Rootstock influence on iron uptake responses in Citrus leaves and their regulation under the Fe paradox effect

Plant material and growth conditions

Our experiment was run in an experimental orchard in the town of Moncada (Valencia) Spain, 38°14′53.57″N; 0°41′46.9″W), which is a subarid region in the Turia River upper basin. The annual average values of the study site are 17.3 °C, 390.1 mm precipitation and 65.8% air humidity (IVIA). Sandy loam soil was the classification for the soil texture within the first 50 cm depth. pH 8.5, 1.63% organic matter (w/w), 40% total calcium carbonate (w/w) and 8% deactive calcium carbonate were the main chemical soil characteristics. A 5–10% elative level of carbonate is considered high.

Measurements were obtained from the leaves of 18-year-old ‘Navelina’ orange trees, Citrus sinensis (L.) Osb. (selection Iniasel 7), which has been grafted onto two citrus rootstocks taken from Cleopatra mandarin × Poncirus trifoliata (L.) Raf. mother plants (J. Forner, IVIA) in as citrus rootstocks breeding programme to assess the new citrus rootstocks’ behaviour when faced with several abiotic disorders. The selected material displayed distinct Fe-chlorosis symptoms: 030146 (non-chlorotic) and 030122 (chlorotic).

In 1993, trees (2.5 × 6 m) were planted with nursery plants (12 months post-grafting). They were arranged six trees per rootstock in a completely randomised experimental plot. Another row of guard trees separated the experimental blocks from one another. Standard cultural practices included fertilising and drip irrigation, mechanical inter-row weed control and inter-tree chemical control. Drip irrigation frequency was modified to seasons, and ranged from once weekly (winter) to five days/week (summer), with 40 L tree⁻¹ per irrigation. Electrical conductivity was 1.4 dS m⁻¹, Cl⁻ 134.8 mg L⁻¹ and NO₃⁻, and water pH was 7.6. Determinations were recorded in spring flush fully expanded leaves taken from the non-fruiting shoots collected in October 2014.

Leaf growth measurements

Random sampling was done using 10 leaves per tree from each tree’s canopy periphery. Leaves were rinsed under a tap, and also with distilled water that contained a non-ionic detergent, and were finally rinsed three times with distilled water. An LI-300 area-meter (LI-COR, Lincoln, NE, USA) was used to measure total leaf area with and a forced draft oven was employed to determine biomass after drying for 48 h at 70 °C to achieve constant dry weight (DW, in g). Measurements are presented as the mean ± SE of the six replicates per scion:rootstock combination. The average value of six technical replicates was taken as being representative of each individual plant.

Chlorophyll concentration and gas exchange parameters

The leaf chlorophyll (Chl) concentration was spectrophotometrically measured (Lambda 25; PerkinElmer, Shelton, CT, USA), as described by Moran & Porath (1980). Leaf discs (Ø = 7 mm; no major veins) were acquired with a cork borer from spring leaves. The samples with 0.06 g of fresh weight (four leaves gave 12 discs) were incubated in 6 mL of n,n-dimethylformamide for 24 h at 4 °C, and were then centrifuged at 6,000g and 4 °C for 15 min. The supernatant was left for 60 min in the presence of Na₂SO₄ and absorbance was measured at 664 and 647 nm (Lambda 25; PerkinElmer, Shelton, CT, USA). The Chl concentration was estimated non-destructively in a SPAD portable apparatus (Minolta Co., Osaka, Japan), and the method was previously calibrated according to Abadía & Abadía (1993).

A portable chlorophyll fluorometer (PAM-2100; Heinz Walz Gmbh, Germany) was employed to record the chlorophyll fluorescence parameters (F_o: minimal fluorescence, F_m: maximal fluorescence and F_v = F_m−F_o: variable fluorescence). After a 30 min dark adaptation period, a saturating pulse of 2,100 μmol quanta m⁻² s⁻¹ for 5 s leaves was used for illuminating purposes, and F_o and F_m measurements were taken.

The photosynthetic activity (net CO₂ assimilation rate, A_CO2) of the single attached leaves was measured between 10:00 and 11:30 h on a sunny day so that measurements could be taken when conditions were relatively stable. Photosynthetically active radiation on was adjusted the leaf surface to a photon flux density of 1,000 μmol m⁻² s⁻¹. Closed gas exchange (CIRAS-2; PP-systems, Hitchin, UK) was used to take measurements. Leaf laminae were completely enclosed in a PLC 6 (U) universal leaf autocuvette inside a closed circuit model, and were left at 25 ± 0.5 °C, with a leaf-to-air vapour deficit of approximately 1.7 Pa. Through the cuvette, the air flow rate was 500–1,500 mL min⁻¹. Three-second intervals were used to take 10 consecutive measurements.

