| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
First published online April 20, 2007; 10.1104/pp.107.097477 Plant Physiology 144:1012-1028 (2007) © 2007 American Society of Plant Biologists OPEN ACCESS ARTICLE
A Reevaluation of the Key Factors That Influence Tomato Fruit Softening and Integrity1,[W],[OA]Department of Plant Biology (M.S., A.J.M., T.I., K.J.N., J.K.C.R.) and Department of Horticulture (C.B.W.), Cornell University, Ithaca, New York 14853; Department of Horticulture and Landscape Architecture, Purdue University, West Lafayette, Indiana 47907 (M.A.J., S.M.G.); College of Horticulture, Northwest Agricultural & Forestry University, Yangling, Shaanxi 712100, China (R.X.); Department of Plant Sciences, University of California, Davis, California 95616 (J.M.L., K.A.S.); Max-Planck-Institut für Molekulare Pflanzenphysiologie, 14476 Potsdam-Golm, Germany (A.R.F., A.L.); and Complex Carbohydrate Research Center and Department of Biochemistry and Molecular Biology, University of Georgia, Athens, Georgia 30602 (M.A.O.)
The softening of fleshy fruits, such as tomato (Solanum lycopersicum), during ripening is generally reported to result principally from disassembly of the primary cell wall and middle lamella. However, unsuccessful attempts to prolong fruit firmness by suppressing the expression of a range of wall-modifying proteins in transgenic tomato fruits do not support such a simple model. ‘Delayed Fruit Deterioration’ (DFD) is a previously unreported tomato cultivar that provides a unique opportunity to assess the contribution of wall metabolism to fruit firmness, since DFD fruits exhibit minimal softening but undergo otherwise normal ripening, unlike all known nonsoftening tomato mutants reported to date. Wall disassembly, reduced intercellular adhesion, and the expression of genes associated with wall degradation were similar in DFD fruit and those of the normally softening ‘Ailsa Craig’. However, ripening DFD fruit showed minimal transpirational water loss and substantially elevated cellular turgor. This allowed an evaluation of the relative contribution and timing of wall disassembly and water loss to fruit softening, which suggested that both processes have a critical influence. Biochemical and biomechanical analyses identified several unusual features of DFD cuticles and the data indicate that, as with wall metabolism, changes in cuticle composition and architecture are an integral and regulated part of the ripening program. A model is proposed in which the cuticle affects the softening of intact tomato fruit both directly, by providing a physical support, and indirectly, by regulating water status.
The ripening of fleshy fruits involves many physiological processes, including the production of aromatic compounds and nutrients, changes in color, and softening of the flesh to an edible texture, which have evolved to attract animals and promote seed dispersal (Giovannoni, 2004  ). The molecular pathways that underlie many of these ripening-related phenomena have been sufficiently characterized to allow the modification of fruit nutritional status or organoleptic characteristics through targeted genetic engineering (Romer et al., 2000; Lewinsohn et al., 2001; Muir et al., 2001; Dharmapuri et al., 2002; Fraser and Bramley, 2004; Davuluri et al., 2005). However, much less is known about the critical molecular determinants of fruit firmness and softening. In part, this reflects the difficulties in evaluating the many physical and sensory attributes that determine texture (Harker et al., 1997; Waldron et al., 2003), a characteristic that, unlike color or aroma, cannot be defined by a quantitative measurement of specific metabolites or by monitoring a particular biosynthetic pathway.
