Agronomy Journal Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 11 April 2006
Published in Agron J 98:536-544 (2006)
DOI: 10.2134/agronj2005.0139
© 2006 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kobata, T.
Right arrow Articles by Ida, K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kobata, T.
Right arrow Articles by Ida, K.
Agricola
Right arrow Articles by Kobata, T.
Right arrow Articles by Ida, K.
Related Collections
Right arrow Rice

Rice

Critical Factors for Grain Filling in Low Grain-Ripening Rice Cultivars

Tohru Kobataa,*, Tomonori Naganob and Kumi Idac

a Faculty of Life and Environmental Science, Shimane Univ., 1060 Nisikawatu-cho, Matsue 690-8504, Japan
b 7-Okita-Ken-Shokuin Syukusha, 20-35 Higashi-cho, Masuda, 698-0004, Shimane, Japan
c 178-3 Umakata-cho, Matsue 690-0024, Japan

* Corresponding author (kobata{at}life.shimane-u.ac.jp)

Received for publication May 12, 2005.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Some improved rice (Oryza sativa L.) cultivars such as the new plant type (NPT) lines of IRRI or Japanese indica x japonica cultivars can attach a greater density of spikelets, and hence have higher yield potential (YP); however, low grain ripening means that their higher YP is not fully used. Our objective was to clarify why their grain ripening is lower than in other cultivars, and whether low assimilate supply to grains or insufficient sink capability of spikelets restricts grain ripening in these cases. IR65564-44-2-2 (NPT) and Akenohoshi (indica x japonica) cultivars yield lower percentages of ripened grain (PRG) and have lower spikelet filling percentages (F%) than other cultivars in field trials. When plant density was halved during the grain-filling period (GFP) to improve plant dry-matter increase (DMI), PRG and F% in Akenohoshi increased, but scarcely changed in IR65564-44-2-2 and other cultivars. Inferior assimilate supply to YP limited the PRG in most cultivars due to close relationships between PRG or F% and DMI/YP; however, in IR65564-44-2-2 increased DMI/YP did not improve PRG and F%. Among the higher YP cultivars, F% in IR65564-44-2-2 started to decrease earlier with accumulated temperature. When panicles detached during the early GFP were cultured in a complete nutrient medium for 1 wk, F% in cultured Akenohoshi and IR65564-44-2-2 was significantly greater than that under field conditions. These results suggest that inferior assimilate supply is the main cause of lower grain ripening in Akenohoshi, whereas the earlier decrease in grain ripening, probably due to higher spikelet nonfertilization, is the factor responsible in IR65564-44-2-2.

Abbreviations: DMI, plant dry-matter increase • F%, spikelet filling percentage • GE, growth efficiency • GFP, grain filling period • NPT, new plant type • PRG, percentage of ripened grain • YP, yield potential


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
IN RECENT DECADES, new rice genotypes have been bred to increase grain yield potential. The NPT line of the International Rice Research Institute (IRRI) is one of these genotypes (Kush and Peng, 1996). First-generation NPT lines were developed at IRRI from tropical japonica subspecies (Peng et al., 2004). The NPT lines are characterized by fewer tillers and a greater density of spikelets. Although these features suggest that the NPT lines have higher potential yield (Peng et al., 1999; Sheehy et al., 2001), actual yields are similar to or less than those of standard improved cultivars (Horie, 2001; Peng et al., 2004). The lower yields result from low grain ripening (Horie, 2001; Sheehy et al., 2001).

The cultivar Akenohoshi was bred in the National Agricultural Research Center for Western Region, Japan, from indica x japonica rice. It is a high-yield variety and carries abundant spikelets, but grain ripening is lower and grain yield is less than that of other high-yield cultivars (Sibayama, 1986; Jiang et al., 1988).

Several hypotheses have been proposed to account for the poor grain ripening of NPT lines and indica x japonica rice, including inferior assimilate accumulation capacity in the panicle or spikelet (Komatsu et al., 1985; Yamagishi et al., 1996; Kush and Peng, 1996). It has also been suggested that morphological impediments such as the arrangement of spikelets or vascular bundle connections for assimilate transport restrict grain ripening in NPT lines and indica x japonica rice. Alternatively, it has been suggested that inadequate assimilate supply during the early grain-filling period limits grain ripening (Jiang et al., 1988; Xu et al., 1997). Assimilate supply is a dominant key factor in determining grain ripening in rice, because grain dry-matter increase capacity is fairly stable, in that grains are sink-dominated organs; as a result, the grain dry matter is highly dependent on the assimilate supply under diverse environmental conditions (Takami et al., 1990; Kobata et al., 2000; Kobata and Uemuki, 2004). Second-generation NPT lines have now been developed by crossing indica with improved tropical japonica to improve the lower grain yield. These second-generation NPT lines have superior yields to those of the first-generation lines (Peng et al., 2004). The new Japanese indica x japonica rice ‘Takanari’, which was bred later than Akenohoshi, has the higher yield potential (Xu et al., 1997). Further improvement in grain ripening is required, however, before the higher yield potential of rice can be realized. Currently, the cause of lower grain ripening in NPT and indica x japonica rice compared with standard improved cultivars is not well understood.

