Grain quality changes and responses to nitrogen fertilizer of japonica rice cultivars released in the Yangtze River Basin from the 1950s to 2000s

Junfei Gu，Jing Chen，Lu Chen，Zhiqin Wang，Hao Zhang，Jianchang Yang*

Key Laboratory of Crop Genetics and Physiology of Jiangsu Province，Co-Innovation Center for Modern Production Technology of Grain Crops，Yangzhou University，Yangzhou，China

Grain quality changes and responses to nitrogen fertilizer of japonica rice cultivars released in the Yangtze River Basin from the 1950s to 2000s

Junfei Gu，Jing Chen，Lu Chen，Zhiqin Wang，Hao Zhang，Jianchang Yang*

Key Laboratory of Crop Genetics and Physiology of Jiangsu Province，Co-Innovation Center for Modern Production Technology of Grain Crops，Yangzhou University，Yangzhou，China

A R T I C L E I N F O

Article history:

Received 23 October 2014

Received in revised form

21 March 2015

Accepted 4 May 2015

Available online 9 May 2015

Breeding

Cultivation technique

Grain quality

Japonica rice

Nitrogenous fertilizer

While the yield potentialofrice has increased butlittle is known aboutthe impactofbreeding on grain quality，especially under different levels of N availability.In order to investigate the integrated effects ofbreeding and Nlevels on rice quality 12 japonica rice cultivars bred in the past 60 years in the Yangtze River Basin were used with three levels of N:0 kg N ha-1，240 kg N ha-1，and 360 kg N ha-1.During the period，milling quality（brown rice percentage，milled rice percentage，and head rice percentage），appearance quality（chalky kernels percentage，chalky size，and chalkiness），and eating and cooking quality（amylose content，gel consistency，peak viscosity，breakdown，and setback）were significantly improved，but the nutritive value of the grain has declined due to a reduction in protein content.Micronutrients，such as Cu，Mg，and S contents，were decreased，and Fe，Mn，Zn，Na，Ca，K，P，Bcontents were increased.These changes in grain quality imply that simultaneous improvements in grain yield and grain quality are possible through selection.Overall，application of N fertilizer decreased grain quality，especially in terms of eating and cooking quality.Under higher N levels，higher protein content was the main reason for deterioration ofgrain quality，although lower amylose content might contribute to improving starch pasting properties.These results suggest that further improvement in grain quality will depend on both breeding and cultivation practices，especially in regard to nitrogen and water management.

?2015 Crop Science Society of China and Institute of Crop Science，CAAS.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.

1.Introduction

Rice（Oryza sativa L.）is the most important staple food in the world，providing more than 21%of the calorific needs of the world population，but up to 76%of the calorific intake of the population in South East Asia［1］.In China，next to yield，improvement in grain qualities is becoming a priority，along with improvements in economic and living standards.Development of both high yield and high quality in rice varieties is essential［2］.

Grain quality traits dictate market value and have a pivotal role in the adoption of new varieties.Rice quality traits encompass milling recovery，physical appearance，cooking and eating qualities，and nutritional value［3，4］.The evaluationcriteria for millingqualitymainlyinclude brown rice percentage，milled rice percentage，and head rice percentage，which reflect the proportion of whole kernels（head rice or head milled rice）and broken kernels produced during milling of rough rice. Appearance quality is mainly determined by grain size，translucency，chalky grain percentage，chalky area and degree of chalkiness.Starch granules in chalky areas of the grain are smaller and less densely packed than the larger tightly packed granules in translucent areas of the grain［5］.Cooking and eating qualities are mostly specified by amylose content，gelatinization temperature，and gel consistency of the grain starch［6-8］. Nutritional value is essential because rice is a main source of dietary protein and micronutrients for most of rice growing countries.In order to improve quality traits along with yield it is necessary to understand the important quality traits，to select for them in a breeding program，and to control them through appropriate nitrogen and water management.

Table 1-Growth duration，plant height，grain weight and the largest application area of the selected japonica rice cultivars.