Measurements on spring fully expanded leaves were taken and are the mean ± SE of the six replicates per scion:rootstock combination. The average value of three technical replicates (four leaves each) for the Chl data was considered to be representative of each individual plant. The average value of six leaves for the F_v/F_m and A_CO2 parameters was taken as being representative of each individual plant.

Catalase and peroxidase activities

Catalase (CAT) and peroxidase (POX) enzyme activities were determined according to Forner-Giner et al. (2010). A Polytron 3100 (Kinematica, Switzerland) was used homogenise fresh sample (2 g) in 10 mL of 10 mM phosphate buffer (pH 6.5) and 2.5% (w/v) insoluble polyvinylpolypyrrolidone (PVPP). The crude extract was then centrifuged for 30 min at 12,000 rpm and 4 °C. The supernatant was collected and utilised to determine soluble POX and CAT activities. Enzyme activities were analysed by measuring the reduction in hydrogen peroxide (H₂O₂), and were spectrophotometrically determined (Lambda 25; PerkinElmer, Shelton, CT, USA) and were taken as fresh weight (U g⁻¹ FW). The molar extinction coefficient was 43.6 M⁻¹ cm⁻¹.

Adding 10 mM H₂O₂ commenced CAT activity (final volume of 2 mL) in a reaction mixture with 50 mM of phosphate buffer (pH 7.0) and 100 μL of supernatant. Reaction monitoring was done for 5 min at 240 nm and room temperature. POX activity (final volume of 2 mL) was established in a reaction mixture. This mixture comprised 50 mM of glycine buffer (pH 4.5), 0.8 mM of H₂O₂, 50 mM of ABTS [2,2-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid)] and 100 μL of enzyme extract. Any change in absorbance was monitored for 5 min at 590 nm and room temperature.

Measurements are presented as the mean ± SE of the six replicates per scion:rootstock combination. The average value of three technical replicates was taken as representing each individual plant.

Cell fluid isolation

Fresh leaves were harvested by using a razor blade to cut petioles at the leaf lamina base. After xylem sap removal, apoplasmic fluid was collected as the methodology described by Dannel, Pfeffer & Marschner (1995) sets out, but with a few modifications. These procedures were followed at 4 °C to minimise apoplasmic fluid composition changes. Leaves (about 10 g of FW) were placed so that all the cut lay in the same direction, were tightened into a roll with plastic foil, placed in a 10 mL syringe and were centrifuged. Eliminating the fluid collected at an initial 1,500g force avoided cytosolic contamination. Apoplasmic fluid was collected after the subsequent centrifugal force for 15 min at 4,500g and 4 °C (Eppendorf Centrifuge 5810R; Eppendorf AG, Hamburg, Germany) and by filtering through a 0.20 μm polyvinyl fluoride (PVF) syringe filters (OlimPeak; Teknokroma, Sant Cugat del Vallés, Spain). Apoplasmic fluid was divided into two subsamples; one was divided into 0.5 mL aliquots, frozen in liquid N₂ and stored at −80 °C until used; the other was employed to take pH measurements (Jenway 3520 Stone; Jenway, Staffordshire, UK). The three apoplasmic fluids combination was considered to represent each individual plant.

Cell sap from centrifuged leaves was collected by freezing in liquid N₂ to rupture cells, storing at −18 °C for approximately 24 h and using a hydraulic press to press leaves after they had been thawed. Cell sap was divided into 1 mL aliquots, frozen in liquid N₂ and stored at −80 °C until used. The three cell sap extractions combination was considered representative of each individual plant.

Total ion extractions

Thirty leaves per tree canopy periphery were randomly sampled to be independently pooled and analysed. Leaves were firstly rinsed with distilled de-ionised water. Then they were frozen in liquid N₂, freeze-dried (Telstar LyoAlfa 6; Terrassa, Barcelona, Spain) and crushed in a hammer-mill. Finally, they were stored at room temperature and in the dark until further analyses were run.

