Energy status of ripening and postharvest senescent fruit of litchi (Litchi chinensis Sonn.)
© Wang et al.; licensee BioMed Central Ltd. 2013
Received: 17 January 2013
Accepted: 18 March 2013
Published: 2 April 2013
Skip to main content
© Wang et al.; licensee BioMed Central Ltd. 2013
Received: 17 January 2013
Accepted: 18 March 2013
Published: 2 April 2013
Recent studies have demonstrated that cellular energy is a key factor switching on ripening and senescence of fruit. However, the factors that influence fruit energy status remain largely unknown.
HPLC profiling showed that ATP abundance increased significantly in developing preharvest litchi fruit and was strongly correlated with fruit fresh weight. In contrast, ATP levels declined significantly during postharvest fruit senescence and were correlated with the decrease in the proportion of edible fruit. The five gene transcripts isolated from the litchi fruit pericarp were highly expressed in vegetative tissues and peaked at 70 days after flowering (DAF) consistent with fruit ADP concentrations, except for uncoupling mitochondrial protein 1 (UCP1), which was predominantly expressed in the root, and ATP synthase beta subunit (AtpB), which was up-regulated significantly before harvest and peaked 2 days after storage. These results indicated that the color-breaker stage at 70 DAF and 2 days after storage may be key turning points in fruit energy metabolism. Transcript abundance of alternative oxidase 1 (AOX1) increased after 2 days of storage to significantly higher levels than those of LcAtpB, and was down-regulated significantly by exogenous ATP. ATP supplementation had no significant effect on transcript abundance of ADP/ATP carrier 1 (AAC1) and slowed the changes in sucrose non-fermenting-1-related kinase 2 (SnRK2) expression, but maintained ATP and energy charge levels, which were correlated with delayed senescence.
Our results suggest that senescence of litchi fruit is closely related with energy. A surge of LcAtpB expression marked the beginning of fruit senescence. The findings may provide a new strategy to extend fruit shelf life by regulating its energy level.
Fruit ripening and senescence are active processes initiated by internal and environmental factors. Considerable evidence suggests that cellular energy supply is a key factor controlling ripening and senescence events, and that aging and browning of postharvest horticultural crops may be related to inadequate supplies and reduced efficiency of cellular energy generation [1–3]. A salient feature initiating the aging process is a decline in adenosine triphosphate (ATP) levels. For instance, increased membrane permeability and enhanced reactive oxygen species (ROS) production by harvested fruit are related to low ATP and energy charge (EC) levels [4, 5]. Preharvest application of boron (B) and calcium (Ca) , and postharvest treatments, including ethylene inhibitors , high oxygen , anaerobic conditions , controlled atmosphere (CA) storage [2, 3, 5, 9], an exogenous carbon source  and ATP [10, 11], can maintain tissue levels of ATP and EC, thereby delaying pericarp browning of litchi and longan and internal flesh browning in 'Conference' pears. In addition, exogenous ATP elevates ATP levels, inhibits ROS accumulation, and maintains unsaturated fatty acid levels and membrane integrity, thus delaying senescence and deterioration of horticultural products [5, 7–10, 12–14]. Therefore, browning and senescence are closely related with cellular energy status, and measures to maintain energy levels can delay aging and inhibit occurrence of browning and deterioration to a certain extent. However, the mechanism contributing to the energy deficit remains unclear owing to the diversity of energy regulatory elements and complexity of energy production, transfer and control. Moreover, a growing body of evidence indicates that cellular energy levels are closely related to the activity, transcription and translation of certain regulatory elements [15–18]. Among these regulatory elements, proteins that synthesize (ATP synthase), dissipate (alternative oxidase, AOX; and mitochondrial inner membrane uncoupling proteins, UCP), transport [adenosine diphosphate (ADP)/ATP carrier, AAC] and regulate (sucrose non-fermenting-1-related kinase, SnRK) ATP have attracted considerable attention.