Research spanning more than 40 years has targeted the causal mechanisms of fruit softening, much of it using tomato (Solanum lycopersicum) ripening as a model system. A decline in fruit firmness typically coincides with dissolution of the middle lamella, resulting in a reduction in intercellular adhesion, depolymerization, and solubilization of hemicellulosic and pectic cell wall polysaccharides and, in some cases, wall swelling (Brummell and Harpster, 2001 Explanations for the lack of progress in identifying the key individual determinants of fruit softening include the possibility that important textural changes associated with wall disassembly are a consequence of numerous enzymes acting in concert on multiple wall structural components, or that the critical enzymatic activity or activities have not yet been identified. However, an alternative explanation is that polysaccharide degradation is not the sole determinant of fruit softening and that other ripening-related physiological processes also play critical roles. We have addressed this latter hypothesis by evaluating a previously uncharacterized tomato cultivar, referred to here as ‘Delayed Fruit Deterioration’ (DFD), whose fruits undergo normal ripening, but remain firm and show no loss of integrity for remarkably extended periods after reaching the fully ripe stage. We report that ripening-related disassembly of the cell wall and middle lamella have similar characteristics in fruits from DFD and the normally softening ‘Ailsa Craig’ (AC), even though DFD fruits typically remain firm for at least 6 months after achieving a fully ripe stage. However, in contrast to AC fruits, those from DFD exhibit minimal water loss and consequent tissue collapse, and it is suggested that this is likely due to specific compositional or ultrastructural characteristics of the fruit cuticle. Evidence is also presented that fruit cuticles from DFD and AC tomatoes undergo distinctly different changes in their biomechanical properties during ripening. Moreover, ripe DFD fruits are highly resistant to infection by opportunistic pathogens, unless the integrity of the cuticle is compromised. Our results suggest that while changes in the polysaccharide components of the cell wall are undoubtedly important for fruit texture, equivalent alterations in cuticle architecture are also an integral element of the ripening program. The data also highlight the mechanistic distinction between a reduction in firmness, or resistance to compression, of intact fruit, and ripening-related textural changes in the pericarp tissue. Both of these phenomena could be referred to as softening, although in most reported analyses, transgenic lines with altered expression of wall-modifying proteins have been evaluated, for obvious commercial reasons, by measuring resistance of the intact fruits to compression. This study indicates that the cuticle has a previously underappreciated influence on intact tomato fruit firmness and ripening physiology, and suggests another biotechnological target to prolong fruit quality, in addition to altering polysaccharide metabolism.
Physiological Characterization of DFD Fruit Development and Softening
DFD is a regionalized cultivar that is grown in specific areas of southern Europe and around the Mediterranean, where it is known to exhibit dramatically delayed softening. As far as we are aware, DFD has not been described in the literature and its genetic background is unknown. Therefore, in the absence of isogenic lines containing the introgressed DFD phenotype, the AC tomato cultivar was used for comparative purposes because it exhibits relatively rapid fruit softening and has similar fruit size, shape, and overall fruit morphology to DFD (Fig. 1
). DFD and AC fruit development and ripening (from approximately 10–50 d after anthesis, as shown in Fig. 1, A and B) were comparable, with both genotypes taking a similar time to reach a fully expanded mature green (MG) stage, although the time taken from the breaker (Br) stage, at the onset of ripening, to the red ripe (RR) stage was approximately 7 d longer for DFD than AC fruits. The peak in ethylene production showed a similar temporal delay, but the DFD fruits exhibited a characteristic climacteric respiratory burst and increase in ethylene synthesis at the Br stage (Fig. 2, A and B
) that was more pronounced than that of AC fruit. This was not surprising, given the substantial variation that has been observed among tomato cultivars (Guillén et al., 2006
After reaching a fully ripe stage, DFD fruits remained visually unchanged for several months, as seen in photographs taken at RR plus 4 months (Fig. 1C) and a time-lapse video spanning 4 months after the RR stage (Supplemental Video S1). This was in direct contrast to AC fruits, which exhibited overripening (tissue degradation and collapse after reaching the fully ripe stage), as is typically seen in most fleshy fruits. Even 7 months after full ripening, the DFD fruits showed little change in external appearance, with no signs of internal desiccation, tissue breakdown, or other morphological changes (Fig. 1D). Storage of the fruits in the light eventually resulted in pigment photobleaching (Fig. 1D), but this did not occur in dark-stored fruits. Compression analysis of intact AC fruits showed a typical loss of fruit firmness from MG to RR, and complete tissue collapse by 2 months after RR (Fig. 3A ). In contrast, MG DFD fruits were statistically firmer than those of AC at the same stage, exhibited minimal softening during ripening, and at the RR stage were statistically firmer than AC Br fruits. Even 8 months after reaching the RR stage, the firmness of DFD fruits was similar to that of AC RR fruits. However, when the firmness of excised pericarp segments with the endocarp facing upwards was measured, no differences were seen between fruits of each cultivar at any ripening stage (Fig. 3B), although the DFD fruits were somewhat firmer at the MG stage. In contrast, the force needed to penetrate the cuticle of intact MG fruits was greater for AC than DFD (Fig. 3C), and while the AC cuticles showed progressive weakening during ripening, those of DFD showed a minimal change, such that the penetration mass for DFD fruits at the RR stage was approximately twice that of AC.