Our objective was to test the hypothesis that shortage of assimilate supply to grain is the primary cause of low grain ripening in NPT and the indica x japonica rice cultivars. To test this hypothesis, we observed whether grain ripening improved if assimilate supply to grains during the grain-filling period was enhanced by thinning treatments in field conditions or by a panicle solution culture.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Yield Trial in Field Sites
Experiment 1
Two IRRI cultivars (IR65564-44-2-2 [NPT] and IR72) and three improved Japanese cultivars (Akenohoshi, Takanari, and Nipponbare) were grown in 2002 in paddy fields at four experimental sites. IR65564-44-2-2 has intermediate yield and PRG among NPT entry cultivars in IRRI dry season trials, and the PRG was lower than the check cultivar IR72 (Peng et al., 1999). Nipponbare is an improved high-yield japonica cultivar bred in 1963. The fields were located at Shimane University, Matsue, Japan (17 m above sea level, 133° E, 35° N); Kyoto University, Kyoto, Japan (20 m above sea level, 135° E, 35° N); Shinsyu University, Ina, Japan (773 m above sea level, 137° E, 35° N); and Yunnan, China (1170 m above sea level, 100° E, 26° N). The four sites experience diverse temperature and radiation conditions. Average mean temperature and solar radiation during the entire growing season is 22.5°C and 15.5 MJ m–2 d–1 at Matsue, 24.8°C and 14.8 MJ m–2 d–1 at Kyoto, 17.6°C and 18.5 MJ m–2 d–1 at Shinsyu, and 23.3°C and 17.7 MJ m–2 d–1 in Yunnan. The Yunnan site is located within one of the highest rice grain yield districts in Asia (Amano et al., 1996; Ying et al., 1998; Peng et al., 1999).

Three-week-old seedlings were transplanted into the paddy fields on 16 May (Matsue), 22 May (Kyoto), 23 May (Shinsyu), and 5 May (Yunnan). Seedlings were planted in rows 0.30 m apart at a spacing of 0.15 m. Fertilizer N [as (NH4)2SO4], P (as CaHPO4), and K (as KCl) were applied at basal dressing rates of 4 g N m–2, 12 g P m–2, and 12 g K m–2. An additional 2 g N m–2 was applied every 20 d after transplanting until 10 d after heading. Total additional N was 8 g N m–2. Nitrogen fertilizer was applied to avoid symptoms of N deficiency. Each cultivar occupied 18 m2 or more of the paddy field at each site, with three replicates.

Experiment 2
Six cultivars, including two IRRI cultivars (IR65564-44-2-2 and IR72) and four Japanese improved cultivars (Akenohoshi, Takanari, Nipponbare, and Koshihikari) were grown in the paddy field at the Shimane University experimental farm between 1997 and 2003. Koshihikari is a popular improved japonica cultivar noted for its fine taste. Three-week-old seedlings were transplanted into paddy fields on 21 May 1997 (Nipponbare), 10 May 2000 (Koshihikari), 22 May 2002 (IR65564-44-2-2, Akenohoshi, and Takanari), and 21 May 2003 (IR72). Seedlings were planted in rows 0.30 m apart at a spacing of 0.15 m. Fertilizer N [as (NH4)2SO4)], P (as CaHPO4), and K (as KCl) were applied at basal dressing rates of 4 g N m–2, 12 g P m–2, and 12 g K m–2 for Nipponbare; 4 g N m–2, 8 g P m–2, and 4 g K m–2 for Koshihikari; and 4 g N m–2, 8 g P m–2, and 4 g K m–2 for IR65564-44-2-2, Akenohoshi, Takanari, and IR72. An additional 2, 4, and 2 g N m–2 were applied to Nipponbare; 0, 4, and 0 g N m–2 to Koshihikari; and 4, 4, and 4 g N m–2 to IR65564-44-2-2, Akenohoshi, Takanari, and IR72 at the tillering, flower initiation, and heading stages, respectively. The amounts of fertilizer and application methods were determined according to conventional procedures intended to give higher yield. Each cultivar occupied 40 m2 of paddy field, with two (1998, 2002, and 2003) or three (2000) replicates. At maturity, plants occupying a 1 by 1 m area (Exp. 1) or 10 plants (Exp. 2) from each replication were harvested and measured for their yield and yield components. The experimental design at each site was a randomized block. Significant differences (0.05 level) by Tukey's test were calculated from the analysis of variance.