F1F0-ATP synthase is a multimeric enzyme that catalyzes the final step of oxidative phosphorylation and photophosphorylation, the synthesis of ATP from ADP and inorganic phosphate (Pi) [19, 20]. The general structure of the core subunits of the enzyme are highly conserved in both prokaryotic and eukaryotic organisms, and are designated as αβγδε . As a novel cell death regulator, subunit β (AtpB) plays a pivotal role in stress responses. Up-regulation of AtpB transcription marks initiation of the aging process [21, 22]. Complex V cannot be assembled successfully and respiration rates are reduced resulting in inhibition of ATP synthesis in the absence of the β subunit . However, it is unknown whether AtpB acts as a senescence marker gene in fruit as in photosynthetic tissues.
AOX and UCP are ubiquitous in plants and have various physiological roles. These include heat production and protection against oxygen free radicals mediating the non-phosphorylating bypasses that affect ATP synthesis [16, 23–26]. In plants UCP and AOX belong to multigene families, of which members exhibit specific organ and temporal expression patterns that respond differentially to stress conditions . AOX expression and activity are generally induced under stress conditions, and consequently AOX is used as a marker for cell reprogramming under stress [28–31]. The alternate oxidation pathway may slow down the aging process and quality deterioration of postharvest horticultural crops . Furthermore, changes in AOX and UCP accumulation in vine-ripened fruit differ significantly from those of postharvest-ripened fruit [32, 33]. This implies that a complementary relationship exists between the two energy dissipation pathways. Several reports have described the complex regulation of AOX and UCP at the translational and posttranslational levels. However, most research on these two genes has focused on climacteric fruit, such as tomato, apple, banana, and mango [34–37], and their roles in non-climacteric fruit remains unclear.
ADP/ATP carrier (AAC) belongs to the high-abundance mitochondrial transporter protein family . It catalyzes a counter-exchange of cytosolic ADP with matrix ATP, enabling mitochondria to supply the cytosol, and subsequently other organelles, with energy . In higher organisms AAC is the core protein of the mitochondrial adenine nucleotide translocator and the main regulator of mitochondrial ATP concentration [15, 40, 41]. As ‘the energy regulator’ in the cell, sucrose non-fermenting-1-related protein kinase (SnRK) can sense the internal energy status by measuring cytosolic ATP/ [adenosine monophosphate (AMP) ] ratios and controlling the expression and phosphorylation of key metabolic enzymes . However, to the best of our knowledge, no report has documented these two genes in the development and senescence of fruit.
Overall, in contrast to the relatively well-known physiological roles of these genes in stress responses, little information is available on how these genes contribute to the energy status of developing and senescing fruit. Studying the transcript level of these genes in relation to cellular energy status is crucial to identify causal agents contributing to energy deficits and their role in fruit ripening and senescence. This knowledge may provide opportunities to develop new strategies to control fruit ripening and the aging process and ultimately extend the storage life of harvested fruit.
Litchi is a non-climacteric, tropical and subtropical fruit with a high commercial value on the international market. Litchi fruit deteriorate rapidly after harvest because of water loss, pericarp browning and rot development . Our previous studies reported that an exogenous ATP supply enhanced antioxidant systems and maintained membrane integrity to delay browning and senescence of litchi fruit [11, 13, 14]. However, the molecular mechanisms underlying these phenomena remain unclear. In the current study, full-length sequences of AtpB, AOX1, UCP1, AAC1 and SnRK2 were cloned from litchi fruit. Transcript abundance of these energy-related genes, respiration intensity and fruit energy status were determined for developing and postharvest-senescent litchi fruit.
