Akihiko Kamoshita

Mapping QTLs for root morphology of a rice population adapted to rainfed lowland conditions

Abstract.In the rainfed lowlands, rice (Oryza sativa L.) develops roots under anaerobic soil conditions with ponded water, prior to exposure to water stress and aerobic soil conditions that arise later in the season. Constitutive root system development in anaerobic soil conditions has been reported to have a positive effect on subsequent expression of adaptive root traits and water extraction during progressive water stress in aerobic soil conditions. We examined quantitative trait loci (QTLs) for constitutive root morphology traits using a mapping population derived from a cross between two rice lines which were well-adapted to rainfed lowland conditions. The effects of phenotyping environment and genetic background on QTLs identification were examined by comparing the experimental data with published results from four other populations. One hundred and eighty-four recombinant inbred lines (RILs) from a lowland indica cross (IR58821/IR52561) were grown under anaerobic conditions in two experiments. Seven traits, categorized into three groups (shoot biomass, deep root morphology, root thickness) were measured during the tillering stage. Though parental lines showed consistent differences in shoot biomass and root morphology traits across the two seasons, genotype-by-environment interaction (G×E) and QTL-by-environment interaction were significant among the progeny. Two, twelve, and eight QTLs for shoot biomass, deep root morphology, and root thickness, respectively, were identified, with LOD scores ranging from 2.0 to 12.8. Phenotypic variation explained by a single QTL ranged from 6% to 30%. Only two QTLs for deep root morphology, in RG256-RG151 in chromosome 2 and in PC75M3-PC11M4 in chromosome 4, were identified in both experiments. Comparison of positions of QTLs across five mapping populations (the current population plus populations from four other studies) revealed that these two QTLs for deep root morphology were only identified in populations that were phenotyped under anaerobic conditions. Fourteen and nine chromosome regions overlapped across different populations as putative QTLs for deep root morphology and root thickness, respectively. PC41M2-PC173M5 in chromosome 2 was identified as an interval that had QTLs for deep root morphology in four mapping populations. The PC75M3-PC11M4 interval in chromosome 4 was identified as a QTL for root thickness in three mapping populations with phenotypic variation explained by a single QTL consistently as large as 20–30%. Three QTLs for deep root morphology were found only in japonica/indica populations but not in IR58821/IR52561. The results identifying chromosome regions that had putative QTLs for deep root morphology and root thickness over different mapping populations indicate potential for marker-assisted selection for these traits.

Genotypic Variation in Response of Rainfed Lowland Rice to Drought and Rewatering : II.Root growth

Abstract Genotypic variation in the root system is a potential source for improving drought tolerance of rainfed lowland rice (Oryza sativa L.). Our work aimed at characterizing both constitutive root traits (those present under well -watered conditions) and adaptive root traits (those developed in response to drought and rewatering) among eight diverse rice genotypes in three sets of greenhouse experiments (experiments 1, 2, and 3). Under well–watered conditions, genotypic variation was observed in root to shoot ratio, root growth rate, specific root length, deep root ratio, root mass per tiller, and root thickness. CT9993 and IR58821 had a high root to shoot ratio, deep and thick root system, and high root mass per tiller. However, CT9993 had a slow root growth rate and short specific root length. In response to drought in experiments 2 and 3, the total amount of assimilate distributed to roots was reduced and roots became thinner, but the proportion of total assimilate supply assigned to deeper layers increased, thereby maintaining deep root mass and increasing specific root length. On rewatering, root to shoot ratio increased, surface roots increased, and roots became thicker. During drought, NSG19, KDML105, Mahsuri, and IR58821 partitioned a larger proportion of assimilate to deep roots and had more deep root branching.