Selected spikelets by specific gravity were counted, oven dried, and weighed. The husks were removed, and oven-dried grain weights (brown rice) determined. Mean single husk weight was estimated by dividing the weight differences between spikelets and grain in the control plants by the spikelet number. The husk weight per hill for each sample was calculated by multiplying the mean husk weight by the number of spikelets, and the grain weight estimated by eliminating the husk weight from the spikelet weight. We ignored changes and differences between fertile and sterile grain in husk weight during the grain-filling period (Kobata et al., 2000). Grain filling was indicated by PRG and F%. The PRG is a ratio of the number of filled to total spikelets, where the number of filled grains is that selected by gravitational method using a 1.06 x 10–3 kg m–3 salt solution. The F% is a measure of grain growth capacity (Murayama, 1982; Horie et al., 1997), and is defined as the ratio of observed grain dry weight (G) to the potential grain yield (YP):

Formula 1[1]
where YP is calculated by multiplying the spikelet number by the dry weight of a single fully ripened grain at harvest, as derived from seed selection by specific gravity (Murayama, 1982). The F% is generally <100 because, even under favorable field conditions, all spikelets cannot ripen due to infertility at flowering (Matsushima, 1959).

Thinning Treatment
Plant densities in parts of the fields in Exp. 2 were halved from the full heading stage to increase assimilation during the grain-filling period through improved irradiance (Kobata et al., 2001; Kobata and Uemuki, 2004). For Nipponbare only, thinning began 10 d after the full heading period. Previous work has shown that when plant density in rice fields was halved during the grain-filling period, dry-matter production in rice increased to approximately double that of control plants, and grain dry-matter increase was dependent on the dry-matter production (Kobata and Moriwaki, 1990).

Culture of Detached Panicles
Plant Materials
IR65564-44-2-2, Nipponbare, and Koshihikari were grown in a paddy field at the Shimane University experimental farm in 2002. Two NPT cultivars (IR65564-44-2-2 and IR66158-38-3-2-1), Akenohoshi, Takanari, IR72, Nipponbare, and Koshihikari were grown in 2004. IR66158-38-3-2-1 was added as a check because it had the lowest PRG among five entry NPT cultivars in IRRI wet season field trials (Peng et al., 1999). In the 2003 plantings, Nipponbare was transplanted on 26 June, Koshihikari on 29 May, and IR65564-44-2-2 on 22 May. In 2004, all cultivars were transplanted on 9 June. Fertilizer N [as (NH4)2SO4], P (as CaHPO4), and K (as KCl) were applied at basal dressing rates of 4 g N m–2, 8 g P m–2, and 4 g K m–2 (2002), and 4 g N m–2, 8 g P m–2, and 8 g K m–2 (2004). An additional 4 g N m–2 (2002) and 2 g N m–2 (2004) were applied at the tillering, flower initiation, and heading stages, except for Koshihikari, for which the additional N at tillering was omitted in both years. All rows planted in 2002 were 0.3 m wide with row spacing of 0.15 m, and those in 2004 were 0.25 m wide with row spacing of 0.15 m. The IR65564-44-2-2 planted in 2002 occupied an area 11 m long by 8 m wide, whereas all other cultivars occupied areas of 5.0 by 2.8 m. In 2004, each cultivar occupied an area of 2 by 1.0 m in a 20-m2 field. Experimental fields where soil conditions and fertility have been carefully equalized and managed were used in both years.

