCR (http://solgenomics.net/tools/insilicopcr/index.pl) and synthesized by Metabion International (Martinsried, Germany). For each primer pair, the primer concentrations were optimized (see Supporting Information Table S1).
Quantitative PCR analysis was carried out using a CFX 96 real-time detection system (Bio-Rad) according to the manufacturer’s directions. The PCR specificity was checked by melting curve analysis. Expected PCR product lengths were confirmed by the profile of agarose gel electrophoresis.
Relative gene expression data were processed using Bio-Rad CFX manager version 1.6 software according to the manufacturer’s instructions. The threshold cycle number (Ct) was determined. The entire experiment was performed in technical triplicates of each reaction and independent biological duplicates using β-tubulin Tub (SGN-U564000) as internal reference for normalization (ΔCt). Nontemplate controls were included. To calculate the average fold change in gene expression, the measured expression level was corrected for PCR efficiency, which was determined using a dilution range of cDNA of 200 to 2 ng, and normalized for nonstored fruits (0 d) adjusting to factor 1 (ΔΔCt). The PCR efficiency of each primer pair was calculated from the slope of a linear regression: E = 10(−1/slope).
Data for tomato fruits within different days of storage were tested by one-way analyses of variance (ANOVAs) followed by Tukey’s honestly significant difference mean separation test for unequal n, or Kruskal–Wallis ANOVA and subsequent multiple comparisons of ranks. One-way ANOVA was only used when the homogeneity of variances was given. Accordingly, comparisons between tomato fruits were tested with Student’s t-test or Mann–Whitney U-test. Spearman’s correlation of ranks was used to estimate the correlation coefficient (R2) of data belonging to days within storage (P <0.05). Statistical analyses were performed with STATISTICA 7.1 (StatSoft, Tulsa, OK, USA).
Microscopic analysis of the stem scar wound-healing process of tomato fruits
The stem scar region of mature green MicroTom fruits was investigated by UV-excited autofluorescence microscopy during 4 d of storage (Figs 1, S1). After harvest, a green autofluorescence signal accumulated beneath the outermost cell layer of the stem scar tissue. After 3 d of storage, this newly formed, brightly fluorescing, tissue completely covered the stem scar wound.
Reduction in water permeance by the newly formed stem scar membrane of tomato fruits
In order to obtain a more detailed view of the function of the stem scar tissue formed during the 6-d period of storage, the permeance for water of harvested MicroTom wild-type fruits and the weight of the enzymatically isolated stem scar tissues were determined (Figs 2a, S2). The permeance for water of mature green MicroTom fruits was reduced to 41% of the initial value after 1 d of storage. After 2 d, the water permeance was decreased by another 11%. During the following 4 d of storage, the permeance for water remained almost constant and ranged between 2.2 × 10−5 and 2.4 × 10−5 m s−1, which was c. 22% of the value measured for nonstored MicroTom fruits immediately after the harvesting process and 71% of the value obtained with MicroTom fruits sealed with paraffin wax at their stem scar region.
Mature green fruits of the MicroTom lecer6 wax mutant – significantly impaired in cuticular water barrier properties (Vogg et al., 2004; Leide et al., 2007) – exhibited a similar pattern of water permeance reduction, but with significantly higher values (Fig. 2b).
This decrease in permeance for water in MicroTom fruits could be prevented by storing the fruits at 4°C instead of 25°C, or by chemical treatments with inhibitors (Fig. S3).
The average weights of the enzymatically isolated stem scar tissues were determined for both MicroTom lines subject to the time of storage (Fig. 2a,b). During 6 d of storage, the stem scar tissue weights increased 2.6-fold for the wild-type and 2.9-fold for the lecer6 mutant, whereas the weights of the stem scar tissues from MicroTom fruits stored at 4°C remained unchanged. A significant negative correlation between the permeance for water and the formation of the stem scar tissue of MicroTom wild-type and MicroTom lecer6 fruits throughout the storage period was found (wild-type, R2 = − 0.79; lecer6, R2 = −0.95; P <0.05, Spearman’s correlation of ranks).