Genotypic Variation in Response of Rainfed Lowland Rice to Drought and Rewatering : I.Growth and water use

Abstract The lack of information on the dynamics of crop growth and water use has limited the capacity for indirect selection through physiological traits that confer drought tolerance in rainfed lowland rice (Oryza sativa L.). Shoot growth and transpiration in response to drought and rewatering were studied among eight diverse rice genotypes in three sets of pot experiments : one under severe stress development after panicle initiation (average transpiration of 15.0 mm d–1 ; experiment 1), and two under slow and progressive stress development during tillering (average transpiration of 2.1 and 7.6 mm d–1 in experiments 2 and 3, respectively). Higher transpiration generally caused a higher plant growth rate in two periods, though there was some contribution of water use efficiency. The first period was soon after ponded water was drained and watering withheld (early drought phase), when soil became aerobic, soil water was still readily available, and transpiration continued at a rate comparable to the well–watered treatment. The second period was after rewatering when transpiration was again not limited by soil water supply. In experiment 1, the effect of plant size before stress imposition was large and genotypic variation for response to drought and rewatering was small, except for KDML105, which tended to show smaller growth during drought and had a more rapid recovery after rewatering. During early drought phase in experiments 2 and 3, genotypes differed in relative amounts of tiller and leaf area production compared with the well-watered treatment. Genotypes with high seedling vigor before stress imposition and during the early drought phase, such as NSG19, KDML105, Mahsuri and IR58821, produced greater root length during the following more severe drought period and had a larger green leaf biomass at the end of the drought period in experiment 3. In these genotypes, transpiration increased sharply and leaf area expanded rapidly after rewatering, which caused superior drought recovery.

Genotypic Variation in Response of Rainfed Lowland Rice to Prolonged Drought and Rewatering

Abstract Duration of the drought period is important for plant response during drought and after rewatering. We hypothesized that, if drought duration is extended, (1) high seedling vigor and rapid development of a deep root system will not be advantageous, and (2) osmotic adjustment will be more important. Six diverse rice (Oryza sativa L.) genotypes were selected from rainfed lowland germplasms to examine the development of a deep root system and osmotic adjustment, and their relationship with biomass production during drought and after rewatering, under two different drought durations (shorter and prolonged) in the greenhouse. NSG19 and KDML105 had greater seedling vigor (larger seedling biomass), developed a deep root system earlier in response to drought, extracted soil water more quickly, and their pre-dawn leaf water potential declined more rapidly during the prolonged drought period. These two genotypes showed superior drought recovery even after a prolonged drought period in which they suffered a greater reduction in transpiration, water use efficiency, and biomass production. The superior recovery ability was associated with larger plant size by the end of the drought period rather than with plant water status during drought, such as osmotic adjustment or leaf water potential. Osmotic adjustment was greater during prolonged drought periods (ca. 0.7 MPa) than during shorter drought periods (ca. 0.5 MPa), and lower osmotic adjustment was mostly associated with a higher leaf water potential. Genotypic variation in osmotic adjustment was observed, but there was no clear relationship between osmotic adjustment and biomass production during drought periods. These patterns of response of rice seedlings to drought and rewatering in the greenhouse should help to explain the patterns of adaptation of rainfed lowland rice in the field. Selection for drought recovery ability should be an advantageous strategy for early season drought.

Genotypic Variation in Response of Rainfed Lowland Rice to Drought and Rewatering. III. Water extraction during the drought period

Abstract Soil water extraction was examined in relation to root system development and leaf osmotic adjustment in a pot experiment with eight rice genotypes (Oryza sativa L.). The time course of cumulative soil water extraction from layers between 5 and 45 cm estimated by time domain reflectometry (TDR) was similar to that of cumulative transpiration estimated from pot weighing for each genotype. The level of the TDR-estimated water extraction was 75% of the cumulative transpiration, and their coefficient of determination between the two was 96%. There was a 5-day difference (18 to 23 d) in transpiring 4 kg of water from the pots among genotypes, and the variation in daily transpiration rate was related to the extraction rate in the top-20 cm soil layers during the early half of the drought period, and to the extraction rate in the below-30 cm soil layers during the latter half of the drought period. The extraction rate in the subsoil was positively correlated with the average root length density at the corresponding depth during the latter half of the drought period, explaining 66% and 58% of the variation around the 30 cm and 40 cm depth, respectively. Mahsuri and IR58821 had higher water extraction rate from these subsoil layers than IR20 and IR62266 during the late drought period. Osmotic adjustment was higher in the genotypes that had a slower rate of transpiration and a lower pre-dawn leaf water potential at the end of the drought period. Among genotypes that extracted water rapidly, KDML105 had the highest osmotic adjustment. IR58821 and CT9993, known to have a deep and thick root system under well-watered conditions, had the lowest levels of osmotic adjustment. This study demonstrated under the simulated rainfed lowland conditions genotypic variation in water extraction rate from the deep soil layers during the late drought period, which was primarily related to proliferation of roots in these layers.