Methods of Panicle Culture
Panicles on the main stems or primary tillers that headed within the late morning of the same day were tagged. Shoots were cut from the stem base in the early evening 5 d after heading, when the flowering in the panicles was almost finished. Culture methods and procedures were derived from Kobata et al. (2001). The stem base was immediately immersed in shallow water and the detached shoots were enclosed in a plastic bag. In the laboratory, each stem was cut off just below the node of the flag leaf, and the flag leaf removed. Cutting was performed in fresh water to protect against capillary discontinuity. The stem below the neck node of the panicle was then sterilized with 25 g L–1 NaOCl solution for 10 min, and subsequently washed with water. The neck of the panicle was inserted through a small hole in a plastic cap into a test tube containing 45 mL of culture solution. The tubes had inside diameter and height of 0.018 and 0.200 m, respectively. In the 2002 experiments, each test tube was then filled with half-strength Murashige & Skoog Plant Salt Mixture (Nippon Seiyaku Co., Tokyo) containing sucrose at concentrations of 0, 20, 40, 60, and 80 g L–1. The same culture solution was used in 2004, but only at a concentration of 60 g L–1, because previous work (Kobata et al., 2001) and the results of the 2002 experiments had shown this to almost cover the optimal range for grain dry-matter increase. Other studies have shown that when amino acids were added to cultural solutions as a N source, grain growth rate did not change or decreased in either wheat (Tritcum aestivum L., Donovan and Lee, 1978) or rice (Kobata et al., 2001). Furthermore, addition of several dominant growth substances to cultural solutions does not affect grain growth rate in rice (Kobata et al., 2002). Hence sucrose concentration should be the major critical factor in grain dry-matter increase in grain solution culture. The experiment was conducted in a complete randomized design with four replications.

The loaded test tubes were housed in a refrigerator at 5°C (Engel, Sawafuji Elec. Co., Tokyo) to prevent microbial contamination of the solution. The upper parts of the panicles were exposed to air in an incubator (NK Systems Biotron, Nippon Ika Co., Tokyo) at 30°C (2002 experiments), and 25 and 30°C (2004). Styrofoam plates were used to form partitions between the test tubes and the upper parts of the panicles. The panicles were continuously exposed to fluorescent light at 84 µmol m–2 s–1. Between 0 and 6 or at most 8 d after the start of culturing, four panicles from each treatment were harvested, dried in an oven at 80°C for 48 h, divided into spikelets and other parts, and weighed. Spikelet dry matter of field-grown plants was also measured at Day 0 and Day 6 or 8. Sugar concentrations in the culture solutions were monitored with a refractometer (N-1E, Atago Co., Tokyo) before and after culturing. The difference in concentration before and after culturing was tested to determine if selective absorption of sucrose by the panicles had occurred. No changes in concentration were observed, and hence the amount of solute absorbed during culturing was estimated from the reduced volume of solution. The F% was calculated in the same manner as in the field trials. Significant differences (0.05 level) by Tukey's test were calculated from the analysis of variance.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Grain Yield and Yield Components in Field Trials
Mean grain yields across the four locations in Exp. 1 ranged between 932 g m–2 for Takanari and 618 g m–2 for IR65564-44-2-2 (Table 1). Mean grain yields across the four cultivars at each location were greatest for Yunnan and least for Shinsyu. A trend of highest yield from Takanari and lowest yield from IR65564-44-2-2 was observed at each location, although yield order among the other cultivars varied depending on location. Takanari and IR72 both had higher spikelet numbers than IR65564-44-2-2 and Nipponbare. One-thousand-grain weights for Takanari, IR65564-44-2-2, and Nipponbare were similar, but for IR72 were 2 g lower than for other cultivars. The PRG for IR65564-44-2-2 (64.8%) was the lowest. The lower PRG resulted in lower grain yield in IR65564-44-2-2, even though spikelet numbers in IR65564-44-2-2 and Nipponbare were similar.


View this table:
[in this window]
[in a new window]
  		Table 1. Grain yield and yield components of four rice cultivars at four different locations at 2002.

 
In Exp. 2, grain yield was highest for Takanari (856 g m–2) and lowest for IR65564-44-2-2 (415 g m–2); Akenohoshi was intermediate between these two cultivars and significantly lower than Takanari in 2002 (Table 2). When grain yield data for six cultivars were included across different years, Takanari yield remained the highest and IR65564-44-2-2 the lowest. Akenohoshi grain yield was similar to the other three cultivars (Nipponbare and Koshihikari), although less than Takanari and greater than IR72 and IR65564-44-2-2. One-thousand-grain weights were similar among all six cultivars, but lowest for IR72. The PRG for IR65564-44-2-2 was clearly the lowest (48%), not only among the three cultivars grown in 2002, but also among all six cultivars across different years. Akenohoshi PRG (72%) was lower than that of the other three Japanese cultivars (89.3% for Nipponbare, 83.4% for Koshihikari, and 87.1% for Takanari) although the difference was not statistically significant. Average standard error (SE) of the six cultivars (4%, Table 2) suggests that mean PRG in Akenohoshi was lower than in the other three Japanese cultivars, as past experiments have also demonstrated (Komatsu et al., 1985; Sibayama, 1986). The lower yield in IR65564-44-2-2 and Akenohoshi compared with Takanari was due to lower PRG.


