Aquaporins are major determinants of water use efficiency of rice plants in the field.

Plant Science 227 (2014) 165–180

Contents lists available at ScienceDirect

Plant Science journal homepage: www.elsevier.com/locate/plantsci

Aquaporins are major determinants of water use efficiency of rice plants in the field Reham M. Nada, Gaber M. Abogadallah ∗ Department of Botany, Faculty of Science, Damietta University, New Damietta 34517, Egypt

a r t i c l e

i n f o

Article history: Received 17 July 2014 Received in revised form 13 August 2014 Accepted 15 August 2014 Available online 22 August 2014 Keywords: Rice Water use efficiency Relative water content Gas exchange Aquaporin expression Root clipping

a b s t r a c t This study aimed at specifying the reasons of unbalanced water relations of rice in the field at midday which results in slowing down photosynthesis and reducing water use efficiency (WUE) in japonica and indica rice under well-watered and droughted conditions. Leaf relative water content (RWC) decreased in the well-watered plants at midday in the field, but more dramatically in the droughted indica (75.6 and 71.4%) than japonica cultivars (85.5 and 80.8%). Gas exchange was measured at three points during the day (9:00, 13:00 and 17:00). Leaf internal CO2 (Ci ) was not depleted when midday stomatal depression was highest indicating that Ci was not limiting to photosynthesis. Most aquaporins were predominantly expressed in leaves suggesting higher water permeability in leaves than in roots. The expression of leaf aquaporins was further induced by drought at 9:00 without comparable responses in roots. The data suggest that aquaporin expression in the root endodermis was limiting to water uptake. Upon removal of the radial barriers to water flow in roots, transpiration increased instantly and photosynthesis increased after 4 h resulting in increasing WUE after 4 h, demonstrating that WUE in rice is largely limited by the inadequate aquaporin expression profiles in roots. © 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Rice is more sensitive to water deficit than other crop plants. In spite of the numerous studies, the physiological basis of rice sensitivity to water deficit is still poorly understood if dissected into processes rather than considered collectively [1]. Even if growing in water-saturated soil, a rice plant shows reduction in stomatal conductance (gs ), rates of transpiration (E) and photosynthesis (A) (collectively known as midday stomatal depression), which are typical symptoms of water deficit in the afternoon where transpiration demand is high on sunny days [2]. Midday stomatal depression has been reported in many species including rice and attributed mainly to reduction in stomatal conductance as a result of plant negative water status [3–5]. Recently, Zhang et al. [6] showed that midday stomatal depression is mainly linked to stem rather than to leaf water potential in sub-tropical trees. The negative water status at midday and subsequent stomatal depression slow down photosynthesis and accelerate photoinhibition which ultimately result in reducing biomass accumulation [7,8]. Midday stomatal depression in well-watered rice plants and their sensitivity to water deficit are attributed to excessive water

∗ Corresponding author. Tel.: +002 057 2403980; fax: +002 057 2403868. E-mail address: [email protected] (G.M. Abogadallah). http://dx.doi.org/10.1016/j.plantsci.2014.08.006 0168-9452/© 2014 Elsevier Ireland Ltd. All rights reserved.

loss by stomatal and non-stomatal transpiration from leaves [9–11] and inadequate water uptake from soil by roots [12–14]. However, recent evidence suggests that rice leaves are not specifically sensitive to water deficit as compared to maize leaves, based on the finding that the sensitivity of leaf elongation rates of different genotypes of rice were similar to that of maize when differences in root system were neutralized [15]. This indicates that the sensitivity of rice to water deficit and hence, the severe midday stomatal depression may be mainly due to its poor root system. According to the composite model of water flow through roots [16], the path of water from soil through roots to leaves includes two sections. The first (radial flow) is the flow of water through different layers of root cells including epidermis, exodermis, cortex and endodermis. The second (axial flow) is the flow of water inside xylem vessels upward to the leaves. Within the radial component of water flow, water could pass through the cell walls (apoplastic), through plasmodesmata (symplastic) or through cell membranes (transmembrane flow). The two latter paths cannot be separated experimentally and are collectively termed as cell-to-cell pathway. Once early metaxylem vessels mature, the axial path of water flow will not be limiting to water movement [17–19]. Rice roots have been found to have high hydraulic resistance compared to maize roots apparently because rice roots have apoplastic barriers to radial water flow as shown by excessive suberin deposition in different cell layers of the root and the

166

R.M. Nada, G.M. Abogadallah / Plant Science 227 (2014) 165–180

well-developed endodermis [13,20]. More specifically, Ranathunge et al. [21] showed that the hydraulic conductance of the outer parts of rice roots (epidermis, exodermis and the outermost cell layer of cortex) was 30 times higher than that of the overall root hydraulic conductance and that the endodermis was limiting to water flow through roots. It has been shown that the contribution of the cellto-cell pathway to radial water flow through roots is negligible under high transpiration. However, under water deficit conditions, more than 40% of the water uptake by roots may be contributed by the cell-to-cell pathway if calculated on leaf elongation rate basis [22]. Even under conditions of high transpiration where bulk water flow through plant root tissues is apoplastic, the root exodermis and endodermis are exceptions where water has to go through the cell-to-cell path [23]. Based on the findings of Ranathunge et al. [21], the cell-to-cell path of water in the endodermis can be viewed as the bottle neck to water uptake by roots in rice and improving the root water uptake, therefore, requires increasing the permeability of the cell-to-cell path to water, precisely in the endodermis. Aquaporins are a family of membrane major intrinsic proteins (MIP) that have been proved to facilitate water movement through membranes in plants [24–26]. They exist in the cell membrane (plasma membrane intrinsic proteins, PIPs), tonoplast (tonoplast intrinsic proteins, TIPs), peribacteroid membranes of N2 -fixing symbiotic root nodules (NIPs) and in the endoplasmic reticulum (small basic intrinsic membranes, SIPs). In rice, 33 aquaporin genes have been characterized, among which there are 11 PIPs and 10 TIPs [27]. PIPs are divided into two subgroups (PIP1s and PIP2s) based on sequence homology. PIP2 and TIP aquaporins have high water transport activity compared to PIP1s [28–30]. Nonetheless, PIP1 aquaporins have been suggested to form heterotetramers with PIP2s with increased water transport activity [31,32]. Some aquaporins have been reported to transport other physiologically important small molecules such as CO2 [33], boron or silicon [25] or H2 O2 [34]. Ohshima et al. [35] have reported that the aquaporin contents of the cell membranes and tonoplasts of the crassulacean acid metabolism (CAM) plant Graptopetalum paraguayense were extremely low compared to those of radish, which could account for the lower water permeability of Graptopetalum membranes. Evidence from numerous studies involving quantifying aquaporin expression and overexpression and/or down-regulation of aquaporin genes strongly supports a prominent role in increasing water permeability of different tissues in rice [36–39] and other plants [34,40,41]. However, the drought tolerance of transgenic plants overexpressing aquaporin genes varied where it increased in some cases [34,38,42] but was not affected or even decreased in some others [43,44]. Moreover, the correlation between the expression of aquaporins and midday stomatal depression, if any in rice or other plants, is yet to be understood. This highlights the need for genome-wide studies on the expression of aquaporins in relation to water status in field-grown plants in order to drive a more informative and precise conclusion on how to manipulate aquaporin genes to improve plant productivity. Sakurai-Ishikawa et al. [45] have shown that most of root aquaporins in rice were upregulated during periods of high transpiration demand (beginning of the light period) probably to maintain a positive plant water status. However, the relevance of this information to plant responses in the field is limited because the authors grew the plants under non-saturating conditions of 370 ␮mole m−2 s−1 light intensity and 75% relative humidity (RH), where stomatal depression did not occur but rather the rates of transpiration increased progressively during the whole light period. Many reports have described the tissue localization and regulation of aquaporin expression in rice in relation to water status in order to dissect their function(s) in the regulation of plant water status during normal and water stressful conditions [10,37,39,46].