The total Fe, K and Ca concentrations in whole leaves were analysed from 0.5 g DW of the leaf tissue burnt for 12 h in a muffle furnace at 550 °C, extracted with 2% nitric acid (Hiperpur Panreac, Fe <1 ppb) in an ultrasonic bath for 30 min at 40 °C (Fungilab S.A., Sant Feliu de Llobregat, Barcelona, Spain) and diluted to 25 mL with ultrapure water. The total Fe concentration in the apoplasmic and cell sap fluids was analysed from 0.5 mL of the extract, which was freeze-dried and treated as formerly described for DW material. Final preparations were stored at 4 °C until used. An AAS model Perkin Elmer Analyst (Waltham, Millipore, Molsheim, France, MA, USA) was used to measure ion concentrations. Ultra-pure water (Ultra Pure Water Systems Milli Q Plus; Millipore, Billerica, MA, USA) was employed in all the analyses. Measurements are presented as the mean ± SE of the six replicates per scion:rootstock combination.

Ferrous analysis

The ferrous concentration in both apoplasmic fluid and cell sap was established by the o-phenanthroline extraction method (Katyal & Sharma, 1980, but with some modifications; Manzanares, Lucena & Garate, 1990; Nikolic & Kastori, 2000; Başar, 2003). Fluid extracts (0.5 mL) were placed inside 10 mL pyrex tubes and Fe²⁺ ions were extracted using 2.5 mL of 1.5% o-phenanthroline in HCl-buffer (pH 3.0). Samples were gently homogenised by vortexing and were left to stand at room temperature for 16 h. Extracts were centrifuged for 5 min at 1,000 rpm and Whatman No. 1 filter paper was used to filter the supernatant. The filtrate’s ferrous (Fe²⁺) content was established by spectrophotometry at the 510 nm wavelength (Lambda 25; PerkinElmer, Shelton, CT, USA). Measurements are presented as the mean ± SE of the six replicates per scion:rootstock combination.

RNA extraction and real-time RT-PCR analysis

Five leaves were randomly sampled per tree canopy periphery and were then washed. Major veins were eliminated and ground in a mortar in N₂. Total leaf tissue RNA (0.1 g) extraction was performed with the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). RNA samples were treated with RNase-free DNase (Qiagen) by column purification, following the manufacturer’s instructions, to remove genomic DNA. RNA quality (OD₂₆₀/OD₂₈₀ ratio) and concentration were evaluated in an ND-1000 full spectrum UV–Vis spectrophotometer (Nanodrop Technologies, Thermo Fisher Scientific, Wilmington, DE, USA). A quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) was run in a LightCycler 2.0 Instrument (Roche, Mannheim, Germany), equipped with version 4.0 of the Light Cycler Software. Reactions contained the following: 2.5 units of MultiScribe Reverse Transcriptase (Applied Biosystems; Roche Molecular Systems, Branchburg, NJ, USA); 1 unit of RNase Inhibitor (Applied Biosystems); 2 μL of LC Fast Start DNA Master PLUS SYBR Green I (Roche Diagnostics GmbH, Mannheim, Germany); 25 ng of total RNA; 0.250 μM of the specific forward and reverse primers of each gene in a total 10 μL volume. Incubations lasted 30 min at 48 °C, followed by, 95 °C for 10 min, and then by 45 cycles at 95 °C for 2 s, 58 °C for 8 s and 72 °C for 8 s. The fluorescent intensity data were obtained in the 72 °C extension step. They were transformed into relative mRNA values with a 10-fold dilution series of an RNA sample taken as the standard curve. The relative mRNA levels were normalised to total RNA amounts, as formerly described (Bustin, 2002; Hashimoto, Beadles-Bohling & Wiren, 2004) and the non-chlorotic sample values were arbitrarily assigned an expression value of 1. The reference gene was actin (Yan et al., 2012). The amplification reactions’ specificity was evaluated by post-amplification dissociation curves and by reaction product sequencing. The resolutions of the curve expressions were confirmed.

A homology search was run with the related genes from the ‘Haploid Clementine Genome, International Citrus Genome Consortium (Goodstein et al., 2012; http://www.phytozome.net/clementine)’ to identify putative genes. Synthetic oligonucleotides were designed to amplify the gene from selected clones, and were sequenced for confirmation as previously mentioned. Table 1 provides details of the forward and reverse primers used. Measurements are the mean ± SE of the six replicates per scion:rootstock combination. At least three RT-PCR reactions were performed with three technical replicates per sample.