Historically，many breeding programs took yield potential as the priority target，particularly in China.Rice yields have increased from 2.0 t ha-1in the 1950s to 6.6 t ha-1in the 2000s［9，10］.In 2006，the average rice grain yield in China was more than 50%higher than the world average［11］.The Yangtze River Basin is the main rice planting area in China，and represents 51.2%of the total rice growing area and 51.3%of the total rice production in the country［12，13］.During the period，the yield potential has significantly improved［2，12］. Farmers often use excessive N fertilizer to maximize grain yield.For example，China's national average N application for rice in 2006 was 193 kg ha-1，more than 90%above the world average［11］.In Jiangsu province the average N application is more than 300 kg ha-1in some high yielding counties［12］. Over-use of N fertilizer also contributes to poor eating and cooking quality of the grain［2，12］.However，little is known about the impact of breeding selection on grain quality of rice，especially under different N levels.Understanding the changes in grain quality of the rice cultivars bred in different decades and their levels of response to of N fertilizer should be beneficial for improving rice quality in both rice breeding and cultivation.

In this experiment we selected twelve representative japonica rice cultivars bred in the area during the last 60 years.Milling characteristics，appearance，cooking and eating quality，and nutritional value of these cultivars were studied.Such a study should provide useful information for achieving a high-quality and high-yield rice production system with inputs from both agronomy and breeding.

Fig.1-Grain yield（A）and harvest index（B）of rice cultivars released in the 1950s，1960s，1970s，1980s，1990s，and 2000s under 0 kg N ha-1（0 N），240 kg N ha-1（240 N），and 360 kg N ha-1（360 N）application levels.Data are averages of two years±1 S.E.（vertical bars）.

Table 2-Significance of variance estimates related to years（Y），cultivars released at different decades（D），cultivars nested within decades［C（D）］，nitrogen level（N），and their interactions on rice grain quality traits.

2.Materials and methods

2.1.Plant materials and cultivation

The experiments were conducted at the research farm of Yangzhou University，Jiangsu province，China（32°30′N，119°30′E）during the rice growing season（May to October）in 2012，and repeated in 2013.The preceding crop was wheat（Triticum aestivum L.）in both years.The soil used in the experiments was a sandy loam（typic fluvaquents，entisols）with 20.4 g kg-1organic matter，106.2 mg kg-1alkali-hydrolysable N，28.5 mg kg-1Olsen-P，and 93.6 mg kg-1exchangeable K.Twelve japonica rice cultivars，Huangkezao，HKZ；Guihuaqiu，GHQ；Jinnanfeng，JNF；Guihuahuang，GHH；Xudao 2，XD2；Liming，LM；Sidao 8，SD8；Yanjing 2，YJ2；Zhengdao 88，ZD88；Huaidao 5，HD5；Huaidao 9，HD9；and Lianjing 7，LJ7，grown in the Yangtze River Basin during the last 60 years were chosen，and classified into six types according to decade of release.The cultivars were chosen because they had large annual planting areas（＞6.67×104ha）. The growth duration and some plant traits of these cultivars are listed in Table 1.To reduce differences in heading time among the cultivars，sowings were at different dates from 8 to 25 May. Twenty-day-old seedlings were manually transplanted into a paddy field with a hill spacing of 13.3 cm×25.0 cm and two seedlings per hill.

The field experiments were a split-block design with 3 replicates，with nitrogen as the main treatment and cultivars as sub-plots.Three main plots，namely low N level（0 kg N ha-1，0 N），medium N level（240 kg N ha-1，240 N），and high N level（360 kg N ha-1，360 N），were applied in each block，leaving a 1.5 m space between plots.A drainage ditch 0.5 m wide was dug to a depth of about 30 cm all around the main plot，and connected to the central drainage channels of the field.Plastic film was inserted into the soil between plots to a depth of 50 cm to form a barrier.Subplots for cultivars were assigned in each main plot.The subplots were 16.0 m2.For medium and high N fertilizer treatments，urea as the N source was applied at pre-transplanting（1 day before transplanting），early tillering（7 days after transplanting）and panicle initiation（about 35 days after transplanting）in the proportions 50%，10%，and 40%，respectively.For all the treatments，phosphorus（30 kg ha-1as single superphosphate）and potassium（40 kg ha-1as KCl）were applied as basal fertilizer 1 day before transplanting.