Nonetheless, the significance of aquaporin expression profiles in relation to whole plant water status, rates of photosynthesis and water use efficiency in the field remains elusive because in most previous studies, data on photosynthesis and water use efficiency in the field were not included. Furthermore, the responsiveness of rice root aquaporins that are expressed in the endodermis to transpiration demand in the field has not been characterized yet. The objectives of the present study were to: (1) investigate the diurnal changes in gas exchange parameters in relation to expression profiles of leaf and root aquaporins in four rice cultivars (two japonica cultivars and two indica cultivars) with contrasting genetic backgrounds grown in the field, (2) implement previous information on tissue localization of aquaporins combined with their water transport activities in order to predict the water permeability of rice roots and leaves and hence, pinpoint the main possible reason(s) of midday stomatal depression, (3) analyse the diurnal changes in leaf soluble and insoluble sugar contents, a possible contributor to midday down-regulation of photosynthesis, as affected by changes in rates of photosynthesis, and (4) test the consequences of removing the barriers to radial water flow in the roots on the rates of transpiration and photosynthesis when stomatal depression is maximum. The data provide evidence that the rates of photosynthesis may be partly affected by inhibited sugar translocation but are independent of leaf internal CO2 (Ci ) and that both rates of photosynthesis and transpiration are tightly linked to the leaf water status which is in turn the product of aquaporin expression profiles in leaves and roots.

2. Materials and methods 2.1. Plant material Four rice cultivars were used in this study, two japonica cultivars (Giza 178 and Sakha 101) and two indica cultivars (IR64 and PSL2). Giza 178 and Sakha 101 were obtained from the Agricultural Research Institute (Giza, Egypt) and IR64 was kindly supplied by the International Rice Research Institute (IRRI, Philippines). PSL2 was obtained from the Ministry of Agriculture of Thailand.

2.2. Plant growth and drought treatment This experiment was carried out in the research field of Botany Department, Damietta University. The seeds of the four rice cultivars were sown outdoors in May (about 12 h photoperiod, 28/24 ◦ C day night temperature, 2850 ␮mole m−2 s−1 maximum light intensity and 45% day RH) in soil prepared by mixing equal volumes of sandy and clay soil, to form a soil layer 30 cm deep. The plants were grown in randomized block design with six blocks for each cultivar. Each block was 2 × 2 m. The plants were watered every day at 8:00–9:00. At the age of 15 days, the plants were thinned to be 25 cm apart. Daily watering continued up to the age of 40 d. Then, water was withheld from three blocks of each cultivar which were used for drought treatment. The remaining three blocks of each cultivar kept receiving water every day and were used as a control. The water content of the soil was monitored several times a day (measured instantly by dividing the weight of a soil sample by its water-saturated weight) so that subsequent measurements and sampling were made when the soil water content was about 4% higher than the soil water content at which the plants started wilting (determined from a preliminary experiment as the weight of soil sample at wilting divided by the saturated soil weight). The soil water content at wilt was considered as the fraction of transpirable soil water, FTSW [47]. All blocks reached the stressful water content after 3 d of withholding water. To maintain the soil water content


at the required value for each cultivar, water was added to soil into 10-cm-deep holes followed by measuring the soil water content.

2.3. Measurement of gas exchange parameters Gas exchange parameters were measured for the control and stressed plants (second leaf) by using LCi-SD gas exchange system (Analytical Development Company Ltd., England) as described previously [48]. Five measurements were made for each treatment from different plants. Gas exchange parameters were measured at 9:00, 13:00 and 17:00 (at light intensities of 2130, 2850 and 1970 ␮mole m−2 s−1 , respectively), i.e. 3 h after the start of light period, at midday and 3 h before the end of the light period. Measurements for each time point lasted for about 90 min. Within each time point, one plant from each treatment was used as a standard for which measurements were repeated every 15 min to make sure that gas exchange measurements did not change due to internal or external factors.

2.4. Measurement of leaf relative water content (RWC), shoot dry weight and total soluble sugars and starch Samples from second leaves were collected in pre-weighed plastic bags; the bags were sealed immediately and weighed to obtain the fresh weight (FW). Leaf samples were placed in distilled water at 4 ◦ C and kept in a fridge overnight. The samples were then weighed (saturated weight, SW). The samples were dried at 60 ◦ C for 2 d and weighed (dry weight, DW). RWC was calculated from the equation: RWC% = [(FW − DW)/(SW − DW)] × 100. Shoot dry weights were determined in samples of whole shoots collected at the end of the experiment. The samples were dried at 60 ◦ C for 2 d and weighed. Total soluble and insoluble sugars were determined as described previously [49,50]. The sugar concentrations were calculated from a standard curve in the range 20–80 ␮g. For both RWC and sugar content measurements, three replicates from three plants were used.

2.5. Harvesting of the plant material For each plant, a sample from the second leaf was collected and frozen immediately in liquid nitrogen. The plant was then removed from the soil and the roots were cleaned by dipping into chilled water for a few seconds. A root sample was then taken as for leaf samples. All samples were stored at −80 ◦ C until further use. 2.6. Quantification of aquaporin genes expression in leaves and roots Total RNA was extracted from 50 mg frozen leaf and root samples using Tri reagent as described previously [51]. To improve RNA quality and yield of root samples, 10% polyvinylpyrrolidone (Fischer Scientific, Belgium) was added to the extraction solvent. Semiquantitative RT-PCR was performed as described previously for rice aquaporins [27] by using the primers in Supplementary Table S1. The PCR products were resolved on 1% agarose gels, stained with ethidium bromide (EtBr) in 1× TAE buffer (Tris–acetic acid–EDTA) and visualized by UVIsave gel documentation system (UVITECH, UK). The band sizes were measured by using ImageStudio v 12.0 software (Li-COR Biosciences, USA). cDNA samples to be loaded into PCR were equalized based on equal 18S rRNA bands as an internal control (housekeeping) gene (Supplementary Fig. S1). For each aquaporin gene, three replicates from different RNA extractions were used.

167

2.7. Root clipping experiment The root clipping treatment was carried out at 13:00 for plants grown in the field. Plants from control blocks (were not subjected to drought at any stage) were removed from soil with most of their roots attached by using a shovel. In 3 L containers, the roots were cleaned gently in tap water and the whole root system was clipped about 3 cm above the root tips under water. Each plant was then held erect in the water where the water surface was exactly at the point of contact between the root and shoot system. Care was taken not to submerge parts of shoot bases. Three plants were used for each cultivar. In addition, three plants from each cultivar were treated similarly except that their roots were not clipped and used as a control. Gas exchange parameters were measured after 30 min as described above. The water in the containers was aerated by brief bubbling and gas exchange measurements were taken again at 17:00. Root and leaf hydraulic conductivity was measured as described by Katsuhara et al. [52].

2.8. Statistical analysis The experimental design is described in Section 2.2. All measurements were replicated as mentioned in each section. To compare between samples, the data were entered into SPSS v 18 and oneway ANOVA was run at significance level of p < 0.05 using least significant difference (LSD) as a comparison method.

3. Results 3.1. Development of drought stress Preliminary experiments showed that Giza 178, Sakha 101, IR64 and PSL2 plants started wilting when the soil water contents were 83.17, 83.76, 85.04 and 86.51%. The FTSW was thus significantly lower for japonica than for indica cultivars. The block design used for growing plants in the field was optimized so that the drought stress developed in relatively short period of time to minimize the production of new biomass under drought stress as indicated by the similar biomass (shoot dry weight) in well-watered and droughted plants at the end of treatment (see below).

3.2. Plant growth The growth of IR64 and PSL2 was significantly higher than that of Giza 178 and Sakha 101 in well-watered plants (Fig. 1). Drought stress did not result in significant reduction in the shoot dry weight as compared to the controls.

3.3. Relative water content (RWC) The RWC was measured at predawn, 9:00 and 13:00. In wellwatered plants, the predawn and 9:00 RWC was statistically similar for all cultivars, where it ranged from 97.9 to 99.7% (Fig. 2). At 13:00, PSL2 showed the lowest RWC compared to other cultivars (p < 0.05). In the droughted plants, predawn RWC was similar to that of the well-watered plants of the four cultivars and did not vary significantly from the corresponding controls. At 9:00, the RWC of the droughted plants was significantly reduced in all cultivars. At 13:00, the RWC was further reduced significantly in indica but not in japonica cultivars. The lowest RWC values were recorded in IR64 and PSL2 at 13:00 (75.6 and 71.4%, respectively).

168


Fig. 1. Dry biomass of shoots of the well-watered and droughted rice plants after 3 d of withholding water. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05.

Fig. 2. Diurnal changes in RWC of leaves of the control and droughted rice plants at predawn (A), 9:00 (B) and 13:00 (C). Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05.