Color parameters (lightness, chroma, and hue angle) of litchi fruit at different development and ripening stages
Days after flowering (DAF)
45.60 ± 1.36
47.93 ± 1.39
51.99 ± 2.77
43.14 ± 2.23
36.48 ± 2.72
32.87 ± 1.09
38.34 ± 0.95
36.03 ± 1.15
34.63 ± 0.94
28.21 ± 1.52
115.19 ± 0.79
109.74 ± 1.37
75.94 ± 6.60
36.67 ± 2.91
34.62 ± 5.50
ATP concentrations were correlated strongly with fruit fresh weight (Figure 1C), and increased rapidly between 50 and 80 DAF before slowing at the final stage of maturation (80–90 DAF) (Figure 3A). ADP concentrations increased during stage I but decreased progressively during stage II (Figure 3B). AMP concentrations decreased gradually throughout fruit development, but increased slightly in the last 10 days before harvest (Figure 3C). Total adenylate levels increased steadily between 50 and 80 DAF, and did not change thereafter (Figure 3D). The EC level of the fruit remained relatively constant in young fruit, and increased markedly as the fruit ripened between 80 and 90 DAF (Figure 3E). The ATP/AMP ratio peaked at 80 DAF, and thereafter declined until maturity (Figure 3F).
Degenerate primers for cloning of energy-related genes in litchi pericarp
Forward primer (5’ to 3’)
Reverse primer (5’ to 3’)
Homologies based on nucleotide sequences for energy-related genes isolated from litchi cv. Huaizhi
Top Arabidopsis BLAST match
Top BLAST match excluding Arabidopsis
AJ235513.2 AtpB Koelreuteria paniculata
EF523518.1 AOX1a Nicotiana glutinosa
XM_002520396.1 UCP Ricinus communis
XM_002531865.1 AAC Ricinus communis
XM_002513909.1 SAPK1 Ricinus communis
The coding region of LcAtpB was 1377 bp in length, encoding a deduced 459-amino acid (aa) sequence, with a predicted MW of 49.90 kDa and a calculated isoelectric point (pI) of 5.12. LcAtpB was 92% homologous with the AtpB from Koelreuteria paniculata, a close relative of litchi. The C-terminal domain of the β subunit contained the highly conserved ‘DELSEED’ motif (Additional files 1 and 2) involved in mechanochemical coupling of ATP synthase . AOX belongs to a multigene family in many plants, including mango, tomato and other horticultural crops. Two types of AOX genes are present in angiosperms , of which isoforms may be differentially expressed in fruit and other tissues, such as in tomato . While in the present study, transcript evidence showed support for a single predicted isoform (LcAOX1) for all candidates in litchi fruit pericarp and consequently only this isoform was detected during experimental validation. LcAOX1 (1032 bp), encoding a 344 aa sequence (predicted MW 39.13 kDa; pI 8.76), showed 71% identity with that of Nicotiana glutinosa. Site-directed mutagenesis has shown that several residues are required for AOX activity . The highly conserved residues were identified in litchi fruit using multiple sequence alignments of AOX proteins from diverse organisms (Additional files 3 and 4). UCP family from diverse organisms are grouped into five subfamilies, plant UCP1 and UCP2 belong to subfamilies II. Moreover, investigations have identified 1 to 6 genes that encode UCPs in different plants . The major feature of UCPs from plants is the presence of three energy transfer protein signatures (ETPS)  conserved in all UCP isoforms (Additional file 5). In the present study, three copies of the ETPS signature were identified in LcUCP1, which contained an open reading frame (ORF) of 915 bp encoding a 305 aa protein, with a predicted MW of 33.23 kDa and a pI of 9.01. A BLAST search of GenBank revealed that LcUCP1 shared 66% identity with AtUCP2 (NM_125287.4) from Arabidopsis thaliana and 65% identity with RcUCP (XM_002520396.1) from Ricinus communis at the protein level. A neighbor-joining cladogram further showed that LcUCP1 was most similar to UCP from R. communis (Additional files 5 and 6). There are four groups of mitochondrial adenine nucleotide carriers presented in the