Genotypic Variations in Response of Lateral Root Development to Fluctuating Soil Moisture in Rice

Abstract Developmental plasticity in lateral roots may be one of the key traits for the growth of rice plants under soil moisture fluctuations. We aimed to examine responses in seminal root system development to changing soil moisture for diverse rice cultivars. Special attention was paid to the two different types of lateral roots ; the generally long, thick L type capable of branching into higher orders, and the non-branching S type. Plants were grown in half-split polyvinyl chloride tubes fixed with transparent acrylic plate for root observation under glasshouse conditions. When plants were grown first under drought conditions, then rewatered, the seminal root system development in terms of dry weight and total length was promoted as compared with plants grown under continuously well-watered conditions in IR AT 109 and Dular, drought tolerant cultivars. Promoted production of L type lateral roots mainly contributed to the development of the longer seminal root system. Plants exposed to soil submergence before they were grown under drought conditions did not show such promoted responses in these two cultivars. However, in KDML 105, a drought tolerant cultivar, the production of especially L type laterals was substantially promoted under drought and rewatered conditions. Honenwase was characterized by the shallow root system and great reduction in root system length when soil moisture becomes limited. These facts show that genotypic variations exist in the plastic response of rice seminal root system and that the L type lateral root plays a key role in manifestation of this plasticity.

Dry Matter Production and Root System Development of Rice Cultivars under Fluctuating Soil Moisture

Abstract Rice plants in the rainfed areas are mostly grown under fluctuating soil moisture. We examined responses in dry matter production, root development and water use to changing soil moisture in diverse rice cultivars. Rice plants were grown in polyvinyl chloride tubes under glasshouse conditions. Progressive drought right after planting greatly inhibited the shoot dry matter production, tiller development, nodal root development and water uptake in all cultivars tested. When the plants experienced soil submergence before being exposed to drought, all the cultivars exhibited higher dry matter production than their well-watered counterparts. Cultivar differences were clearly noted in the growth responses to rewatering after these plants were droughted. With well–watered control as basis, IRAT 109 and KDML 105 plants increased efficiency in converting available dry matter to increase their total root length by means of enhanced lateral root development. In the latter, however, the dry weight of roots also increased and so did root water uptake. In Dular, droughted plants did not show a clear response in terms of root development and water uptake to rewatering while its shoot growth was much more severely inhibited than the other cultivars. These findings suggest that phenotypic plasticity in the root system structure exhibited by promoted lateral root development and new nodal root production play a key role in the growth of rice under changing moisture level in the soil.

Evaluating the resistance of six rice cultivars to drought: restriction of deep rooting and the use of raised beds

Soil water deficits reduce rice (Oryza sativa L.) productivity under upland field conditions. In this study, we constructed screening facilities to evaluate the performance of rice cultivars under drought conditions and to assess the roles of deep roots. Two experiments were conducted with six rice cultivars, including drought-tolerant and drought-susceptible cultivars, grown in two root environments: a root-restricted treatment that restricted rooting depth with water-permeable sheets, and a raised bed that reduced water availability in the surface soil by inserting a gravel layer between the topsoil and subsoil layers to interrupt capillary transport of water. In the root-restricted treatment, in which root growth was restricted to the surface 25-cm layer, leaf water potential decreased faster in cultivars with a large canopy during drought stress, and there was little difference in panicle weight among cultivars. With a normal (unrestricted) root environment, the deepest-rooting cultivar (‘IRAT109’) maintained higher leaf water potential during drought, although panicle weight under drought stress was affected by yield potential as well as by deep rooting. Under the intermittent drought stress in the raised bed, deep-rooting cultivars accumulated more nitrogen and produced more biomass, and the difference in panicle weight between deep-rooting drought-tolerant and shallow-rooting drought-susceptible cultivars was magnified by the raised bed compared with the yield differences under drought in a normal root environment. These results demonstrate that the drought screening facilities we developed can help to identify superior cultivars under upland field conditions without time-consuming measurement of deep root systems.