Functional characterization of the newly formed stem scar membrane as a barrier of nutrient outflow of tomato fruits
To characterize the importance of the stem scar tissue as a barrier for nutrient elution, the ion leakage of mature green MicroTom fruits was determined. MicroTom fruits were freshly harvested or stored for 4 d before incubation in deionized water. As controls, MicroTom fruits, which were freshly harvested and sealed with paraffin wax at their stem scar region immediately after harvest, were analyzed.
During the course of incubation, the electrical conductivity of the solution that enclosed freshly harvested, nonstored MicroTom wild-type fruits increased successively up to a factor of 53.7 (Fig. 3a). Starting at a level of 5.2 ± 0.4 μS cm−1 (mean ± SE) on the first day of incubation, the conductivity averaged 276.5 ± 23.5 μS cm−1 on the fourth day. During the same incubation period, the solution of stored or paraffin wax-treated fruits showed an increase of only c. 11.6-fold.
Comparable values for electrical conductivity, as found for MicroTom wild-type fruits, were detected for MicroTom lecer6 fruits (Fig. 3b). The incubation solution of stored or paraffin wax-sealed lecer6 mutant fruits exhibited an elevated conductivity of c. 14.0-fold on the fourth day of incubation, whereas nonstored fruits caused a 38.0-fold increase in conductivity.
Decline in microbial infection by the newly formed stem scar membrane of tomato fruits
To further characterize the functional properties of the wound-induced stem scar tissue, particularly as a potential portal of microbial entry during storage, a bioassay employing the tomato fruit pathogenic fungus F. solani was established. For this purpose, the infection rate of F. solani was recorded subject to different periods of fruit storage before inoculation (Fig. 4c,d). Although 65% of freshly harvested and subsequently inoculated mature green MicroTom wild-type fruits showed a mycelial in-growth of F. solani, only 38% of the MicroTom fruits stored for 4 d before fungal inoculation exhibited a visible fungal penetration through the stem scar region.
In addition, toluidine blue infiltration with stored and nonstored MicroTom wild-type fruits demonstrated the efficient barrier properties of the stem scar wound, preventing infiltration after 4 d of storage, whereas freshly harvested fruits absorbed blue-colored aqueous solution through the stem scar region (Fig. 4a,b).
Chemical analysis of the enzymatically isolated stem scar membrane from stored tomato fruits
Wound-healed stem scar tissues were isolated after enzymatic digestion and chloroform extraction (Fig. S4). The membranous layers mainly consisted of saturated and unsaturated alkanols, alkanoic acids, ω-hydroxyalkanoic acids and α,ω-alkandioic acids, as aliphatic compounds with chain lengths of C16–C30, and ferulic acid (Fig. 5a). Even-numbered, saturated alkanols, alkanoic acids, ω-hydroxyalkanoic acids and α,ω-alkandioic acids were found for all chain lengths. In addition, for the chain length of C18, corresponding mono- and di-unsaturated components were identified. Aliphatics with chain lengths of C22 represented the most prominent proportion (23%) within the isolated membranous layers, followed by C18 and C16 compounds. However, α,ω-alkandioic acids constituted the major compound class, with a proportion of 29%. When comparing the stem scar tissues of wound-healed MicroTom fruits with the suberized tissue of wounded MicroTom fruit cuticles, with respect to their chemical nature, both tissues exhibited a rather similar composition with only minor differences (Fig. 5b).
Gene expression analysis of the stem scar tissue during the storage of tomato fruits
A custom-made tomato microarray was established for the preliminary screening of genes putatively involved in cuticle biosynthesis and for some genes known to be stress responsive. In addition, these genes may also play a role in the biosynthesis of the barrier biopolymer suberin. To compare changes in relative transcript abundance during the wound-healing process, the ratio of the expression signals between nonstored and stored fruits was determined (Table S2). The lipid transfer protein transcript Tsw12, ABA and environmental stress inducible transcript Tas14, ABA stress ripening transcript Asr4 and early-responsive to dehydration transcript Erd7 showed differential expression patterns during the first days of fruit storage.