Fig.2-Brown rice percentage（A），milling rice percentage（B）and head rice percentage（C）of rice cultivars from the 1950s（Huangkezao，HKZ；Guihuaqiu，GHQ），1960s（Jinnanfeng，JNF；Guihuahuang，GHH），1970s（Xudao 2，XD2；Liming，LM），1980s（Sidao 8，SD8；Yanjing 2，YJ2），1990s（Zhengdao 88，ZD88；Huaidao 5，HD5），and 2000s（Huaidao 9，HD9；Lianjing 7，LJ7）under 0 kg N ha-1（0 N），240 kg N ha-1（240 N），and 360 kg N ha-1（360 N）application levels.Data are average of two years with±1 S.E.（vertical bars）.

2.2.Harvesting and grain quality measurement Plants were hand-harvested on 12-15 October.Plants in two rows on each side of the plot were discarded to avoid border effects.Grain yield was determined from a harvest area of 6.0 m2in each plot and adjusted to 14%moisture.Aboveground biomass and yield components，i.e.number of panicles per square meter，number of spikelets per panicle，percentage of filled grains，and grain weight，were determined on plants from 0.6 m2（excluding the border rows）sampled randomly from each plot.The percentage of filled grains was defined as the filled grains（specific gravity≥1.06 g cm-3）as a percentage of the total number of spikelets.

About 500 g of grains harvested at physiological maturity from each plot was dried at 40°Cin a forced-air oven for quality analysis according to the Rice Quality Measurement Standards［14］.Grain samples of 150 g were twice passed through a Dehusker，polished and separated into broken and unbroken grains.The brown rice rate，milled rice rate，and head rice rate were expressed as percentages of total grain weights.Head rice refers to the kernels that remain at three-fourths or more of their normal length［4］.The lengths and breadths of 10 milled grains from each sample were measured using a vernier micrometer.Chalkiness was evaluated visually on 100 milled grains per plot.Grains containing 20%of more of white belly，white center，white back，or a combination of these were considered chalky kernels.Chalkiness area was expressed as percentage ofthe totalarea ofa kernel.The degree ofchalkiness was calculated from chalky kernels and chalky size，i.e.the degree of chalkiness（%）=chalky kernels×chalky size.

Representative milled rice samples were oven-dried to constant weight at60°C，and were then ground with a stainless steel grinder（FW-100，China）with a 0.25 mm sieve［15］in order to further prepare them for subsequent analyses，viz.，amylose content，protein content，starch viscosity and elemental concentrations.To determine amylose content，flour samples weregelatinized using 1 mol L-1NaOH.Amylose was determined colorimetrically using the amylose-iodine reaction.Amylose content was calculated by using amylose/amylopectin mixed standards［16］.Protein content was measured with a grain analyzer（Infratec 1241，F(xiàn)oss，Denmark）.

Fig.3-Chalky kernel percentage（A），chalky area（B），chalkiness（C），and grain length/width ratio（D）of rice cultivars from the 1950s to 2000s.See caption to Fig.2 for other details.

As starch paste parameters generated from a Rapid Visco-Analyzer（RVA）reflect starch gelatinization，disintegration，swelling and gelling ability，they are often used to evaluate quality ［17，18］.In this experiment，rice starch viscosity characters were evaluated using a RVA（Model RVA-3D；Newport Scientific，Sydney，Australia）as described by Han et al.［3］. Viscosity values were recorded as centipoises（cP）.The original components of starch viscosity characters include peak viscosity，hot viscosity，and cool（final）viscosity.Secondary components，such as breakdown and setback，were calculated from the original components.A breakdown value was calculated by subtracting the hot viscosity from peak viscosity，a setback value was calculated by subtracting peak viscosity from cool viscosity，and a consistency value was calculated by subtracting hot viscosity from cool viscosity.

