Physiologia Plantarum 154: 210–222. 2015
© 2014 Scandinavian Plant Physiology Society, ISSN 0031-9317
Functional, histological and biomechanical characterization of wheat water-mutant leaves Agata Rascioa,* , Nicoletta Rasciob , Michele Rinaldia and Massimiliano Valentinic a
Consiglio per la Ricerca e la Sperimentazione in Agricoltura, Centro di Ricerca per la Cerealicoltura S.S. 673 Km 25,200, 71122 Foggia, Italy Dipartimento di Biologia, Università degli Studi di Padova, viale Colombo 3, 35121 Padova, Italy c Consiglio per la Ricerca e la Sperimentazione in Agricoltura – Centro di Ricerca per la Nutrizione e gli Alimenti, Via Ardeatina 546, 00178 Rome, Italy b
Correspondence *Corresponding author, e-mail: [email protected] Received 26 May 2014; revised 22 July 2014 doi:10.1111/ppl.12280
A wheat (Triticum turgidum subsp. durum) mutant, generated with sodium azide from wild-type (WT) cv. ‘Trinakria’, differs in its water affinity of dry leaves, and was designated as a water-mutant. Compared with the WT, water-mutant leaves have lower rates of water uptake, while stomatal and cuticular transpiration do not differ. The nuclear magnetic resonance proton signals used for image reconstruction of leaf cross sections showed differences between these genotypes for the T1 proton spin–density and the T2 proton spin–spin relaxation time. Structural and histochemical analyses at midrib level showed that the water-mutant has thinner leaves, with more and smaller cells per unit area of mesophyll and sclerenchyma, and has altered staining patterns of lignin and pectin-like substances. Stress–strain curves to examine the rheological properties of the leaves showed a biphasic trend, which reveals that the tensile strength at break load and the elastic modulus of the second phase of the water-mutant are significantly higher than for the WT. These data support the proposal of interrelationships among local biophysical properties of the leaf, the microscopic water structure, the rheological properties and the water flux rate across the leaf. This water-mutant can be used for analysis of the genetic basis of these differences, and for identification of gene(s) that govern these traits.
Introduction The water in plant tissues has biophysical properties that are different from those of pure water. It can be ‘free’ or ‘bound’, as seen from analyses of its thermodynamic properties (Vertucci and Leopold 1987, Rascio et al. 1992), from the intensities of the sample-intrinsic nuclear magnetic resonance (NMR) parameters (Iwaya-Inoue et al. 2004, Sardans et al. 2010), and from the relationship between tissue-water content and potential, as shown by pressure–volume (PV) curves (Lamont and Lamont 2000). Leaves are chemically heterogeneous matrices that are composed of molecules that have very different affinities for water: lignin is hydrophobic, mucilage is
hygroscopic and pectins and hemicelluloses have gelling properties. Together with the solutes and the cellular compartments, the numbers and types of charged groups on these macromolecular surfaces affect the biophysical properties of the tissue water, while the tissue integrity can also be a quantitatively relevant determinant for the water status. Indeed, the volume of the non-osmotically active solution of fresh wheat leaves [i.e. bound water content estimated through PV curves] is linearly correlated with, and double, the total water adsorbed by the hydrophilic and hydrophobic groups of dry leaves (Rascio et al. 2005). Water properties change with the plant abiotic stress (Rascio et al. 1998, Mercado et al. 2004, Capitani et al. 2009), and these reflect the degree of drought
Abbreviations – NMR, nuclear magnetic resonance; PV, pressure–volume; WT, wild type.
210
Physiol. Plant. 154, 2015
or desiccation adaptation of the plant (Clayton-Green 1983, Hájek and Beckett 2008). However, the causes and functional consequences of these phenomena remain to be fully elucidated. The cell wall pectins have a high content of arabinose, and thus they can retain more water and the flexibility of the cell wall during dehydration (Moore et al. 2008a, 2008b). The water–polyelectrolyte interactions are responsible for the biomechanical characteristics of leaves (Köhler and Spatz 2002, Balsamo et al. 2005), and for cell growth and adhesion (Leboeuf et al. 2005, Evered et al. 2007). The diffuse electrical double layer of the negatively charged polymers of the xylem pore surfaces can reduce the water mobility, and hence the flux of water (van Doorn et al. 2011). Alternatively, the swelling and shrinking of the pectin network that is driven by its affinity for water (Zsivanovits et al. 2004) and by ions (Zimmermann 1978) regulates the hydraulic conductance of the xylem through the so-called ionic effect (Zwieniecki et al. 2007). Comparisons of ion-mediated increases in water fluxes show large variability for the intensity of this phenomenon across different species (Nardini et al. 2011), which has also been ascribed to the relevant influence of vessel length and distribution (across different plant species) on the water flux (Gascó et al. 2007). A new opportunity to analyze the interactions among the leaf biomechanical properties, microscopic features, water binding strength and macroscopic water flux across the leaf has been offered by the development of a wheat (Triticum turgidum subsp. durum) water-mutant (previously named 364), which has an altered tissue water affinity (Rascio et al. 1999). This mutant and its wild type (WT) are less genetically distant than plants of different species, so the attribution of the changes in water properties or water fluxes to non-architecture-related traits is little biased by differences in leaf anatomy. In this study, the water uptake rates of fresh leaves, the histological and histochemical characteristics and rheological properties of the leaves of the WT and the water-mutant were analyzed. NMR imaging of their proton status and distribution were also investigated through different pulse sequences and changes in their imaging parameters for their proton spin density (T1 ) signals and their proton spin–spin relaxation time (T2 ) signals.