View this table:
[in this window]
[in a new window]
  		Table 2. Yield and yield components of seven rice cultivars in control and thinned plants at Matsue. In thinned plots, plant density was halved during the grain-filling stage.

 
The field experiments, therefore, suggested that grain ripening in IR65564-44-2-2 and Akenohoshi was inferior to other most improved cultivars under diverse cultivation conditions, as also shown by past reports (Xu et al., 1997; Jiang et al., 1988; Horie, 2001; Peng et al., 2004).

The Effect of Thinning on Grain Ripening
After thinning in Exp. 2, whole-plant dry weights increased to between 109 and 121% of the unthinned plots (Fig. 1 ). Thinning effects on whole-plant dry weight during the late grain-filling period became obvious in all cultivars except Takanari. Increased grain dry weight due to thinning was clearly observed for Akenohoshi, with continuous gains up to 40 d after the full heading date; however, grain dry-matter increase in IR65564-44-2-2 slowed about 20 d after full heading, while whole-plant dry matter increased continuously. The lesser decrease of straw dry weight in thinned IR65564-44-2-2 plants from 20 d after full heading also reflected an earlier reduction in grain growth. In the other cultivars, thinning had little effect on grain dry-matter increase, whereas whole-plant dry matter increased.


Figure 1
View larger version (35K):
[in this window]
[in a new window]
  		Fig. 1. Dry matter weight of the grain, straw, and whole plant (aboveground parts) per hill during the grain-filling period for normal density (filled circles and solid lines) and thinned (open circles and dotted lines) plants of six rice cultivars grown in Matsue. Akenohoshi and Takanari (japonica x indica) and IR65564-44-2-2 ("New Plant Type") data are from 2002, IR72 (indica) from 2003, Nipponbare (japonica) from 1998, and Koshihikari (japonica) from 2000. In the thinned plots, plant densities were halved during the grain-filling period. Data are the mean ± standard error of two (1998, 2002, and 2003) or three (2000) replications.

 
In Akenohoshi, IR65564-44-2-2, and IR72, F% in unthinned plants 30 d after the full heading period was lower than other cultivars (Fig. 2 ). When plant density was halved, only in Akenohoshi did F% increase significantly over the control 30 and 40 d after the full heading date. In thinned IR65564-44-2-2 and Takanari, F% was a little greater than in the control 20 and 30 d after full heading; following that, rates were similar. In both control and thinned IR65564-44-2-2 plants, F% reached a ceiling 30 d after the full heading date.


Figure 2
View larger version (26K):
[in this window]
[in a new window]
  		Fig. 2. Spikelet filling percentage (F%) in the control and thinned plants of six rice cultivars during the grain-filling period. The F% is the ratio of the grain dry weight to the dry weight of one ripened grain at harvest x number of spikelets. Vertical bars indicate significant LSD (P < 0.05) between control and thinned plants by Tukey's test. Each data point is the mean ± standard error of two or three replicates.

 
When plant density was halved, grain yield per hill for Akenohoshi became similar to the three Japanese cultivars, and an increase of PRG in Akenohoshi yield components was the highest (116% of control; Table 2); however, PRG in thinned IR65564-44-2-2 plants and the other four cultivars increased by only 106% or less relative to unthinned control plants. Therefore, although thinning improved PRG and F% in Akenohoshi, in IR65564-44-2-2 and other cultivars these parameters scarcely changed.

The YP in the six cultivars, as calculated from spikelet numbers and single ripened grain weights (Table 2), varied from 27 g hill–1 (706 g m–2 at 14% moisture content) in Nipponbare to 38 g hill–1 (981 g m–2) in Takanari. The current assimilate supply capacity over potential grain dry-matter increase can be evaluated by current DMI/YP. The DMI/YP was calculated for the periods between 0 and 10, 0 and 20, and 0 and 30 d after the full heading date (Fig. 1). When all data in each term of the grain-ripening period are combined, the relationship between PGR and DMI/PY forms an exponential curve with a ceiling of 91% (Fig. 3 ). The curve fit for the term between 0 and 10 d after the full heading period was the best among the three terms; however, data points for thinned IR65564-44-2-2 in the 0- to 20- and 0- to 30-d terms lie well away from the curves. If these two points are omitted, the curve fits improve and are similar to that of the 0- to 10-d term (Fig. 3). This suggests that, in most cultivars, PRG is highly affected by whether assimilate supply to grain capacity is adequate, and hence the lower PRG in Akenohoshi resulted from lower assimilate supply to grain capacity. In Koshihikari, Takanari, and Nipponbare, PRG reached the ceiling and was not improved even if DMI/YP was increased by thinning; however, in IR65564-44-2-2 and IR72, even if DMI/YP increased from the middle grain-filling period onward, PRG was not improved.