In addition to physiochemicalcomposition analysis，cooking and eating quality was also evaluated with a rice taste analyzer STA1A，which is a convenient and accurate tool for evaluation of rice cooking and eating quality［19］.

Exactly 1 g ofpowdered grains was decomposed in a mixture of 8 mL of HNO3and 2 mL of H2O2.The ash was thendecomposed using a microwave ashing device（CEM 2000，Kamp，Lintfort，Germany）and diluted in 10 mL of aqua regia. Cu，F(xiàn)e，Mn，Zn，Na，Ca，K，Mg，S，P，and B were measured by ICP-OES（Thermo Jarrell Ash，Trace Scan，F(xiàn)ranklin，USA）.

Fig.4-Amylose content（A），gel consistency（B），and overall palatability（C）of rice cultivars from the 1950s to 2000s.See caption to Fig.2 for other details.

2.3.Statistical analysis

Analysis of variance was performed using the SAS/STAT statistical analysis package（version 6.12，SAS Institute，Cary，NC）.A GLM analysis of variance procedure was run to identify significantsources ofvariance，which were partitioned between effects due to year（Y），cultivars released in different decades（D），cultivars nested within decades［C（D）］，nitrogen fertilizer（N），and their interactions（Y×D，Y×N，D×N，and Y×D×N）and residual error.Means were tested by least significant difference at P=0.05（LSD0.05）.

3.Results

3.1.Grain yield and quality differences

Differences in grain yield and harvest index among the 12 cultivars bred are shown in Fig.1.Grain yield progressively increased from the 1950s（3.13 t ha-1）to 2000s（9.44 t ha-1）across different N levels with an average yearly increase of 4%. Grain yield increased by 35.7%from the 1950s to 1960s，31.2% from the 1960s to 1970s，17.3%from the 1970s to 1980s，17.4% from the 1980s to 1990s，and 23.1%from the 1990s to 2000s. Simultaneously，average plant height decreased from 117.9 cm in the 1950s to 92.56 cm in the 1990s，but increased to 100.6 cm in super rice cultivars（2000s）.Harvest index increased from 36.5%in the 1950s to 48.9%in the 2000s（Fig.1）.N fertilizersignificantly increased grain yield and harvestindex exceptthat yields of some early decade cultivars dropped at high N levels（i.e.360 kg N ha-1）due to lodging.

Fig.5-The value of peak viscosity（A），hot viscosity（B），cool viscosity（C），breakdown（D），setback（E），and consistency（F）in the RVA profile of rice cultivars from the 1950s to 2000s.See caption to Fig.2 for other details.

Table 2 summarizes computed F-values for differences in grain quality.Differences in quality traits between the two years were not significant and average data across the years are used in the following analyses.There were significant differences in all quality traits among cultivars（Table 2）.Even within the same decade significant differences were detected among cultivars.Nitrogen fertilizer had significant impacts on grain quality traits except for milled rice percentage，head rice percentage，length/width ratio，and gel consistency.Generally，no significant interactions were observed between year，cultivars，and nitrogen fertilizer application（Table 2）.

3.2.Milling quality

Milling quality was evaluated by three milling traits:brown rice percentage（BRP），milled rice percentage（MRP），and head rice percentage（HRP）.BRP responded positively to nitrogen levels（Fig.2-A）.On average，BRP increased from 82.79%at the 0 N level to 83.81%at the 240 N level，and to 84.83%at the 360 N level.No clear tendency was attributable to breeding. BRP increased by 1.8%per year from the 1950s to 2000s，with the highest increase in the 1970s（Fig.2-A）.

MRP and HRP were not influenced by N fertilizer application（Table 2），while significant genetic differences were observed in MRP and HRP（Fig.2-B，C）.Generally MRP and HRP were improved during breeding regardless of N levels.For example，HRP increased on average by 68.7%from 39.78%in the 1950s to 67.11%in the 2000s，with more variation in the early decades（Fig.2-C）.Compared with HRP，the increase in MRP attributed to breeding was less and showed the highest value in 1980s（Fig.2-B）.