Materials and methods Plant material The water-mutant of the durum wheat (T. turgidum subsp. durum) cv. ‘Trinakria’ was previously referred
Physiol. Plant. 154, 2015
to as the 364 mutant (Rascio et al. 1999), and it was identified by screening for ‘differences for water sorption’ (DWS) mutations in selfed M4 progenies of seeds, mutagenized using sodium azide. This genotype was chosen here from those generated with sodium azide from the WT cv. Trinakria, because it showed different properties to the others (such as 354 or 422) that were identified at the same time. In particular it had higher affinity for strongly bound water (with respect to the WT) across different environmental conditions (Rascio et al. 1999). Therefore, the water binding mode of these water-mutant plants should be constitutive. After vernalization at 4∘ C for 1 week, seeds of the WT and the non-isogenic water-mutant that were reproduced by self-pollination over 15 generations were sown in plastic pots in medium-textured soil, sand and peat (6:3:1). The soil was fertilized at pre-sowing with 18 g m−2 ammonium sulfate and mineral superphosphate. The maximum water capacity of the soil was 31% of the dry weight, and the plants were kept well watered, with water loss restored every day to about 80% soil capacity. Initial investigations were performed in a growth chamber for the NMR analysis. After completion of the NMR, the same genotypes were sown in a greenhouse, with the procedure designed to explore the biomechanical, macroscopic and microscopic properties and the water fluxes of the leaves. NMR imaging The WT and water-mutant seeds were sown in 2.6-l pots according to a completely randomized design with replicates. Ten seeds were sown for each genotype, and two pots represented a replicate. The controlled temperature, light and relative humidity cycles in the growth chamber were as previously reported (Rascio et al. 1992). Starting from the observation that the water binding mode of this water-mutant should not change with progressive dehydration, NMR analysis was performed on well-hydrated leaves. At heading stage, one fully expanded flag leaf was collected from each replicate and brought to full turgor using a 2-h incubation in distilled water at 20∘ C. The NMR imaging was carried out on two replicates, using a Bruker AVANCE 300 MHz spectrometer (Bruker Biospin, Milan, Italy), equipped with a cylindrical bird-cage single-tuned nucleus (1 H) coil probe head with an inner diameter of 70.0 mm. The proton signal was monitored and used for NMR image reconstruction. Gradient-echo (Bruker library: m-gefi-ortho) and multislice-multiecho (Bruker library: m-msme-ortho) 211
were used. Gradient-echo generates echoes by applying gradient pulses, and the signal is directly proportional to the water content; the image obtained is known as the T1 , or proton spin–density-weighted, image. These images were acquired according to the following parameters: field of view, 12.0 × 12.0 mm; matrix size, 256 × 256 pixels; spectral width, 100.0 kHz; echo time, 3.927 ms; repetition time, 60.0 ms; number of scans, 4; slice thickness, 2 mm; and excitation pulse, sinc3. For the multislice-multiecho T2 , or proton spin–spin relaxation-time-weighted, image (which produces echoes via a spin-echo-based pulse sequence), the characteristics were: field of view, 12.0 × 12.0 mm; matrix size, 256 × 256 pixels; spectral width, 100.0 kHz; echo time, 25 ms; repetition time, 4000.0 ms; number of echoes and images, 2; number of scans and dummy scans, 1; slice thickness, 2.0 mm; and excitation pulse, sinc3. The data were processed to obtain image matrices of 256 × 256 pixels in size, with a field view of 12.0 × 12.0 mm. The processing modes were FT-MODE and complex-FFT, respectively, and spike elimination was allowed. The data acquired were then processed as maximum intensity pixels, to accurately determine the signal intensity from the T1 -weighted and T2 -weighted images of the leaf profiles. Two-dimensional graphs of pixel intensities as a function of horizontal distance through standard image selection were constructed, using IMAGEJ, a public domain software (National Institutes of Health, Bethesda, MD). Leaf selection Three pots (160 l each) were sown, with 150 seeds for each genotype. After germination, the plants were thinned to 130 per pot. For the WT, the heading stage
A
B
began 126 days after sowing, and for the water-mutant, it began 132 days after sowing. To confirm that both genotypes had completed leaf expansion, two measurements of leaf area were made during heading: the first when the whole spike was visible and the flag leaf was not completely expanded (Fig. 1A), and the second at the end of heading, to collect the fully expanded flag leaf (Fig. 1B). To perform the biomechanical and microscopic analysis of samples with the same growth stage, care was taken to collect 10 flag leaves that were fully expanded. Water uptake rate Three expanded and non-expanded flag leaves were weighed immediately after excision, and then scanned to measure the leaf area. Subsequently, they were dehydrated to a fixed water potential, as the rate of water uptake is strongly related to the water deficit of the plant tissues (Tyree et al. 1981). To standardize the dehydration intensity and avoid tissue damage due to leaf pressurization, a Scholander pressure chamber was not used. Solutions of 0.4 M sucrose or mannitol provided sufficient water deficit so that the patterns of water uptake could be traced, with 2 h required to reach osmotic equilibrium (Fig. 2A). Sucrose was used, although it is believed to enter the cell membranes, because based on our preliminary experiments, the water uptake was not affected by the type of osmoticum used. Following this dehydration, the leaves were thoroughly dried (i.e. patted with paper towels) and weighed. On the basis of several preliminary experiments, the leaf rehydration was carried out in a thermostatic bath at 20∘ C, in the dark and without stirring of the medium, as this does not affect the leaf permeability (Graziani and Livne 1971). The rehydration curves of the
WT
water–mutant
Fig. 1. Left: Developmental stage of the flag leaf collection. (A) Visible head with flag leaf not fully expanded. (B) Heading completed with flag leaf fully expanded. Right: Morphology of the WT and water-mutant cv. Trinakria plants (as indicated), at the third-leaf stage.
