JXB Advance Access originally published online on September 22, 2006
Journal of Experimental Botany 2007 58(2):169-175; doi:10.1093/jxb/erl101
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Whole plant responses, key processes, and adaptation to drought stress: the case of rice
1International Rice Research Institute, Los Baños, Philippines
2Laiyang Agricultural University, Shandong, China
3Chinese Academy of Agricultural Science, Beijing, China
* Present address and to whom correspondence should be sent: Pioneer Hi-Bred International, 18285 Co. Rd, 96, Woodland, CA 95616, USA. E-mail: Renee.Lafitte{at}pioneer.com
Received 16 February 2006; Accepted 5 July 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Most high-yielding rice cultivars developed for irrigated conditions, including the widely grown lowland variety IR64, are highly susceptible to drought stress. This limits their adoption in rainfed rice environments where there is a risk of water shortage during the growing season. Mapping studies using lowland-by-upland rice populations have provided limited information about the genetic basis of variation in yield under drought. One approach to simultaneously improve and understand rice drought tolerance is to generate backcross populations, select superior lines in managed stress environments, and then evaluate which features of the selected lines differ from the recurrent parent. This approach was been taken with IR64, using a range of tolerant and susceptible cultivars as donor parents. Yields of the selected lines measured across 13 widely contracting water environments were generally greater than IR64, but genotype-by-environment effects were large. Traits expected to vary between IR64 and selected lines are plant height, because many donors were not semi-dwarf types, and maturity, because selection in a terminal stress environment is expected to favour earliness. In these experiments it was found that some lines that performed better under upland drought were indeed taller than IR64, but that shorter lines with good yield under drought could also be identified. In trials where drought stress developed in previously flooded (lowland) fields, height was not associated with performance. There was little change in maturity with selection. Other notable differences between IR64 and the selected backcross lines were in their responses to applied ABA and ethylene in greenhouse experiments at the vegetative stage and in leaf rolling observed under chronic upland stress in the field. These observations are consistent with the hypothesis that adaptive responses to drought can effectively allow for improved performance across a broad range of water environments. The results indicate that the yield of IR64 under drought can be significantly improved by backcrossing with selection under stress. In target environments where drought is infrequent but significant in certain years, improved IR64 with greater drought tolerance would be a valuable option for farmers.
Key words: Adaptation, drought stress, rice, whole plant response
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Drought is a major problem that limits the adoption of high-yielding rice varieties (HYVs) in drought-prone rainfed rice environments, where high sensitivity to even short periods of water deficit constitutes a risk that farmers cannot afford to take. This limits the use of a technology that will return significant benefits to farmers in the more common years of adequate rainfall, because the chance of failure in one year out of every four or five cannot be absorbed by the limited resource base available to small-scale producers. In environments where severe drought is less frequent and farmers already use improved HYVs, farmers are reluctant to adopt drought-tolerant varieties that have lower yield potential than their current variety in favourable years. For both situations, there is a need for highly productive rice cultivars with improved stability during moderate drought events. Prospects for combining drought tolerance with high yield potential under favourable conditions seem promising, because the genetic correlation between yield in stress and non-stress environments is usually positive (Atlin et al., 2004).
The cultivar IR64 is widely grown in irrigated areas of tropical Asia (Narciso and Hossain, 2002). It is a semi-dwarf type derived from extensive intercrossing of improved lines, and it is favoured by farmers because of its good yield potential, satisfactory grain type, and resistance to numerous biotic stresses (Khush, 1995). IR64 is moderately susceptible to drought, with yield reduction being particularly dramatic when stress occurs around flowering (Wade et al., 1999). Because of its wide adaptability and semi-dwarf habit, IR64 has been used as a parent in a range of studies designed to clarify the molecular basis of complex traits. It was used in the widely-studied doubled-haploid population with Azucena, which has been evaluated for plant height and maturity (Li et al., 2003), yield components (Courtois et al., 1995), leaf traits under drought (Courtois et al., 2000), root traits (Yadav et al., 1997), and yield under drought (Lafitte et al., 2002). IR64 was also used as a recurrent parent in the international molecular breeding programme, a project to introgress alleles from a broad range of novel sources into the recipient genome (Yu et al., 2003).
