Metabolic Contributors to Drought-enhanced Accumulation of Sugars and Acids in Oranges

The nature of sink strength in orange fruit and changes occurring during drought stress were investigated. Potted trees of ‘Hamlin’ orange [Citrus sinensis (L.) Osbeck] grafted on Troyer citrange [Citrus sinensis x Poncirus trifoliata (L.) Raf.] were irrigated using a microsprinkler system creating either well-watered or water-stressed conditions, as determined by stem water potentials. Fruit were harvested every other week from trees of both wellwatered and drought-stressed treatments during the final stage of fruit development when sugars accumulate rapidly. Fruit quality indices and activities of sucrose synthase (SuSy), invertase, sucrose-P synthase, sucrose-P phosphatase, VATPase, and V-PPase were measured. Acids and soluble sugar concentrations were elevated in drought-stressed fruit, whereas juice pH decreased in those same fruit. Results indicate that increased sink strength in fruit from stressed trees was accompanied by an increase in SuSy activity and lowered juice pH. The remaining enzymes examined in this experiment showed no changes in activity between control and treated fruit, as was the case for plasmalemma and tonoplast sucrose carriers. Based on the present data, we conclude that SuSy and vacuolar pH are the predominant factors controlling photoassimilate accumulation in orange fruit under enhanced sink conditions brought about by imposition of a mild drought stress. phosphatase (SPP, EC 3.1.3.24), although their exact functions in a sink organ are uncertain (Wendler et al., 1990). Membranebound sucrose carriers at the tonoplast and the plasmalemma may also be important in sink strength, though often their roles are overlooked in sink studies. An additional mechanism exists for sucrose cleavage in highly acidic fruit. For example, the vacuoles of most citrus (Citrus L. sp.) juice cells are highly acidic, with pHs of 3 or lower (Echeverria and Burns, 1989; Sinclair, 1984). Such a low pH permits nonenzymatic acid hydrolysis of sucrose into glucose and fructose in vitro (Wienen and Shallenberger, 1988) and in vivo (Echeverria and Burns, 1989). Hydronium-mediated cleavage is dependent on temperature, pH, sucrose concentration, and water content (Schoebel et al., 1969). In plants, vacuolar pH is determined by the activities of the tonoplast-bound proton-pumping enzymes such as V-ATPase (EC 3.6.1.34) and V-PPase (EC 3.6.1.1). In addition to inducing sucrose hydrolysis, lower vacuolar pH also boosts the efficiency of the sucrose carrier at the tonoplast to sequester sucrose into the vacuole. During mild drought stress, fruit of several species (including citrus) accumulate higher levels of soluble carbohydrates (Ebel et al., 1993; Yakushiji et al., 1996) without affecting fruit yield. Therefore, controlled drought stress provides an experimental tool to better understand the factors controlling fruit sink strength. Understanding these critical determinants of sink strength could aid in future breeding and molecular work with fruit crops while, more importantly, allowing better understanding of the mechanisms at work in sink strength and stress physiology of plants. The objective of this study was to identify factors involved in increased soluble carbohydrate accumulation in orange (Citrus sinensis) fruit in response to drought stress. Materials and Methods PLANT MATERIAL. Uniform 3-year-old potted ‘Hamlin’ orange trees grafted on ‘Troyer’ citrange (Citrus sinensis x Poncirus trifoliata) in 26-L pots were purchased in June 1999 from a local nursery. Trees were cultivated outdoors at the University of Movement of assimilated carbon into a particular plant organ is determined by its sink strength, and by the capacity of the plant for photoassimilate production. Sink strength refers to the ability of a particular organ to attract photoassimilates (Ho, 1988). Most plant organs are sinks at some time during development, especially during tissue elongation and expansion, when fixed carbon provides the growing tissue with energy for metabolism and solutes to regulate osmotic potential. In addition, many sinks accumulate carbohydrates, as in the