3 1. Introduction The Japanese apricot, mume (Prunus mume Sieb. et Zucc.) fruit, is widely distributed among the different cli- mate regions of East Asia, and is harvested at the yellow- peel stage for pickles called “umeboshi” or at green-peel stage for its juice. Mume is known to be a climacteric fruit that produces large amounts of ethylene as it ripens (Ina- ba and Nakamura, 1981; Koyakumaru, 1997; Mita et al,. 1999), making it difficult to preserve for long periods of time as the characteristic taste and firmness is quickly lost. The reason for the degradation of the fruits and measures to prevent this degradation have never been thoroughly investigated. Investigating harvest timing and improving conditions for fruit quality preservation during commer- cial processes may make the fruit more common. NMR spectroscopy is a promising tool for exploring new aspects of plant science, especially given that there are various types of plant materials in which the physical states of cell-as- sociated water change naturally occur, e.g. in the maturation process of red raspberry (Williamson et al., 1992), goose- berry (Williamson et al., 1993), olives (Gussoni et al., 1993), kiwifruit (Callaghan et al., 1994), coconuts (Jagannathan et al., 1995), cherry tomatoes (Ishida et al., 1994), cherry fruits (Ishida et al., 1997), in floral malformation in mangos (Usha et al., 1994) and maturation in apples (Wang et al., 1988). In addition, the detection of water in different subcellular organ- elles is often used to understand the changing water distri- bution among plant components. It has been suggested that multi-exponential 1H-NMR relaxation times (T1 and T2) in plant tissues reflect water in different plant cell compartments (Burke et al., 1974; Stout et al., 1978; Gusta et al., 1979; Bacic and Ratkovic, 1984; Hills and Duce, 1990; Isobe et al., 1999), and can be ascribed to several different causes: cellu- lar heterogeneity and subcellular compartmentation (Belton Ontogenetic changes of the water status and accumulated soluble compounds in developing and ripening mume (Prunus mume) fruit measured by 1H-NMR analysis G. Watanabe(1), Y. Ishibashi(2)*, M. Iwaya-Inoue(2) (1) Agricultural Development Total Center, Kagoshima Prefecture, Ibusuki, 810-0013 Kagoshima, Japan. (2) Laboratory of Crop Science, Department of Plant Resources, Faculty of Agriculture, Kyushu Univer- sity, Hakozaki, 812-8581 Fukuoka, Japan. Abbreviations: CPMG= Carr-Percell- Meiboom-Gill; FAA= formalin acetic acid alcohol; NMR= nuclear magnetic resonance; T1= spin-lattice relaxation times; T2= spin-spin relaxation times. Key words: Histological observation, membrane integrity, mobility of water, NMR relaxation times (T1, T2), water content. Abstract: The physiological changes of intact mume (Prunus mume Sieb. et Zucc. cv. Rinshu) fruit tissues were examined by measuring the physical states of cell-associated water in the fruit tissues with developing and ripening using 1H-NMR spec- troscopy. We found that the water molecules in mume fruit tissues existed in several different compartments with different mobilities. Additionally, spectral recovery in the water proton indicated reverse relationships between the pericarps and seeds at the immature and mature stages. In the pericarp tissues, the longest T1 and longer T2 markedly increased, while those in the seeds decreased. From these results, the change in the water status with growth stage had reverse trajectories in the peri- carp and seed of the fruit. In the pericarp tissues, both water uptake and dry weight prominently increased with ripening. The epidermis and inner parenchymal cells of the pericarp tissues remarkably enlarged as a sigmoidal growth curve. Membrane permeability, indicating a loss of membrane integrity, increased in the pericarp tissues. The elongation in the fully vacuolated cells and changes in the membrane permeability in the pericarp tissues with ripening correlated to the longest T1. In contrast, the high mobility of water in the seeds began to decrease with maturation, while oil began to accumulate. Thus, the mobility of water, as analyzed in this study, is considered to reflect the results of physiological changes such as cellular heterogeneity and spatial arrangements both in the pericarp and in seed tissues for mume fruit with development and ripening. Adv. Hort. Sci., 2015 29(1): 3-12 * Corresponding author: yushi@agr.kyush-u.ac.jp Received for publication 17 September 2014 Accepted for publication 28 December 2014 4 Adv. Hort. Sci., 2015 29(1): 3-12 and Ratcliffe, 1985; Snaar and Van, 1992; Kumamoto et al., 1998; Iwaya-Inoue et al., 2004 a, b). The movement of water is controlled by cellular organization, such as compartmen- talization by membrane structures (Tanner, 1978, 1983), and water is ordered by macromolecules, such as proteins and polymers of organic compounds (Hazlewood, 1995). Thus, the dynamic states of water in cells closely correlate with the organic properties of macromolecular structures. The objectives of the present research were (1) to ex- amine the ontogenetic changes in the state of water by 1H-NMR spectroscopy, water content, membrane perme- ability and histochemical observation during development and ripening of Japanese apricot fruit, and (2) to study the interrelationships among them. 2. Materials and Methods Plant materials