IR64 carries the sd-1 gene, which confers a semi-dwarf phenotype. The effect of sd-1 is to reduce the amount of gibberellic acid in the plant through a defect in GA-20 oxidase (Spielmeyer et al., 2002). This results in the characteristic short stature of modern improved rice cultivars, which in turn allows application of higher levels of inputs without increased lodging. Under drought, plant abscisic acid (ABA) content increases, and many GA-dependent processes are affected by GA-ABA antagonism (Yazaki et al., 2004). Plants with already low levels of GA might tend to be disproportionately affected by drought-induced ABA accumulation. This evidence, along with the observation that few semi-dwarf lowland rice varieties show appreciable levels of drought tolerance, leads to the hypothesis that the sd-1 gene may limit drought tolerance in modern lowland cultivars.
The objective of this study was to evaluate drought tolerance, measured as grain yield in water-limited field environments, in backcross lines of IR64 selected under reproductive stage drought stress. In addition, possible characteristics responsible for improved drought tolerance, such as variation in growth habit, maturity, water relations, and yield under different drought scenarios and the response to selected growth regulators at the vegetative stage, were examined.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Germplasm
A program was established to massively introgress alleles from over 100 donors, some tall and some semi-dwarf, into IR64 by backcrossing (Yu et al., 2003; Li et al., 2005). These donors were selected to represent the breadth of cultivated rice germplasm, and they are not necessarily drought-tolerant. Bulks of backcrossed progeny were subjected to selection under drought stress in the BC2F2 generation (Lafitte et al., 2006). Selected plants were advanced and over 1000 IR64-derived F4 families were evaluated under drought (386 lines screened under both lowland and upland drought, 146 screened under upland drought only, 524 under lowland drought only) in the 2002 dry season and 494 lines were re-screened under rainfed upland conditions in the 2002 wet season. On the basis of these results, a selected subset of 57 lines with superior lines derived from crosses with 28 donors was further tested for yield under contrasting water levels in 2003 (Table 1). Those lines that performed well across stress levels were further evaluated for yield in subsequent seasons, physiological traits, and seedling response to growth regulators.
|
Field experiments
All field experiments were conducted at the International Rice Research Institute (14°11' N, 121°15' E, 21 masl), as described elsewhere (Lafitte et al., 2006). Seeds were either sown directly into dry soil (upland experiments) or plants were grown in a seedbed for 18–24 d before being transplanted into the field (lowland experiments). Plant spacing in lowland experiments was 20 cm by 25 cm, with 2–3 seedlings per hill. In upland (aerobic) experiments, rows were 25 cm apart and 5 seeds were sown at 10 cm spacing within the row. Plot size ranged from 0.3 m2 to 3 m2. Fertilizer was applied as a rate of 40 kg ha–1 N, P, and K as a basal treatments, followed by a topdressing of 60 kg N ha–1 in two applications. Recommended plant weed and insect control measures were used for each experiment. Lowland control treatments received full irrigation – paddies were flooded with 2–5 cm standing water for the entire season as soon as the transplanted seedlings were established. Drought treatments experienced a period of water deficit when the soil water content declined below saturation. The degree of drought ranged from mild (aerobic soil near field capacity for most of the season) to severe. Water deficit was considered to be the primary factor limiting performance in these experiments.
Lines were assessed based on their performance relative to the recurrent parent IR64 in each environment. In 2003, 57 lines were tested. Thirty-six of these lines were selected for further evaluation in 2004 and 2005 on the basis of superior performance compared with IR64 in one or more environments in 2003. Characteristics measured included plant height, anthesis date, biomass production, and grain yield and its components. In 2005, field measurements of leaf water potential and stomatal conductance were included.
Greenhouse experiments
Seedling growth of 36 selected lines was measured at two stages: (i) shortly after germination on plexiglass slant plates with a filter paper support, and (ii) in fully heterotrophic seedlings in a hydroponic system. Seeds of each line were pregerminated on wet filter paper in a Petri dish at a constant temperature of 30 °C for 48 h before being placed on slant plates, or were allowed to develop for 72 h before being placed in foam collars over a basin of Yoshida's nutrient solution at pH 5.5. Each line was replicated two to three times in each experiment, and experiments were also duplicated in time. On the slant plates, each line was represented by 10 uniformly germinated seeds. The recurrent parent IR64 was included on each individual slant plate, and data for lines on each plate were standardized relative to IR64 in order to correct for any plate-to-plate variation. Compounds tested included abscisic acid (ABA), gibberellic acid (GA), indoleacetic acid (IAA), and EthephonTM, at rates and times as indicated in Table 2. For seedlings on the slant plates, data were collected on shoot and root length and dry mass. In older plants grown in basins, leaf number and the length of each leaf sheath were also measured.
|
A subset of 13 lines was grown in tall pots in the greenhouse during the wet season of 2004 for measurement of water use and leaf growth. Pots were 1 m tall with diameter of 20 cm, and each contained 30 kg of potting mixture (2 parts soil: 1 part aged coir dust, with 100 g complete fertilizer). Pots were oversown and thinned to one seedling. Each line was sown in eight pots. Four pots were well watered throughout the experiment. For the stress treatment, four pots were irrigated normally for 45 d, and then water was withheld until panicle emergence in the check, IR64 (
78 d after sowing). The stress treatment was rewatered and irrigation was continued until maturity. Data collected during the stress period included leaf rolling and leaf drying scores, daily water loss, relative water content (RWC; measured weekly), chlorophyll estimated with a SPAD meter (twice weekly), and stomatal conductance (weekly between 11.00–14:00 h; Li-Cor 1600). At the end of the stress period, predawn shoot water potential was measured using a Soil Moisture pressure chamber. Above-ground plant parts were harvested at the end of the stress, and tiller number, panicle number, and plant height were measured. Samples were dried and weighed. The soil column was extracted from the pot and divided into 15 cm sections to a depth of 60 cm, and a lower section from 60–90 cm. Samples were carefully washed, dried, and weighed.
Statistical analysis
All field experiments were arranged in incomplete blocks. Analysis was conducted with SAS procedure PROC MIXED, with incomplete block as a random factor (SAS, 1996). In cases where trends were observed in the field, plot location was used as a covariate. Results from locations were combined using the pattern analysis module in IRRISTAT (IRRI, 2005). The slant plate and hydroponic experiments were analysed separately for each growth regulator. Experiments were arranged as split plots with the presence or absence of the growth regulator as the main effect and line as the sub-plot treatment. The greenhouse pot experiment was also handled as a split-plot, with water level the main treatment. Data were analysed with SAS PROC GLM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Water treatments resulted in marked yield differences among experiments within a season (Table 1). Significant differences in grain yield were observed among cultivars in 12 out of the 13 experiments. The introgression lines developed using the advanced backcross approach exhibited variable responses to stress. Over two-thirds of the lines evaluated in all sites yielded significantly more than IR64 (P <0.05) in an analysis across the 13 sites (Table 3). Rice is an inbreeding crop, and in the second backcross generation a yield penalty of about 10% is expected relative to a pure line. When this adjustment is made to the yield of IR64, all selected lines yielded significantly more than IR64 across environments. Analysis of yield in the water-limited sites only (removing the four flooded lowland sites) indicated that 40% of the lines yielded significantly more than IR64. If adjustment is made for the anticipated yield penalty in BC2 lines, all but five lines out-yielded IR64.
|
Pattern analysis of the full data set indicated that the lowland environments interacted differently with genotypes than stress environments, and that wet season environments differed from dry season environments. The ordination biplot (Fig. 1) shows that genotypes interacted differently with lowland control and stress or wet season environments in 2003. In the 2004 dry season, lines interacted similarly with the well-watered upland site A4 and the lowland site L4, and this was in direct contrast to patterns of interaction observed in the stress environment S4 and the other low-yielding environments that clustered with S4 (Fig. 2). Site L5 was omitted in this analysis because the unusually low yield of IR64 in that location indicated possible disease problems. Most of the variation in the genotype-by-environment (GxE) analysis was due to environmental effects, which accounted for over 90% of the total variance. The direct effect of genotype was small (1%), and about 8% of the total variation was due to interactions between genotype and environment. When only stress environments (non-flooded) were considered, the very low-yielding environments grouped together and they did not interact strongly with genotypes; environments S3, A3, and R3 accounted for most of the GxE observed (data not shown). In the GxE analysis, group 65 was low-yielding in both all sites and in water-limited sites; group 61 was low-yielding when only water-limited environments were considered (group identification in Table 3).
|
|
In addition to yield, plant height, maturity, and tiller number were monitored in all experiments. Correlations among these traits are presented for 2003, when the number of lines tested was largest (Table 4). Correlations between yield and plant height and yield and maturity were inconsistent across those environments. Correlations between maturity and plant height were consistently negative. Tall lines yielded poorly in rainfed experiments, which were similar to environments where traditional upland rice is grown, but no significant association between yield and height was found in other trials. Early maturity was beneficial in the terminal lowland stress, but was negative in the wet season rainfed environment when stress occurred during tillering. Lines with superior yield under stress were identified with heights similar to IR64, indicating that it is possible to combine the semi-dwarf habit with improved yield under drought.
|
A set of 36 lines that differed from IR64 for yield in one or more environments was evaluated for additional characteristics in the field experiments conducted in 2005. In the flooded control treatment, many lines out-yielded IR64. The main yield component affected in IR64 in the control plots was spikelet fertility (data not shown). Superior yield of lines was seldom observed in lowland control treatments in 2003 and 2004, and may reflect disease pressure in 2005 because the base, IR64 is quite susceptible to rice tungro virus. Because the effects of tungro were not seen until quite late in this experiment, however, it was still possible to ascertain that introgression lines did not differ from IR64 in tiller number, and that only three lines flowered slightly but significantly earlier (2–5 d) than IR64 in control conditions in this season (data not shown). In water-limited conditions, a number of additional differences were observed between IR64 and the selected introgression lines. For example, in the upland field, half of the lines produced more tillers than IR64, even though no differences in this trait were observed in the flooded control (Table 3). In that experiment a third of the lines flowered later than IR64, while no lines showed later maturity in the flooded field. In the upland experiment, 30 of the lines tested showed significantly more leaf rolling (measured just prior to flowering, 10 d after irrigation) than IR64, 24 lines had lower stomatal conductance, and 14 lines maintained greater SPAD readings than IR64. No consistent differences among lines were observed for leaf relative water content or leaf water potential; several lines exhibited greater values than IR64 and several had lower values.
On the slant plates, a range of seedling responses to growth regulators were observed. Relative to the effect of each chemical on IR64, 10 lines showed significantly less effect of ABA on shoot growth, eight lines showed less effect of GA on root elongation, and in the case of IAA, the chemical increased shoot length in four lines and decreased it in seven lines, while there was no effect on IR64 (Table 3). These results suggested some difference between the lines and IR64 in terms of sensitivity to growth regulators, but the results were quite inconsistent. The only association observed with interaction groups was that shoot growth was less affected by ABA in three of the four lines in group 56.
In the hydroponic system, ABA reduced root length of young plants more in the selected lines than in IR64. The root:shoot ratio was thus largely unaffected in the lines, but it increased in IR64, leading to significant differences from IR64 in the response of the root:shoot relationship to ABA in 29 of the lines (Table 3). This was the most consistent difference observed in the selected lines. Ethephon also affected growth of the lines differently in hydroponics. This chemical tended to increase tiller number in IR64, but in 15 of the selected lines, the effect was significantly less. Concomitantly, total shoot weight of IR64 decreased with the application of ethephon, but seven of the lines were even more sensitive than IR64 for this trait. These results suggest differences in the sensitivity of some selected lines to ethylene. These differences were not associated with interaction groups.
The subset of 14 lines evaluated in the greenhouse did not exhibit consistent differences in water use. Some lines differed from IR64 in root growth below 15 cm when the plants were stressed, but the responses were not consistent (data not shown). Three lines produced significantly greater root mass below 15 cm with stress, while seven lines produced less root mass. Root growth in IR64 and the remaining lines was not significantly affected by the drought treatment.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
It was possible to improve yield across a range of environments by selecting for seed set under terminal stress in BC2 populations and then screening under managed drought stress for grain yield. Dual goals of good yield potential in flooded environments and yield under drought were not incompatible in these materials. Some of the IR64-type lines that yielded better under stress did exhibit a loss in yield potential in favourable environments, but other lines were found to have equal or even better performance than IR64 in favourable sites. Thus, improving drought tolerance in a background of a semi-dwarf modern variety is not necessarily linked to losses in yield potential.
The interaction between line and water environment was large, confirming that testing across a range of water levels is essential to identify superior lines. Because of the limited interaction of genotypes with environments in the sites with very low yields (C3, V3, C4, and S4), it is concluded that check yields in a testing environment need to be greater than 0.5 t ha–1 to evaluate line differences effectively. Crossover interactions, which result in a re-ranking of genotypes, were observed mostly above the yield level of 2.5 t ha–1.
There was no consistent relationship between the donor genotype and the interaction of line with environment. Molecular markers are being used to pinpoint the introgressions that lead to the differences in performance in selected lines (Li et al., 2005). Preliminary analysis indicates that many QTL with small effects underlie the observed yield differences, and that different segments from a given donor may be present in different introgression lines. The relationship between yield and maturity differed among experiments, and there was no evidence to support the idea that lines with greater drought tolerance tended to be earlier or later than IR64.
It was difficult to identify a clear physiological mechanism associated with improved yield under drought. There was no evidence of improved water status in the selected lines measured as improved water potential or RWC in the upland field experiment or the greenhouse pot experiment. No consistent variation was observed in water use or root characteristics in the subset of lines evaluated in pots in the greenhouse. This study's finding that root traits did not appear to change with selection are not inconsistent with the QTL evidence, which has ample examples of root QTL that are apparently unrelated to yield QTL and vice-versa. QTL have been identified in an IR64 population for traits related to root depth and distribution (Shen et al., 2001; Yadav et al., 1997), but these have not been related to grain yield under stress (Lafitte et al., 2002). In another upland-by-lowland mapping population, one of the many reported root-related QTL did coincide with grain yield (Babu et al., 2003). In a third example, however, in an upland-by-upland population, yield QTL did not associate with root QTL reported in previous studies (Lafitte et al., 2004). The most consistent changes observed in selected lines in the present study were the effect of ABA on the ratio of root to shoot length, measured in the vegetative stage in the hydroponic system, and leaf rolling, measured after 10 d of water exclusion in the upland field. QTL have been identified for leaf rolling with drought in the field in the IR64/Azucena mapping population (Courtois et al., 2000), but it is generally a trait that is negatively correlated with yield. This suggests a different mechanism underlying leaf rolling in that population, perhaps associated with poor rooting, which was not important in these drought-selected introgression lines.
Improved grain yield under drought was observed in lines carrying the sd-1 gene, as demonstrated by the lack of correlation between yield and plant height. This supports the hypothesis that improved drought tolerance is compatible with the semi-dwarf plant type. The biological function of the sd-1 gene is to reduce the amount of gibberellin in the plant (Spielmeyer et al., 2002), which could have an impact on drought tolerance through interactions with ABA (Yazaki et al., 2004). It is intriguing that the most consistent difference between lines and IR64 in response to growth substances was the lack of change in the root:shoot length in the presence of ABA. It is possible that our selection protocol favoured insensitivity to ABA, a mechanism that is compatible with grain yield under stress in the low-GA sd-1 phenotype. Insensitivity to ABA could also have led to the observed phenomenon of greater leaf rolling and superior tiller development in selected lines in the upland environment. These results also suggested a possible alteration in ethylene response in the selected lines; this could also be related to interactions with ABA (Tanaka et al., 2005). These results are consistent with the hypothesis that plant response to drought, or adaptation, is a key element of stress tolerance (Trewavas, 2003). This type of mechanism may be more suitable than constitutive structural changes for drought tolerance in crop plants, where the goal is to stabilize yield in poor years while maintaining yield potential in favourable ones.
It was possible to improve drought tolerance incrementally in a broadly adapted, popular lowland rice cultivar through selection under managed drought stress. This was achieved using donors that were not considered drought-tolerant (Lafitte et al., 2006). While it was possible to improve yield significantly through this process, the level of tolerance achieved still falls considerably below that observed in cultivars developed for aerobic or drought-prone areas (Lafitte et al., 2002). Use of highly tolerant, agronomically superior donors might result in greater gains than were achieved in these studies (Atlin, 2003). The benefit to farmers of adopting a drought-tolerant cultivar depends on the frequency of drought years, the advantage of the tolerant cultivar in drought years, and any yield penalty associated with the drought-tolerant cultivar in favourable years. The frequent crossovers observed in the range of 2 t ha–1 indicates that for environments where drought frequently reduces yield below that level, it may be more suitable to use other strategies to develop cultivars with tolerance to more extreme drought.
|
Acknowledgements |
|---|
The authors gratefully acknowledge support from the Rockefeller Foundation. Excellent technical support was provided by R Torres, L Holongbayan, N Turingan, and G Dimayuga.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Atlin G. (2003) Improving drought tolerance by selecting for yield. In Fischer K (Ed.), et al. Breeding rice for drought-prone environments IRRI pp. 14–22.
Atlin G, Lafitte R, Venuprasad R, Kumar R. (2004) Heritability of rice yield under reproductive-stage drought stress, correlations across stress levels, and effects of selection: implications for drought tolerance breeding. Resilient crops for water-limited environments. Abstracts of an international workshop. 24–28 May 2004, Cuernavaca, MexicoMexico CIMMYT.
Babu RC, Nguyen BD, Chamarerk V, et al. (2003) Genetic analysis of drought resistance in rice by molecular markers: Association between secondary traits and field performance. Crop Science 43 1457–1469.
Courtois B, Huang N, Guiderdoni E. IRRI. (1995) RFLP mapping of genes controlling yeild components and plant height in an indicaxjaponica doubled haploid population. The international rice research conference , et al. Los Baños, Philippines IRRI 963–976.
Courtois B, McLaren G, Sinha PK, Prasad K, Yadav R, Shen L. (2000) Mapping QTLs associated with drought avoidance in upland rice. Molecular Breeding 6 55–66.
International Rice Research Institute. (2005) IRRISTAT for WindowsManila, Philippines International Rice Research Institute.
Khush G. (1995) Modern varieties: their real contribution to food supply and equity. GeoJournal 35 275–284.[CrossRef]
Lafitte H, Li ZK, Vijayakumar C, et al. (2006) Improvement of rice drought tolerance through backcross breeding: evaluation of donors and selection in drought nurseries. Field Crops Research 97 77–96.[CrossRef]
Lafitte HR, Courtois B, Arraudeau M. (2002) Genetic improvement of rice in aerobic systems: progress from yield to genes. Field Crops Research 75 171–190.[CrossRef]
Lafitte HR, Price AH, Courtois B. (2004) Yield response to water deficit in an upland rice mapping population: associations among traits and genetic markers. Theoretical and Applied Genetics 109 1237–1246.[CrossRef][Web of Science][Medline]
Li ZK, Fu BY, Gao YM, et al. (2005) Genome-wide introgression lines and their use in genetic and molecular dissection of complex phenotypes in rice (Oryza sativa L.). Plant Molecular Biology 59 33–52.[Medline]
Li ZK, Yu SB, Lafitte HR, et al. (2003) QTLxenvironment interactions in rice. I. Heading date and plant height. Theoretical and Applied Genetics 108 141–153.[CrossRef][Web of Science][Medline]
Narciso J and Hossain M. (2002) World Rice Statistics [Online]. Available by IRRI http://www.irri.org/science/ricestat (posted 2002).
SAS Institute. (1996) SAS system for Windows NT. Release 6.12Cary, NC, USA SAS Institute.
Shen L, Courtois B, McNally KL, Robin S, Li Z. (2001) Evaluation of near-isogenic lines of rice introgressed with QTLs for root depth through marker-aided selection. Theoretical and Applied Genetics 103 75–83.[CrossRef]
Spielmeyer W, Ellis M, Chandler P. (2002) Semidwarf (sd-1), ‘green revolution’ rice, contains a defective gibberellin 20-oxidase gene. Proceedings of the National Academy of Sciences, USA 99 9043–9048.
Tanaka Y, Sano T, Tamaoki M, Nakajima N, Kondo N, Hasezawa S. (2005) Ethylene inhibits abscisic acid-induced stomatal closure in Arabidopsis. Plant Physiology 138 2337–2343.
Trewavas A. (2003) Aspects of plant intelligence. Annals of Botany 92 1–20.
Wade LJ, McLaren CG, Quintana L, et al. (1999) Genotype by environment interactions across diverse rainfed lowland rice environments. Field Crops Research 64 35–50.[CrossRef]
Yadav R, Courtois B, Huang N, McLaren G. (1997) Mapping genes controlling root morphology and root distribution on a doubled-haploid population of rice. Theoretical and Applied Genetics 94 619–632.[CrossRef]
Yazaki J, Shimatani Z, Hashimoto A, et al. (2004) Transcriptional profiling of genes responsive to abscisic acid and gibberellin in rice: phenotyping and comparative analysis between rice and Arabidopsis. Physiological Genomics 17 87–100.
Yu SB, Xu WJ, Vijayakumar CHM, et al. (2003) Molecular diversity and multilocus organization of the parental lines used in the International Rice Molecular Breeding Program. Theoretical and Applied Genetics 108 131–140.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Saito and K. Futakuchi Genotypic Variation in Epidermal Conductance and its Associated Traits among Oryza sativa and O. glaberrima Cultivars and their Interspecific Progenies Crop Sci., December 30, 2009; 50(1): 227 - 234. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||