five eukaryotic clades, among which the ADP/ATP carriers are the best known. Furthermore, three mitochondrial AAC genes: AtAAC1, AtAAC2, and AtAAC3 were presented in the genome of A. thaliana. In the present study, LcAAC1 gene isolated from litchi fruit pericarp contained an ORF of 1266 bp encoding a 422 aa protein (predicted MW 46.00 kDa; pI 9.79), and exhibited 71% identity with AAC2 from A. thaliana and 75% identity to AAC from R. communis. LcAAC1 contained a highly conserved RRRMMM signature motif (Additional files 7 and 8) . On the basis of sequence similarity, domain structure and cellular function, the plant SnRK family can be divided into three subfamilies: SnRK1, SnRK2 and SnRK3. Among these subfamilies, SnRK2 is unique to plants and is involved in responses to environmental stresses . In the present study, SnRK cDNA isolated from litchi pericarp was 1394 bp in length with a 1014 bp ORF encoding a predicted polypeptide of 338 aa (MW 38.00 kDa; pI 5.36). The deduced amino acid sequence showed homology with counterpart SnRK2 family members, namely R. communis, Glycine max, Zea mays, and A. thaliana. Similar to other SnRK2 genes, the N-terminal catalytic domain was highly conserved, containing an ATP-binding site (DIGSGNFGVAKLVRDTWTDELLAVK) and protein kinase activating signature (QICHRDLKLENTLL), whereas the C-terminal regulatory domain contained an acidic ‘patch’, which is essential for kinase activity (Additional file 9). In a neighbor-joining cladogram LcSnRK2 and its counterparts OsSAPK3, ZmSnRK2.2, GmSARK2, and RcSAPK1 were clustered in subclass 2 (Additional file 10) as reported by Zhang et al. .
Expression of LcAtpB was up-regulated by 114% at 80–90 DAF (Figure 4F), which was consistent with changes in ATP concentration (Figure 3A). Increased transcript abundance of LcAOX1 peaked at 14-fold at 70 DAF but significant differences in transcript abundance at other stages of fruit development were not observed (Figure 4G). Abundance of LcUCP1 declined by 43% between 50 and 60 DAF before increasing by 363% at 70 DAF and thereafter declining rapidly by 48% at 90 DAF (Figure 4H). LcAAC1 and LcSnRK2 exhibited similar changes in transcript levels (Figure 4I and J). Transcription of these two genes increased significantly between 50 and 70 DAF before decreasing to the lowest levels between 70 and 80 DAF, but recovered partly in the final stage of maturation (80–90 DAF).
Respiration intensity fluctuated and was significantly blocked by ATP treatment (Figure 5E). The oxygen consumption rate peaked after 2 days of storage and was enhanced significantly by ATP application (Figure 5F). In the first 2 days of storage ρV alt/V t decreased significantly, then rose slowly during the medial period of storage and increased sharply by the end of storage; exogenous ATP treatment significantly slowed down these effects (Figure 5G). In the ATP-treated fruit ρValt/Vt was relatively stable throughout the storage process, which indicated that stability in ρValt/Vt played an important role in maintenance of fruit storability.
Changes in energy levels may be a key factor in switching on fruit ripening and senescence. Initiation of senescence is characterized by a reduced ATP level . In the present study, ATP, EC, and total adenylate levels and the ATP/AMP ratio increased rapidly, whereas the AMP concentration decreased gradually, during fruit development (50–80 DAF) (Figure 3). ATP is produced mainly by oxidation of carbohydrates and is regulated by transcription of key proteins catalyzing nucleotide synthesis . The respiratory rate of pericarp tissue of litchi fruit declined during fruit development, whilst the cyanide-resistant respiration to total respiration ratio (ρV alt/V t) and energy production were enhanced (Figures 2 and 3). A declining respiratory rate and enhanced alternative respiration is an overflow protection mechanism to prevent ROS production .