3.4. Diurnal changes in gas exchange as affected by drought stress 3.4.1. Rates of photosynthesis (A) In well-watered plants of the four cultivars, the rate of photosynthesis (A) was highest at 9:00 and decreased significantly and progressively down to 17:00, except that it was statistically similar in PSL2 at 9:00 and at 13:00 (Fig. 3A–C). A was also significantly lower in Sakha 101 and PSL2 than in Giza 178 and IR64 at 9:00 (Fig. 3A). At 13:00, the highest A was recorded in PSL2 (p < 0.05) (Fig. 3B). At 17:00, A was significantly lower in IR64 and PSL2 (indica cultivars) than in Giza 178 and Sakha 101 (japonica cultivars) (Fig. 3C). In the droughted plants of all cultivars, A was significantly lower than in the corresponding controls at 9:00 (Fig. 3A). At 13:00, A of the droughted plants was similar to those of the controls except in PSL2 which had significantly reduced A (Fig. 3B). At 17:00, A was significantly reduced by drought as compared to the control except in Giza 178 (Fig. 3C).

Fig. 3. Diurnal changes in gas exchange in leaves of well-watered (black bars) and droughted (grey bars) rice plants. A, B, C are rates of photosynthesis (A). D, E, F are rates of transpiration (E). G, H, I are leaf internal CO2 concentrations. J, K, L are leaf level water use efficiency (A/E). Bars are means of five replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05.

3.4.2. Rates of transpiration (E) The rate of transpiration (E) of the well-watered plants of all cultivars was highest at 9:00 and decreased progressively and significantly during the day down to 17:00 except in PSL2 where it was similar at 9:00 and 13:00 (Fig. 3D–F). In the droughted plants, E was significantly reduced (to similar levels in all cultivars) compared to the controls at 9:00 and 13:00 except in PSL2 where E remained similar to the corresponding controls. The lowest E at 13:00 was recorded in Giza 178. At 17:00, E was significantly lower than that at 9:00 in all cultivars with the lowest value in Giza 178 (Fig. 3F). Compared to the controls at 17:00, E was reduced by drought in the japonica but not in indica rice.


169

the controls in all cultivars at 9:00 and 13:00 Except in PSL2 where it was significantly increased by drought stress at 13:00 (Fig. 4A and B), but significantly increased in the droughted plants of all cultivars (compared to the controls) except in Sakha 101 at 17:00 (Fig. 4C). 3.5.2. Insoluble sugar contents In well-watered plants, the insoluble sugar contents of all cultivars were statistically similar at 9:00 but were significantly higher in japonica than in indica cultivars at 13:00 and 17:00 (Fig. 4D–F). Drought stress did not change the insoluble sugar contents at 9:00 and 13:00 except in Giza 178 and Sakha 101 where there was a significant decrease at 13:00 (Fig. 4D and E). At 17:00, the japonica cultivars showed significantly lower insoluble sugar contents compared to the controls, but these contents were similar to those of the droughted indica cultivars (Fig. 4F). 3.6. Diurnal changes in leaf and root aquaporin expression in response to drought We selected eight PIPs (PIP1;1, PIP1;2, PIP1;3, PIP2;1, PIP2;2, PIP2;4, PIP2;6 and PIP2;7) and four TIPs (TIP1;1, TIP1;2, TIP2;2 and TIP4;1) for this study essentially because they have been reported to be abundantly expressed in rice [27]. All of the PIP and TIP genes were expressed in leaves but only five PIP (PIP1;1, PIP1;2, PIP1;3, PIP2;2 and PIP2;4) and three TIP (TIP1;1, TIP1;2 and TIP2;2) genes were expressed in roots. One-way ANOVA was carried out for each aquaporin (leaves and roots) in order to compare the diurnal changes in leaves and roots as well as response to drought of each aquaporin. The data in Fig. 5 through 16 are summarized below. Fig. 4. Diurnal accumulation of soluble (A, B, C) and insoluble sugars (D, E, F) in the control and droughted leaves of rice cultivars. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05.

3.4.3. Leaf internal CO2 concentrations (Ci ) The Ci of the well-watered plants remained unchanged at all times except that it was significantly higher in Sakha 101 and PSL2 at 17:00 (Fig. 3G–I). In the droughted plants, Ci was significantly lower than in the corresponding controls in Sakha 101 and IR64 at 9:00 and in all cultivars except PSL2 at 13:00 but increased at 17:00 up to the control values except in Giza 178 and PSL2. 3.4.4. Leaf level water use efficiency (WUE: A/E) In well-watered plants, WUE was significantly higher in japonica than in indica rice at 9:00 (Fig. 3J). At 13:00, Sakha 101 showed similar WUE as indica cultivars, whereas Giza 178 maintained significantly higher WUE than all cultivars (Fig. 3K). At 17:00, WUE was significantly higher in Giza 178 and IR64 than Sakha 101 (Fig. 3L). In droughted plants, the WUE was significantly increased as compared to the corresponding controls in the japonica cultivars but remained unchanged in IR64 and significantly decreased in PSL2 at 9:00 and 13:00 (Fig. 3J and K). At 17:00, WUE significantly increased in Giza 178 compared to the control but was significantly reduced in Sakha 101, IR64 and PSL2 (Fig. 3L). 3.5. Diurnal changes in soluble and insoluble sugar contents of leaves as affected by drought stress 3.5.1. Soluble sugar contents In well-watered plants, the soluble sugar contents of leaves were similar in all cultivars at 9:00 (Fig. 4A) and increased significantly at 13:00 only in the japonica cultivars (Fig. 4B). At 17:00, the japonica cultivars showed similar soluble sugar contents to those at 9:00 but indica cultivars had significantly lower contents (Fig. 4C). In the droughted plants, the soluble sugar contents remained similar to

3.6.1. PIP1 aquaporins In well-watered leaves, PIP1;1, PIP1;2 were consistently expressed at highest levels in all cultivars at 9:00 and then the expression levels decreased at 13:00 and/or 17:00 except in Giza 178 which maintained a constant expression level for PIP1;1 at all times (Fig. 5A–C and Fig. 6A–C). Contrarily, PIP1;3 showed the greatest expression levels at 13:00 and/or 17:00 in all cultivars (Fig. 7A–C). In the roots, no comparable pattern to leaves was observed. At 9:00, PIP1;1 showed largely similar transcript abundance to those in leaves except in PSL2 in which the root had higher levels than the leaves (Fig. 5A and D). At 13:00 and 17:00, PIP1;1 was down-regulated in the japonica but not in indica cultivars (Fig. 5E and F). A largely similar diurnal pattern was observed for PIP1;2 and PIP1;3 in the roots although the expression level in roots relative to that in leaves varied in some cultivars (Fig. 6D–F). In general, the expression levels of PIP1s in the leaves were similar to, higher or lower than those in the roots. In response to drought, the most consistent response was found in leaves at 9:00 where PIP1;1 and PIP1;2 were significantly induced by drought in indica but not in japonica cultivars (Figs. 5 and 6) whereas PIP1;3 was induced by drought in leaves of all cultivars but to higher levels in indica cultivars (Fig. 7). The greatest induction of PIP1;3 was observed in the indica cultivars. PIP1s did not respond consistently to drought at 13:00 and 17:00 where their expression was induced, reduced or even unchanged (Figs. 5–7). In droughted roots, no consistent response was found at 9:00 that matches that of leaves and the same was also found at 13:00 and 17:00 (Figs. 5–7). In most cases, at least at 9:00, the expression levels of PIP1s were significantly lower than in the corresponding leaves. 3.6.2. PIP2 aquaporins In well-watered plants, PIP2;1 transcripts were detected in leaves but not in roots (Fig. 8), where the highest expression level was observed at 13:00 (except in Sakha 101). PIP2;1 was induced by drought at 9:00 in all cultivars except Sakha 101, but was inhibited

170


Fig. 5. Diurnal expression levels of PIP1;1 in leaves (A, B, C) and roots (D, E, F) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. One-way ANOVA was run for data including leaves and roots. In gel labels: C is control and D is drought.

at 13:00 in all cultivars except in Sakha 101. No consistent response was found at 17:00. In well-watered plants, PIP2;2 and PIP2;4 showed the highest expression levels at 9:00 in all cultivars and were then downregulated at 13:00 and/or 17:00 (Figs. 9 and 10). In the roots, the expression levels of PIP2;2 and PIP2;4 were significantly lower than in the corresponding leaves in all cultivars at 9:00 and at 13:00 and 17:00 with minor exceptions in IR64 and PSL2. PIP2;2 and PIP2;4 were consistently induced by drought in leaves of all cultivars at 9:00 with the greatest induction in indica cultivars (Figs. 9 and 10). At 13:00 and 17:00, both aquaporins were down-regulated compared to 9:00 in all cultivars. In the droughted roots, the transcript abundance of PIP2;2 and PIP2;4 was lower than in the corresponding droughted leaves in all cultivars at all times (Figs. 9 and 10). PIP2;6 and PIP2;7 expression was detected in leaves but not in roots. PIP2;6 showed neither diurnal expression pattern nor consistent response to drought (Fig. 11), whereas PIP2;7 showed its lowest expression levels at 9:00 and was then induced at 13:00 and 17:00 in all cultivars (Fig. 12). PIP2;7 was upregulated by drought in all cultivars at 9:00 and only in indica cultivars at 13:00, but no comparable response was found at 17:00. Generally in all cultivars at least at 9:00, the expression levels of PIP2 aquaporins under drought were either greatly and significantly lower in the roots than in the corresponding leaves (PIP2;2 and PIP2;4) or totally undetectable in roots (PIP2;1, PIP2;6 and PIP2;7).