Figure 3
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Percentage of ripened grain and whole-plant dry-matter increase (DMI) per yield potential (YP) in six rice cultivars. The DMI/YP was calculated for each different grain-filling term (0–10, 0–20, and 0–30 d) after the full heading date. Closed symbols indicate unthinned control plants and open symbols thinned plants. Dotted curves and their equations exclude data for thinned IR65564-44-2-2. Each data point is the mean of two or three replicates.

 
The role of reserved assimilate in the stem as an additional assimilate source was not evaluated in this analysis because current assimilation should be the dominant assimilate during grain filling in most high-yield cultivars (Evans, 1996).

The F% changed during the grain-filling period, depending on DMP/YP (Fig. 4 ). The F% in IR65564-44-2-2 and Akenohoshi, in both of which DMP/YP was lower than other cultivars, did not reach the values observed in Takanari, Nipponbare, and Koshihikari.


Figure 4
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Spikelet filling percentage vs. whole-plant dry-matter increase (DMI) per yield potential (YP) during the grain-filling period of control and thinned plants of six cultivars. Closed symbols indicate unthinned control plants and open symbols thinned plants. Each data point is the mean of two or three replicates.

 
The F% was approximated using a logistic equation based on accumulated temperature, and incorporating a ceiling (Kobata and Uemuki, 2004). The relationship between F% and accumulated temperature after the full heading period was calculated from temperature data measured at a neighboring weather station. The F% curves for Koshihikari, Nipponbare, and Takanari control plants have high ceilings, although the slope of the Takanari curve was lower than that of the two other cultivars (Fig. 5 ). The slope of the Akenohoshi curve was a little lower than Takanari. The F% in IR72 increased earlier than other cultivars, but flattened out at a lower ceiling. The slope for IR65564-44-2-2 was similar to or a little steeper than that of Akenohoshi, but the ceiling was lower. When plants were thinned during the grain-ripening period, F% in all cultivars was increased (Fig. 5). The F% curves can be divided into four groups. The first group has a steeper slope and a higher ceiling (Nipponbare and Koshihikari), the second group has a lower slope and a similar high ceiling (Takanari and Akenohoshi), and the third group has a lower slope and a lower ceiling (IR65564-44-2-2). In contrast, the fourth group has a steep slope and the lowest ceiling (IR72). The F% under thinning treatments can indicate the potential grain growth (Kobata et al., 2000; Kobata and Uemuki, 2004). The final F% in Akenohoshi was able to reach that of Takanari, and their ceilings were similar to that of Nipponbare and Koshihikari. Although the slope of the curve for IR65564-44-2-2 was same as in Takanari and Akenohoshi, however, its ceiling was considerably lower than these two cultivars. This suggests that F% in Akenohoshi can increase to the same level as Takanari if assimilate supply increases, but the lower ceilings of IR65564-44-2-2 and IR72 are scarcely improved. Although the reason for the lower F% ceiling in IR65564-44-2-2 and IR72 cannot be determined from our results, spikelet infertility is probably the dominant cause. Higher rates of spikelet nonfertilization have been observed in NPT cultivars than in other cultivars such as Takanari (Horie, 2001; Kobayasi et al., 2003). In IR72, infertile spikelets occasionally comprise 20% of total spikelets, compared with a rate of only 5% in Koshihikari grown in Matsue (unpublished data, 2005). Hence the F% lower ceiling in IR65564-44-2-2 may result from higher spikelet nonfertilization.


Figure 5
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5. Spikelet filling percentage (F%) in normal-density and thinned plants vs. accumulated temperature (AT)after the full heading date. For the normal-density plants, respective factors and coefficients for the fitted curves of the form F% = a/[1 + b exp(–cAT)] are 91, 104.0, 0.012, and 0.976 (‘Nipponbare’ and ‘Koshihikar’); 91, 23.4, 0.006, and 0.997 (‘Takanari’); 90, 39.4, 0.006, and 0.977 (‘Akenohoshi’); 73, 94.5, 0.009, and 1.000 (‘IR65564-44-2-2’); and 73, 16.2, 0.010, and 0.989 (‘IR72’) for a, b, c, and r2, respectively. For the thinned plants, factors and coefficients are 91, 798.8, 0.017, and 0.980 (Nipponbare and Koshihikar); 91, 119.5, 0.009, and 0.989 (Takanari and Akenohoshi); 78, 85.3, 0.010, and 1.000 (‘IR65564-44-2-2’); and 69, 191.3, 0.022, 0.998 (‘IR72’) for a, b, c, and r2, respectively.