We isolated total RNAs from the root, stem, leaf, pericarp, aril and seed from fruit 80 DAF for real-time RT-PCR analysis. Five gene transcripts were predominantly expressed in the root, stem, and leaf at this specific developmental stage and increased significantly in developmental stage I except for LcUCP1, which was predominantly expressed only in the root, and LcAtpB, which was up-regulated by the end of developmental stage II. The LcAtpB expression level rose continuously during fruit development and maturation consistent with the increased ATP concentration (Figure 3A). Transcript abundance of LcAOX1, LcUCP1, LcAAC1, and LcSnRK2 reached peaks at 70 DAF in the rank order LcAOX1 > LcSnRK2 > LcAAC1 > LcUCP1 (Figure 4F–J). Their transcription peaks coincided with the peak in ADP concentration (Figure 3D). Transcription peaks of AOX1b and UCP were also observed in mango fruit at the turning stage , which indicated that the color-breaker stage may be a key turning point in fruit energy metabolism. LcAOX1, LcUCP1, LcAAC1 and LcSnRK2 transcript levels significantly decreased after 70 DAF, allowing a surge in ATP synthesis. Under the combined effects of AtpB, AOX1, UCP1, AAC1, and SnRK2, the EC level remained relatively constant during fruit development and ripening. LcAOX1 expression spiked at 70 DAF, which probably reduced levels of ROS generated by high rates of fruit photosynthesis. Recent studies indicate that the SnRK2 family plays key roles in stress responses, such as hyperosmotic stress and abscisic acid (ABA) signaling. ABA is involved in the maturation of many climacteric and non-climacteric fruit. Commonly, endogenous ABA concentrations increase during fruit maturation, whereas exogenous application of ABA accelerates fruit ripening or maturation . Sugar signaling also regulates senescence in a complex network with other signals  resulting from biotic or abiotic stress. SnRK2 has been implicated in stress and ABA-mediated signaling pathways . Enhanced transcript abundance of LcSnRK2 at 70 DAF might be a consequence of increased ABA and sugar accumulation during maturation of litchi fruit .
The levels of ATP, total adenylate, and EC, and the ATP/AMP ratio significantly decreased, but the AMP concentration significantly increased, in stored harvested fruit (Figure 6). Similar results were reported for apples, pears, cut flowers, and other horticultural products [1–3]. ATP content decreased significantly in ‘Jonagold’ apples during controlled atmosphere storage. Brown heart disease of ‘Conference’ pears during storage was partly attributed to energy depletion . As the petals of carnation cut flowers wilt (a senescence symptom), ATP synthesis decreases [1, 2]. ATP concentrations were negatively correlated with the browning index and proportion of edible fruit, which was consistent with our previous work . ATP and EC levels in horticultural crops can be maintained by appropriate handling, thus effectively postponing senescence. Application of ethylene inhibitors, high oxygen, anaerobic conditions, exogenous sucrose, or ATP can maintain the ATP content and EC levels in litchi, longan, tulip and carnation flowers, thereby delaying pericarp browning and extending flowering longevity [1, 3, 4, 7, 10]. In the present study, the decrease in ATP, total adenylate, and EC levels and the ATP/AMP ratio were significantly slowed by exogenous ATP. At the same time, the disease and browning indices were lower and commodity rates were higher in ATP-treated fruit (Figure 5, Additional file 11). These data support the hypothesis that intracellular energy depletion or ATP deficiency induce senescence of horticultural crops. Respiratory intensity decreased in harvested litchi fruit but increased rapidly during 2–4 days in storage. Respiration rates might be activated by ADP accumulated during the first 2 days of fruit storage (Figure 6B). Energy levels, oxygen consumption rate, and ρV alt/V t were increased by exogenous ATP treatment. A reduced respiration rate is apparently favorable to extend the storage life of litchi fruit. The highly flexible nature of respiratory metabolism in plants allows survival under varying and often stressful environmental and nutritional conditions . In the present study, exogenous ATP treatment extended the storage life of litchi fruit by reducing respiration intensity, which differed from the results of Klotz et al.  that suggested adenylates did not limit respiration in stored sugarbeet roots.
LcAtpB transcription was up-regulated in harvested litchi fruit, which might restrain the rapid decline in ATP level during 0–4 days of storage. Exogenous ATP treatment inhibited LcAtpB expression, possibly by feedback inhibition by ATP. Chivasa  reported a new role for AtpB as a pro-cell death protein. In the current study, in preharvest (80–90 DAF) or postharvest (0–2 days in storage) litchi fruit, the rapid increase of LcAtpB expression followed by rapid senescence and decline in fruit quality suggested that the increase in LcAtpB transcript level marked initiation of the aging process. During abiotic stresses, such as heat shock and hypoxia, AMPKs (SnRK analogues) switch off ATP-consuming processes and activate ATP-generating catabolic pathways through direct enzyme regulation and transcriptional regulation . SnRK may sense ATP deficiency and trigger downstream transcription of AtpB. In the present study, the transcript abundance of LcAtpB was significantly increased in postharvest litchi fruit, which might be regulated by LcSnRK2.
This study investigated energy characteristics and the level of five transcripts that are involved in the modulation of ATP levels during preharvest development and postharvest senescence of litchi fruit. Measurements were also recorded after ATP treatment of harvested fruits. Senescence and deterioration of litchi fruit during storage was closely related to ATP content and EC levels. Dramatic increase in LcAtpB expression either in planta or postharvest suggested that LcAtpB expression may mark the onset of senescence of litchi fruit. The energy status in litchi fruit seems to be controlled by the combined effects of genes responsible for energy production, dissipation, transfer and regulation, in which LcAtpB and LcAOX1 play a decisive role. Given their importance for energy regulation, these genes could be used to generate markers for the breeding of new litchi cultivars that show increased ATP production or fruit with a longer postharvest storage life.
Seven 10-year-old litchi (Litchi chinensis Sonn. cv. Huaizhi) trees were chosen for these experiments from a commercial orchard in Guangzhou, China, in 2010 and 2011. Ten fruit on each tree were used to measure fruit diameter every 10 days over a period of 40 days beginning at 50 DAF (31 May, 2010; 2 June, 2011) and ending at 90 DAF (11 July, 2010; 12 July, 2011). At each stage, fruit were randomly sampled from different parts of the canopy between 9 and 11 a.m. Fresh weights of individual fruit were recorded, and respiration rate was measured using fresh fruit. The pericarp was collected, frozen in liquid nitrogen and stored at -80°C for RNA extraction. All trees used in the above studies were maintained in accordance with commercial litchi production practices.
Litchi fruit at about 80% maturity (80–82 DAF) were collected from the same orchard. In order to examine the expression pattern of energy-related gene transcripts, a variety of tissues (root, stem, leaf, pericarp, aril and seed) were collected at this maturation stage in 2012. Fruit were selected for uniformity of shape, color, and absence of blemishes or disease. The fruits were surface sterilized in 0.5% sodium hypochlorite solution for 5 s before rinsing twice in sterile distilled water, and then placed in 4.2 L glass desiccators containing sterile distilled water or 1.0 mmol L-1 ATP solution (200 fruit L-1). A partial vacuum at 75 kPa was gently applied for 3 min. The vacuum-infiltrated fruits were air-dried, then packed in 0.015-mm-thick polyethylene bags (15 pieces/bag, 12 bags/treatment), and stored at 25°C and 85–90% relative humidity. Thirty fruit from different treatments were randomly sampled every 2 days after storage. Pericarp browning and disease incidence were recorded, and respiration rate and alternative oxidative respiratory activity were tested using fresh fruit.
The pericarp color was measured with a Konica Minolta CR-400 Chroma Meter (Minolta, Japan) using 15 fruits immediately after picking. The meter was calibrated using the manufacturer’s standard white plate. Color was recorded in the L*, C* and h color space, and these values were measured randomly at the site opposite to the fruit suture (Table 1). L* indicated the lightness ranging from black (L* = 0) to white (L* = 100), chroma values (C*) represented the color saturation that varied from dull to vivid, lower and higher values, respectively, and hue angle (h*) referred to a color wheel, with red at an angle of 0°, yellow at 90°, green at 180°, and blue at 270°, in accordance with the method of McGuire .