212
Physiol. Plant. 154, 2015
B
0.0
15.0 12.5
-0.5
Water uptake (% fresh weight)
Water loss (% fresh weight)
A
-1.0 -1.5
10.0 7.5 5.0 2.5
-2.0 0
0.0
50 100 150 200
0
50
100 150 200
Time (min) Fig. 2. (A) Representative dehydration kinetics of whole leaves of the WT (solid line) and the water-mutant (broken line) for changes in weight with time after immersion in 0.4 M sucrose. (B) Representative rehydration kinetics of whole leaves of the WT (solid line) and the water-mutant (broken line) for changes in weight with time after immersion in distilled water. Temperature, 20∘ C.
Stomatal and cuticular transpiration To analyze the stomatal and cuticular transpiration, fully expanded leaves were cut early in the morning and rehydrated for 1.5 h in distilled water in darkness at 20∘ C. They were then blotted dry, and their saturated mass was measured. The leaves were scanned to determine their surface areas. Isotherm water-loss curves were constructed according to the water loss with time, at 20∘ C. The leaves were put on an electronic balance and maintained at the prefixed temperature in a climate chamber that was closed by a transparent door. The air humidity within the balance was controlled using silica gel. Irradiance was provided by fluorescent lamps, at 40 μmol PAR photons m−2 s−1 (400–700 nm), for stomatal opening. The water loss was determined by recording the weight
Physiol. Plant. 154, 2015
wild type Water flow (mg H2O cm–1)
detached leaves (Fig. 2B) had the biphasic kinetics that is widespread among vascular plants (Tyree et al. 1981, Sack et al. 2004, Zwieniecki et al. 2007). The linear phase during which the water uptake was at its maximum rate lasted about 75 min, and full turgor was attained within 100 min. To estimate the maximum uptake rate, the whole leaf weight increase was determined after 30 min of incubation in deionized water. To ensure that no water remained on the leaf surfaces and to quantify any residual water, a pre-weighed thin towel was pressed onto the leaf surfaces and then re-weighed. No further appreciable traces of humidity were observed here. The dry leaf weights were determined gravimetrically by heating the leaves in an oven at 90∘ C for 3 days. The water flux rate is expressed as percentages of the fresh weight increase per minute. The leaf-specific mass was calculated as the ratio of the dry weight to the leaf area.
water–mutant
5
5
4
4
3
3
2
2
1
1
0
0
0
25
50
75
0
25
50
75
Time (min) Fig. 3. Dehydration curves for the WT and water-mutant leaves (as indicated) at 20∘ C. The curvilinear interpolation of the data points from three different curves (distinguished by different symbols) is shown.