 
Grain-Ripening Capability under Solution Culture
The relationship between F% and sucrose concentration in Nipponbare, Koshihikari, and IR65564-44-2-2 suggests that a sucrose concentration of 40 to 60 g L–1 was optimal for spikelet dry weight increase (Fig. 6 ). It has been suggested that the sucrose concentration in the sieve tube of the panicle neck in rice during the grain-filling period is 200 g L–1 (Hayashi and Chino, 1990); hence, the optimal concentration of sucrose in our culture tubes was lower than that in the sieve tube. The critical sucrose concentration in liquid culture for grain dry weight in rice is thus similar to or slightly greater than that in wheat (40 g L–1; Singh and Jenner, 1983) and soybean [Glycine max (L.) Merr., 35–40 g L–1; Egli et al., 1989]. A sucrose concentration of 60 g L–1 was used as a standard concentration of cultural solution in the 2004 experiment because this concentration was able to approximately cover the peak of F% in all three cultivars.


Figure 6
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 6. Spikelet filling percentage (F%) of panicles of three rice cultivars incubated for 7 to 8 d under different sucrose concentrations of culture solution. Data are the mean ± standard error of four replications. Dotted lines indicate F% in field-grown plants on the last day of the incubation. Panicle incubation was performed at 30°C under 24-h illuminated conditions.

 
The F% in the field-grown plants on the final day of panicle incubation was similar to or a little lower than that of incubated Koshihikari and Nipponbare plants; however, the incubated F% in IR65564-44-2-2 was significantly greater than that of the field-grown plants (Fig. 6). At the beginning of panicle culture, F% in all cultivars was very small and was ignored. The upper ranges of F% (20–30%) encompass the linear phase of the grain-filling period of the field experiment (Fig. 2). This suggests that F% during the early active grain-filling period in IR65564-44-2-2 can increase if abundant sucrose is supplied to the panicles. The F% potential of IR65564-44-2-2 is thus not fully utilized under field conditions.

Grain dry matter and F% are affected by temperature (Kobata and Uemuki 2004). Therefore, if the temperature condition of incubated panicles differs greatly from the field conditions, F% capacity under incubated and field conditions cannot be directly compared. Ranges in mean and maximum temperature in the field coinciding with the terms of panicle culture were 26 to 27°C and 30 to 32°C, respectively. The highest rate of grain dry matter increase in rice is observed at late midday, when temperature reaches daily maximum (Matsushima et al., 1957; Matsushima, 1959). Although the mean field temperature was a little less (3–4°C) than the incubated temperature, the mean maximum temperature at which maximum grain growth is expected was similar. Spikelet dry matter in Nipponbare increased by 25% when the temperature of the panicle culture was elevated from 20 to 30°C (Kobata et al., 2001). Therefore, if mean temperature is used for comparison between incubated and field-grown plants, the acceleration of F% in incubated Nipponbare plants by higher temperature (3°C) can therefore be expected to be <2% in the F%. The effect of temperature contrast between the incubated and field-grown plants can thus be ignored, even though the temperature response of F% in the other two cultivars is not yet known.

When panicles of seven cultivars were incubated in 2004 at 25 and 30°C, F% in incubated IR65564-44-2-2, IR66158-38-3-2-1, and Akenohoshi was significantly greater than that in field-grown equivalents (Fig. 7 ); however, for Nipponbare, Koshihikari, Takanari, and IR72, F% in incubated panicles on the final day of panicle incubation did not significantly differ from that in their field-grown equivalents. No consistent differences in F% were observed between the two temperature conditions (25 and 30°C) investigated. The ranges in mean and maximum field temperature during the term coincident with the incubation period in six cultivars varied between 24 and 27°C (25 ± 1°C, the mean of six cultivars and SD) and between 28 and 31°C (29 ± 1°C), respectively. Hence the two temperatures used in the incubation room in which the panicles were housed almost spanned the ranges of the field temperatures. Cultivars with lower F% in the field than for incubated plants resulted in lower PRG at plant maturity (correlation between PRG and the ratio of F% in the field to F% in the incubated plants, r = 0.647, P < 0.05). Hence, low F% due to lack of assimilate supply during the active grain-filling period seemed to affect PRG.