Fe chelate reductase (FC-R) activity

Ferric chelate reductase activity was established by measuring the formation of Fe²⁺ and the BPDS (bathophenanthroline-disulfonic acid disodium salt hydrate) complex from FeEDTA (De la Guardia & Alcántara, 1996). Leaf discs without major veins (Ø = 7 mm) were obtained from spring leaves using a cork borer. The samples that contained 0.1 g FW (20 discs from five leaves) were placed inside 10 mL glass test tubes and were washed in 5 mL of 0.5 mM CaSO₄ for 30 min. Then, they were incubated in 5 mL of assay medium (0.5 mM of CaSO₄, 1 mM of KCl, 50 mM of potassium phosphate buffer, pH 7.0, 0.1 mM of FeEDTA and 0.3 mM of BPDS). Rubber stoppers were used to seal tubes, which were completely covered with aluminium foil to avoid light entering. Blanks with no leaf disks were run under the same conditions with FeEDTA and in the presence of BPDS to calculate the degree of photochemical reduction. The control tubes with the plant material, but with no Fe source, were included to evaluate the release of reducing compounds by leaf disks (Larbi et al., 2001). Incubation was run at 23 °C for 24 h in the darkness. Absorbance was determined at 535 nm by spectrophotometry (Lambda 25; PerkinElmer, Shelton, CT, USA). BPDS formed a stable water-soluble red complex with Fe²⁺, and only a weak complex with Fe³⁺. The quantity of reduced Fe was calculated by the Fe²⁺-(BPDS)₃ complex concentration with an extinction coefficient of 22.14 mM⁻¹ cm⁻¹. Finally, it was expressed as nmol Fe²⁺ reduced g⁻¹ leaf FW h⁻¹.

Measurements are the mean ± SE of the six replicates per scion:rootstock combination. The average value of the three technical replicates was taken to represent each individual plant.

Extraction and determination of LMWOAs

Total LMWOAs were extracted in accordance with Veberic et al. (2010), but with some modifications. Leaf samples (0.05 g DW) were homogenised (Kinematica Polytron PT 3100; Kinematica, Lucerne, Switzerland) at 15,000g for 30 min in 2 mL of H₂SO₄ 0.1%, and were then centrifuged for 30 min at 7,500g and 4 °C (Eppendorf Centrifuge 5810R; Eppendorf AG, Hamburg, Germany). The supernatant was collected and filtered through a 0.45 μm PVF syringe filter (OlimPeak; Teknokroma, Sant Cugat del Vallés, Spain). Final extracts were divided into 0.6 mL aliquots, frozen in liquid N₂ and stored at −80 °C until used.

The LMWOAs (malate and citrate) in both the total leaf and apoplast fluid were established by HPLC in an Alliance liquid chromatographic system (Waters, Barcelona, Spain), equipped with a 2,695 separation module coupled to a 2,996 photodiode array detector and a ZQ2000 mass detector. Data acquisition was performed with the Empower 2 software (Waters, Barcelona, Spain). Sample temperature and column temperature were 5 and 35 °C, respectively. Other values were: capillary voltage, 3.0 kV; cone voltage, 23 V; source temperature, 100 °C; desolvation temperature, 200 °C; desolvation gas flow, 400 L h^–1. Full data acquisition was done using electrospray ion negative conditions by scanning from 100 to 400 uma in the centroid mode. Chromatographic separation involved the use of an ICSep ICE-COREGEL 87H3 column (Transgenomic, Glasgow, UK), an ICSep ICE-COREGEL 87H guard kit and an automatic injector. An isocratic mobile phase of 0.008 N H₂SO₄ solution was the solvent system. The total run time at 0.6 mL min^–1 was 20 min, and the injection volume was 5 μL. Compounds were identified by comparing their retention times, UV–Vis spectra and mass spectrum data to the corresponding authentic standards. Standards were run daily with the samples utilised for validation. Concentrations were established by an external calibration curve with citric and acid malic acid, purchased from Sigma (Sigma Co., Barcelona, Spain). Solvents and water were of HPLC-MS grade.

Statistical analyses

The obtained data underwent an ANOVA. Data distributions were checked for normality and statistical analyses were done by ANOVA with version 5.1 of Statgraphics Plus (Statistical Graphics, Englewood Cliffs, NJ, USA). The mean values were compared at the 95% confidence level by the least significant differences (LSD) method.

Fe-deficiency symptoms in leaves

The fully expanded leaves of the ‘Navelina’ orange trees grafted onto rootstocks 030146 and 030122 displayed striking differences in the non-chlorotic and chlorotic Fe-chlorosis symptoms, respectively (Fig. 1), and some leaf growth parameters (Table 2). The samples that exhibited Fe-deficiency symptoms were smaller than the non-chlorotic leaves on both DW and area bases (36.6% and 26.6%, respectively).