Cell Wall Analysis, Wall Disassembly, and Dissolution of the Middle Lamella
Changes in the amounts of total wall material, based on dry weight (Fig. 4A
), uronic acids (Fig. 4B), cellulose (Fig. 4C), and cell wall neutral sugar composition (Fig. 4D), showed typical ripening-related trends in both AC and DFD fruits. Similarly, size exclusion chromatographic (SEC) separation of water-soluble pectins from MG and RR fruits of both cultivars showed evidence of characteristic ripening-related pectin depolymerization (Fig. 4, B and E). A fractionation of chelator-soluble pectins suggested similar patterns of pectin modification in DFD and AC ripe fruits (data not shown). To further contrast ripening-related wall metabolism in the DFD and AC fruits, the expression of a range of genes encoding proteins involved in cell wall modification and disassembly, including PG, two expansins, and a xyloglucan endotransglucosylase/hydrolase, was examined by northern-blot analysis (Fig. 4F). All the genes showed similar qualitative patterns of transcript accumulation and while some quantitative differences were apparent, they were within the range of typical intercultivar variation that we have previously observed (data not shown). Importantly, the expression of these genes was not dramatically repressed, as is the case for ripening-impaired mutants such as rin and nor, where expression is generally undetectable (Maclachlan and Brady, 1994
An examination of the fruit pericarp tissues by light microscopy (Supplemental Fig. S1, A and B) and transmission electron microscopy (TEM; Supplemental Fig. S1, C and D) showed that both AC (Supplemental Fig. S1, A and C) and DFD (Supplemental Fig. S1, B and D) ripe fruits exhibited characteristic cell-cell separation, indicating degradation of the primary wall and middle lamella and resulting in the formation of enlarged cell junction zones and cells that are interconnected only through plasmodesmata (Supplemental Fig. S1, C and D).
Even though fruit cell wall breakdown and cell separation showed similar trends in DFD and AC, the pericarp of DFD, but not AC, exhibited substantial swelling during ripening (Fig. 5A ). Microscopic analysis of three zones of the pericarp, (1–3, corresponding to outer, middle, and inner, respectively; Fig. 5B) showed that cell size throughout the pericarp was similar in AC and DFD fruits at the MG stage (Fig. 5, B and C), but that by the RR stage, cells in the DFD pericarp had undergone a substantial increase in cell size. For example, cells in the DFD inner pericarp (zone 3; Fig. 5C) showed, on average, at least 4-fold greater increases in mean cell cross-sectional area during ripening. In contrast, while cell size in RR AC fruits was on average somewhat greater than in MG fruits, this was only statistically significant in zone 3 and the extent of the increase was far less than that seen with DFD (Fig. 5, B and C).
An increase in cell size is indicative of a positive hydrostatic (turgor) pressure that would be required to drive cell expansion. To determine whether this was the case, cellular turgor in the outer pericarp cells of AC and DFD fruits at different ripening stages was measured using a pressure microprobe (Fig. 6A , inset). Cellular turgor in AC fruits declined prior to the onset of ripening at the MG stage and decreased linearly throughout ripening, reaching minimal values at the overripe (OR) stage (Fig. 6A). The turgor values are comparable to those previously reported for other tomato cultivars (Shackel et al., 1991 ). The initial cellular turgor pressure and rate of linear decrease were similar in DFD fruits at the MG stage and the onset of ripening. However, in DFD, the rate of decrease slowed early in ripening and showed little further decrease after a midripe stage. An analysis of covariance indicated a significant (P = 0.016) difference in turgor of OR fruit between AC (0.09 MPa) and DFD (0.03 MPa) fruit, after adjustment to the same mean color level. A decline in cellular turgor may be associated with water loss from the fruits, which in turn may be an important factor that contributes to the extremely delayed softening of DFD fruits. Transpirational water loss from AC and DFD fruit was assessed by measuring weight reduction over a 3 month period in fruits starting at the RR stage (Fig. 6B). AC fruits showed a high rate of water loss, while the loss was minimal in DFD fruits. After 3 months the AC fruits typically exhibited extensive desiccation and splitting, while the DFD fruits showed no apparent change, similar to the fruits shown in Figure 1C.