Figure 7
View larger version (56K):
[in this window]
[in a new window]
  		Fig. 7. Spikelet filling percentage (F%) of seven rice cultivars incubated under two temperature conditions (25 and 30°C) for 6 to 8 d, using 60 g L–1 sucrose culture solution. Field indicates F% in field-grown plants on the last day of the incubation. Panicle incubation was performed under 24-h illuminated conditions. Data are the mean ± standard error of four replications. Vertical bars indicate significant LSD (P < 0.05) by Tukey's test.

 
Growth efficiency (GE) from supplied assimilate to grains may affect grain dry-matter increase through assimilate loss during the grain-ripening processes. The relationship between absorbed sucrose and increase of spikelets was compared for data across different sucrose concentrations in cultural solutions for Nipponbare, Koshihikari, and IR65564-44-2-2 in 2002. During the incubation, the weight of panicle organs other than spikelets increased far less than the average spikelet dry weight (data not shown). Thus, most of the panicle dry-matter increase is due to the increase of the spikelets. The amount of sucrose absorbed could be estimated from the amount of solution culture absorbed, because sucrose concentration in the solution scarcely changed during panicle incubation (data not shown). This has also been demonstrated by previous work (Kobata et al., 2001). Examination of the relationships between the increase in average spikelet dry weight and absorbed sucrose for three cultivars shows that their slopes (GE) were similar (0.66–0.70). Generally, 70% of absorbed sucrose is converted into panicle dry matter increase, and the remaining 30% is used for respiration (Fig. 8 ). The GE of spikelets in rice grown under irrigated conditions was found to be 0.77 (Kobata and Takami, 1986) and values in almost all starch grain crops lie between 0.7 and 0.8 (Amthor 1989). The GE value determined here for these three cultivars is thus appropriate. This suggests that GE differs little between IR65564-44-2-2 and the other cultivars, and thus assimilate loss during the grain-filling process does not lower the grain dry-matter increase in IR65564-44-2-2.


Figure 8
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 8. Relationship between spikelet dry-matter increase and absorbed sucrose during the incubation period in three rice cultivars incubated under different sucrose concentrations. The slope indicates spikelet growth efficiency. Relationships were calculated from the data in Fig. 6. Each point is the mean of four replications.

 
These results suggest that the low-ripening cultivars IR65564-44-2-2 and Akenohoshi can ripen during the active grain-filling period if abundant sucrose is supplied to panicles as in the solution culture method, and that the lower GE does not result in low grain ripening.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
We conclude that low grain ripening in Akenohoshi mostly results from a shortage of assimilate supply to spikelets, because grain ripening was improved when assimilate was increased by thinning treatments or by a culture solution. The main cause of the lower grain ripening in IR65564-44-2-2 was earlier cessation of grain filling, probably due to higher spikelet nonfertilization and partially to lower assimilate supply during the early grain-filling period. The grain dry matter in cultured IR65564-44-2-2 was able to increase during the early grain-filling period if adequate assimilate was supplied in solution. Further investigation of spikelet nonfertilization may help to overcome the lower grain ripening in NPT cultivars. Therefore, morphological impediments such as inadequate assimilate transport systems to the grains, which restricts the flow of substrate from source to sink, do not seem to be responsible for low grain ripening in these cultivars. Amounts of assimilate flow to grains or available capacity size of spikelets are the dominant factors affecting grain ripening, as suggested by Milthorpe and Moorby (1969), Gifford and Evans (1981), and Takami et al. (1990).


   ACKNOWLEDGMENTS
 
Our thanks to Dr. Takeshi Horie, project leader of the Asian Rice Network, and members Dr. Tatsuhiko Shiraiwa (Kyoto site), Naoto Inoue and Motoyuki Hagiwara (Shinshyu), Keisuke Katsura (Yunnan), and Masao Ohnishi, Kazuhiro Kobayasi, and Shohei Yasumura (Shimane) for their data collection and contributions, and to Mr. Naoya Uemuki for supplying data from the Shimane site. We thank Dr. Barry Roser for reading our manuscript.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kobata, T.
Right arrow Articles by Ida, K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kobata, T.
Right arrow Articles by Ida, K.
Agricola
Right arrow Articles by Kobata, T.
Right arrow Articles by Ida, K.
Related Collections
Right arrow Rice

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome