In rice, gene expression profiling under different stresses have been carried out earlier through microarrays and SAGE (Kawasaki et al., 2001; Rabbani et al., 2003,;. Yazaki et al., 2003, 2004; Walia et al., 2005; Zhou et al., 2005; Jiao et al., 2005; Zhou, 2007; Wang, et al., 2007)). The results reported so far were mostly derived from different stress treatments, array formats, species, tissue types, and time courses, which make it difficult for direct comparison among studies. Also, most of the stress experiments on expression profiles were conducted under laboratory, growth chamber or field grown plants subjected to stress in laboratory conditions, usually by inducing stress with exogenous compounds such as abscisic acid (ABA) or gibberellin (GA) genes (Yazaki et al., 2003, 2004; Rabbani et al., 2003), salt (Kawasaki et al., 2001; Rabbani et al., 2003; Walia et al., 2005), organ specific gene expression under high salinity (Jiao et al., 2005) and drought (Zhou et al., 2007), comparison with different cultivars (Wang et al., 2007) and other environmental stresses such as cold and drought (Rabbani et al., 2003; Zhou et al., 2005).
However, such laboratory, growth chamber grown plants or field grown plants subjected to stress under laboratory conditions significantly differ from the field stress conditions and therefore the resultant plant response may not truly represent the actual stress conditions prevailing on field. This has been extensively reviewed by Mittler, 2006 and the need for utilizing data from multiple stress experiments for developing plants resistant to stress has been focused. The review also focuses on evaluating farmers experiences in stress management and kinds of stress that a plant is subjected prior to taking up research on stress resistances has been of significant importance. In the present study, to bridge the gap between the available data on drought stress and the gene regulation under field drought conditions has been paid attention. The outcome of this data will give new leads for breeders to design strategies to develop plants resistant to drought without much loss in yield. In the present study experiments were conducted under different field capacities using a rainout shelter to minimizing the effects of drought induction due to rains. The microarray analysis is thus expected to throw light on stress responsive gene expression in field drought conditions. Till now, to our knowledge there have been no reports on gene expression in rice under field drought conditions. Further, there is no data available on gene expression profiling during various stages of plant growth in leaf and panicle tissues under different levels of water stress. Here, we have focused on generating stress responsive gene (SRG) expression profiles in field drought conditions through the entire life cycle of rice plant.
Optimal drought simulation was achieved by conducting a drought stress experiment and validated by a series of dry runs. The main reason for evaluating gene expression under drought in field condition is as drought itself is not limited to water stress but various other factors, such as heat, humidity, hot winds, and long day lengths effect the gene regulation. All these are likely to trigger common pathways that are also a part of drought stress response and therefore significantly affect the expression profiles.
In the present study, field drought stress was initiated from 31 days after germination and transplantation to the field at 100%, 60%, 40% and 15% field capacities designated as A1, A2, A3, and A4. It took 7 days for the onset of the actual drought for the experimental plots to reach to their designated field capacities from 100% FC. Samples were collected from 7 days after drought stress to 33 days after drought stress to evaluate different drought stress regimes based on the developmental stages of the plant. At all the tested stress levels the data points were Hierarchically grouped and clustered sets of genes were examined and an approximate number of clusters to be formed were estimated. Finding a group of genes showing a similar pattern of gene expression by means of expression data from microarray analysis is the basis for identification of genes and pathways controlling a specific trait. We have followed unsupervised classification and clustered the genes hierarchically after normalizing expression data to obtain a set of co-regulated genes. Analysis of data revealed that there are at least 3 to 4 clusters which show differential regulation under each stress regime. The annotated genes falling in these clusters interestingly include a number of novel genes from N22 library along with some known candidate genes for drought tolerance which were earlier identified in our study (Markandeya et al., 2007). Expression pattern of these genes from 7 days after drought stress to 33 days after drought stress with their fold changes were evaluated in the present study. In particular, it is observed that many genes of unknown function are also co-regulated in an interesting pattern along with known genes.
Overall design: Drought Stress Induction Under Field Conditions:
Simulation of drought under field conditions for gene expression profiling was done by constructing an experimental field plot with a rain out shelter. The experimental plots were constructed by excavation of soil up to 5 feet deep of 13 feet width and 16 feet length. A sand bed of 1 foot was laid across the length in a slanting manner such that the end of the plot has just 1 inch sand to allow drain out and prevent seepage. Eight plots (3 feet X 4 feet) were partitioned and constructed with concrete walls with water-proof agents to avoid seepage of water across plots. The inner walls were of 5 feet and the outer walls facing the exterior soil apart from the black clay soil were of 3 feet in depth. Below this, 3 feet wall, 6 inch sand was put until the base (5 feet) runs into the bottom 1 foot sand. Also, these blocks are raised enough to avoid accidental overflow from the control plots. These plots were filled to 5 feet depth with soil mixture in which the rice crop is usually cultivated with the recommended amount of farm yard manure. Rainout shelter was constructed to cover the plots during rainy days to keep the plants from rain during experiment. The drought stress induction was monitored through high-precision digital irrigation controllers (Netafim Inc, Israel). Irrigation was given taking plot A1 as 100% field capacity, which is a control plot in this experiment, and irrigation was set up in the other plots with moisture probes/timer controllers.
Rice Seedlings Growth, Transplantation and Stress Treatments Under Field Condition:
Rice seeds were surface sterilized, washed and germinated in a growth chamber. Fifteen-day-old seedlings were transplanted in the field for drought stress induction. The transplanted seedlings were grown in 100% field capacity with 2cm standing water until day 30. Field drought stress was initiated from day 31 at 100%, 60%, 40% and 15% field capacities designated as A1, A2, A3, and A4, respectively. It took 8 days for the onset of the actual drought for the experimental plots to reach to their designated field capacities from 100% FC. Samples were collected at various stages and drought stress regimes based on the developmental stages of the plant.
Phenotypic Response to Field Drought Stress:
Rice seedlings were subjected to field drought stress at various Field Capacities (FC), i.e., 60% FC (A2), 40% FC (A3) and 15% FC (A4) compared to that of the control grown at 100% FC (A1), through controlled irrigation of experimental plots. This protocol allowed us to study the response of rice plants to the desired stress levels. The field drought stress experimental set up is basically a rain out shelter with an electronically controlled irrigation system. Drought stress significantly reduced plant height, biomass and yield. Considerable variation was observed in the phenotypic response to water stress across the various stress regimes. The Relative Water Content (RWC) of the plants, measured in leaf samples periodically across the developmental stages, was in the range of 92 to 28% at various stress levels and periods.
Under mild stress (A2), plants showed, on an average, 7.5 cm (~6%) reduction in plant height by 63 DAG (Days After Germination). Similarly, there was a significant reduction (25.8%) in tiller number and about 5-fold spikelet sterility leading to a 7% yield loss, compared to that of the control. Under moderate stress conditions (A3), there was a reduction in plant height (18%), tillering (45%) and a high percentage of seed sterility (60%), leading to an yield loss of about 54%. Although milking was observed in the initial stages in all the seeds, seed setting was drastically poor, leading to a significant reduction in the yield. At about 70 DAG, the stressed seedlings started begin to lodge and eventually get desiccated by 75 DAG. Under severe stress (A4), booting was three days earlier on an average, which is a characteristic drought escaping mechanism stressing rice. Other observed drought stress responses include (a) a reduction in RWC (28%), (b) a reduction in tiller number (75.8%), and (c) stunted growth (40% reduction) compared to the control. There was complete spikelet sterility resulting in total yield loss. The plants began to lodge on 67 DAG and completely desiccated by 70 DAG.
Monitoring Expression Profiles of Stress Responsive Genes:
To monitor gene expression profiling, the target samples were prepared from field drought stressed plants under A2, A3 and A4 stresses at vegetative (DAG 45, 52), pre-anthesis (DAG 55), post-anthesis and seed setting stages (DAG 63, 67 & 71). A sampling of 5 leaves per plot were used to obtain high-quality RNA samples for labeling and hybridization for all the above targets.
Set A1 is not included in this submission.
Less...