Japanese apricot (Prunus mume Sieb. et Zucc, cv. Rins- hu) fruits cultivated in the orchard at the University Farm of Kyushu University were harvested at each stage (a total of four times) from April to June. The fruits were defined as belonging to four ripening stages as follows (Fig. 1 and Ta- ble 1): Stage 1, fruits at approximately 2.0 g in fresh weight and 1.5 cm in transverse diameter, peels green, and seeds immature and 0.1 g in fresh weight; Stage 2, fruits at 15.0 g in fresh weight and 3.0 cm in transverse diameter, and seeds 0.5 g in fresh weight; Stage 3, fruit at their maximum size, seeds 0.7 g in fresh weight; Stage 4, fruit at maximum size and 45.0 g in fresh weight, 4.0 cm in transverse diameter, the firmness in the pericarp softening, the peels yellow, and seeds hard with a fresh weight of 0.8 g. Fruits were sepa- rated into pericarp (including epicarps) and seed, and both tissues were used for experiments. Tissue water contents were determined by obtaining the weight loss after drying in an oven at 90°C for 20 h. Eight to ten fruits were used in each stage in the following experiments. 1H-NMR analysis An NMR spectroscope with a magnet operating at 89.5 MHz for 1H (JEOL EX 90A) was used for the measure- ment of spin-lattice relaxation times (T1), 1H-NMR spin- spin relaxation times (T2) and 1H-NMR spectra. A piece of pericarp or seed of an intact Japanese apricot fruit was placed in an NMR tube (8 mm in diameter) which was then placed in an outer glass tube (10 cm in diameter) contain- ing 99.8% D 2 O as an internal lock signal; spectroscopic measurement at 25±1°C was then undertaken. The repeti- tion time was 15 s with four accumulation transients for each tissue. The decay between scans was always greater than five times T1. Table 1 - Characteristics of mume fruit tested Characteristics Stage 1 2 3 4 Harvested day April 9 ∼ 24 May 6 ∼ 12 May 29 ∼ June 4 June 13 ∼ 30 Skin color green green green yellow Endocarp hard hard soft soft Size (cm) Longitudinal diameter 1.6±0.1 3.2±0.4 3.9±0.1 4.2±0.1 Transverse diameter 1.3±0.1 2.9±0.1 3.5±0.9 4.2±0.1 Weight (g) Whole (z) 1.7±0.2 14.8±1.1 32.8±3.0 45.8±2.4 Pericarp 1.1±0.3 11.2±1.1 27.7±2.4 35.6±2.6 Seed 0.1±0.0 0.5±0.1 0.8±0.1 0.7±0.1 Values represent the mean of eight to ten fruits±SE. (z) The weight of the stone is not included. Fig.1 - Mume (Prunus mume Sieb.) fruit during four ripening stages. P=pericarp; S=seed. S P P S S P P S Stage 1 Stage 2 Stage 3 Stage 4 Fig. 1 - Mume (Prunus mume Sieb.) fruit during four ripening stages. P= pericarp; S= seed. 5 Watanabe et al., Characteristic of developing and ripening of mume fruit measured by 1H-NMR analysis T1 measurements were determined by the inversion re- covery (180°- τ -90°pulse sequence) method. τ is the time between the radio-frequency pulses in the sequence, and the indicated angles are between the average direction of the original proton spins and that induced by the radio-frequen- cy pulse (Farrar and Becker, 1971). Twenty-two values of pulse interval (τ) ranging from 0.001 s to 15 s were used to acquire a T1 data set. T1 was determined from the slope of ln (M0 - M) /2M0 versus τ, where M is the amplitude of the FID of the water proton signal following the 90° pulse at τ, and M0 is the limiting value of M. The existence of water components with different T1 values was revealed from spectral recovery and semi-log plots of signal intensity of 1H-NMR according to Ishida et al. (1994). A graphical method was used to determine the relaxation time of water and the percentage of the fraction (Hazlewood and Nichlos, 1969; Belton and Packer, 1974). Short T1 values including those below 0.1 s were considered from spectral recovery ranging in the pulse intervals of 0.03 s and 0.05 s. T2s measurements were determined by the CPMG (Carr-Percell-Meiboom-Gill) method from the slope of ln M/M0 versus t, where M0 is the magnetization ampli- tude of the water proton signal occurring at time τ after the initial 90° pulse in the CPMG (90°-τ-180°-2τ-180°- 2τ---) pulse sequence. t = 2nτ, where n is the number of refocusing pulses (25 points from 2 to 4000 loops) and τ was 0.001s. T2 was also determined from a semi-log plot of signal intensity as in the case of T1. Histological observations Materials were fixed in FAA (formalin acetic acid al- cohol; 80% ethanol: 100% acetic acid: formalin = 90:5:5), and 20-μm sections were cut using a microtome (Cryostat HM500-OM, Microm Co. Ltd) at -20°C. The sections were stained with 0.01% Ruthenium red for pectic substances in the middle lamella (Fig. 2A) and by Sudan III for lipid sub- stances (Fig. 2B), respectively. They were then subjected to microscopic observations (Axiphot, Carl Zeiss Co. Ltd). Leakage of electrolytes Pericarp tissues of Japanese apricot fruit (about 3.0 g) were cut into pieces of about 2-mm square. These small pieces were immersed in distilled water (50 ml) and stirred at 180 cycles/min. The extent of leakage of electrolytes was determined with an electrolytes conductivity meter (Toa conductivity meter, Model CM-20E, Toa Electronics Ltd.) and expressed as the percentage of the total electrolytes in each sample measured after samples were killed by a cycle of freezing and thawing (Iwaya-Inoue et al., 2004 a, b). 