Total Fe and Fe²⁺ concentrations

The Fe concentration in the total leaf tissue in both samples was similar when referenced to biomass (about 63 μg g⁻¹ of DW; Table 2). This parameter differed significantly between both chlorotic states in relation to leaf area. The non-chlorotic leaves had an Fe total concentration of 1.10 μg Fe cm⁻², and this value lowered by 20% when analysed in the chlorotic samples.

A similar total tissue pattern was maintained when analysing the Fe concentration in the apoplast and cell sap fluids (Fig. 2). On the area basis, both leaf types accumulated more total Fe in the leaf apoplast than in cell sap (4.1- and 3.6-fold respectively for non-chlorotic and chlorotic leaves). However, significant differences between the samples with different chlorotic symptoms were observed. The Fe concentration in the chlorotic leaves was respectively 20.3% and 10.3% lower in the apoplast and cell sap than in the non-chlorotic leaves.

Figure 2 illustrates the Fe accumulation in the active form (Fe²⁺) in the apoplast and cell sap extracts of leaf tissue (expressed on the area basis). In quantitative terms, the quantity of Fe²⁺ was substantially lower than total Fe, but marked differences between samples appeared. The non-chlorotic leaves had 2.5- and 2.8-fold higher Fe²⁺ concentrations than the chlorotic ones. The main Fe²⁺ accumulation in leaves with no Fe-deficiency symptoms was observed in the cell sap extract.

Photosynthetic parameters and antioxidant activities

Chlorotic leaves displayed marked reductions in Chl a and b concentrations (60.7% and 48.2%, respectively) versus the non-chlorotic samples (Table 3), which corresponded to a lowering of 24.8% in the Chl a/b ratio for the former. The SPAD value in the chlorotic leaves was also lower (62.1% decrease) than in green leaves. According to these results, Fe-deficiency symptoms caused major changes to the chlorophyll fluorescence parameters versus with the non-chlorotic leaves, and the maximum quantum PSII yield lowered by 18.1% (F_v/F_m; Table 3). The net photosynthetic rate (A_CO2) lowered by 16.4% in the chlorotic leaves versus the non-chlorotic ones (Table 3).

Figure 3 depicts the activity of CAT and POX enzymes in relation to the antioxidant response to Fe-chlorosis in leaves. CAT and POX enzyme activities were respectively 5.5- and 7.8-fold higher in the non-chlorotic leaves than in the chlorotic ones.

Fe uptake responses in leaves

In molecular terms, the expression level of FC-R gene FRO2 dropped by 41% in the chlorotic leaves versus the non-chlorotic ones (Fig. 4). The activity of this enzyme strongly paralleled gene behaviour and was 1.3-fold lower in the leaves with chlorosis symptoms than in those without them.

The chlorotic leaves showed increases of 31.8% and 94.2% of mRNA transcripts abundance respectively for the H⁺-ATPase HA1 and iron transporter IRT1 genes (Fig. 5). The proton extrusion between the leaf symplast and apoplast, established by the pH measurements in leaf apoplast fluid, gave values of 6.87 and 6.98 in the non-chlorotic and chlorotic leaves, respectively (Fig. 6). The results were not significantly different despite the apparently lower pH in the former. The ratio between the K⁺ and Ca²⁺ concentrations (K⁺/Ca²⁺) in whole leaves, in relation to H⁺ excretion across the plasma membrane, was 2.2-fold higher in the chlorotic leaves than in those samples that lacked Fe-chlorosis symptoms (Fig. 6).

LMWOA concentrations

Figure 7 shows the LMWOA concentration in the total leaf tissue and apoplast fraction. The major LMWOAs obtained in whole leaf homogenates were citrate and malate, and accounted for >90% of the total LMWOA pool (data not shown). In the total leaf tissue, citrate and malate concentrations were 25.7% and 124.3% higher in the non-chlorotic leaves than in the chlorotic ones. Malate was the major LMWOA in both samples, but the relation between malate and citrate was stronger in the non-chlorotic leaves than in the chlorotic ones (respectively 1.8- and 1.3-fold). Despite the accumulated quantity of LMWOA in the apoplast extract being much smaller than in the total leaf, it displayed a similar tendency to that described in the total organ tissue. Therefore, both citrate and malate concentrations were 65.0% higher in the leaves without Fe-deficiency symptoms than in the chlorotic ones.