3.6.3. TIP aquaporins In well-watered leaves, TIP1;1, TIP1;2 and TIP4;1 were expressed at highest levels at 13:00 and/or 17:00 in all cultivars (except TIP1;2 in PSL2) (Figs. 13, 14 and 16). TIP2;2 showed no diurnal

regulation (Fig. 15). In the roots of well-watered plants, the expression of TIP1;1, TIP1;2 and TIP2;2 was either equal (in a few cases) or significantly and greatly lower than in the corresponding leaves in all cultivars at all times, whereas transcripts of TIP4;1 were detected only in leaves in all cultivars (Fig. 16). In droughted leaves, TIP1;1 and TIP1;2 were strongly induced by drought in all cultivars at 9:00 but not at 13:00 or 17:00 (Figs. 13–14). Moreover, TIP1;2 was upregulated to significantly greater levels in indica than in japonica cultivars. Conversely in roots, TIP1;1 did not respond similarly to drought in any cultivar whereas TIP1;2 was up- or down-regulated under drought. TIP2;2 showed no consistent response to drought in leaves or roots, but in all cultivars its expression level was higher in leaves than in roots (Fig. 15). TIP4;1 was expressed in only leaves of all cultivars where it had the lowest transcript abundance at 9:00 and was then strongly induced at 13:00 and 17:00 in all cultivars (Fig. 16). TIP4;1 did not respond to drought except in PSL2 at 13:00 and 17:00 and in Giza 178 at 17:00 where it was significantly upregulated. In general, the transcript abundance of TIP aquaporins under drought was either greatly and significantly higher in leaves than in the corresponding roots at least at 9:00 in all cultivars (TIP1;1, TIP1;2 and TIP2;2) or even totally undetectable in roots (TIP4;1).

3.7. Effect of root clipping on gas exchange and WUE Removal of radial barriers to water movement through roots was carried out by clipping the roots 3 cm above the root tips at 13:00 for plants growing in the field that were not previously subjected to drought. No immediate changes were observed in A but E increased significantly in all cultivars (Fig. 17A and C). After 4 h, A


171

Fig. 6. Diurnal expression levels of PIP1;2 in leaves (A, B, C) and roots (D, E, F) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. One-way ANOVA was run for data including leaves and roots. In gel labels: C is control and D is drought. ND means non-detectable.

was significantly enhanced in all cultivars (up to 2-fold in Giza 178 and 3-fold in PSL2) and E was also higher than in the corresponding controls although E was reduced both in the control and clipped plants as compared to those at 13:00 (statistics not shown) (Fig. 17B and D). Consequently, WUE decreased significantly at 13:00 but increased significantly at 17:00 in all cultivars compared to the control plants (Fig. 17E and F). 4. Discussion Japonica rice has been reported to have higher water use efficiency [53], lower sensitivity to photoinhibition [54,55] and lower fraction of transpirable soil water, FTSW [56] compared to indica rice. In the present study, japonica rice showed significantly lower FTSW (see results, Section 3.1) suggesting higher WUE and/or drought tolerance. Giza 178 is a drought-tolerant cultivar compared to Sakha 101, and PSL2 has inherent higher growth rate compared to IR64 (unpublished data). We thus expected variable performance in terms of growth, gas exchange and aquaporin expression profiles in the four cultivars under contrasting conditions of water availability and attempted, based on the data, to specify the major contributor(s) to midday stomatal depression and WUE of rice in the field. Data in Fig. 1 show that growth of well-watered plants in the field in terms of dry weight was significantly higher in indica than in the japonica cultivars. Drought stress caused no significant change in dry mass accumulation in any cultivar (Fig. 1A and B). This does not necessarily mean that drought stress did not slow down plant growth but rather the biomass differences

were too small to be detected because drought stress developed over a relatively short period of time. This is supported by the reduced plant biomass observed when the droughted plants were maintained under drought for further 5 d (data not shown). Consequently, the contribution of the biomass fraction produced under drought (expected to have more developed apoplastic barriers and higher resistance to water flow [57,58]) to overall plant permeability to water would be minimum, and hence changes in plant permeability to water would be mainly due to aquaporin activity. With PSL2 as an exception, which maintained high E at midday day resulting in dramatically lower RWC compared to other cultivars (Fig. 2) and suggesting a more extreme anisohydric behaviour in PSL2 [59–61], the well-watered plants of all cultivars showed significantly lower E and hence RWC at midday compared to at 9:00 or predawn, respectively (Fig. 2 and Fig. 3D–F). This suggests imbalance between water uptake and loss in all cultivars, a situation which was more intensified in PSL2 as evidenced by its lower RWC at midday (Fig. 2C). This imbalance is suggested to have resulted from excessive water loss by leaves or inadequate water uptake by roots or both. Drought stress of all cultivars resulted in further significant reductions in the RWC and E (except in PSL2) at 13:00 compared to the corresponding controls which had already undergone midday stomatal depression (Fig. 2 and Fig. 3D–F). Nonetheless, if E in the droughted plants is compared at 9:00 and 13:00, it appears that the droughted plants did not suffer from midday stomatal depression and consequently, the diurnal drop in RWC in the droughted plants was not as sharp as in the well-watered ones (Fig. 3D and

172



E). This suggests that a situation that is similar to midday stomatal depression (in terms of E) in the well-watered plants existed in the droughted plants at 9:00. However, the inhibition of A was not as much as that for E at 9:00 at least in the japonica cultivars presumably because A was augmented by sufficient Ci (see below). As a result, the WUE at leaf level was enhanced in the droughted japonica cultivars at 9:00 and 13:00, to less extent in IR64 but decreased in PSL2 (Fig. 3J–L). The decrease in WUE of PSL2 under drought was obviously because of excessive transpiration.

The data in Fig. 3G–I suggest that leaf Ci was not limiting to photosynthesis in the well-watered plants since leaf Ci was not depleted at midday in any cultivar compared to 9:00 but even increased to higher levels in Sakha 101 and PSL2 at 17:00 when A was minimal. This indicates that midday inhibition of photosynthesis was mainly due to factor(s) other than CO2 availability [4] and contrasts with plants that maintain high carboxylation efficiency during the day, such as maize in which Ci was depleted when A was highest [62]. In droughted plants of all cultivars, leaf

Fig. 8. Diurnal expression levels of PIP2;1 in leaves (A, B, C) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. In gel labels: C is control and D is drought.


173


Ci was reduced compared to the control in all cultivars except in PSL2 at 13:00 (Fig. 3H). Nonetheless, midday reduction in A in the droughted plants as compared to the controls was only observed in PSL2 (Fig. 3B). This suggests that leaf Ci was not limiting to photosynthesis in the droughted plants and particularly in PSL2. It can be concluded here that inhibition of A in all cultivars at midday and even more at 17:00 was dependent on the plant water status but independent from leaf Ci . This conclusion is further supported by the leaf Ci contents at 17:00 in the control and droughted plants where Ci in all treatments (except the droughted Giza 178) was statistically similar to or even higher than those at 9:00 (where A was maximum) although A was minimum at 17:00. Furthermore, the decrease in leaf level WUE in indica compared to japonica cultivars at 9:00 and 13:00 was mainly attributed to higher E in the indica rice. Data in Fig. 4 show clearly that the well-watered plants of japonica cultivars had higher soluble and insoluble sugar contents in their leaves at 13:00 and 17:00 if compared to the indica cultivars. Previous reports have demonstrated that rice genotypes with early vigour had lower carbohydrate contents in their leaves because carbon is consumed for daily growth and hence had carbon source-limited growth [63,64]. This may apply to the indica cultivars used in this study since they showed higher growth rates as shown in Fig. 1. High soluble sugar contents in the leaves have been reported in many drought-stressed plants and attributed either to inhibition of phloem transport or sequestration of sugars away from the transport sites [5]. Insoluble sugars in the leaf are a temporary sink to sucrose that cannot be exported [65].

The presence of higher soluble and insoluble sugar contents in leaves of well-watered japonica cultivars at 13:00 and 17:00 suggests that the photosynthates were in excess of sink requirements and/or transport capacity [47,62,66,67]. However, no specific inhibition of A was observed in japonica cultivars as compared to indica cultivars. Under drought, the increase in soluble sugar contents of indica cultivars at 17:00 simultaneously with the absence of change in the insoluble sugar contents in spite of the dramatic inhibition of A (Fig. 3C) suggests more severe sink limitation presumably as a result of severe growth inhibition that was observed over the next 2 weeks following re-watering (data not shown). If this was true then, the depletion of insoluble sugars in the droughted japonica cultivars at 17:00 compared to the controls (Fig. 4I) would indicate less inhibitory effect of drought on growth. The data discussed so far strongly suggest that indica cultivars have lower WUE mainly due to higher rates of E and are more sensitive to water deficit (as indicated by higher FTSW and lower RWC at midday) than japonica cultivars. Moreover, since the stomatal conductance and transpiration rates have been considered as indicators of the whole plant hydraulic conductance [68], and given that rice roots are limiting to plant hydraulic conductance [69], the severe drops in E and gs (Supplementary Fig. S2) in all cultivars and particularly in indica rice suggest imbalanced water relations caused by increased water permeability in leaves and/or high resistance to water flow in roots. More specifically, measurements of root and leaf hydraulic conductivity (Supplementary Fig. S3) showed that stomatal conductance and transpiration rates were

174



more related to root rather than to leaf hydraulic conductivity. Although measurement of hydraulic conductivity in the present study showed that roots have lower water permeability than leaves under drought, it cannot be ruled out how much of this permeability is attributed to aquaporins because we were unable to apply mercurials to field-grown plants in order to measure cell-to-cell water permeability. Consequently, rates of transpiration and stomatal conductance were presented as indicators of root hydraulic

conductivity and data on hydraulic conductivity were presented as supplementary data for the sake of simplicity. Rice aquaporins have been reported to be expressed in leaves and roots. Leaf aquaporins are predominantly expressed in mesophyll cells and root aquaporins are most abundant in exodermis and endodermis [70]. Because the endodermis is most limiting to water flow through roots (see Introduction), the abundance of aquaporins in this tissue suggests a crucial role in maintaining efficient water

Fig. 11. Diurnal expression levels of PIP2;6 in leaves (A, B, C) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. In gel labels: C is control and D is drought. ND means non-detectable.


175

Fig. 12. Diurnal expression levels of PIP2;7 in leaves (A, B, C) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. In gel labels: C is control and D is drought. ND means non-detectable.

uptake by roots particularly under high transpiration demand. If this is true, then the relative abundance of aquaporins with high water transport activity should be in favour of root. PIP2 and TIP2 aquaporins have been shown to have high water transport activities than PIP1 and TIP1 aquaporins in rice [27,70], in maize [28,31] and in radish [30]. However, PIP1 aquaporins have been suggested to interact with PIP2s forming heterotetramers with increased water transport activity and were thus suggested to regulate the water

transport activity of other aquaporins [31,32]. Sakurai et al. [70] reported that PIP1;1, PIP1;2, PIP1;3, PIP2;1, PIP2;3 and TIP2;1 are preferentially expressed in roots while PIP2;2 and TIP1;2 are leaf specific. However, different aquaporin expression profiles in rice have been reported [45,71], presumably because of the different rice cultivars and/or experimental conditions, but generally, more aquaporins were root specific. A discrepancy appears here as to if aquaporins increase the tissue conductance to water and given

Fig. 13. Diurnal expression levels of TIP1;1 in leaves (A, B, C) and roots (D, E, F) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. One-way ANOVA was run for data including leaves and roots. In gel labels: C is control and D is drought. ND means non-detectable.

176



that more of them are root specific, rice should not be that sensitive to water deficit because water uptake by roots would exceed water loss by leaves if root/shoot ration is close to unity. Possible reasons for this discrepancy could be that aquaporin regulation varied among rice cultivars and that growth conditions were too mild to exhaust the water uptake and transport system (since most previous reports have used plants grown at relatively low light intensities and high RH that did not suffer from midday stomatal depression) thereby triggering effective aquaporin response. Maurel et al. [26] pointed out that the significance of aquaporin expression profiles has to be specified under conditions similar to those experienced by plants in the field. To our knowledge, aquaporin expression has not been studied yet in field-grown rice plants in relation to midday stomatal depression. We selected eight PIP and four TIP aquaporins for this study (Figs. 5–16) based on previous reports, which showed that they have high water transport activities and are expressed abundantly in rice tissues (see above). Information summarized in Section 3.6 (see Results) clearly indicates that many aquaporins had high expression levels in leaves at 9:00 in well-watered plants and were strongly induced by drought in leaves of all cultivars. The induction was even stronger in leaves of indica compared to japonica cultivars. The consistent expression profiles of most aquaporins in all cultivars in the wellwatered plants and the strong induction of them by drought at 9:00 where transpiration demand is most intense (Fig. 3) confirm that they are regulated by transpiration demand rather than simply by circadian rhythm [45]. More importantly, the strong induction of aquaporins by drought in the leaves at 9:00 in all cultivars which was even stronger in indica rice, not coupled by comparable induction in root aquaporins suggests that leaf permeability to water

(and hence transpiration) is increased while that of root (and hence water uptake) is diminished in all cultivars and even to greater extent in indica rice, a situation which explains the dramatic reduction in RWC of all cultivars and particularly indica cultivars at midday (Fig. 2). The diminished expression of PIP2;1, PIP2;2 and PIP2;4, which have high water transport activity [27,70] in roots although they are expressed in leaves to considerably high levels, would further intensify the shortage of water supply to leaves at 9:00 (or later during the day, see below) and cause rapid decline in plant water budget resulting in stomatal depression at midday. Interestingly, two aquaporin expression profiles can be distinguished in well-watered leaves of all field-grown cultivars, one at 9:00, in which aquaporins with high water transport activities (PIP1 PIP1;2PIP2;2 and PIP2;4) showed the highest transcript abundance with no comparable response in roots, and another at 13:00 (which largely resembles that at 17:00), in which these aquaporins were down-regulated simultaneously with midday stomatal depression. A question arises as to which of these two profiles has caused stomatal depression. Addressing this question has to take into consideration that aquaporin proteins do not necessarily peak at the same time with mRNA. A lag time of about 4 h has been reported for many aquaporin proteins in rice where rates of transpiration (but not hydraulic conductance) peaked simultaneously with aquaporin proteins but not with mRNA in rice leaves [45]. If this is true then, the mRNA profile observed at 9:00 would take effect (by translating mRNA into functional aquaporins) 4 h later, i.e. at 13:00 and the rates of transpiration observed at 9:00 would also be the product of predawn mRNA expression profiles which were largely similar to those at 17:00 (data not shown). However, it should be taken into consideration that most of water flow under conditions of high transpiration is apoplastic [22] and hence the


177


contribution of aquaporins to E at 9:00 could be less than that at 13:00 and 17:00. Furthermore, the more extreme upregulation of aquaporins at 9:00 in the droughted leaves (Figs. 7, 9, 10, 12–14) apparently has resulted in more negative water status at 13:00. Differences in aquaporin expression profiles could account for the significantly higher rates of transpiration in the indica rice at least at a certain point during the day (for example in the droughted PSL2 at 9:00 and at 13:00, Fig. 3D and E) combined with the greater drop in RWC (particularly at 17:00) compared to the japonica cultivars. Figs. 5–7 show that PIP1s where induced by drought to

higher levels in leaves of PSL2 compared to the japonica cultivars. Figs. 8–12 also show that PIP2 aquaporins were more strongly upregulated by drought in leaves of PSL2 than in the japonica cultivars. Furthermore, Figs. 13–15 show that TIP aquaporins were expressed to higher levels under drought in leaves of PSL2 compared to japonica cultivars. No comparable induction of aquaporins was observed in roots of PSL2. A relatively similar situation was observed, though to less extent in IR64 which is an indica cultivar. The stronger induction of PIP1s in PSL2, which were assigned a regulatory role in water transport [31,32] where they form

Fig. 16. Diurnal expression levels of TIP4;1 in leaves (A, B, C) of the well-watered and droughted rice plants. Bars are means of three replicates ±SE. Bars ± SE not sharing small letters are significantly different at p < 0.05. In gel labels: C is control and D is drought. ND means non-detectable.

178


Fig. 17. Rates of photosynthesis (A, B), transpiration (C, D) and leaf level water use efficiency (E, F) in the control and root-clipped plants at 13:00 and 17:00. Bars are means of five replicates ±SE. Bars ± SE labelled with asterisk are significantly different from the corresponding control at p < 0.05.

heterotetramers with PIP2 aquaporins (which were also strongly induced in PSL2), suggests a dramatically increased osmotic water transport under drought in leaves but not in roots of this cultivar. This could explain the increased transpiration and also the greater drop in RWC in PSL2 as a result of the imbalance between water loss by leaves and uptake by roots. One other possible cause of higher E in PSL2 in spite of the low RWC could be the increased expression and activity of TIP aquaporins in leaves that facilitate cell-to-cell water flow and thereby allow transpiration even at low RWC, i.e. render the plant more anisohydric [41,72]. Although anisohydry has been viewed as an adaptive feature to drought stress in tomato [41], it does not seem to be similarly adaptive in rice since TIP aquaporins were not equally induced in roots where the high E in leaves was not coupled with high water uptake by roots. Although

drought stress in the present work did not bring about detectable changes in growth (Fig. 1), subsequent growth of PSL2 over the next 2 weeks after re-watering was more severely inhibited than in other cultivars (data not shown). It can be concluded here that the negative water status in rice and in particular indica cultivars could be attributed to the imbalanced water transport in leaves and roots due to strong induction of most aquaporins in leaves but not in roots and hence inadequate water uptake by roots. Previous studies on expression of aquaporins in rice have reported that most aquaporins are down-regulated in leaves under drought stress. For example, Nguyen et al. [10] found that only OsTIP3;1 was induced by drought in rice whereas all other aquaporins including PIP1s, PIP2s, TIPs and NIPs were either downregulated or unaffected in the Dongjin rice cultivar. Conversely, Kuwagata et al. [73] have reported that many (10 out of 13 studied genes) aquaporins (including PIP1s, PIP2s and TIPs) were upregulated in the leaves of plants grown in dry air in the Akitakomachi rice cultivar although the extent of expression induction was not as great as in the present study. The systemic and strong induction of most aquaporin gene expression in leaves of the studied rice cultivars and more specifically in indica cultivars also contrasts to previous reports. It appears that the growth conditions are the dominant factor determining the aquaporin expression profiles in a given rice cultivar. To our knowledge, all previous studies have reported aquaporin gene expression in rice plants grown under controlled conditions (in contrast to data from field-grown plants) where the plants were not exposed to transpiration demand that is sufficient to exhaust the water transport system and induce a clear response. Interestingly, when the four rice cultivars were grown in a climate chamber and exposed to drought, the aquaporin expression profiles were found to be largely similar to those in the literature (unpublished data). This finding strongly suggests that it is quite feasible and realistic to focus future research on rice photosynthesis and water relations on field-grown plants. In the light of our data compared to previous studies, further work is necessary to investigate the quantities of aquaporin proteins in the field. It has been reported previously that aquaporin protein quantities do not necessarily peak in time with mRNA [45,74,75]. However, this finding has been based on data from plants grown under controlled conditions, where plant responses would differ substantially from those in the field. Data clearly show that the root and specifically the radial path [21,23] (see Introduction) is a major constraint to water uptake and consequently to photosynthesis, WUE and plant growth. When the radial barriers to water flow were removed by clipping the roots of all cultivars in the field at midday, instant increase in transpiration was observed indicating rapid increase in water uptake by roots (Fig. 17C and D) although there was no comparable increase in photosynthesis. This increase in water uptake (up to 53.15% increase in root hydraulic conductivity in Sakha 101, Supplementary table S2) is apparently not the product of root aquaporin activity but rather seems to be due to exposure of root xylem vessels to the external solution. After 4 h, E in all cultivars remained higher than in the control and there was also an increase in A (Fig. 17A and B), thus resulting in a decrease in WUE at 13:00 but significant increase after 4 h (Fig. 17E and F). However, over the next few days, A and E of the treated plants were similar or in some cases lower than in the control plants presumably due to nutrient imbalance (data not shown). Sasaki and Hoshikawa [76] reported that the growth of rice plants in terms of leaf sheath length was significantly improved after 7 d when the roots were pruned 2 mm above the root tips. Furthermore, Steudle et al. [77] demonstrated that puncturing the root of maize radially by making 18–60 ␮m pores at 70–90 mm from the root tip substantially decreased the root pressure (a feature that developed by the restricted permeability of endodermis) and increased the root hydraulic conductance by a factor of 10


compared to the control. Our results demonstrate that removal of radial barriers improves the water status of plants suffering from midday stomatal depression, suggesting that the unbalanced aquaporin regulation in leaves and roots and in particular, the inadequate activity of root aquaporins was a major contributor to plant negative water status at midday. We conclude that midday reduction in photosynthesis in rice is mainly due to unbalanced expression of aquaporins in leaves and roots resulting in rapid depletion of leaf water and subsequent inhibition of photosynthesis. Our conclusion is further supported by the increase in rates of A and E and subsequent increase in WUE after removal of radial barriers to water flow in roots. Acknowledgement This project was supported financially by the Science and Technology Development Fund (STDF), Egypt, Grant number 3871. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.plantsci. 2014.08.006. References [1] F. Tardieu, R. Tuberosa, Dissection and modelling of abiotic stress tolerance in plants, Curr. Opin. Plant Biol. 13 (2010) 206–212. [2] K. Ishihara, K. Saito, Diurnal course of photosynthesis, transpiration and diffusive conductance in the single leaf of the rice plants grown in the paddy field under submerged condition, Japn. J. Crop Sci. 56 (1987) 8–17. [3] T. Hirasawa, Y. Iida, K. Ishihara, Dominant factors in reduction of photosynthetic rate affected by air humidity and leaf water potential in rice plants, Japn. J. Crop Sci. 58 (1989) 383–389. [4] T.L. Pons, R.A.M. Welschen, Midday depression of net photosynthesis in the tropical rainforest tree Eperua grandiflora: contributions of stomatal and internal conductances, respiration and rubisco functioning, Tree Physiol. 23 (2003) 937–947. [5] W.P. Quick, M.M. Chaves, M. Wendler, The effect of water stress on photosynthetic carbon metabolism in 4 species grown under field conditions, Plant Cell Environ. 15 (1992) 25–35. [6] Y.J. Zhang, F.C. Meinzer, J.H. Qi, G. Goldstein, K.F. Cao, Midday stomatal conductance is more related to stem rather than leaf water status in subtropical deciduous and evergreen broadleaf tree species, Plant Cell Environ. 36 (2013) 149–158. [7] K. Iseki, K. Homma, T. Irie, T. Endo, T. Shiraiwa, The long-term changes in midday photoinhibition in rice (Oryza sativa L.) growing under fluctuating soil water conditions, Plant Prod. Sci. 16 (2013) 287–294. [8] N. Murata, S. Takahashi, Y. Nishiyama, S.I. Allakhverdiev, Photoinhibition of photosystem II under environmental stress, Biochim. Biophys. Acta 1767 (2007) 414–421. [9] S. Fukai, M. Cooper, Development of drought-resistant cultivars using physiomorphological traits in rice, Field Crop Res. 40 (1995) 67–86. [10] M.X. Nguyen, S. Moon, K.H. Jung, Genome-wide expression analysis of rice aquaporin genes and development of a functional gene network mediated by aquaporin expression in roots, Planta 238 (2013) 669–681. [11] V.C. Tanguilig, E.B. Yambao, J.C. OToole, S.K. De Datta, Water stress effects on leaf elongation, leaf water potential transpiration and nutrient uptake of rice, maize and soybean, Plant Soil 103 (1987) 155–168. [12] J.E. Cairns, A. Audebert, C.E. Mullins, A.H. Price, Mapping quantitative trait loci associated with root growth in upland rice (Oryza sativa L.) exposed to soil water deficit in fields with contrasting soil properties, Field Crops Res. 114 (2009) 108–118. [13] N. Miyamoto, E. Steudle, T. Hirasawa, R. Lafitte, Hydraulic conductivity of rice roots, J. Exp. Bot. 52 (2001) 1835–1846. [14] R. Serraj, A. Kumar, K.L. McNally, I. Slamet-Loedin, R. Bruskiewich, R. Mauleon, J. Cairns, R.J. Hijmans, Improvement of drought resistance in rice, Adv. Agron. 103 (2009) 41–99. [15] B. Parent, B. Suard, R. Serraj, F. Tardieu, Rice leaf growth and water potential are resilient to evaporative demand and soil water deficit once the effects of root system are neutralized, Plant Cell Environ. 33 (2010) 1256–1267. [16] E. Steudle, C.A. Peterson, How does water get through roots, J. Exp. Bot. 49 (1998) 775–788. [17] E. Steudle, Water uptake by roots: effects of water deficits, J. Exp. Bot. 51 (2000) 1531–1542. [18] E. Steudle, Water uptake by plant roots: an integration of views, Plant Soil 226 (2000) 45–56.

179

[19] E. Steudle, J. Frensch, Water transport in roots: role of the apoplast, Plant Soil 187 (1996) 67–79. [20] L. Schreiber, R. Franke, K. Hartmann, K. Ranathunge, E. Steudle, The chemical composition of suberin in apoplastic barriers affects radial hydraulic conductivity differently in the roots of rice (Oryza sativa L. cv. IR64) and corn (Zea mays L. cv. Helix), J. Exp. Bot. 56 (2005) 1427–1436. [21] K. Ranathunge, E. Steudle, R. Lafitte, Control of water uptake by rice (Oryza sativa L.): role of the outer part of the root, Planta (2003) 193–205. [22] Z. Lu, P.M. Neumann, Water stress inhibits hydraulic conductance and leaf growth in rice seedlings but not the transport of water via mercury-sensitive water channels in the root, Plant Physiol. 120 (1999) 143–151. [23] H.M. Zimmermann, K. Hartmann, L. Schreiber, E. Steudle, Chemical composition of apoplastic transport barriers in relation to radial hydraulic conductivity of corn roots (Zea mays L.), Planta 210 (2000) 302–311. [24] C. Maurel, Aquaporins and water permeability of plant membranes, Annu. Rev. Plant. Physiol. Plant Mol. Biol. 48 (1997) 399–429. [25] C. Maurel, Plant aquaporins: novel functions and regulation properties, FEBS Lett. 581 (2007) 2227–2236. [26] C. Maurel, L. Verdoucq, D.T. Luu, V. Santoni, Plant aquaporins: membrane channels with multiple integrated functions, Annu. Rev. Plant Biol. 59 (2008) 595–624. [27] J. Sakurai, F. Ishikawa, T. Yamaguchi, M. Uemura, M. Maeshima, Identification of 33 rice aquaporin genes and analysis of their expression and function, Plant Cell Physiol. 1546 (2005) 1568–1577. [28] F. Chaumont, F. Barrieu, R. Jung, M.J. Chrispeels, Plasma membrane intrinsic proteins from maize cluster in two sequence subgroups with differential aquaporin activity, Plant Physiol. 122 (2000) 1025–1034. [29] T. Matsumoto, H.L. Lian, W.A. Su, D. Tanaka, C.W. Liu, I. Iwasaki, Y. Kitagawa, Role of the aquaporin PIP1 subfamily in the chilling tolerance of rice, Plant Cell Physiol. 50 (2009) 216–229. [30] S. Suga, M. Maeshima, Water channel activity of radish plasma membrane aquaporins heterologously expressed in yeast and their modification by sitedirected mutagenesis, Plant Cell Physiol. 45 (2004) 823–830. [31] K. Fetter, V. Van Wider, M. Moshelion, F. Chaumont, Interactions between plasma membrane aquaporins modulate their water channel activity, Plant Cell 16 (2004) 215–228. [32] E. Zelazny, J.W. Borst, M. Muylaert, H. Batoko, M.A. Hemminga, F. Chaumont, FRET imaging in living maize cells reveals that plasma membrane aquaporins interact to regulate their subcellular localization, Proc. Natl. Acad. Sci. USA 104 (2007) 12359–12364. [33] Y.T. Hanba, H. Kogami, I. Terashima, The effect of internal CO2 conductance on leaf carbon isotope ratio, Isotopes Environ. Health Stud. 39 (2003) 5–13. [34] L.L. Wang, A.P. Chen, N.Q. Zhong, N. Liu, X.M. Wu, F. Wang, C.L. Yang, M.F. Romero, G.X. Xia, The Thellungiella salsuginea tonoplast aquaporin TsTIP1;2 functions in protection against multiple abiotic stresses, Plant Cell Physiol. 55 (2014) 148–161. [35] Y. Ohshima, I. Iwasaki, S. Suga, M. Murakami, K. Inoue, M. Maeshima, Low aquaporin content and low osmotic water permeability of the plasma and vacuolar membranes of a CAM plant Graptopetalum paraguayense: comparison with radish, Plant Cell Physiol. 42 (2001) 1119–1129. [36] M. Katsuhara, K. Koshio, M. Shibasaka, M. Hayashi, T. Hayakawa, K. Kasamo, Over-expression of a barley aquaporin increased the shoot/root ratio and raised salt sensitivity in transgenic rice plants, Plant Cell Physiol. 44 (2003) 1378–1383. [37] G.W. Li, Y.H. Peng, X. Yu, M.H. Zhang, W.M. Cai, W.N. Sun, W.A. Su, Transport functions and expression analysis of vacuolar membrane aquaporins in response to various stresses in rice, J. Plant Physiol. 165 (2008) 1879–1888. [38] H.L. Lian, X. Yu, Q. Ye, X. Ding, Y. Kitagawa, S.S. Kwak, W.A. Su, Z.C. Tang, The role of aquaporin RWC3 in drought avoidance in rice, Plant Cell Physiol. 45 (2004) 481–489. [39] H.L. Lian, X. Yu, D. Lane, W.N. Sun, Z.C. Tang, W.A. Su, Upland rice and lowland rice exhibited different PIP expression under water deficit and ABA treatment, Cell Res. 16 (2006) 651–660. [40] W. Hu, Q. Yuan, Y. Wang, R. Cai, X. Deng, J. Wang, S. Zhou, M. Chen, L. Chen, C. Huang, Z. Ma, G. Yang, G. He, Overexpression of a wheat aquaporin gene, TaAQP8, enhances salt stress tolerance in transgenic tobacco, Plant Cell Physiol. 53 (2012) 2127–2141. [41] N. Sade, B.J. Vinocur, A. Diber, A. Shati, G. Ronen, H. Nissan, R. Wallach, H. Karchi, M. Moshelion, Improving plant stress tolerance and yield production: is the tonoplast aquaporin Sl TIP2;2 a key to isohydric to anisohydric conversion, New Phytol. 181 (2009) 651–661. [42] Q. Yu, Y. Hu, J. Li, Q. Wua, Z. Lin, Sense and antisense expression of plasma membrane aquaporin BnPIP1 from Brassica napus in tobacco and its effects on plant drought resistance, Plant Sci. 169 (2005) 647–656. [43] R. Aharon, Y. Shahak, S. Wininger, R. Bendov, Y. Kapulnik, G. Galili, Overexpression of a plasma membrane aquaporin in transgenic tobacco improves plant vigour under favourable growth conditions but not under drought or salt stress, Plant Cell 15 (2003) 439–447. [44] A. Tsuchihira, Y.T. Hanba, N. Kato, T. Doi, T. Kawazu, M. Maeshima, Effect of overexpression of radish plasma membrane aquaporins on water use efficiency, photosynthesis and growth of Eucalyptus trees, Tree Physiol. 30 (2010) 417–430. [45] J. Sakurai-ishikawa, M. Murai-Hatano, H. Hayashi, A. Ahamed, F. Fukushi, T. Matsumoto, Y. Kitagawa, Transpiration from shoots triggers diurnal changes in root aquaporin expression, Plant Cell Environ. 34 (2011) 1150–1163.

180


[46] L. Guo, Z.Y. Wang, H. Lin, W.E. Cui, J. Chen, M. Liu, Z.L. Chen, L.J. Qu, H. Gu, Expression and functional analysis of the rice plasma membrane intrinsic protein gene family, Cell Res. 16 (2006) 277–286. [47] M. Rebolledo, D. Luquet, A. Clément-Vidal, L. Rouan, M. Dingkuhn, Phenomics of rice early vigour and drought response: are sugar related and morphogenetic traits relevant? Rice (N Y) 5 (2012) 22–36. [48] G.M. Abogadallah, Sensitivity of Trifolium alexandrinum L. to salt stress is related to the lack of long-term stress-induced gene expression, Plant Sci. 178 (2010) 491–500. [49] R. Rose, C.L. Rose, S.K. Omi, K.R. Forry, D.M. Durall, W.L. Bigg, Starch determination by perchloric acid vs enzymes: evaluating the accuracy and precision of six colorimetric methods, J. Agric. Food Chem. 39 (1991) 2–11. [50] U. Schulter, R.M.M. Crawford, Long-term anoxia tolerance in leaves of Acorus calamus L. and Iris pseudacorus L, J. Exp. Bot. 52 (2001) 2213–2225. [51] P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction, Anal. Biochem. 162 (1987) 156–159. [52] M. Katsuhara, K. Koshio, M. Shibasaka, K. Kasamo, Expression of an aquaporin at night in relation to the growth and root water permeability in barley seedlings, Soil Sci. Plant Nutr. 49 (2003) 883–888. [53] S. Peng, R.C. Laza, G.S. Khush, A.L. Sanico, R.M. Visperas, F.V. Garcia1, Transpiration efficiencies of indica and improved tropical japonica rice grown under irrigated conditions, Euphytica 103 (1998) 103–108. [54] B.H. Ji, Z.G. Li, M.Z. Ge, H.Y. Tong, D.M. Jiao, Traits related to photoinhibition of photosynthesis in Indica and Japonica subspecies of rice and their reciprocal cross F1 hybrids, Acta Phytophysiol. Sinica 20 (1994) 8–16. [55] D. Jiao, B. Ji, Photoinhibition in indica and japonica subspecies of rice (Oryza sativa) and their reciprocal F1 hybrids, Aust. J. Plant Physiol. 28 (2001) 299–306. [56] R. Serraj, V.R.P. Gowda, Y.S. Guan, H.Y. He, S. Impa, D.C. Liu, R.C. Mabesa, R. Sellamuthu, R. Torres, Drought-resistant rice: physiological framework for an integrated research strategy, in: R. Serraj, J. Bennett, B. Hardy (Eds.), Drought Frontiers in Rice Crop Improvement for Increased Rainfed Production, International Rice Research institute, Manila, Philippines, 2008, pp. 35–59. [57] D.E. Enstone, C.A. Peterson, Effects of exposure to humid air on epidermal viability and suberin deposition in maize (Zea mays L.) roots, Plant Cell Environ. 21 (1998) 837–844. [58] G.B. North, P.S. Nobel, Radial hydraulic conductivity of individual root tissues of Opuntia ficus-indica (L.) Miller as soil moisture varies, Ann. Bot. 77 (1996) 133–142. [59] H. Boonjung, S. Fukai, Effects of soil water deficit at different growth stages on rice growth and yield under upland conditions, 1. Growth during drought, Field Crops Res. 48 (1996) 37–45. [60] B. Jongdee, S. Fukai, M. Cooper, Leaf water potential and osmotic adjustment as physiological traits to improve drought tolerance in rice, Field Crop Res. 76 (2002) 153–163. [61] V. Sibounheuang, J. Basnayake, S. Fukai, Genotypic consistency in the expression of leaf water potential in rice (Oryza sativa L.), Field Crop Res. 97 (2006) 142–154. [62] Y. Peng, C. Li, F.B. Fritschi, Diurnal dynamics of maize leaf photosynthesis and carbohydrate concentrations in response to differential N availability, Environ. Exp. Bot. 99 (2014) 18–27.

[63] D. Luquet, J. Soulié, M. Rebolledo, L. Rouan, A. Clément-Vidal, M. Dingkuhn, Developmental dynamics and early growth vigour in rice 2. Modelling genetic diversity using Ecomeristem, J. Agron. Crop Sci. 198 (2012) 385–398. [64] M.C. Rebolledo, D. Luquet, B. Courtois, A. Henry, J.C. Soulié, L. Rouan, M. Dingkuhn, Can early vigour occur in combination with drought tolerance and efficient water use in rice genotypes? Funct. Plant Biol. 40 (2013) 582–594. [65] M.J. Paul, C.H. Foyer, Sink regulation of photosynthesis, J. Exp. Bot. 360 (2001) 1383–1400. [66] A.J. Gordon, Diurnal patterns of photosynthate allocation and partitioning among sinks, in: J. Cronshaw, W.J. Lucas, R.T. Giaquinta (Eds.), Phloem Transport, vol. 1, Alan RLiss, Incorporation, New York, NY, 1986, pp. 499–517. [67] G.H. Heichel, R.H. Delaney, H.T. Cralle, Carbon assimilation, partitioning, and utilization, in: A.A. Hanson, D.K. Barnes, R.R. Hill (Eds.), Alfalfa and Alfalfa Improvement ASA-CSSA-SSSA, Madison, American Society of Agronomy, USA, 1988, pp. 195–228. [68] T. Hirasawa, M. Tsuchida, K. Ishihara, Relationship between resistance to water transport and exudation rate and the effect of the resistance on the midday depression of stomatal aperture in rice plants, Japn. J. Crop Sci. 61 (1992) 143–152. [69] T. Hirasawa, T. Gotou, K. Ishihara, On resistance to water transport from roots to the leaves at the different positions on a stem in rice plants, Japn. J. Crop Sci. 61 (1992) 153–158. [70] J. Sakurai, A. Ahamed, M. Murai, M. Maeshima, M. Uemura, Tissue and cellspecific localization of rice aquaporins and their water transport activities, Plant Cell Physiol. 49 (2008) 30–39. [71] H.T. Nguyen, R.C. Babu, A. Blum, Breeding for drought resistance in rice: physiology and molecular genetics considerations, Crop Sci. 37 (1997) 1426–1434. [72] Y. Peng, W. Lin, W. Cai, R. Arora, Overexpression of a Panax ginseng tonoplast aquaporin alters salt tolerance, drought tolerance and cold acclimation ability in transgenic Arabidopsis plants, Planta 226 (2007) 729–740. [73] T. Kuwagata, J. Ishikawa-Sakurai, H. Hayashi, K. Nagasuga, K. Fukushi, A. Ahamed, K. Takasugi, M. Katsuhara, M. Murai-Hatano, Influence of low air humidity and low root temperature on water uptake, growth and aquaporin expression in rice plants, Plant Cell Physiol. 53 (2013) 1418–1431. [74] Y. Boursiac, S. Chen, D.T. Luu, M. Sorieul, N. van den Dries, C. Maurel, Early effects of salinity on water transport in Arabidopsis roots. Molecular and cellular features of aquaporin expression, Plant Physiol. 139 (2005) 790–805. [75] B. Muries, M. Faize, M. Carvajal, M.C. Martínez-Ballesta, Identification and differential induction of the expression of aquaporins by salinity in broccoli plants, Mol. Biosyst. 7 (2011) 1322–1335. [76] R. Sasaki, K. Hoshikawa, The role of crown roots from coleoptilar node in the rooting and development of transplanted rice nursling seedlings, Japn. J. Crop Sci. 66 (1997) 259–267. [77] E. Steudle, M. Murrmann, C.A. Peterson, Transport of water and solutes across maize roots modified by puncturing the endodermis. Further evidence for the composite transport model of the root, Plant Physiol. 103 (1993) 335–349.

Aquaporins in developing rice grains.

Legumes are different: Leaf nitrogen, photosynthesis, and water use efficiency.

A single SNP in NRT1.1B has a major impact on nitrogen use efficiency in rice.

The Interactions of Aquaporins and Mineral Nutrients in Higher Plants.

Structural determinants of the hydrogen peroxide permeability of aquaporins.

Rice performance and water use efficiency under plastic mulching with drip irrigation.

Corrigendum: Ectopic expression of Arabidopsis Target of Rapamycin (AtTOR) improves water-use efficiency and yield potential in rice.

Ectopic expression of Arabidopsis Target of Rapamycin (AtTOR) improves water-use efficiency and yield potential in rice.

Photosynthesis and chloroplast genes are involved in water-use efficiency in common bean.

Nutrient availability and nutrient use efficiency in plants growing in the transition zone between land and water.

A hairy-leaf gene, BLANKET LEAF, of wild Oryza nivara increases photosynthetic water use efficiency in rice.

Sealing rice field boundaries in Bangladesh: a pilot study demonstrating reductions in water use, arsenic loading to field soils, and methane emissions from irrigation water.

Strategies for improving potassium use efficiency in plants.

Metaproteomic identification of diazotrophic methanotrophs and their localization in root tissues of field-grown rice plants.

Are Aquaporins the Missing Transmembrane Osmosensors?

Diurnal patterns of canopy photosynthesis, evapotranspiration and water use efficiency in chickpea (Cicer arietinum L.) under field conditions.

Renal aquaporins and water balance disorders.

The rice production practices of high yield and high nitrogen use efficiency in Jiangsu, China.

The impact of water use fees on dispatching and water requirements for water-cooled power plants in Texas.

Multiple abiotic stimuli are integrated in the regulation of rice gene expression under field conditions.

Transcription enhancers as major determinants of SV40 polyomavirus growth efficiency and host cell tropism.

Cardiovascular risk factors are major determinants of thrombotic risk in patients with the lupus anticoagulant.

The carbon bonus of organic nitrogen enhances nitrogen use efficiency of plants.

Transgenic indica rice plants.