3. Results Histochemical characteristics of fruit tissues with ripening Mume fruit are characterized by four stages (Fig. 1 and Table 1). The histochemical changes of the mume fruit tis- sues with ripening are shown in figure 2. The individual pericarps and seeds are mainly comprised of the parenchy- mal tissue, beneath the epidermal tissue and around the vas- cular tissue. Pectic substances stained by Ruthenium red are abundant in the middle lamella of cell walls in the pericarp parenchymal tissues from the small green fruit to the large green fruit (Fig. 2A, Stages 3 and 4). The components in the cell wall of pericarp tissues began to change with matura- Fig. 2A Photomicrographs of cross-sections of the pericarp tissues in mume fruit with ripening (A～D). Parenchyma tissues beneath epidermis. A: Stage 1; B: Stage 2; C: Stage 3; D: Stage 4. (E～ H) Vascular bundles. E: Stage 1; F: Stage 2; G: Stage 3; H: Stage 4. epi, epidermis; v.b., vasucular bundle. Tissues were stained with Ruthenium Photomicrographs of cross-sections of the pericarp tissues in mume. Bars indicate 50μm. A B C D E F G H epi. v.b Stage 1 2 3 4 v.b. v.b v.b epi. epi. epi. Fig. 2A) Photomicrographs of cross-sections of the pericarp tissues in mume fruit with ripening (A ∼ D). Parenchymal tissues beneath epidermis. A= Stage 1; B= Stage 2; C= Stage 3; D= Stage 4. (E ∼ H) Vascular bundles. E= Stage 1; F= Stage 2; G= Stage 3; H= Stage 4. epi= epidermis; v.b.= vascular bundle. Tissues were stained with Ruthenium. Photomicrographs of cross-sections of the pericarp tissues in mume. Bars indicate 50 μm. 6 Adv. Hort. Sci., 2015 29(1): 3-12 tion of the fruit (Stage 3). As the peel of the fruit assumes a yellow color, pectic substances in the cell wall structures in the parenchymal tissue collapse, except for the epidermis and vascular bundles (Fig. 2A, Stage 4). On the other hand, lipids in the seeds stained with Sudan III were observed in the epidermis of seed tissues, indicating that the cells of the epidermis are abundant in suberin (Fig. 2B). Moreover, oil bodies were also observed in the parenchyma of the cotyle- dons in the matured and ripened fruit seeds (Stages 3 and 4). Water uptake in relation to cell enlargement In the pericarp tissues, both water uptake and dry weight prominently increased during ripening, especially from the large green-fruit stage to the matured-fruit stage (Fig. 3). Vacuoles occupy a large part of individual cells in the tissues at these stages (data not shown). The epi- dermis and inner parenchymal cells of the pericarp tissues remarkably enlarge during ripening in a sigmoidal growth curve (Table 2). The enlargement of cells in the pericarp region closely correlated with that for fruit diameter, ac- companied by a marked increase in water uptake and dry matter accumulation (Tables 1 and 2). In seed tissues, wa- ter uptake did not change, while dry weight remarkably increased in both the matured and ripened fruit (Fig. 2). Therefore, the seed water content remarkably decreased in the mature stages (Fig. 4, Stage 3). Membrane integrity in the pericarp tissues Changes in the membrane permeability as well as the cell wall integrity of mume fruit with ripening were con- sidered. Leakage of electrolytes from the pericarp tissues increased during ripening, and 90% of the total electro- lytes were found to have leaked at fruit maturation (Fig. 4, Stage 3). Thus, membrane permeability, indicating a loss of membrane integrity, arises in the pericarp tissues. Tissue specificity of 1H-NMR spectra with various pulse intervals between 180  and 90° The physiological changes of intact tissues were exam- ined by measuring the physical states of cell-associated Table 2 - Changes in cell size of pericarp during ripening of mume fruit Cell size (μm) Stage 1 2 3 4 Epidermis Long side×short side 25.1× 9.9 39.0×12.3 38.9×22.9 49.5×49.5 Parenchyma Small cells Long side×short side 44.7×18.6 97.9×54.4 119.7×68.9 142.9×102.9 Larage cells Diameter 61.4 145.5 181.2 243.9 Values represent the mean of ten cells of the pericarp tissues in three fruits. A B C D Fig. 2B Photomicrographs of cross-sections of the seed tissues in mume fruit with ripening (A～D). Parenchyma tissues beneath epidermis. A: Stage 1; B: Stage 2; C: Stage 3; D: Stage 4. s.c, seed coat; co., cotyledon. v.b; vasucular bundle. Tissues were stained with Sudan Ⅲ. Bars indicate 50μm. s.c co. v.b. s.c co.co. s.c co. s.c s.c 1 2 3 4 Stage Fig. 2B - Photomicrographs of cross-sections of the seed tissues in mume fruit with ripening (A ∼ D). Parenchymal tissues beneath epidermis. A= Stage 1; B= Stage 2; C= Stage 3; D= Stage 4. s.c.= seed coat; co.= cotyledon. v.b= vascular bundle. Tissues were stained with Sudan III. Bars indicate 50 μm. 7 Watanabe et al., Characteristic of developing and ripening of mume fruit measured by 1H-NMR analysis water in fruit tissues with ripening using 1H-NMR spec- troscopy. 1H-NMR spectra of both intact pericarp and seed in the fruit tissues were examined by spectral recovery of 1H-NMR with the inversion recovery method. The pulse interval was varied between 180o and τ- 90o pulses. An 1H- NMR spectral peak was observed at 4.8 ppm of chemi- cal shift, which corresponds to the 1H nuclei of water. 