every 5–10 min, over 60 min. The difference in the fresh mass between successive measurements was attributed to the amount of water lost over that time period. At the end of the experiments, the excised leaves were oven-dried. From the slopes for the first and the final portions of the water loss curves (Fig. 3), the rates of the stomatal water flow and the cuticular water flow were estimated, respectively. Tensile test The tensile strength was determined using a digital instrument (DINF-ONE, DAS, Rome, Italy). The expanded flag leaves were rinsed for 2 h in de-ionized water before the stretch test. Fully hydrated standard leaf strips (4.0 × 0.3 cm) were obtained using a double-edged blade. The strips were cut starting at 2 cm above the 213
basal part of the leaves, and following parallel to the primary veins. The specimens were clamped using two clips separated by 3.0 cm. The rate of stretching was 5 mm min−1 . The force and displacement were recorded at intervals of 1 s, and the biphasic force–displacement curves were constructed. The inflexion slope and the minimal slope were calculated from the angular coefficients of the regression lines before and after the ‘hump’, respectively. The samples were stretched to the break point. Maximum elongation and leaf tensile strength to failure were determined, i.e. the total force necessary to achieve catastrophic failure. Micromorphometric analysis For each individual sample, 1-cm wide transverse sections (1 cm above the basal limit of the leaf) were obtained and fixed in formalin/acetic acid: 50% ethyl alcohol, 10% formaldehyde, 5% acetic acid and 35% distilled water. For long-term preservation, the fixed samples were kept in 3:1 (v/v) aqueous acetic alcohol solution. Leaf sections were cut with a hand microtome. The sections were stained with phloroglucinol [1 g phloroglucine in 50 ml ethanol and 25 ml of 36% (v/v) hydrochloric acid for 5 min]. After staining, at least three transverse sections from each leaf sample were examined under 40× or 100× magnification, using a Leitz microscope (Dialux20, Oberkochen, Germany). A digital camera (Coolpix 5200; Nikon, Tokyo, Japan) was used to obtain images. Micrographs were subjected to morphometric analysis of the epidermis, mesophyll and vascular area using IMAGEJ. The following were measured at the leaf midrib: leaf cross-sectional thickness, area of mestome sheath, phloem area, number and area of vessels, thickness of abaxial and adaxial sclerenchyma, mean cell area of adaxial sclerenchyma and adjacent mesophyll, number of cells of adaxial sclerenchyma, number of cells per unit area of adaxial sclerenchyma and epidermis thickness. Light microscopy and cytochemical tests For the light microscopy and cytochemical tests, the fixed samples were dehydrated through a graded series of ethyl alcohol and then embedded in London Resin White (London Resin Company, Woking, UK). Thin sections (2 μm) were cut with an ultramicrotome (Ultracut; Reichert-Jung, Vienna, Austria). The sections were examined under a light and epifluorescence microscope (5000B; Leica, Wetzlar, Germany) fitted with an excitation filter (BP 365/12), a chromatic bean splitter (FT 395) and a barrier filter (LP395), and 214
equipped with a digital camera (DFC425C; Leica, Wetzlar, Germany). The following different types of staining were used: Toluidine blue: thin sections were stained for 10 min with the general dye: 1% toluidine blue and 1% sodium tetraborate (in equal volumes). Ruthenium red: thin sections were stained for 30 min with an aqueous solution (1:5000) of ruthenium red, which is considered specific for pectic substances (Jensen 1962). Acidified toluidine blue: Thin sections were stained for 30 min with acidified toluidine blue (0.05% toluidine blue in 0.02 M phosphate buffer, pH 4.4). This test is diagnostic for acidic polysaccharides, such as pectic acids, which are recognizable by their characteristic red-purple staining. Lignin and some polyphenols turn this dye green (Feder and O’Brien 1968). Auramine-O: Thin sections were stained for 10 min with the fluorochrome auramine-O (0.02% auramine-O in 0.05 Tris–HCl buffer, pH 7.2) and examined under the epifluorescence microscope. The auramine-O fluorescence under UV light excitation specifically reveals lipid-like materials (cutin, lignin and suberin) (Helsop-Harrison 1977). All of the histochemical reactions were repeated several times on sections from WT and water-mutant leaves.
Results The water-mutant has little effect on the overall shape of the plants, as they showed only slightly reduced growth of the aerial parts, beginning at the second- to third-leaf stage, as compared with the WT (Fig. 1, right). The non-expanded flag leaves had a smaller area and less dry matter per unit of leaf area compared with the expanded leaves. The mean area of the water-mutant leaves was less than for the WT, although they had greater specific dry matter content (Table 1). The significance of the genotype × leaf type interaction indicates that the differences in the areas of the expanded and non-expanded leaves of the water-mutant were not significant, and hence they stop expanding before those of the WT (although they continue to accumulate more dry matter). Water uptake rate The water uptake rate after 30 min of leaf incubation at 20∘ C was significantly lower for the water-mutant leaves (Table 1). Linear regression of the leaf-specific mass per unit of leaf area vs water uptake rate showed significant and negative correlation (Fig. 4). Examination of the data point distribution around the regression line appears to
Physiol. Plant. 154, 2015
Table 1. ANOVA of leaf area, specific mass and rate of water uptake of not expanded and fully expanded leaves of WT and water-mutant, as determined at the beginning and at the end of heading, respectively. Values are means of four determinations. Significance of F (Fisher test) values is shown. ns, not significant. Genotype Leaf characteristic 2
Area (cm ) Specific mass (mg cm−2 ) Water uptake rate (mg g−1 FW min−1 )
LSM (mg cm–2)
5
P (Fisher test)
Leaf status
WT
Water-mutant
Genotype
Leaf type
Interaction
Non-expanded Expanded Non-expanded Expanded Non-expanded Expanded