Growth of Three Rice Cultivars (Oryza sativa L.) under Upland Conditions with Different Levels of Water Supply

Abstract Upland rice production has great potential as a water-saving form of agriculture if yield can be increased and stabilized across a range of environments with different levels of water supply. The objective of this study was to clarify the effects of water supply and plant characteristics on grain yield of rice (Oryza sativa L.) grown under upland conditions. We compared grain yield (ranging from 346-685 g m-2) and yield components of three rice cultivars (‘Yumeno-hatamochi’, YHM; ‘Lemont’, LMT; ‘Nipponbare’, NPB) grown under upland conditions with three water regimes (rain-fed, RU; irrigated, IU; and water deficit during the panicle-formation stage, WD) with those of rice grown under flooded lowland (FL) conditions (ranging from 394-649 g m-2) from 2001 to 2003 at Nishitokyo, Japan. Grain yield and each yield component of NPB in RU were comparable to those in FL when there was ample rain during the 40 days before heading in 2003. However, grain yield of NPB decreased with decreasing water supply during the period of 20-40 days before heading under upland conditions (r = 0.93) as a result of reduced number of spikelets per unit area and reduced harvest index. Water productivity (grain yield per unit water supply) in rice in RU and IU ranged from 0.43 to 1.05 kg m-3 in the three cultivars across the 3 years, and was more than twice the corresponding value in FL. We found a cultivar – water regime interaction for grain yield within each year and a cultivar × environment interaction across all the 5 upland conditions in 2002 and 2003. In FL, NPB and LMT had higher yields than YHM, while LMT had the highest yield under all upland conditions and NPB grain yield under the suboptimal upland environments (i.e. RU and IU in 2002) decreased to the largest extent compared with that under optimal upland environment, i.e. IU in 2003 among the three cultivars. The reasons for the highest grain yield of LMT across upland conditions were maintenance of large panicle and high harvest index. Maximum yield was lowest in YHM. In WD, yield potential and growth recovery, rather than crop growth during water stress, affected the cultivar ranking in terms of grain yield. We conclude that water supply during panicle development is important for maintenance of high yield and that a high potential yield and harvest index, as well as yield stability under different water regimes, are important putative plant characters for developing new elite varieties for water-saving upland rice production.

Effect of Planting Density on Grain Yield and Water Productivity of Rice (Oryza sativa L.) Grown in Flooded and Non-flooded Fields in Japan

Abstract The effect of planting density on grain yield and water productivity was evaluated in rice (Oryza sativa L.) grown in non-flooded lowland fields in Japan in comparison with flooded fields. One rice cultivar, IR24 was grown both in flooded and non-flooded lowland fields in 2001 and 2002, and only in flooded field in 2003, with different planting densities ranging from 5.6 to 44 hills m-2. Another rice cultivar, Dontokoi was also grown in 2001. Straw mulching treatment was added in non-flooded field in 2002. In non-flooded fields, standing water disappeared from 36 and 8 days after transplanting until maturity in 2001 and 2002, respectively, the mean water content of surface soil during non-flooded period was 72 % g g-1 on a dry basis and 63 % v v-1. Grain yield in flooded fields (637 g m-2; average of 2001 and 2002) was higher than that in non-flooded fields (467 g m-2; average of 2001 and 2002), due to larger spikelet number per panicle in both years, larger 1000 grain weight in 2001, and higher percentage of ripened grains in 2002. Straw mulching tended to increase sink size but reduced percentage of ripened grains, resulting in no yield advantage in 2002. Water productivity in non-flooded fields (0.34 kg m-3; average of 2001 and 2002) was significantly higher than that in flooded fields (0.14 kg m-3). Grain yield increased with higher planting density in flooded fields in 2001 and 2003. In non-flooded fields, however, the effects of planting density on grain yield were little or marginal in both cultivars, due to the trade-off relationship between panicle number and spikelet number per panicle. This study showed that higher planting density would result in higher grain yield in favourable flooded fields, but is not advantageous for higher grain yield under non-flooded lowland fields in Japan in improved cultivars with relatively high tillering and yielding abilities.