1H- NMR spectra of both the pericarp and seed tissues were not symmetrical, indicating that the peak consists of com- ponents with different chemical shifts and various recov- ery times (Fig. 5). In pericarp tissues of the small green fruit, major spec- tral recoveries were observed in the pulse intervals of 0.3 s and 0.5 s, while in the ripened fruit, these were 0.9 s and 1.5 s (Fig. 5A and B, Stage 1). On the other hand, in seeds of the small green fruit, several peaks of spectral recovery were mainly observed at pulse intervals of 0.8 s and 1.0 s, Fig.3. Changes in electrolyte leakage of pericarp during ripening of mume fruit. Values represent the mean of three fruit ± SE. 0 10 20 30 40 50 60 70 80 90 100 1 Stage E le c tr o ly te l e a ka ge ( % ) 1 2 3 4 Fig. 4 - Changes in electrolyte leakage of pericarp during ripening of mume fruit. Values represent the means of three fruits ± SE. Fig. 4A Changes in the water content of pericarp (□）and seed (■） for mume fruit with ripening. Values represent the mean of eight to ten fruit ± SE. W at e r c o n te n t (g H 2 O g - 1 D W ) 0 2 4 6 8 10 12 14 16 18 11 2 3 4 Stage Fig. 3A - Changes in the water content of pericarp ()and seed () for mume fruit with ripening. Values represent the mean of eight to ten fruits ± SE. R el at iv e va lu e Fig. 4B Relative value in the dry weight of pericarp (□）and seed (■） for mume fruit with ripening. Relative value of pericarp and seed at Stage 1 , respectively, is 1.0. 0 5 10 15 20 25 30 35 1 Stage 1 2 3 4 Fig. 3B - Changes in the dry weight of pericarp ()and seed () for mume fruit with ripening. Values represent the mean of eight to ten fruits ± SE. Fig. 5 - Spectral recovery of 1H-NMR of mume fruit with inversion recovery method. Τ: pulse interval. A= Spectra of pericarp tissues at Stage 1; B= Spectra of pericarp tissues at Stage 4; C= Spectra of seed tissues at Stage 1; D= Spectra of seed tissues at Stage 4. 1; D= Spectra of seed tissues at Stage 4. 8 Adv. Hort. Sci., 2015 29(1): 3-12 while they occurred between 0.05 s and 0.2 s in ripened fruit (Fig. 5C and D, Stage 4). The individual values of the T1 components can be obtained by pulse interval (τ) at null point, T1 = null / ln2. Thus T1 values that are due to major spectral recovery times in pericarp tissues became longer in the ripened fruit, while the tendency was reversed in the seed tissues. Changes of T1 components in the fruit tissues The relaxation times and estimated amounts of the individual water fractions calculated from semi-logarith- mic plots of 1H-NMR signal intensities in the fruit tissues with ripening processes are listed in Table 3. The spin- lattice NMR relaxation time (T1) describes the process of realigning the magnetic moment with the external mag- netic field. Cellular water exists in two to three components, which are shown by NMR relaxation times; in plant tissues, these water components consist of three states of water: i.e. free water, loosely bound water, and tightly bound water (Iwaya-Inoue and Nonami, 2003). The three compartmen- talizations of water originally identified as the vacuole, cytoplasm, and cell wall/extracellular space (apoplast) are reflected by the different relaxation times in the parenchy- mal tissues of apples (Snaar and Van As, 1992; Hills and Remigereau, 1997). Thus, differences in the relaxation times (T1, T2) of biological tissues can be interpreted as differences between the ratio of free water to bound water. Semi-logarithmic plots of 1H-NMR signal intensity with the inversion recovery method of mume fruit were multi- exponential. The long T1 fraction could be understood as highly mobile water (free water) derived from the vacu- oles, while the short T1 fraction with restricted mobility represents loosely bound and bound water from the cyto- plasm and the apoplastic region, respectively. In the pericarp tissues, T1 values of the longest water component markedly increased from 0.8 to 2.0 s from the small green fruit to the large green fruit (Stages 1 and 2). The major fraction of the longest T1 was about 80% and the ratio was constant during the ripening stage. The frac- tion of the shortest component of T1, at 0.4 s was about 10% and was also constant through all stages. In addition, vacuoles stained by neutral red were observed in a large proportion of individual cells in the pericarp tissues with ripening (data not shown). Thus, water in the pericarps was considered to increase the amount of free water with ripening. On the other hand, in the seeds of the small green and large green fruits, T1 values and the fractions of the longest component were about 2.0 s and 75%, respectively (Table 3, Stages 1 and 2). T1 values of the longest wa- ter component in the seeds markedly decreased from 2.0 to 0.4 s from the large green-fruit to matured-fruit stages (Stage 2 and 3). The T1 values did not change further for the ripened fruit. However, the fraction ratio of the longest water component in the seed decreased from 45 to 30 % during that stage (Stage 4). Table 4 - Changes in components of 1H-NMR spin-spin relaxation times (T2)s during ripening of Japanese apricot fruit Stage 1 2 3 4 T2 values (ms) Fraction (%) T2 values (ms) Fraction (%) T2 values (ms) Fraction (%) T2 values (ms) Fraction (%) Pericarp 151±12 28.1±6.0 289±17 26.4±1.2 394±34 25.9±0.9 404±38 26.9±6.7 211±10 71.9±6.0 442±10 73.6±1.2 731±26 74.1±0.9 713±14 73.1±6.7 Seed 143±11 12.8±2.8 205±19 16.7±6.3 36± 2 16.7±2.8 27± 2 44.8±2.8 323±24 87.2±2.8 317±14 83.3±6.3 56± 2 83.3±2.8 47± 6 55.2±2.8 Values represent the means of eight to ten fruit±SE. Table 3 - Changes in components of 1H-NMR spin-lattice relaxation times (T1)s during ripening of mume fruit Stage 1 2 3 4 T1 values (ms) Fraction (%) T1 values (ms) Fraction (%) T1 values (ms) Fraction (%) T1 values (ms) Fraction (%) Pericarp 412±13 12.1±3.2 416± 1 12.1±0.5 413± 5 12.3±0.8 424±13 11.3±1.7 603±55 7.4±3.4 1171±44 9.2±2.4 1332±63 11.9±1.1 1723±28 10.2±2.1 758±54 80.5±1.6 1367±34 78.7±1.7 1619±91 75.8±2.1 2057±94 78.5±1.1 Seed 349±11 13.5±4.5 785± 6 12.2±1.1 181± 3 20.2±1.1 138± 2 18.5±1.2 1414±28 13.4±4.3 1443±25 15.9±1.3 321±49 33.7±3.6 193± 3 50.6±1.2 2216±15 73.0±0.4 2040±19 75.6±0.6 443±66 46.1±2.6 370± 2 30.9±0.8 Values represent the means of eight to ten samples±SE.. 9 Watanabe et al., Characteristic of developing and ripening of mume fruit measured by 1H-NMR analysis Changes of the T2 components in the fruit tissues The spin-spin NMR relaxation time (T2) describes the time-dependent decay of NMR signal due to the dephasing process of the individual spins with respect to each other. The water component estimated by T2 was divided into two fractions (Table 4). The regions with long T2 values in Glycine max seed tissues had high concentrations of free water, while the regions with short T2 values had high con- centrations of loosely bound and bound water (Ishida et al., 1987). In the pericarps of the mume fruit, the fractions in T2 comprised about 70% through the four stages. T2 values of both long and short components increased until fruit maturation (Stage 3); however, they did not prolong further at the ripened-fruit stage (Stage 4). By contrast, T2 values of the longer water component in seeds mark- edly decreased from about 300 ms to 60 ms from the large green fruit to the matured fruit (Stage 2 to Stage 3). 4. Discussion and Conclusions Relationship between NMR relaxation time T1 and the characteristics in mume fruits during development and ripening Nuclear magnetic resonance (NMR) spectroscopy is a useful technique to follow physiological changes with re- spect to the state of water in developing Japanese apricot fruits. Water in living tissues is known to consist of several components with regard to the relaxation of the magne- tized protons (Hazlewood et al., 1969; Hazlewood, 1995; Isobe et al., 1999; Iwaya-Inoue, 2004 a, b). In developing Japanese apricot fruits, three components with different T1 values were distinguished on the semi-logarithmic plots of the recovery of the 1H-NMR signal (Fig. 5 and Table 3). Multicomponent water fractions are similar to those reported for other plant tissues (Stout et al., 1978; Gusta et al., 1979; Isobe et al., 1999). Referring to the line width of 31P-NMR signals (Kano et al., 1990; Taka- gishi et al., 1991), the longest component is ascribed to that of vacuoles containing small molecules, such as met- abolic intermediates, secondary products and inorganic ions, and the middle component to that of cytoplasm in plant tissues. The water with the shortest T1 is consid- ered to be exchangeable water around macromolecules, such as starches, proteins and strings of macromolecules in vesicles or between cell walls (Rorschach and Hazle- wood, 1986). Large amounts of highly mobile water were detected in the seeds of small green and large green fruits. The mobil- ity of exchangeable water in the seed tissues became high- er as fruit developed. However, the mobility became low at fruit maturation; thereafter, it continued to be low. On the other hand, the mobile water in the pericarp of small green fruit was low. This suggests that concentrations of cell components that bind water were high in these tissues. Thereafter, the mobile water in the pericarp increased with enlargement and even more so upon coloring of the fruit (Figs. 1 and 5, Tables 1 and 3). These trends agree with previous reports (Ishida et al., 1989; 1994; 1997). Changes in the mobility of water are considered to be the result of physiological changes in fruit growth including seed de- velopment and maturation, which is the most important natural function of the fruits (Crane, 1964). Relaxation times are strongly influenced by the avail- ability of the water and the presence of macromolecules to which water molecules can be “bound”. It was shown that T1 closely correlated with water content in developing and maturing rice grains (Funaba et al., 2006), in azalea buds subjected to low-temperature stress (Kaku et al., 1984), and in the heat-tolerant and heat-sensitive cultivars rice grains subjected to high-temperature stress (Tanaka et al., 2009). In the pericarp tissues, the changes in water contents were not correlated with the values of the longest T1 com- ponent with ripening (Fig. 3A, Tables 3). However, our results suggest that other factors may contribute to the motional restriction of water. A similar tendency was ob- served in sweet potato tubers exposed to cold stress (Iwa- ya-Inoue et al., 2004 b). We found that membrane permeability, indicating a loss of membrane integrity, increased in mume pericarp tissues during ripening. An increase in membrane perme- ability also has been reported in ripening wild-type to- mato fruit, but not in the ripening-inhibited (rin) mutant (Poovaiah et al., 1975). Moreover, the permeability of the plasma membrane increased and the membrane-lipid com- position changed in ripening Malus domestica fruit (Lurie et al., 1987) and during the development and senescence of Cucumis melo fruit (Lester and Stein, 1993). This coin- cided with our results from the electrolyte leakage study for mume fruit. In wood plants (i.e. galled leaves invaded by insects), a higher ratio of ion leakage was strongly as- sociated with prolongation of the NMR relaxation times (Kaku and Iwaya-Inoue, 1990). In addition, it was dem- onstrated that the relaxation times strongly related to the size and geometry of the vacuolated cells in mushroom tis- sues during postharvest senescence (Donker et al., 1997). In our experiment, elongation in fully vacuolated cells and changes in membrane permeability in the pericarp tissues may have contributed to the correlation of the longest T1 components with fruit ripening. However, in seed tissues, a considerable decrease in the longest T1 components in seeds of mume fruit was accom- panied with markedly decreased water content, from 14 to 1 g H 2 O/g dry weight from Stage 2 to Stage 4 (Fig. 3A and Table 3). Phenomena such as cold acclimation in plant tis- sues correlated with the decrease both in T1 and in water content (Burke et al., 1974; Kaku et al., 1984; Fennell et al., 1996; Yoshida et al., 1997). The decrease in water content of the seed tissues during maturation is considered to be as- sociated with a reduction in free water, perhaps due to an in- crease of dry matter, that is, an increase in the accumulation of cellular substances (Fig. 2A and B and Table 3). 10 Adv. Hort. Sci., 2015 29(1): 3-12 Relationship between NMR relaxation time T2 and charac- teristics in mume fruits during development and ripening The larger fraction of the two, with the long T2, can still be assigned to vacuolar water, whereas the small frac- tion with short T2 represents water from the cytoplasm and perhaps the contribution of water inside the cell wall and extracellular water (Scheenen et al., 2002). The amount of water or T2 in the pericarp and seed changed inversely according to the progression of growth stages. Mobile water in the pericarp of small green fruit was low (Table 4). Thereafter, the mobile water in the peri- carp increased with enlargement, but did not change fur- ther at the coloring fruit stage. Thus, the changes in water contents were not correlated with the values of a longer T2 component with ripening (Fig. 3A and Table 4). During the stage at which the fruit peel became yellow, however, pectic substances in the cell wall structures in the parenchymal tissue collapsed except for in the epidermis and vascular bundles (Figs. 1 and 2A, Stage 4). Kaneko et al. (1989) reported changes in the pectic substance components during mume fruit ripening. The cellular changes, such as the changes of polysaccharide components or pectic substances, specific to the pericarp tissue ripening characteristics, may contribute to a more ordered state of water, thereby resulting in unchanged re- laxation time. In previous reports, McCarthy et al., (1995) measured a decrease in T2 in bruised regions of apples. In addition, marked shortening of T2 values compared with T1 values in the fruit tissues may reflect a correlation be- tween the relaxation times and compartment size of cross- linked polymer gels (Murase and Watanabe, 1989). On the other hand, in seed tissues, a considerable de- crease in the longer T2 components in seeds of mume fruit is accompanied with markedly decreased water content from Stage 2 to Stage 4 (Fig. 3A and Table 4). Water status can provide useful information about the characteristics of mume fruit development and ripening When the seed of a small green mume fruit was gellified, the mobility of water was high. Thereafter, when the color inside the seed of a small green mume fruit turned milky white, the mobility of water, water derived from the vacuole and exchangeable water around macromolecules, became low (Fig. 1 and Tables 3 and 4). The decrease in water con- tent of the seed tissues during maturation is associated with a reduction in free water, perhaps due to an increase of dry matter, that is, an increase in the accumulation of cellular substances such as oil bodies (Fig. 2B and Fig. 3B). Japanese apricot fruit pericarp tissues may act as a sort of storehouse where photosynthates are temporarily accu- mulated for further transport into the seed during seed for- mation. Therefore, seed formation, which is the primary object of fruit growth, may require larger amounts of en- ergy than are available in the pericarp. The parenchymal tissues of the pericarps in matured mume fruit began to collapse in the yellow-peel stage (Figs. 1 and 2A), prob- ably because the pericarp tissues are no longer necessary to physiological functions. These are considered to be the result of a loss of membrane integrity and efflux of pec- tic substances in the cell wall structures (Figs. 2A and 4). Regardless of the constant high water content, the level of high-mobility water, i.e. water derived from vacuoles and exchangeable water, was high (Fig. 3A, Tables 3 and 4). This increase is considered to result from the fact that the increase of the mobile water by cell wall breakdown is greater than the decrease of the free water by cellular accumulation. In addition, the enlargement of cells in the pericarp region closely correlates with increases in fruit diameter, water uptake, and dry matter accumulation. Changes in the fully-vacuolated-cell volume were closely correlated with changes in the values of the longest T1 component during ripening (Tables 2, 3 and 4). Thus, the water compartments and the mobility of water analyzed in this study are considered to reflect the result of physi- ological changes such as cellular heterogeneity and spatial arrangements both in the pericarp and in seed tissues for mume fruit with development and ripening. In conclusion, the analysis of water components de- rived from 1H-NMR spectroscopy provides useful infor- mation about the characteristics of fruit ripening. Acknowledgements This work was supported in part by a Grant-in-Aid (No.11460016) for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan to M. I. I. We are grateful to Drs. H. Kano and M. Koizumi of the National Institute of Agrobiological Resources and Dr. Ishida of the National Food Research Institute for their advice regarding NMR techniques and for their helpful suggestions. We also thank Kyushu Matsushita Denki Co. Ltd. for allowing us the use of their NMR spectroscope, and Dr. N. Tamura of Fukuoka Women’s University for the analysis of the NMR spectra. We wish to express our special thanks to Dr. A. Wakana of our university and the Kasuga family of an orchard located in Fukuoka city for providing us the mume fruits. References BACIC G., RATKOVIC S., 1984 - Water exchange in plant tis- sue studied by proton NMR in the presence of paramagnetic centers. - Biophys. J., 45: 767-776. BELTON P.S., PACKER K.J., 1974 - Pulsed NMR studies of water in striated muscle. III. The effects of water content. - Biochim. Biophys. Acta, 354: 305-314. BELTON P.S., RATCLIFFE R.G., 1985 - NMR and compart- mentation in biological tissue. - Prog. NMR Spectroscopy, 17: 241-279. BURKE M.J., BRYANT R.G., WEISER C.J., 1974 - Nuclear magnetic resonance of water in cold acclimating red osier dogwood stem. - Plant Physiol., 54: 392-398. 11 Watanabe et al., Characteristic of developing and ripening of mume fruit measured by 1H-NMR analysis CALLAGHAN P.T., CLARK C.J., FORDE L.C., 1994 - Use of static dynamic NMR microscopy to investigate the origins of contrast in images of biological tissues. - Biophys. Chem., 50: 225-235. CRANE J.C., 1964 - Growth substances in fruit setting and de- velopment. - Annu. Rev. Plant Physiol., 15: 303-326. DONKER H.C.W., VAN AS H., SNIJDER H.J., EDZES H.T., 1997 - Quantitative 1H-NMR imaging of water in white but- ton mushrooms (Agaricus bisporus). - Magn. Reson. Imag., 15(1): 113-121. FARRAR T.C., BECKER E.D., 1971 - Pulse and Fourier Trans- form NMR. - Academic Press, New York, USA, pp. 115. FENNELL A., WAKE C., MOLITOR P., 1996 - Use of 1H-NMR to determine grape bud water state during the photoperiodic induction of dormancy. - J. Amer. Soc. Hort. Sci., 121(6): 1112-1116. FUNABA M., ISHIBASHI Y., MOLLA A.H., IWATANANI K., IWAYA-INOUE M., 2006 - Influence of low/high tempera- ture on water status in developing and maturing rice grains. - Plant Prod. Sci., 9(4): 347-354. GUSSONI M., GRECO F., CONSONNI R., MOLINARI H., ZANNONI G., BIANCHI G., ZETTA L., 1993 - Applica- tion of NMR microscopy to the histochemistry study of olives (Olea europea, L.). - Magn. Reson. Imag., 11(2): 259-268. GUSTA L.V., FOWLER D.B., CHEN P., RUSSELL D.B., STOUT D.G., 1979 - A nuclear magnetic resonance study of water in cold acclimating cereals. - Plant Physiol., 63: 627-634. HAZLEWOOD C.F., 1995 - Water movement and diffusion in tissues: diffusion in biological tissues, pp. 123-126. - In: BI- HAN DL (ed.) Diffusion and perfusion magnetic resonance imaging. - Raver Press, New York, USA, pp. 398. HAZLEWOOD C.F., NICHOLS B.L., CHAMBERLAIN N.F., 1969 - Evidence for the existence of a minimum of two phases of ordered water in skeletal muscle. - Nature, 222: 747-750. HILLS B.P., DUCE S.L., 1990 - The influence of chemical and diffusive exchange on water proton transverse relaxation in plant tissues. - Magn. Reson. Imag., 8: 321-331. HILLS B.P., REMIGEREAU B., 1997 - NMR studies of chang- es in subcellular water compartimentation in parenchyma apple tissue during drying and freezing. - Int. J. Food Sci. Tech., 32: 51-61. INABA A., NAKAMURA R., 1981 - Ripening characteristics of Japanese apricot (Mume, Prunus mume Sieb. et Zucc.) fruits on and of the tree. - J. Jpn. Soc. Hort. Sci., 49(4): 601-607. ISHIDA N., KANO H.T., KOBAYASHI T., HAMAGUCHI H., YHOSHIDA T., 1987 - Estimation of biological activities by NMR in soybean seeds during maturation. - Agr. Biol. Chem., 51(2): 301-307. ISHIDA N., KOBAYASHI T., KOIZUMI M., KANO H., 1989 - 1H-NMR imaging of tomato fruits. - Agr. Biol. Chem., 53: 2363-2367. ISHIDA N., KOIZUMI M., KANO H., 1994 - Ontogenetic changes in water in cherry tomato fruits measured by nucle- ar magnetic resonance imaging. - Sci. Hort., 57: 335-346. ISHIDA N., OGAWA H., KOIZUMI M., KANO H., 1997 - Ontogenetic changes of the water status and accumulated soluble compounds in growing cherry fruits studied by NMR imaging. - Magn. Reson. Chem., 35: 22-28. ISOBE S., ISHIDA N., KOIZUMI M., KANO H., HAZLE- WOOD C.F., 1999 - Effect of electric field on physical states of cell-associated water in germinating morning glory seeds observed by 1H-NMR. - Biochim. Biophys. Acta, 1426: 17-31. IWAYA-INOUE M., MATSUI R., FUKUYAMA M., 2004 a - Cold- or Heat-Tolerance of leaves and roots in perennial ryegrass determined by 1H-NMR. - Plant Prod. Sci., 7(2): 118-128. IWAYA-INOUE M., MATSUI R., SULTANA N., SAITOU K., SAKAGUCHI K., FUKUYAMA M., 2004 b - 1H-NMR method enables early identification of degeneration in the quality of sweet potato tubers. - J. Agron. & Crop Sci., 190: 65-72. IWAYA-INOUE M., NONAMI H., 2003 - Effects of trehalose on flower senescence from the view point of physical states of water. - Environ. Cont. Biol., 41: 3-15. JAGANNATHAN N.R., GOVINDARAJU V., RAGHUNA- THAN P., 1995 - In vivo magnetic resonance study of his- tochemistry of coconut (Cocos nucifera). - Magn. Reson. Imag., 13: 885-892. KAKU S, IWAYA-INOUE M., GUSTA L.V., 1984 - Relation- ships of nuclear magnetic resonance relaxation time to water content and cold hardiness in flower buds of evergreen aza- lea. - Plant Cell Physiol., 25: 875-882. KAKU S., IWAYA-INOUE M., 1990 - Factors affecting the pro- longation of NMR relaxation times of water protons in leaves of woody plants affected by formation of insect galls. - Plant Cell Physiol., 31: 627-637. KANEKO H., OTA K., KAWANO K., MAEDA Y., 1989 - Re- lationships between various parameters, especially organic acid content and picking time, on the processing of Plums into Ume-Zuke. - Nippon Eiyo. Syokuryo Gakkaishi, 42: 179-184 (In Japanese with English summary). KANO H., ISHIDA H.., KOBAYASHI T., KOIZUMI M., 1990 - 1H-NMR imaging analysis of changes of free water distribu- tion in barley and soybean seeds during maturation. - Jpn. J. Crop Sci., 59: 503-509. KOYAKUMARU T., 1997 - Effects of temperature and ethyl- ene removing agents on respiration of mature-green mume (Prunus mume Sieb. et Zucc.) fruit held under air and con- trolled atmospheres. - J. Jpn. Soc. Hort. Sci., 66: 409-418. KUMAMOTO Y., WATANABE G., IWAYA-INOUE M., 1998 - Changes of bound water and free water components traced by 1H-NMR in immature and mature fruit. - Cryobiol. and Cryotech., 44: 29-37. LESTER G., STEIN E., 1993 - Plasma membrane physicochem- ical changes during maturation and postharvest storage of muskmelon fruit. - J. Amer. Soc. Hort. Sci., 118(2): 223-227. LURIE S., SONEGO L., BEN-ARIE R., 1987 - Permeability, microviscosity and chemical changes in the plasma mem- brane during storage of apple fruit. - Sci. Hort., 32: 73-38. MCCARTHY M.J., ZION B., CHEN P., ABLETT S., DARKE A.H., LILLFORD P.J., 1995 - Diamagnetic susceptibility changes in apple tissues after bruishing. - J. Sci. Food Agr., 67(1): 13-20. MITA S., KIRITA C., KATO M., HYODO H., 1999 - Expres- sion of ACC synthase is enhanced earlier than that of ACC oxidase during fruit ripening of mume (Prunus mume). - Physiol. Plant., 107: 319-328. 12 Adv. Hort. Sci., 2015 29(1): 3-12 MURASE N., WATANABE T., 1989 - Nuclear magnetic relax- ation studies of the compartmentalized water in cross linked polymer gels. - Magn. Reson. Med., 9: 1-7. POOVAIAH B.W., MIZRAHI Y., DOSTAL H.C., CHERRY J.H., LEOPOLD A.C., 1975 - Water permeability during to- mato fruit development in normal and in non-ripening mu- tant. - Plant Physiol., 56: 813-815. RORSCHACH H.E., HAZLEWOOD C.F., 1986 - Protein dy- namics and the NMR relaxation time T1 of water in biologi- cal systems. - J. Magn. Reson., 70: 79-88 SCHEENEN T.W.J., HEEMSKERK A.M., DE JAGER P.A., VERGELDT F.J., VAN AS H., 2002 - Functional imaging of plants: a nuclear magnetic resonance study of a cucumber plant. - Biophys. J., 82: 481-492. SNAAR J.E.M., VAN AS H., 1992 - Probing water compart- ments and membrane permeability in plant cells by 1H NMR relaxation measurements. - Biophys. J., 63(6): 1654-1658. STOUT D.G., STEPONKUS P.L., COTTS R.M., 1978 - Nucle- ar magnetic resonance relaxation times and plasmalemma water exchange in ivy bark. - J. Plant Physiol., 62: 636-641. TAKAGISHI H., SHIRATA K., ISHIDA N., KOBAYASHI T., KOIZUMI M., KANO H., 1991 - Broadening of the 31P- NMR signal of vacuole-associated inorganic phosphate by NaCl in wild-turf roots. - J. Plant Physiol., 138: 511-515. TANAKA K., ONISHI R., MIYAZAKI M., ISHIBASHI Y., YUASA T., IWAYA-INOUE M., 2009 - Changes in NMR relaxation of rice grains, kernel quality and physicochemical properties in response to a high temperature after flower- ing in heat-tolerant and heat-sensitive rice cultivars. - Plant Prod. Sci., 12(2): 185-192. TANNER J.E., 1978 - Transient diffusion in a system partitioned by permeable barriers; application to NMR measurements with a pulsed field gradient. - J. Chem. Phys., 69: 1748-1954. TANNER J.E., 1983 - Intracellular diffusion of water. - Arch. Biochim. Biophys, 224: 416-428. USHA K., GAMBHIR P.N., SHARMA H.C., GOSWAMI A.M., SINGH B., 1994 - Relationships of molecular mobility of water with floral malformation in mango as assessed by nu- clear magnetic resonance. - Sci. Hort., 59: 291-295. WANG S.Y., WANG P.C., FAUST M., 1988 - Non-destructive detection of watercore in apple with nuclear magnetic reso- nance imaging. - Sci. Hort., 35: 227-234. WILLIAMSON B., GOODMAN B.A., CHUDEK J.A., 1992 - Nucelear magnetic resonace (NMR) macroimaging of ripen- ing red raspberry fruits. - New Phytol., 120: 21-28. WILLIAMSON B., GOODMAN B.A., CHUDEK J.A., 1993 - The structure of mature gooseberry fruits revealed non- invasively by NMR microscopy. - Micron., 24(4): 377-383. YOSHIDA M., ABE J., MORIYAMA M., SHIMOKAWA S., NAKAMURA Y., 1997 - Seasonal changes in the physical state of crown water associated with freezing tolerance in winter wheat. - Physiol. Plant., 99: 363-370.