27.8 40.4 2.7 2.9 1.5 1.7
23.6 23.5 3.7 4.2 0.5 0.8
© 2014 Scandinavian Plant Physiology Society, ISSN 0031-9317
Functional, histological and biomechanical characterization of wheat water-mutant leaves Agata Rascioa,* , Nicoletta Rasciob , Michele Rinaldia and Massimiliano Valentinic a
Consiglio per la Ricerca e la Sperimentazione in Agricoltura, Centro di Ricerca per la Cerealicoltura S.S. 673 Km 25,200, 71122 Foggia, Italy Dipartimento di Biologia, Università degli Studi di Padova, viale Colombo 3, 35121 Padova, Italy c Consiglio per la Ricerca e la Sperimentazione in Agricoltura – Centro di Ricerca per la Nutrizione e gli Alimenti, Via Ardeatina 546, 00178 Rome, Italy b
Correspondence *Corresponding author, e-mail: [email protected] Received 26 May 2014; revised 22 July 2014 doi:10.1111/ppl.12280
A wheat (Triticum turgidum subsp. durum) mutant, generated with sodium azide from wild-type (WT) cv. ‘Trinakria’, differs in its water affinity of dry leaves, and was designated as a water-mutant. Compared with the WT, water-mutant leaves have lower rates of water uptake, while stomatal and cuticular transpiration do not differ. The nuclear magnetic resonance proton signals used for image reconstruction of leaf cross sections showed differences between these genotypes for the T1 proton spin–density and the T2 proton spin–spin relaxation time. Structural and histochemical analyses at midrib level showed that the water-mutant has thinner leaves, with more and smaller cells per unit area of mesophyll and sclerenchyma, and has altered staining patterns of lignin and pectin-like substances. Stress–strain curves to examine the rheological properties of the leaves showed a biphasic trend, which reveals that the tensile strength at break load and the elastic modulus of the second phase of the water-mutant are significantly higher than for the WT. These data support the proposal of interrelationships among local biophysical properties of the leaf, the microscopic water structure, the rheological properties and the water flux rate across the leaf. This water-mutant can be used for analysis of the genetic basis of these differences, and for identification of gene(s) that govern these traits.
Introduction The water in plant tissues has biophysical properties that are different from those of pure water. It can be ‘free’ or ‘bound’, as seen from analyses of its thermodynamic properties (Vertucci and Leopold 1987, Rascio et al. 1992), from the intensities of the sample-intrinsic nuclear magnetic resonance (NMR) parameters (Iwaya-Inoue et al. 2004, Sardans et al. 2010), and from the relationship between tissue-water content and potential, as shown by pressure–volume (PV) curves (Lamont and Lamont 2000). Leaves are chemically heterogeneous matrices that are composed of molecules that have very different affinities for water: lignin is hydrophobic, mucilage is
hygroscopic and pectins and hemicelluloses have gelling properties. Together with the solutes and the cellular compartments, the numbers and types of charged groups on these macromolecular surfaces affect the biophysical properties of the tissue water, while the tissue integrity can also be a quantitatively relevant determinant for the water status. Indeed, the volume of the non-osmotically active solution of fresh wheat leaves [i.e. bound water content estimated through PV curves] is linearly correlated with, and double, the total water adsorbed by the hydrophilic and hydrophobic groups of dry leaves (Rascio et al. 2005). Water properties change with the plant abiotic stress (Rascio et al. 1998, Mercado et al. 2004, Capitani et al. 2009), and these reflect the degree of drought
Abbreviations – NMR, nuclear magnetic resonance; PV, pressure–volume; WT, wild type.
210
Physiol. Plant. 154, 2015
or desiccation adaptation of the plant (Clayton-Green 1983, Hájek and Beckett 2008). However, the causes and functional consequences of these phenomena remain to be fully elucidated. The cell wall pectins have a high content of arabinose, and thus they can retain more water and the flexibility of the cell wall during dehydration (Moore et al. 2008a, 2008b). The water–polyelectrolyte interactions are responsible for the biomechanical characteristics of leaves (Köhler and Spatz 2002, Balsamo et al. 2005), and for cell growth and adhesion (Leboeuf et al. 2005, Evered et al. 2007). The diffuse electrical double layer of the negatively charged polymers of the xylem pore surfaces can reduce the water mobility, and hence the flux of water (van Doorn et al. 2011). Alternatively, the swelling and shrinking of the pectin network that is driven by its affinity for water (Zsivanovits et al. 2004) and by ions (Zimmermann 1978) regulates the hydraulic conductance of the xylem through the so-called ionic effect (Zwieniecki et al. 2007). Comparisons of ion-mediated increases in water fluxes show large variability for the intensity of this phenomenon across different species (Nardini et al. 2011), which has also been ascribed to the relevant influence of vessel length and distribution (across different plant species) on the water flux (Gascó et al. 2007). A new opportunity to analyze the interactions among the leaf biomechanical properties, microscopic features, water binding strength and macroscopic water flux across the leaf has been offered by the development of a wheat (Triticum turgidum subsp. durum) water-mutant (previously named 364), which has an altered tissue water affinity (Rascio et al. 1999). This mutant and its wild type (WT) are less genetically distant than plants of different species, so the attribution of the changes in water properties or water fluxes to non-architecture-related traits is little biased by differences in leaf anatomy. In this study, the water uptake rates of fresh leaves, the histological and histochemical characteristics and rheological properties of the leaves of the WT and the water-mutant were analyzed. NMR imaging of their proton status and distribution were also investigated through different pulse sequences and changes in their imaging parameters for their proton spin density (T1 ) signals and their proton spin–spin relaxation time (T2 ) signals.
Materials and methods Plant material The water-mutant of the durum wheat (T. turgidum subsp. durum) cv. ‘Trinakria’ was previously referred
Physiol. Plant. 154, 2015
to as the 364 mutant (Rascio et al. 1999), and it was identified by screening for ‘differences for water sorption’ (DWS) mutations in selfed M4 progenies of seeds, mutagenized using sodium azide. This genotype was chosen here from those generated with sodium azide from the WT cv. Trinakria, because it showed different properties to the others (such as 354 or 422) that were identified at the same time. In particular it had higher affinity for strongly bound water (with respect to the WT) across different environmental conditions (Rascio et al. 1999). Therefore, the water binding mode of these water-mutant plants should be constitutive. After vernalization at 4∘ C for 1 week, seeds of the WT and the non-isogenic water-mutant that were reproduced by self-pollination over 15 generations were sown in plastic pots in medium-textured soil, sand and peat (6:3:1). The soil was fertilized at pre-sowing with 18 g m−2 ammonium sulfate and mineral superphosphate. The maximum water capacity of the soil was 31% of the dry weight, and the plants were kept well watered, with water loss restored every day to about 80% soil capacity. Initial investigations were performed in a growth chamber for the NMR analysis. After completion of the NMR, the same genotypes were sown in a greenhouse, with the procedure designed to explore the biomechanical, macroscopic and microscopic properties and the water fluxes of the leaves. NMR imaging The WT and water-mutant seeds were sown in 2.6-l pots according to a completely randomized design with replicates. Ten seeds were sown for each genotype, and two pots represented a replicate. The controlled temperature, light and relative humidity cycles in the growth chamber were as previously reported (Rascio et al. 1992). Starting from the observation that the water binding mode of this water-mutant should not change with progressive dehydration, NMR analysis was performed on well-hydrated leaves. At heading stage, one fully expanded flag leaf was collected from each replicate and brought to full turgor using a 2-h incubation in distilled water at 20∘ C. The NMR imaging was carried out on two replicates, using a Bruker AVANCE 300 MHz spectrometer (Bruker Biospin, Milan, Italy), equipped with a cylindrical bird-cage single-tuned nucleus (1 H) coil probe head with an inner diameter of 70.0 mm. The proton signal was monitored and used for NMR image reconstruction. Gradient-echo (Bruker library: m-gefi-ortho) and multislice-multiecho (Bruker library: m-msme-ortho) 211
were used. Gradient-echo generates echoes by applying gradient pulses, and the signal is directly proportional to the water content; the image obtained is known as the T1 , or proton spin–density-weighted, image. These images were acquired according to the following parameters: field of view, 12.0 × 12.0 mm; matrix size, 256 × 256 pixels; spectral width, 100.0 kHz; echo time, 3.927 ms; repetition time, 60.0 ms; number of scans, 4; slice thickness, 2 mm; and excitation pulse, sinc3. For the multislice-multiecho T2 , or proton spin–spin relaxation-time-weighted, image (which produces echoes via a spin-echo-based pulse sequence), the characteristics were: field of view, 12.0 × 12.0 mm; matrix size, 256 × 256 pixels; spectral width, 100.0 kHz; echo time, 25 ms; repetition time, 4000.0 ms; number of echoes and images, 2; number of scans and dummy scans, 1; slice thickness, 2.0 mm; and excitation pulse, sinc3. The data were processed to obtain image matrices of 256 × 256 pixels in size, with a field view of 12.0 × 12.0 mm. The processing modes were FT-MODE and complex-FFT, respectively, and spike elimination was allowed. The data acquired were then processed as maximum intensity pixels, to accurately determine the signal intensity from the T1 -weighted and T2 -weighted images of the leaf profiles. Two-dimensional graphs of pixel intensities as a function of horizontal distance through standard image selection were constructed, using IMAGEJ, a public domain software (National Institutes of Health, Bethesda, MD). Leaf selection Three pots (160 l each) were sown, with 150 seeds for each genotype. After germination, the plants were thinned to 130 per pot. For the WT, the heading stage
A
B
began 126 days after sowing, and for the water-mutant, it began 132 days after sowing. To confirm that both genotypes had completed leaf expansion, two measurements of leaf area were made during heading: the first when the whole spike was visible and the flag leaf was not completely expanded (Fig. 1A), and the second at the end of heading, to collect the fully expanded flag leaf (Fig. 1B). To perform the biomechanical and microscopic analysis of samples with the same growth stage, care was taken to collect 10 flag leaves that were fully expanded. Water uptake rate Three expanded and non-expanded flag leaves were weighed immediately after excision, and then scanned to measure the leaf area. Subsequently, they were dehydrated to a fixed water potential, as the rate of water uptake is strongly related to the water deficit of the plant tissues (Tyree et al. 1981). To standardize the dehydration intensity and avoid tissue damage due to leaf pressurization, a Scholander pressure chamber was not used. Solutions of 0.4 M sucrose or mannitol provided sufficient water deficit so that the patterns of water uptake could be traced, with 2 h required to reach osmotic equilibrium (Fig. 2A). Sucrose was used, although it is believed to enter the cell membranes, because based on our preliminary experiments, the water uptake was not affected by the type of osmoticum used. Following this dehydration, the leaves were thoroughly dried (i.e. patted with paper towels) and weighed. On the basis of several preliminary experiments, the leaf rehydration was carried out in a thermostatic bath at 20∘ C, in the dark and without stirring of the medium, as this does not affect the leaf permeability (Graziani and Livne 1971). The rehydration curves of the
WT
water–mutant
Fig. 1. Left: Developmental stage of the flag leaf collection. (A) Visible head with flag leaf not fully expanded. (B) Heading completed with flag leaf fully expanded. Right: Morphology of the WT and water-mutant cv. Trinakria plants (as indicated), at the third-leaf stage.