Microscopic Analysis of Cuticles
The cuticle provides the principal barrier to water loss in plant tissues (Jeffree, 2006
TEM analysis of the AC and DFD cuticles indicated no differences in ultrastructure in either the cuticular layer or periclinal and anticlinal cell walls of the outer epidermal cells (Fig. 7, E and F) and SEM imaging revealed no patterns of external wax accumulation, such as wax crystals, in either cultivar. However, during SEM analysis, focusing of the electron beam on the fruit surface consistently induced rapid peeling of a thin, outer membranous layer in the DFD samples (highlighted with an arrow in Fig. 7H), but not in AC fruits (Fig. 7G).
Enzymatically isolated cuticles from AC and DFD fruits at the MG and RR stages were evaluated with an Instron analyzer to determine their extensibility (strain to failure), work of fracture (energy required to induce failure), and viscoelasticity (creep rates under constant applied forces) under uniaxial loading. These tests revealed significant differences between the mechanical behaviors of the cuticles from the two cultivars, and changes in cuticle mechanical properties attending the ripening of each cultivar (Fig. 8 ). At the MG stage, the yield stress (i.e. the force required to induce material failure) was 33% higher in DFD than AC cuticles. Although the mean yield stress value increased during ripening in AC, there was no statistically significant change in DFD cuticles (Fig. 8A). In contrast, the AC cuticles had greater yield strains than those of DFD at the MG stage, indicating greater extensibility, and while the yield strain of AC cuticles decreased during ripening, those of DFD became markedly larger, indicating an increase in their extensibility, with a mean value approximately 30% greater than AC at the RR stage (Fig. 8B). Thus, the tensile elastic modulus Et (i.e. the ability to resist tensile forces) of isolated cuticles differed between the two cultivars and changed during the maturation of each cultivar. At the MG stage, the Et of DFD cuticles was significantly larger than that of AC, while at the RR stage, the Et of AC was larger than that of DFD. Moreover, during the MG to RR transition, the Et of AC cuticles increased, whereas that of DFD cuticles decreased. Preliminary tests to determine viscoelastic behavior (e.g. creep tests and cyclical loading-unloading tests) indicated that DFD cuticles at the RR stage deform plastically more rapidly and require smaller forces than do those of AC at the same developmental stage (data not shown).
Cuticle Composition To determine if the cuticle chemical composition was different in AC and DFD fruits, the levels of component cuticular waxes and cutin monomers at the MG and RR stages were assayed. The total wax amount increased during ripening in both cultivars (Table I ), but was higher in DFD than AC fruits at both the MG and RR stages (33% and 36%, respectively). In RR fruits, the most significant proportional differences were apparent in the alkadienes, which showed a more than 4-fold increase in the mean percentage levels in DFD than AC, and n-alkanes, with lower levels in DFD (bold in Table I). Interestingly, the total and proportional amount of alkadienes increased between the MG and RR stages more than any other constituent class (Table I): In DFD fruits, the total amount of alkadienes increased nearly 50-fold from MG to RR, corresponding to a 28-fold proportional increase, while the equivalent increase in AC fruit was only 8-fold, which represents a 5-fold proportional increase. Individual wax constituents showed relatively little variation, other than substantially greater levels of tritriacontadiene in the DFD RR cuticles (bold in Table II ).
Unlike the waxes, the total amounts of cutin monomers per unit fruit surface area were similar in MG and RR stages of AC fruits, but increased substantially (84%) during ripening in the DFD fruits (bold in Table III ). Accordingly, while the amounts of cutin monomers were similar in AC and DFD MG fruits, the amount of cutin in DFD RR cuticles was approximately double that of AC at the same stage. The relative proportions of the monomeric constituents generally showed no significant change during ripening and no qualitative differences between AC and DFD at either ripening stage were detected. A notable exception was an increase in the proportion of 9,10,18-triOH octadecanoic acid in AC cuticles and a concomitant decrease in 9,10,18-triOH octadecenoic acid (bold in Table III). This change was not detected in DFD cuticles, which also contained higher levels of p-coumaric acid (Table III).