Extraction of ATP, ADP and AMP were conducted with the method of Ozogul et al.  with a minor modification. Litchi fruit pericarp (2 g) was ground in liquid nitrogen and extracted with 6 mL of 0.6 mol L-1 perchloric acid. The homogenate was centrifuged at 4°C and 16,000 × g for 10 min. The supernatant (3 mL) was quickly neutralized to pH 6.5–6.8 using 1 mol L-1 potassium hydroxide. The solution was diluted to 4 mL and passed through a 0.45 μm filter. The filtrate was then stored at -30°C until analyzed. ATP, ADP and AMP measurements were performed in accordance with the method of Liu et al.  with a Waters 2695 HPLC (Waters, Inc., Milford, MA, USA) using a Pinnacle ll-C18 column (4.6 × 250 mm) and an ultraviolet (UV) detector at 254 nm. Mobile phase A consisted of 0.06 mol L-1 dipotassium hydrogen phosphate and 0.04 mol L-1 potassium dihydrogen phosphate dissolved in deionized water and adjusted to pH 7.0 with 0.1 mol L-1 potassium hydroxide. Mobile phase B was 100% acetonitrile. Elution was carried out by a linear gradient program with 75–100% A and 0–25% B for 7 min. The flow rate was 1.2 mL min-1. Sample aliquots of 5 μL were injected into the HPLC. ATP, ADP and AMP concentrations were calculated according to the external standard curve and expressed on a fresh weight (FW) basis. Energy charge was calculated as: ([ATP] + 0.5 × [ADP])/([ATP] + [ADP] + [AMP]).
Pericarp browning was assessed after 0, 2, 4 and 6 days of storage using the following scale by an experienced technician : 0, no browning (excellent quality); I, slight browning; II, <1/4 browning; III, 1/4 to 1/2 browning; and IV, >1/2 browning (poor quality). The browning index was calculated as: Σ (browning scale × percentage of corresponding fruit within each class).
Disease development was monitored using 30 randomly selected fruits and recorded as the ratio of mean proportion (%) of the fruit surface with fungal growth to the total number of fruit. The rating scale described above for browning was also applied for disease severity . The proportion of edible fruit was calculated as the percentage of the total number of fruit exhibiting a browning scale value ≤1.
Membrane permeability was expressed as the relative electrolyte leakage. Fifteen discs (10 mm in diameter) from 30 fruits were washed three times in deionized water, dried with filter paper and then incubated for 30 min in 20 mL of 0.3 mol L-1 mannitol solution at 25°C. Initial electrolyte leakage rates were determined using a conductivity meter (model DDS-11A; Shanghai Scientific Instruments, Shanghai, China). Total electrolyte leakage was then determined after boiling for 20 min and cooling rapidly to 25°C. The relative leakage was expressed as the percentage of the initial electrolyte to the total electrolyte.
The respiratory activities of intact litchi fruit were measured by infrared gas analysis. A sample of 10 litchi fruit was weighed before being sealed in an easy-lock food container (2.2 L in volume) at 25°C, through which CO2-free air was pumped. Increases in CO2 concentration in the container were monitored for 5 min by passing the air stream through an infrared gas analyzer (Li-6262 CO2/H2O analyzer, LI-COR, Inc., Lincoln, NE, USA). Respiration rate was expressed in terms of carbon dioxide output per hour per kilogram of fresh weight (mg CO2 kg-1 h-1). The analysis was carried out using five replicate fruit samples.
The oxygen consumption rate of fruit pericarp was measured using a Clark-type Oxygraph oxygen electrode (Hansatech Instruments, King’s Lynn, UK) at 25°C by the method of Bingham and Farrar  with some modifications. Ten pericarp discs (5 mm in diameter) were dipped in distilled water before surface drying with filter paper. Potassium phosphate buffer (65 mmol L-1, pH 6.8) was used as the respiratory medium. The total respiration rate (V t) was determined in the absence of any respiratory inhibitors. The alternative respiratory pathway (ARP) rate (ρV alt) was calculated from V t minus the respiration rate in the presence of 25 mmol L-1 benzohydroxamate. The contribution of ARP to total respiration was expressed as ρV alt/V t. The oxygen consumption rate was expressed as nmol min-1 cm-2.
Total RNA was extracted from litchi fruit pericarp using the hot borate method of Won and Wilkins , with some modifications as described by Kuang et al. . Extracted RNA was digested with DNase I (TaKaRa Bio, Inc., Otsu, Shiga, Japan) and then confirmed by PCR to be free of detectable amounts of DNA. RNA concentration was determined by measuring UV absorbance at 260 nm with a SmartSpec Plus spectrophotometer (Bio-Rad, Hercules, CA, USA). The integrity and quality of the RNA were verified by agarose gel electrophoresis and calculation of the A 260 /A 280 ratio.
To isolate fragments of energy-related genes, degenerated oligonucleotides were constructed based on conserved sequence elements of AtpB, AOX, UCP, AAC, and SnRK from other species (Table 2). Using these oligonucleotides, amplification of fragments was carried out on a mixture of cDNA obtained from pericarp of litchi fruit stored for 0 and 2 days, because the genes displayed high expression at these stages in a preliminary analysis. The product of the first-strand cDNA was subjected to PCR amplification. The PCR product of the predicted size was purified, cloned into the pMD20-T vector (TaKaRa), and then transformed into E. coli DH5a cells (TaKaRa) in accordance with the manufacturer’s protocol. Plasmid DNA isolated from positive E. coli cells was digested with EcoRI and HindIII. The inserted DNA was sequenced by Sangon Biotech Co., Shanghai, China. The DNA fragments were extended by 3′- and 5′-RACE using the 3′-Full RACE Core Set version 2.0 and the 5′ RACE Kit (TaKaRa) with gene-specific primers. PCR products from 3′- or 5′-RACE were sequenced, and their sequence information was used to design gene-specific primers for full-length amplification. Full-length clones from each gene were sequenced in double strands. The resulting sequences were designated as LcAtpB, LcAOX1, LcUCP1, LcAAC1, and LcSnRK2, and were deposited in GenBank (http://www.ncbi.nlm.nih.gov) with the accession numbers JQ349005 (LcAtpB), JQ349006 (LcAOX1), JQ349009 (LcUCP1), JQ349007 (LcAAC1), and JQ349008 (LcSnRK2) (Table 3).
Primers used for real-time PCR analysis
Forward primer (5’ to 3’)
Reverse primer (5’ to 3’)
Identification of nucleotide sequences from RT-PCR clones was carried out with the NCBI BLAST program (http://www.ncbi.nlm.nih.gov/BLAST). Alignments were made using Clustal X and Jalview software, and cladograms were constructed by the neighbor-joining method using the MEGA program and visualized with TreeView software. The theoretical isoelectric point (pI) and mass value for mature peptides were calculated using the PeptideMass program (http://us.expasy.org/tools/peptidemass.html).
The experiment was repeated twice during two consecutive growing seasons. Experiments for respiration intensity were repeated six times, whereas other experiments were repeated three times. The results are presented as the mean values ± standard deviation. Significant differences among different developmental stages, different storage times, and those between ATP-treated fruit and the untreated control were determined by one-way analysis of variance (ANOVA) using SPSS® version 13.0 (SPSS, Inc., Chicago, IL, USA). Statistical differences were assessed with a significance level of 5%. Graphs were drawn with SigmaPlot 9.0 and Excel 2003.
This study was supported by the National Basic Research Program of China (973 program; No. 2013CB127100), National Key Technologies R&D Program (Grant No. 2012BAD38B03), National Natural Science Foundation of China (Grant Nos. 30972039 and 31272216), Guangdong Natural Science Foundation (Grant No. S2011020001156), and the Fund of Key Laboratory of Plant Resources Conservation and Sustainable Utilization, South China Botanical Garden, Chinese Academy of Sciences.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.