212
Physiol. Plant. 154, 2015
B
0.0
15.0 12.5
-0.5
Water uptake (% fresh weight)
Water loss (% fresh weight)
A
-1.0 -1.5
10.0 7.5 5.0 2.5
-2.0 0
0.0
50 100 150 200
0
50
100 150 200
Time (min) Fig. 2. (A) Representative dehydration kinetics of whole leaves of the WT (solid line) and the water-mutant (broken line) for changes in weight with time after immersion in 0.4 M sucrose. (B) Representative rehydration kinetics of whole leaves of the WT (solid line) and the water-mutant (broken line) for changes in weight with time after immersion in distilled water. Temperature, 20∘ C.
Stomatal and cuticular transpiration To analyze the stomatal and cuticular transpiration, fully expanded leaves were cut early in the morning and rehydrated for 1.5 h in distilled water in darkness at 20∘ C. They were then blotted dry, and their saturated mass was measured. The leaves were scanned to determine their surface areas. Isotherm water-loss curves were constructed according to the water loss with time, at 20∘ C. The leaves were put on an electronic balance and maintained at the prefixed temperature in a climate chamber that was closed by a transparent door. The air humidity within the balance was controlled using silica gel. Irradiance was provided by fluorescent lamps, at 40 μmol PAR photons m−2 s−1 (400–700 nm), for stomatal opening. The water loss was determined by recording the weight
Physiol. Plant. 154, 2015
wild type Water flow (mg H2O cm–1)
detached leaves (Fig. 2B) had the biphasic kinetics that is widespread among vascular plants (Tyree et al. 1981, Sack et al. 2004, Zwieniecki et al. 2007). The linear phase during which the water uptake was at its maximum rate lasted about 75 min, and full turgor was attained within 100 min. To estimate the maximum uptake rate, the whole leaf weight increase was determined after 30 min of incubation in deionized water. To ensure that no water remained on the leaf surfaces and to quantify any residual water, a pre-weighed thin towel was pressed onto the leaf surfaces and then re-weighed. No further appreciable traces of humidity were observed here. The dry leaf weights were determined gravimetrically by heating the leaves in an oven at 90∘ C for 3 days. The water flux rate is expressed as percentages of the fresh weight increase per minute. The leaf-specific mass was calculated as the ratio of the dry weight to the leaf area.
water–mutant
5
5
4
4
3
3
2
2
1
1
0
0
0
25
50
75
0
25
50
75
Time (min) Fig. 3. Dehydration curves for the WT and water-mutant leaves (as indicated) at 20∘ C. The curvilinear interpolation of the data points from three different curves (distinguished by different symbols) is shown.