case of storage tissues. Fixed carbon is transported through the phloem mostly in the form of the disaccharide sucrose. For adequate sink strength, cells must cleave sucrose (Ho et al., 1991; Sung et al., 1989) or effectively sequester it from the symplast and into the vacuole (Bennett et al., 1986). Therefore, sink strength is determined by the ability of a sink to metabolize or to store sucrose. Metabolic conversion and sequestration of sucrose in plant cells can be accomplished through a variety of enzymatic and cellular mechanisms. Two enzymes (invertase, EC 3.2.1.26 and sucrose synthase, SuSy, EC 2.4.1.13) can catalyze the conversion of sucrose to hexoses or to other metabolic intermediates. Invertase catalyzes the irreversible hydrolysis of sucrose to glucose and fructose, whereas SuSy catalyzes the reversible interconversion of sucrose and uridine-5'-diphosphate (UDP) to uridine-5'diphosphoglucose (UDPG) and fructose. Cleaving sucrose allows the sink to amplify the existing gradient of sugar between the sink and phloem, allowing for continuation of sucrose movement toward the sink cell. In some tissues, sucrose-synthesizing enzymes may also play a part in a cycle of sucrose breakdown and resynthesis that controls the cytosolic sucrose concentration (Wendler et al., 1990). Therefore, related sucrose-synthesizing enzymes that may be involved in sink strength include sucrosephosphate synthase (SPS, EC 2.4.1.14) and sucrose-phosphate Florida, Citrus Research and Education Center, Lake Alfred. Pots were placed on small boards positioned atop black landscaping cloth to prevent weed growth. Each tree was selected for the presence of 10 to 15 fruit and equivalent canopy size. Pest and disease problems were managed throughout the season as they appeared. A systemic pesticide (AdMire 2F, 1-[(6-chloro-3pyridinyl) methyl]-N-nitro-2-imidazolidinimine; Bayer Co., Kansas City, Mo.) was applied to the soil about once a month and the trees were sprayed periodically with oil and soapy water for mite control. The trees were fertilized every 2 weeks with a 20N– 8.8P–16.6K water soluble fertilizer (Peters 20–20–20; Spectrum Brands, St. Louis, Mo.). All trees were trimmed and pruned during summer and early fall to develop similar leaf to fruit ratios and fruit sizes before experimentation. TREATMENTS. As a preliminary trial in late August through early September 1999, five trees were set aside and monitored for drought stress according to specific volumes of water applied each day. Water volumes ranged between 200 and 1000 mL. Stem water potential of these five trees was measured every 2 d using a laboratory assembled Scholander-type pressure chamber. From water potential data and environmental conditions, we found that 1 L·d constituted well-watered plants under all conditions, while 500 mL or less imposed drought stress. Irrigation to droughtstressed trees was modified according to environmental fluctuations such as humid or rainy weather. Fifty trees were randomly assigned to two groups: control (1 L·d) and drought-stress (≈500 mL·d) treatments. A microsprinkler system was designed and output was monitored frequently to ensure appropriate water distribution to each tree in both treatments. On occasions, when rainfall was forecasted, plastic bags were placed under the tree pots, then slid up and tied around the trunks to prevent moisture from entering the pots of stressed trees. Holes were cut into the bottom of the bags as a preventive measure should leaking into the bag occur while also limiting buildup of condensation. This was adequate to prevent significant alteration of the soil water status of drought-stressed trees. The experiment began 25 Oct. 1999 and continued until 24 Dec. 1999, with five trees remaining stressed through January 2000. Water status of treatment and control trees was monitored by measurement of stem water potential every 3 to 7 d using the same pressure bomb described previously. Two leaves from each of five trees were selected for similar solar angle of incidence and canopy position, wrapped in black plastic shortly before sunrise, and then covered with a foil wrap to prevent solar heating. Several hours were allowed for equilibration of the leaves with stem xylem potential before measurement in the pressure chamber. FRUIT ANALYSIS. Fruit were analyzed every other week beginning in mid-October through the end of December 1999, with one sample in mid-January 2000. Ten fruit were used per treatment per sampling point, divided into two replications of five fruit each. Five fruit were randomly selected for a replicate sampling, one from each of five trees, on the morning data were to be taken. Harvest occurred over 4 d, taking fruit from drought stress or control trees on alternate mornings. Fruit for one sampling point were harvested from the same five control trees, then the same five drought-stressed trees, such that each tree contributed two fruit. After sampling, those trees were excluded from the experiment with the consideration that the sink to source ratio might be altered by fruit picking. Collected fruit were taken immediately from the field to the lab and used for fruit quality analyses and enzyme assays. Each fruit was weighed and cross-sectioned at the equator. One half of the fruit was juiced at a time on a hand-held juicer, and 15 mL of this juice was added immediately to 30 mL of a cold stirring buffer containing 0.5 M 3-(N-morpholino)propanesulfonic acid (MOPS, pH 8.0), 1.5% w/v polyvinylpyrrolidone (PVP-40), 250 mM sucrose, 7.5 mM ethylenediaminetetraacetic acid (EDTA), 0.1% w/v bovine serum albumin (BSA), and 14 mM mercaptoethanol. An additional 15 mL of juice was placed in a large centrifuge tube. The remaining half of the fruit was juiced and 15 mL of juice added to 45 mL of stirring, buffered extract. The remaining juice was treated as with the other half, added to the same test tube. The buffer/juice mixture from each fruit (60 mL total) was then filtered through a cheesecloth/nylon mesh into a cold beaker on ice. This procedure was repeated for all five fruit, mixing the buffer/juice solutions together for a total of 300 mL and stirring well after addition of 1 mL of phenylmethylsulfonyl fluoride (PMSF) solution (100 mM PMSF in 95% ethanol, and 5% 2-propanol). The buffered juice extract (pH ≈ 7.2) was used subsequently for enzyme assays and isolation of cell compartment membranes as described below. Enzyme assays were conducted using subsamples of buffered extract remaining from the membrane isolation procedure (see below). Buffered juice was centrifuged at 77,000 gn for 30 min at 4 °C. After centrifugation, 2.5 mL of the supernatant solution were desalted through a Sephadex PD-10 column preequilibrated with a buffer containing 10 mM 4(-2-hydroxyethyl)-1piperazeneethanesulfonic acid ( HEPES, pH 7.0), 2 mM dithiothreitol (DTT), and 1 mM MgCl2. Three milliliters were collected for enzyme assays. Juice pH and titratable acids were measured individually for each of the five unbuffered fruit-juice extracts. Juice pH (15 mL) was measured using a pH meter (model HI-9219; Hannah Instruments Inc., Woonsocket, R.I.) and titrated to an end point pH of 8.0 using standard alkali solution (0.3125 N NaOH). Total soluble carbohydrate concentration was measured using an ABBE refractometer (American Optical Corp., Buffalo, N.Y.) at 20 °C after compensation for organic acids. In citrus fruit, ≈90% of the soluble solids is the sum of sugars and organic acids. MEMBRANE ISOLATION. Isolation and purification of ‘Hamlin’ vesicle membranes was conducted using the protocol described by Echeverria et al. (1997), with a few modifications. The cold juice/buffer solution was adjusted to pH 7.0 with 3.0 N KOH and 250 mL was immediately centrifuged at 113,000 gn for 1.75 h at 4 °C. The pellet containing the microsomal fraction was resuspended in buffer containing 50 mM HEPES (pH 7.6), 100 mM KCl, 150 mM sucrose and 2 mM DTT, and centrifuged at 134,000 gn for 1 h. The resulting pellet was washed and resuspended again into 3 mL of this same buffer. The microsomal fraction was loaded on a sucrose gradient (8%, 17%, 26%, and 38% sucrose) and centrifuged for 45 min at 110,000 gn. Membrane fractions were removed from each sucrose interface into separate tubes and diluted with a buffer solution of 10 mM HEPES (pH 7.6), 20 mM KCl, and 2 mM DTT. These vesicle samples were centrifuged at 113,000 gn for 1 h, resuspended into a storage buffer of 10 mM 1,3bis[tris(hydroxymethyl)methylamino]propane (BTP)/2-(4morpholino)ethanesulfonic acid (MES, pH 7.5), 250 mM sorbitol, 2 mM DTT, and stored at –80 °C. Protein concentration in vesicles and juice extract was determined using the Coomassie Blue method described by Bradford (1976) and BSA as standard. SUCROSE SYNTHASE. Sucrose synthase was assayed in the synthetic direction in a solution containing 100 mM HEPES (pH 7.2), 2 mM UDPG, 10 mM fructose, 2 mM MgCl2, and 100 μL protein extract in a final volume of 500 μL. Aliquots of 100 μL were taken at 0, 20, 40, and 60 min, mixed with 100 μL of 30% KOH and boiled for 10 min. Sucrose produced was determined following the anthrone method of van Handel (1968). Rates of sucrose breakdown were calculated using the ratio of synthetic/ breakdown activity estimated previously for acid lime SuSy of 8:1 (Echeverria, 1992) and corroborated for other Citrus cultivars (unpublished data). SUCROSE PHOSPHATE SYNTHASE. The reaction mixture contained 100 mM HEPES (pH 7.5), 7.5 mM UDPG, 7.5 mM fructose6-phosphate (F-6-P), 37.5 mM glucose-6-phosphate (G-6-P), 18 mM MgCl2, 1.0 mM EDTA, and 100 μL enzyme extract in a final volume of 500 μL. SPS reaction was terminated and analyzed for sucrose and sucrose-6-P (S-6-P) as for sucrose synthase. SUCROSE PHOSPHATE PHOSPHATASE. Activity of SPP was assayed in a solution containing 100 mM MES (pH 6.5), 10 mM MgCl2, 2 mM S-6-P, and 100 μL enzyme extract in a total volume of 500 μL. Aliquots of 50 μL were taken at specific times and mixed with 250 μL of 7.2% SDS to stop the reaction. Released Pi was determined at 850 nm by the method of Chifflett et al. (1988). INVERTASES. Both acid and alkaline invertases were assayed using either BTP/MES (pH 7.5) or sodium acetate (pH 5.0) buffer. In a total reaction volume of 2.5 mL, 0.5 mL enzyme extract, 100 mM buffer, and 100 mM sucrose were combined and incubated. Aliquots of 500 μL were placed into 1 mL of phosphate buffer (100 mM, pH 7.0), and boiled for 2 min. Glucose was determined at 450 nm by the glucose oxidase method of Kilburn and Taylor (1969). V-ATPase. V-ATPase activity was determined using 75 μL tonoplast vesicles, 50 mM BTP/MES (pH 7.5), 250 mM sorbitol, 50 mM KCl, BSA at 0.4 mg·mL, 2 mM DTT, 10 μM gramicidin, 3 mM ATP, and 3 mM MgSO4 adjusted to a total volume of 500 μL. Aliquots of 50 μL were removed from the assay at determined times and product determined as for SPP. V-PPase. V-PPase activity was assayed as for V-ATPase except that ATP and MgSO4 were replaced with 1 mM NaPPi and 1 mM MgSO4. Assays for Pi were conducted as with ATPase and SPP. SUCROSE UPTAKE EXPERIMENTS. Assays for sucrose uptake into either tonoplast or plasmalemma vesicles contained the following: 50 μg tonoplast or plasmalemma vesicle, 50 mM BTP/MES at pH 7.5 or 5.5, 250 mM sorbitol, BSA at 0.4 mg·mL, 2 mM DTT, and 2 mM C-sucrose (5.18 × 10 Bq·μmol) as described by Echeverria et al. (1997). Control experiments contained 100 μM KCl and 10 μM nigericin. Assays were run for 30 min with aliquots (100 μL) taken at 0, 10, 20, and 30 min. Aliquots were removed at the indicated times and washed on prerinsed cellulose nitrate filters (pore size 0.22 μm, 25 mm in diameter, Whatman Intl. LTD., Maidstone, United Kingdom). After vacuum (625 mm Hg) was applied, the vesicles were washed with 5 mL storage buffer (Echeverria et al., 1997). Radioactivity retained in the vesicles was determined by scintillation spectroscopy after placing the filter discs in 5 mL of ScintiVerse BD SX 18-4 (Fisher Scientific, Pittsburgh, Pa.). Control vesicle samples with nigericin accumulated low amounts of labeled sucrose which corresponded to the values calculated using vesicle void volume (Echeverria et al., 1997). STATISTICAL ANALYSIS. Leaf and fruit juice data were subjected to analyses of variance procedures. Student’s t test (paired) was employed for determining statistical significance with enzyme data when differences in means were recorded. All data were calculated during linear stages of catalytic activity. Data points for enzyme activities were done in triplicate for each experimental sample.