In order to corroborate the differences obtained in gene expression with respect to hybridization signal intensities, an independent, quantitative PCR approach was performed for the stem scar tissue. Quantitative PCR analysis was performed with gene-specific primer pairs for β-tubulin Tub as reference for normalization and for elongation factor 1-αX14449 as an internal control (Fig. 6f).
The expression levels for the genes Tsw12, Tas14 and Asr4 were induced after 1 d of storage (Fig. 6a–c). The transcript level of the Tsw12 gene was elevated by 105.2-fold, whereas the transcript amounts of the Tas14 and Asr4 genes increased by 25.6-fold and 20.1-fold, respectively. Nevertheless, a decrease in gene expression for these transcripts was found from days 1–4 of storage. Reductions to 16.7-fold for Tsw12 transcripts and to 16.2-fold for Tas14 transcripts were detected compared with the initial levels immediately after harvest. Moreover, the Asr4 transcript level decreased to 8.0-fold of the initial value during the final days of storage (day 2 to day 4).
The transcript level of the Erd7 gene was reduced in the stem scar tissue of stored fruits (Fig. 6d). During the course of the first day of storage, the transcript amount of the Erd7 gene decreased by a factor of 5.0 and remained at this reduced level during the 4 d spanning storage. On the last day of storage, the Erd7 transcript level accounted for 33% of the initial value immediately after fruit harvest.
The transcriptional regulation of the Cer6 gene in the stem scar region was also investigated by quantitative PCR analysis (Fig. 6e). At the beginning of storage, the transcript amount decreased to 0.6-fold of the initial value and, afterwards, showed a slight increase by a factor of 1.7 on day 4 of storage.
Influence of endogenous ABA on wound healing of the stem scar tissue during storage of tomato fruits
The ABA content in the stem scar region of mature green MicroTom wild-type fruits was determined. The stem scar tissue of freshly harvested, nonstored fruits had an ABA content of 6.7 ± 0.6 nmol g−1 FW. One day after harvest, ABA accumulated significantly (3.7-fold) and remained at an elevated level (c. 2.7-fold) up to day 4 after harvest (Fig. 7).
To further specify the stem scar tissue, the expression of the 9-cis-epoxycarotenoid dioxygenase gene Not, the molybdenum cofactor sulfurase gene Flc and the aldehyde oxidase gene Sit, which encode enzymes that catalyze steps in ABA biosynthesis, were analyzed. The gene expression levels showed only a slight transcriptional activation on day 1 (1.6-fold) for the Not gene and on days 3 and 4 (1.5-fold to 2.2-fold) for the Flc gene (Fig. 8a–c). However, the transcript levels of Sit increased by a factor of 5.9 during the first day of storage and remained at an elevated level of c. 4.6-fold during the course of the 4-d storage, thus reflecting the pattern of ABA accumulation during fruit storage.
In order to investigate the influence of ABA on the wound-healing processes, mature green fruits of tomato lines with a defect in ABA metabolism were analyzed. The tomato lines notabilis, flacca and sitiens– functionally defective in 9-cis-epoxycarotenoid dioxygenase, molybdenum cofactor sulfurase and aldehyde oxidase, respectively – in the genetic background of cv AC and cv RR, were selected.
With regard to the ABA content of the stem scar tissue during fruit storage, a similar pattern as in the MicroTom wild-type was found for wild-type fruits of RR and AC, but with a reduced ABA content of c. 3.9-fold (Fig. 9a,b). Within the first 2 d of storage, the ABA level increased significantly by 3.7-fold for RR wild-type and 3.1-fold for AC wild-type. Following the time course of storage, the ABA content was reduced slightly, but an elevated ABA content of 3.3-fold to 2.5-fold remained. The ABA amounts of the ABA-deficient mutants flacca, sitiens and notabilis were not enhanced significantly on wounding (Fig. 9c–e).
To functionally characterize the wound-induced stem scar tissue under conditions of low endogenous ABA concentrations, the water permeance and stem scar tissue weights were determined for mature green fruits of RR and AC wild-type and their ABA-deficient mutants RR flacca, RR sitiens and AC notabilis during the 6-d period of storage (Fig. 10a–e). Freshly harvested, wild-type fruits of RR exhibited a permeance for water of (6.0 ± 1.6) × 10−5 m s−1, whereas AC fruits showed a water permeance of (5.7 ± 1.1) × 10−5 m s−1.
The water permeance of RR and AC wild-type fruits continuously fell to about one-third of the initial value within 6 d of storage. The weights of enzymatically isolated stem scar tissues increased by factors of 3.3 and 2.0 when compared with nonstored fruits (Fig. 10a,b). The storage of fruits at 4°C prevented this increase. A significant negative correlation was found between the permeance for water and the weight of the stem scar tissues throughout the storage of RR and AC wild-type fruits (RR, R2 = − 0.96; AC, R2 = − 0.96; P <0.05, Spearman’s correlation of ranks).
Immediately after harvest, flacca and sitiens mutant fruits exhibited a comparable water permeance to RR wild-type fruits (Fig. 10c–e). During storage, the permeance for water decreased significantly for flacca to 47% and for sitiens to only c. 74% of the initial value. A similar pattern was observed for notabilis with a reduction to 59% of the initial value. A slight weight increase (1.3-fold) of the stem scar tissues became apparent after 6 d of storage. The decline in water permeance did not correlate with the slightly increased weights of the isolated stem scar tissues of the stored ABA-deficient mutant fruits.
Wound-healing process of tomato fruits
A fast and efficient wound closure stops excessive water loss and prevents pathogenic organisms from entering through wounded plant surfaces. Suberization is supposed to be a common plant response to wounding and plays an important role in the formation of a protective, impervious layer (Kolattukudy, 1981). Suberin is a specific lipophilic biopolymer characterized by the deposition of a polyaromatic domain associated with the cell wall, and a polyaliphatic domain thought to be deposited between the cell wall and plasma membrane (Bernards, 2002; Lulai et al., 2008). The aromatic domain primarily consists of p-coumaric and ferulic acids, which are presumably involved in covalently linking the aromatic domain to the cell wall (Bernards, 2002; Santos & Graça, 2006), whereas the aliphatic domain of suberin comprises long-chain aliphatic compounds, such as alkandioic acids, hydroxyalkanoic acids, alkanoic acids and alkanols (Kolattukudy et al., 1975; Holloway, 1983). It still remains hypothetical how these monomeric units are assembled at the macromolecular level (Bernards, 2002).
Tomato fruits are predominantly harvested by removal from their pedicel (Poole & McLeod, 1994). Consequently, a stem scar wound is generated, which causes significant water loss and makes the fruit amenable to post-harvest microbial infections. Hence, a rapid biological response is crucial for tomato fruits to reduce dehydration, as well as the efflux of nutrients, and to restrain sites of microbial invasion, aiming to ensure proper ripening and/or a prolonged shelf life. The compositional nature, time course, regulation and efficiency of these wound-healing processes of tomato fruits have so far not been investigated in detail.
Suberized stem scar membrane
Our analyses confirmed the presence of aromatic and aliphatic components characteristic of the biopolyester suberin in the newly formed stem scar membrane, with the exception of glycerol (Moire et al., 1999) which was not analyzed because of experimental limitations. The detected components could not have originated from the cuticular membrane because the cuticle surrounding the stem scar tissue was removed completely before chemical analysis. The abundance of ω-hydroxyalkanoic acids, α,ω-alkandioic acids and a significant proportion of C18 unsaturated aliphatic compounds is in agreement with the suberin monomer composition reported from tomato fruit and other plant sources (Dean & Kolattukudy, 1976; Zeier & Schreiber, 1998). Compositional changes at the stem scar during the time course of suberization remain to be determined.