In a separate experiment, the levels of flavonoids and the flavonoid precursor naringenin chalcone were measured in total pericarp tissue minus the cuticle and outer epidermal cell layers, and in enzymatically isolated cuticles of MG and RR fruit from both cultivars. Naringenin chalcone was barely detectable in AC total pericarp extracts at the RR stage and not at all in DFD pericarp, and while levels were approximately 300-fold higher in isolated RR AC cuticles, it was not detected in DFD cuticles. The flavonoids kaempferol and quercetin were detected in MG cuticles from both cultivars and the amount of kaempferol was 2- to 3-fold greater in DFD than AC.
During storage and overripening it was often observed that the AC fruits became infected by opportunistic fungal pathogens, while intact DFD fruits never succumbed to infection, even following prolonged storage in high humidity conditions (Fig. 9A ). To test the resistance of the DFD fruits to fungal infection in a more controlled environment, spores of the fungus Botrytis cinerea were applied at various titers to RR fruits from AC (Fig. 9, B and E), a commercial tomato cultivar (Fig. 9, C and F), and DFD (Fig. 9, D and G), either with a syringe through a small puncture in the fruit cuticle (Fig. 9, B–D), or ectopically (Fig. 9, E–G). Fruit from AC and the commercial tomato cultivar were consistently infected by both ectopic application and through the wound, whereas DFD fruits were only infected when the cuticle was damaged, upon which infection occurred at the same rate and to the same extent as in the other cultivars.
DFD Provides a Unique Source of Genetic Material to Dissect Fruit Softening
One of the advantages of tomato as an experimental model to study fleshy fruit softening is the availability of pleiotropic nonripening mutants, such as rin, nor, alcobaça, and colorless nonripening, which are impaired in many ripening-related processes and exhibit delayed or impaired softening (Kopeliovitch et al., 1980
DFD fruits exhibit a substantial climacteric burst of ethylene and carbon dioxide production at the onset of ripening and undergo normal ripening, as determined by assessing color, soluble sugars (Fig. 2), metabolite content, and aroma (data not shown), in contrast to all previously reported pleiotropic tomato mutants. However, DFD fruits show remarkably small changes in their firmness when compared with the well studied experimental AC and indeed, as far as we are aware, with all previously reported tomato cultivars. While some tomato mutants show delayed softening for up to several weeks (Mutschler, 1984
The common assumption that fruit softening (applying the most commonly used definition, of resistance of intact fruits to compression) is primarily, or even exclusively, the result of wall disassembly (Brownleader et al., 1999
It has been suggested that fruit firmness, in common with the biomechanical properties of most plant tissues, is influenced by cellular turgor pressure (Shackel et al., 1991
The notable difference in DFD fruit water status and puncture analysis of the fruit skin (Fig. 3C) suggested an association with the cuticle, which acts as the primary barrier to transpirational water loss from aerial tissues (Riederer and Schreiber, 2001
No gross differences in DFD and AC cuticle ultrastructure (Fig. 7) were seen and cuticle thickness was not statistically different between the two cultivars. This was not unexpected, since there is no correlation between cuticle thickness and water permeability in leaves (Riederer and Schreiber, 2001
Our data suggest that the DFD cuticle is stiffer and stronger than that of AC at the MG stage (Fig. 8). Since breaking stresses are calculated by normalizing applied mechanical forces with respect to the cross-sectional areas on which they act, the differences in breaking stresses between cultivars and developmental stages must reflect differences in the cuticle (ultra) structure or biochemical composition (Niklas, 1992
MG and RR DFD fruit have approximately 30% more wax than AC and a substantially increased level of n-alkadienes at the RR stage, which are inferred to be n-tritricontadiene (n-C33-6,9-diene) from previous analyses (Bauer et al., 2004a
DFD is unlike previously reported tomato mutants in that loss of fruit firmness is largely uncoupled from other aspects of ripening, thus providing a unique opportunity to examine the factors that contribute to fruit softening. Our results suggest that multiple coordinated processes are involved, including disassembly of polysaccharide networks in the primary wall and middle lamella and transpirational water/turgor loss. The minimal water loss and maintenance of cellular turgor in DFD fruit further allow an assessment of the relative contribution and timing of these two processes. In this regard, the turgor and fruit firmness data indicate that softening of intact fruits results from an early decline in cellular turgor coincident with early changes in wall architecture, presumably as a result of wall relaxation. A second component is then provided by substantial water transpiration, which occurs in parallel with continued wall degradation and a reduction in intercellular adhesion.
The cuticle itself is also thought to have an important influence on the biomechanical properties of ripening fruit (Petracek and Bukovac, 1995
While the goal of this study was not to identify the components of the cuticle that are responsible for water permeance, DFD provides a valuable experimental system to address this question. Two features of DFD cuticle composition are particularly noteworthy and may provide some indication as the basis for the drastically reduced water permeance. First, the cutin matrix appears to be substantially denser and the consequent change in cuticle architecture and macromolecular packing may influence water permeability. Second, while the cuticles of most ripe tomatoes contain high levels of the flavonoid precursor naringenin chalcone (Bauer et al., 2004b DFD is of obvious interest for agricultural biotechnology, a field that has to date focused almost exclusively on cell wall metabolism in an attempt to alter fruit softening and prolong shelf life. Our study reveals the clear distinction between complex textural changes in the pericarp flesh (sometimes called softening), which would be directly affected by wall metabolism, and the reduction in firmness of intact fruits, which is influenced by multiple factors, including the cuticle. The lack of progress in extending intact fruit firmness and quality by primarily targeting wall metabolism may partly reflect an underappreciation of this difference. Future research will be directed toward identifying the molecular elements that underlie the drastic reduction in water loss and apparently provide resistance to postharvest pathogens. Preliminary genetic analysis using a Solanum pimpinellifolium mapping population has revealed the existence of a single genetic locus that makes a major contribution to the DFD trait (data not shown) and the functional characterization of a candidate gene that cosegregates with that locus is currently in progress. In addition, as a control, the DFD trait is being introgressed into AC for further detailed phenotypic characterization.
Plant Materials
Tomato fruit (Solanum lycopersicum; AC and DFD) plants were greenhouse grown at Cornell under 16 h light, 8 h dark conditions, using standard practices. Fruits were harvested at the following stages: MG, Br, turning, pink (Pi), light red, and RR based on external color, as described in Lashbrook et al. (1994)
Six AC and DFD fruits were harvested at the MG stage and kept at 20°C. CO2 and ethylene production of each fruit was measured by sealing individual fruit of known weight for 2 h in 475 mL glass jars fitted with sampling septa, on a daily basis for 9 or 21 d for AC or DFD, respectively. CO2 concentrations in 1 mL samples of the headspace of each jar were measured using a Fisher Gas Partitioner (model 1200) gas chromatograph (GC; Fisher Scientific) equipped with a thermal conductivity detector. Ethylene concentrations were measured using a Hewlett-Packard 5710A GC (Hewlett-Packard) equipped with a flame ionization detector. Fruit color was measured using a Minolta CR300 color meter (Konica Minolta), expressing color change using the a* chromaticity coordinate.
Firmness measurements were made, based on compression of AC and DFD intact fruits (minimum of five replicate fruits), at MG, Br, Pi, RR, and RR plus 2 or 8 months. Each fruit was tested four times at equidistant points along the equatorial plane of the fruit with a 50 mm wide P50 DIA compression plate controlled by a Stable Microsystems Texture Analyzer (TA-XT2i; Stable Micro Systems), loading at 1 mm s–1 and compressed to a vertical displacement of 1 mm. Firmness was defined as the response force to a 0.05 N applied force. Fruits at the RR stage were detached, stored at room temperature, and analyzed after 60, 90, and 120 d for DFD and after 60 d for AC. To compare flesh firmness with compression of whole fruits, pericarp discs (1.5 cm diameter) were excised from the same fruit set with a cork borer and placed cuticle face up and tested as above. To determine the maximum force required to perforate the cuticle, segments of tomato pericarp were placed cuticle face down on a base plate with a vertical groove (3 mm wide) and the 1 mm probe applied until the cuticle was penetrated (40 replicates for each fruit stage). Statistical analysis was performed using SAS software (SAS Institute) and the Tukey-Kramer multiple comparison test. The time-lapse video (Supplemental Video S1) was generated by taking a digital photograph (0.25 s exposure) of two AC fruits and one DFD fruit harvested at the MG stage and left to ripen at room temperature, every 10 min for approximately 4 months. Images were edited and compressed using Avid Xpress video editing software (Avid Technology).
Cell walls were prepared from 50 g of diced outer pericarp of AC and DFD fruit (MG and RR stages) by boiling the pericarp pieces in 95% ethanol for 30 min to prevent autolytic activity, as described in Huysamer et al. (1997)
For the SEC analysis the AIS was treated with water and 50 mM cyclohexane diamine tetraacetic acid (CDTA; Rose et al., 1998
Total RNA was isolated from pericarp fruit tissues as described in Wan and Wilkins (1994)
Prior to microscopy pericarp thickness of AC and DFD fruits (15 replicates) was measured with calipers. To evaluate the pericarp cell size differences, pericarp from MG and RR stages were hand sectioned and viewed with a light microscope. Cell areas in the outer, middle, and inner pericarp zones (see Fig. 5) were calculated and statistical analyses performed using Image-Pro Plus software (Media Cybernetics).
For light microscopy visualization of cuticles, cuticular membranes were peeled from the fruits and fixed in formaldehyde:acetic acid 1:1 (v/v) in 18 volumes of 70% alcohol (formaldehyde-acetic acid) for 48 h, followed by dehydration in an alcohol series (50%–10%) and a water wash. Cuticle samples were frozen at –35°C and 10 µm sections were cut with a microtome and stained with 0.05% of toluidine blue. For TEM analysis of cuticles and cell separation, pericarp from AC and DFD RR fruits was excised from three replicate fruits at the equator at the RR stage and fixed for 1 h at room temperature in primary fixative, containing 2.5% (v/v) glutaraldehyde and 2% (v/v) formaldehyde in 0.05 M phosphate buffer (PB), pH 6.8 (Karnovsky, 1965 For scanning electron microscopy analysis, pericarp from AC and DFD RR fruits was freeze dried in a VirTis Freeze mobile (VirTis Company) at –40°C, mounted on aluminum stubs, and sputter coated with gold/palladium using a BalTec SCD 050 coater (Balzers Union). Microscopic observations were made with a Hitachi S4500 scanning electron microscope (Hitachi High Technologies). Images were acquired using Princeton Gamma Tech Imix software (Princeton Gamma-Tech Instruments).
Measurements of the turgor pressure of outer pericarp cells from detached intact AC and DFD fruits during ripening were performed as described in Shackel et al. (1991)
Ten fruits from AC and DFD were detached at the RR stage and kept at room temperature for 3 months. Water loss per unit fruit surface area was calculated after measuring the weight decrease over time and measuring fruit dimensions.
Rectangular strips of cuticle were removed from MG and RR AC, and DFD fruits using two parallel razor blades bonded to a metal block to give uniform segments with a width of 6.25 mm and incubated in a mixture of cellulase and pectinase (0.2% m/v and 20% m/v, respectively; Sigma) in sodium citrate buffer (50 mM, pH 4.0), 1 mM NaN3 for 7 to 10 d at 37°C, followed by washes in sodium citrate buffer and dried at room temperature. A minimum of eight samples from each ripening stage was tested in unaxial tension using a model 4502 Instron Testing machine (Instron) and a 2.0 mm s–1 strain rate (see Matas et al., 2004
Cuticles were enzymatically isolated (Schönherr and Riederer, 1986
All cutin monomers were identified from electron impact mass spectrometry of the methyl ester trimethylsilyl derivatives on the basis of the published spectra (Eglinton and Hunneman, 1968
Based on Jenks et al. (1995) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
RESULTS
-amylase (from porcine pancreas, Sigma-Aldrich) and pullulanase (from Bacillus acidopullulyticus, Novozymes Biologicals) in 20 mL of Tris-HCl pH 7.0, 0.1% NaN3, for 24 h. Neutral sugar composition of the AIS was determined by gas-liquid chromatography as in Campbell et al. (1990)
20, respectively. 
-pentadecalactone were added as internal standards. After 16 h the reaction was cooled, adjusted to pH 4 to 5 with acetic acid, followed by two dichloromethane extractions (10 mL) to remove methyl ester monomers (Bonaventure et al., 2004
-amyrin coeluted during the above GC analysis, an additional GC analysis was performed for each sample using a temperature program to separate these constituents. The GC was programmed with an initial temperature of 80°C and increased at 15°C min–1 to 220°C, held for 65 min, then increased 15°C min–1 to 320°C, then for held 10 min. A r