every 5–10 min, over 60 min. The difference in the fresh mass between successive measurements was attributed to the amount of water lost over that time period. At the end of the experiments, the excised leaves were oven-dried. From the slopes for the first and the final portions of the water loss curves (Fig. 3), the rates of the stomatal water flow and the cuticular water flow were estimated, respectively. Tensile test The tensile strength was determined using a digital instrument (DINF-ONE, DAS, Rome, Italy). The expanded flag leaves were rinsed for 2 h in de-ionized water before the stretch test. Fully hydrated standard leaf strips (4.0 × 0.3 cm) were obtained using a double-edged blade. The strips were cut starting at 2 cm above the 213
basal part of the leaves, and following parallel to the primary veins. The specimens were clamped using two clips separated by 3.0 cm. The rate of stretching was 5 mm min−1 . The force and displacement were recorded at intervals of 1 s, and the biphasic force–displacement curves were constructed. The inflexion slope and the minimal slope were calculated from the angular coefficients of the regression lines before and after the ‘hump’, respectively. The samples were stretched to the break point. Maximum elongation and leaf tensile strength to failure were determined, i.e. the total force necessary to achieve catastrophic failure. Micromorphometric analysis For each individual sample, 1-cm wide transverse sections (1 cm above the basal limit of the leaf) were obtained and fixed in formalin/acetic acid: 50% ethyl alcohol, 10% formaldehyde, 5% acetic acid and 35% distilled water. For long-term preservation, the fixed samples were kept in 3:1 (v/v) aqueous acetic alcohol solution. Leaf sections were cut with a hand microtome. The sections were stained with phloroglucinol [1 g phloroglucine in 50 ml ethanol and 25 ml of 36% (v/v) hydrochloric acid for 5 min]. After staining, at least three transverse sections from each leaf sample were examined under 40× or 100× magnification, using a Leitz microscope (Dialux20, Oberkochen, Germany). A digital camera (Coolpix 5200; Nikon, Tokyo, Japan) was used to obtain images. Micrographs were subjected to morphometric analysis of the epidermis, mesophyll and vascular area using IMAGEJ. The following were measured at the leaf midrib: leaf cross-sectional thickness, area of mestome sheath, phloem area, number and area of vessels, thickness of abaxial and adaxial sclerenchyma, mean cell area of adaxial sclerenchyma and adjacent mesophyll, number of cells of adaxial sclerenchyma, number of cells per unit area of adaxial sclerenchyma and epidermis thickness. Light microscopy and cytochemical tests For the light microscopy and cytochemical tests, the fixed samples were dehydrated through a graded series of ethyl alcohol and then embedded in London Resin White (London Resin Company, Woking, UK). Thin sections (2 μm) were cut with an ultramicrotome (Ultracut; Reichert-Jung, Vienna, Austria). The sections were examined under a light and epifluorescence microscope (5000B; Leica, Wetzlar, Germany) fitted with an excitation filter (BP 365/12), a chromatic bean splitter (FT 395) and a barrier filter (LP395), and 214
equipped with a digital camera (DFC425C; Leica, Wetzlar, Germany). The following different types of staining were used: Toluidine blue: thin sections were stained for 10 min with the general dye: 1% toluidine blue and 1% sodium tetraborate (in equal volumes). Ruthenium red: thin sections were stained for 30 min with an aqueous solution (1:5000) of ruthenium red, which is considered specific for pectic substances (Jensen 1962). Acidified toluidine blue: Thin sections were stained for 30 min with acidified toluidine blue (0.05% toluidine blue in 0.02 M phosphate buffer, pH 4.4). This test is diagnostic for acidic polysaccharides, such as pectic acids, which are recognizable by their characteristic red-purple staining. Lignin and some polyphenols turn this dye green (Feder and O’Brien 1968). Auramine-O: Thin sections were stained for 10 min with the fluorochrome auramine-O (0.02% auramine-O in 0.05 Tris–HCl buffer, pH 7.2) and examined under the epifluorescence microscope. The auramine-O fluorescence under UV light excitation specifically reveals lipid-like materials (cutin, lignin and suberin) (Helsop-Harrison 1977). All of the histochemical reactions were repeated several times on sections from WT and water-mutant leaves.
Results The water-mutant has little effect on the overall shape of the plants, as they showed only slightly reduced growth of the aerial parts, beginning at the second- to third-leaf stage, as compared with the WT (Fig. 1, right). The non-expanded flag leaves had a smaller area and less dry matter per unit of leaf area compared with the expanded leaves. The mean area of the water-mutant leaves was less than for the WT, although they had greater specific dry matter content (Table 1). The significance of the genotype × leaf type interaction indicates that the differences in the areas of the expanded and non-expanded leaves of the water-mutant were not significant, and hence they stop expanding before those of the WT (although they continue to accumulate more dry matter). Water uptake rate The water uptake rate after 30 min of leaf incubation at 20∘ C was significantly lower for the water-mutant leaves (Table 1). Linear regression of the leaf-specific mass per unit of leaf area vs water uptake rate showed significant and negative correlation (Fig. 4). Examination of the data point distribution around the regression line appears to
Physiol. Plant. 154, 2015
Table 1. ANOVA of leaf area, specific mass and rate of water uptake of not expanded and fully expanded leaves of WT and water-mutant, as determined at the beginning and at the end of heading, respectively. Values are means of four determinations. Significance of F (Fisher test) values is shown. ns, not significant. Genotype Leaf characteristic 2
Area (cm ) Specific mass (mg cm−2 ) Water uptake rate (mg g−1 FW min−1 )
LSM (mg cm–2)
5
P (Fisher test)
Leaf status
WT
Water-mutant
Genotype
Leaf type
Interaction
Non-expanded Expanded Non-expanded Expanded Non-expanded Expanded
27.8 40.4 2.7 2.9 1.5 1.7
23.6 23.5 3.7 4.2 0.5 0.8