Soluble wax-like substances integrated into the suberin matrix have been shown to be largely responsible for the formation of a water-loss barrier of native potato tuber periderm (Soliday et al., 1979; Schreiber et al., 2005). In potato wound periderm, the permeability was, on average, 100-fold higher than that of the native periderm, although the total amounts of wax-like substances and suberin biopolymer were c. 50–60% of the amounts of the native periderm without any pronounced qualitative differences in chemical composition. These results show that the water barrier properties of suberized tissue cannot be inferred from the occurrence of suberin with depositions of wax-like substances (Schreiber et al., 2005). From the suberized stem scar tissue of tomato fruits, only small amounts of chloroform-soluble aliphatic compounds, C20, C22, C24 alkanols and alkanoic acids – probably embedded in the suberin polymer matrix – were detected. These soluble compounds showed a completely different composition from waxes embedded in the tomato fruit cutin matrix. Previous studies have shown that the main difference between potato periderm and other suberized plant tissues is the significant fraction of aliphatic wax compounds sorbed to the suberin biopolymer of potato, whereas this fraction is basically missing in suberized tissues of other plants (Schreiber et al., 2005; Efetova et al., 2007). Accordingly, our results suggest that soluble wax-like compounds are presumably not the major determinants for the formation of an effective barrier at the stem scar tissue of tomato fruits, restricting water loss, which, however, remains to be elucidated.
Wound healing and water loss of tomato fruits
According to Cameron & Yang (1982), transpiration through the tomato fruit stem scar accounts for c. 67% of the whole-fruit water loss. This is in full agreement with the results provided in the present study. The stem scar tissue of tomato fruit is also known to account for c. 97% of the fruit gas exchange (Cameron & Yang, 1982). It has long been recognized that a gas-tight sealing of stem scar tissue causes reduced ripening rates and extends the storage life of tomato fruits (Brooks, 1937; Yang & Shewfelt, 1999). Thus, post-harvest suberization of the stem scar wound should at least reduce gas exchange to a level still sufficient to allow for a nonretarded ripening of tomato fruit.
As a result of their comparably small surface area and fragility, the enzymatically isolated stem scar tissues were not suitable for the direct determination of water permeability. Our results demonstrate an effective reduction in fruit water loss as a result of the efficient sealing of the stem scar wound by suberization, reflected by the permeance for water and the ion flows out of the fruit, approximating those of tomato fruits artificially sealed with paraffin wax immediately after harvest. The dependence on active suberin biosynthesis to attain efficient closure of the stem scar wound is demonstrated by the low-temperature storage experiment.
Function of the stem scar membrane as a microbial barrier
The greatly reduced intrusion of actively growing mycelium of the post-harvest tomato fruit pathogen F. solani (Amadioha & Uchendu, 2003) as a result of prolonged fruit storage indicates a role for suberized stem scar tissue simply as a mechanical barrier and/or for prevention of entry by accumulating antibiotic/antimicrobial compounds. Although the aliphatic domain of potato tuber suberin has been shown to specifically provide resistance against the fungal pathogen Fusarium sambucinum (Lulai & Corsini, 1998), suggesting a chemical mode of action, the toluidine blue infiltration experiment corroborated the effective wound closure. The newly formed stem scar barrier membrane prevents the seepage of water through the stem scar wound and, consequently, reduces the risk of infiltration with microbial contaminants. From the present data, it can be suggested that the pretreatment of traditionally harvested tomato fruits to accelerate the suberization process at the stem scar wound might reduce significantly microbial infection, improve the quality and extend the shelf life of tomato fruits. This pretreatment of harvested tomato fruits includes conditions of moderate temperature and reduced humidity for a 2-d period of storage before washing and further processing. Likewise, a prestorage period of kiwifruit (Actinidia deliciosa) has been shown to markedly reduce stem scar infection by the fungal pathogen Botrytis cinerea (Pennycook & Manning, 1992; Bautista-Baños et al., 1997).
Nevertheless, there are also indications that soluble compounds associated with the suberin biopolymer, such as phenolics or wax-like components, may themselves act as antifungal agents (Kolattukudy, 1984; Thomas et al., 2007). Similarly, H2O2-generating systems associated with suberization might additionally affect microbial growth, resulting in an overall reduced microbial amenability of the tomato fruit stem scar during the wound-healing process.
Impact of the CER6 β-ketoacyl-coenzyme A synthase
Fatty acid elongases catalyze the elongation of the carbon chain of C16 and C18 fatty acids to different lengths as found in cuticular wax biosynthesis (Domergue et al., 1998). MicroTom lecer6 is defective in a β-ketoacyl-coenzyme A synthase involved in very-long-chain fatty acid elongation, resulting in a largely increased cuticular permeance for water of tomato fruits (Vogg et al., 2004; Leide et al., 2007). Despite the overall elevated weights of the stem scar tissue of the MicroTom lecer6 mutant, probably as a result of a generally enlarged stem scar area, the time course and efficiency of wound healing were not modified in this mutant. An induction of tomato Cer6 gene expression was not found at the transcriptional level of the stem scar tissue during tomato fruit storage. Hence, our results indicate that the CER6 enzyme is not required for the biosynthesis of suberin and associated wax-like compounds in tomato fruits. These findings are not consistent with the expression of the KCS6 protein of the potato periderm, which shares a high homology with the CER6 protein of tomato (Serra et al., 2009). As stem scar wound healing was not modified in the MicroTom lecer6 mutant, our results indicate that enzymes other than CER6 might be substantially responsible for fatty acid elongation in tomato fruit suberin biosynthesis.
Relationship between gene expression and wound healing
So far, only a few genes involved in the biosynthesis of the suberin biopolymer have been identified (Ranathunge et al., 2011). Interestingly, genes found to be differentially expressed in the stem scar tissue during wound healing are related to desiccation and regulation by ABA.
Some genes encoding nonspecific lipid transfer proteins, such as the Tsw12 gene, are inducible by ABA, and thus mediate in plant responses to environmental stress conditions (Torres-Schumann et al., 1992; Yubero-Serrano et al., 2003). Lipid transfer proteins are known to bind a variety of hydrophobic fatty acids and lipids (Kader, 1996; Blein et al., 2002), potentially also providing monomers for suberin biosynthesis. Therefore, increased transcript levels of Tsw12 may indicate a post-harvest stimulation of transport processes of lipophilic compounds in the stem scar tissue. The formation of suberized cell layers, in turn, prevents excessive water loss from the detached tomato fruit, as generally suggested for plant organ abscission zones (Bleecker & Patterson, 1997; Roberts et al., 2002).
In the present study, we also expected to identify the dehydrin gene Tas14 as being induced in the early stage of fruit storage on wounding at the stem scar tissue. Dehydrins have the ability to bind to lipid vesicles that contain acidic phospholipids, scavenge hydroxyl radicals or display protective activity towards lipid membranes against peroxidation (Close, 1996; Richard et al., 2000). The Tas14 gene has also been described to accumulate in response to ABA or environmental stress (Godoy et al., 1990; del Mar Parra et al., 1996).
Strikingly, the expression pattern of the Asr4 gene was similar to that of the highly induced Tsw12 gene and the moderately up-regulated Tas14 gene. The ABA, water stress and ripening-inducible Asr gene family has been reported to be up-regulated under different environmental stress conditions in an ABA-dependent manner and during the process of tomato fruit ripening (Dóczi et al., 2005; Fischer et al., 2011). Accumulating functional evidence suggests that Asr genes have several functions in tomato plant adaptation to desiccation (Maskin et al., 2001; Frankel et al., 2006), probably effective in transcriptional regulation of post-harvest changes in tomato fruits.
The only tomato gene detected in the present study that showed a decrease in expression after fruit harvest was the Erd7 gene encoding an early-responsive to dehydration protein. In Arabidopsis, the Erd7 gene has been shown to be induced by ABA and dehydration, and repressed by rehydration (Seki et al., 2002; Oono et al., 2003). Hence, the down-regulation of the tomato Erd7 gene might be an early indication for desiccation stress recovery, probably because of the fairly efficient closure of the stem scar wound already 1 d after harvest. However, as the senescence-related Erd7 gene has also been reported to be up-regulated under high light stress in Arabidopsis (Kimura et al., 2003), post-harvest storage in darkness might have caused the subsequent decrease in gene expression.
These genes display a rapid, although to some extent, transient reaction in response to tomato fruit wounding as a consequence of the harvesting process, suggesting a significant role in the mediation of plant stress responses and/or suberization. The precise function of these enzymes has not been characterized to date. The effect of wounding on transcript accumulation in tomato fruits could be linked to a higher water loss rate through more open surfaces arising from the mechanical treatment and with subsequent closure caused by the wound-healing process.
Importance of endogenous ABA in wound healing
The gene expression analysis results of the present study support the idea that ABA could play a pivotal role in the stem scar wound-healing process of tomato fruits, although the induction of genes putatively involved in wound healing does not prove the direct involvement of ABA in stem scar suberization of tomato fruits. As documented previously for tomato leaves, ABA accumulation was also found in the stem scar tissue after harvest (Herde et al., 1999). Furthermore, during the wound-healing process, an induction at the transcriptional level was found for the Sit gene encoding an aldehyde oxidase, which is known to be the catalyst of the abscisic aldehyde oxidation to ABA – the final step of ABA biosynthesis in tomato plants (Taylor et al., 1988; Seo et al., 2000). The application of tomato mutants with reduced ABA levels and the assessment of ABA contents of the stem scar tissue during wound healing, in combination with the functional characterization of the wound-healing process, provided appropriate tools to investigate in vivo the role of ABA in the stem scar suberization of tomato fruits. The regulatory involvement of ABA in the wound-healing processes of tomato fruits has not been demonstrated unequivocally by reduced ABA concentrations and the determination of the resulting functional effects. A rapid and transient boost in ABA content has also been reported for potato tubers after wounding (Lulai et al., 2008; Kumar et al., 2010).
The decreased ABA contents in the stem scar tissues of RR flacca, RR sitiens and AC notabilis were accompanied by a retarded and attenuated stem scar wound-healing response in the ABA-defective tomato fruits. These results unambiguously indicate that ABA is involved as a mediating agent in the wound-healing processes of the stem scar tissue of tomato fruits. The presence of basal ABA levels in combination with readily synthesized ABA significantly accelerates the suberization response after stem scar wounding. Lulai et al. (2008) showed that ABA plays a similar role in wounded potato tubers. Nevertheless, the specific role of ABA in the regulation/promotion of wound-induced gene expression at the stem scar tissue of tomato fruits remains to be elucidated. Future experimental approaches need to address the functional involvement and interplay of other regulatory molecules, such as jasmonic acid, salicylic acid, auxin, ethylene and nitric oxide (León et al., 2001; París et al., 2007; Lulai et al., 2011), in the wound responses at the tomato fruit stem scar tissue.
The authors are indebted to Lena Reinhardt for excellent technical assistance and Jutta Winkler-Steinbeck for plant care. We are grateful to Günther Bahnweg (GSF-National Research Center for Environment and Health, Institute of Biochemical Plant Pathology, Neuherberg, Germany) for providing us with F. solani and Avraham A. Levy (Weizmann Institute of Science, Rehovot, Israel) for the lecer6 mutant. This work was supported by the Deutsche Forschungsgemeinschaft (grant no. VO 934-1-3) and by the Sonderforschungsbereich 567.