3.3.Appearance quality

Fig.3 summarizes four main appearance quality properties，including chalky kernel percentage（CKP），chalky size（CS），chalkiness，and length/width ratio（LWR）.As shown in Fig.4，CKP，CS and chalkiness were significantly increased when nitrogen fertilizerwas applied regardless ofcultivars.On average，CKP，CS and chalkiness increased by 21.4%，25.3%，and 39.5% from 0 N to 240 N，respectively，and by 24.1%，29.5%，and 56.5% from 240 N to 360 N（Fig.3）.

Cultivars bred during different decades showed substantial decreases in CKP，CS and chalkiness.For example，CKP was on average decreased from 51.6%to 43.3%，42.5%，29.7%，23.5%，and 16.3%during the 1950s，1960s，1970s，1980s，1990s，and 2000s，respectively.For CS and chalkiness，similar tendencies were observed regardless of N application level.When compared with modern cultivars there was more genetic variation in appearance quality among old cultivars（Fig.3）.No significant differences in LWR were observed across N fertilizer levels（Table 2）.

3.4.Cooking and eating quality

Cooking/eating quality is largely determined by physical and chemical characteristics of the starch in the endosperm:that is，amylose content（AC），gel consistency（GC），and starch pasting properties.There were significant variations among N fertilizer levels and cultivars（Table 2；Fig.4）.At each N level，AC was first decreased rapidly，and then slightly increased with the breeding progress.AC was comparatively lower at a higher N level with average 19.1%，16.7%，and 15.3%at 0 N，240 N，and 360 N levels，respectively（Fig.4-A）.

Across N levels，GC steadily increased from 39.65 to 59.41 mm，with the highest values in the 1970s（70.9 mm）. Application of N significantly reduced GC from 69.11 mm at 0 N to 60.9 mm at 240 N and 56.37 mm at 360 N，a declining trend with increasing N levels.

The Rapid Visco Analyzer（RVA）is a useful tool for rapid assessment of eating and cooking quality of rice.The starch pasting properties analyzed by RVA are shown in Fig.6，including peak viscosity，hot viscosity，cool viscosity，breakdown，setback，and consistency.These starch pasting parameters reflects the texture and stickiness of cooked rice. Application of N fertilizer significantly decreased peak viscosities across all cultivars（Fig.5-A）.Similar patterns of response to N fertilizer were obtained for hot viscosity（Fig.5-B），cool viscosity（Fig.5-C），and breakdown values（Fig.5-D）.In contrast，N fertilizer increased the value of setback（Fig.5-E），but there was no significant effect of N on consistency.No obvious trend was found in starch pasting properties as a consequence of breeding.Compared to 1950s cultivars peak viscosity and breakdown values of modern japonica cultivars were significantly increased by 5.7%and 130.3%，respectively，and setback value was significantly decreased by 141.7%.

In addition to biochemical and biophysical analysis，cooking and eating quality can also be measured by sensory assessment. A rice taste analyzer STA1A was used to rate the overall palatability（Fig.4-C）.Palatability steadily improved with breeding.Compared to 1950s cultivars，overall palatability of modern rice cultivars increased by 20.7%，17.6%，and 22.9%with 0 N，240 N，and 360 N levels，respectively.N applications significantly reduced overall palatability regardless of cultivars.On average，overall palatability value dropped from 73.31 at 0 N to 67.07 at 240 N and 64.07 at 360 N.

3.5.Nutritive quality

Regardless of N application rates，protein content in milled rice was significantly reduced during breeding progress（Fig.6）.It decreased from 8.8%in the 1950s to 7.4%in the 2000s at the 0 N level，from 9.55%in the 1950s to 8.35%in the 2000s at the 240 N level，and from 10.35%in the 1950s to 9.2%in the 2000s at the 360 N level of fertilization.However，the application of N fertilizer significantly increased the protein content（Fig.6；Table 2）.The average protein contents were 7.7%，9.0%，and 9.7% at 0 N，240 N，and 360 N levels，respectively.

Under higher N levels，all micronutrient contents were lower（Fig.7）.This may due to enhanced carbon assimilation under high Napplication，as high Nwould increase the concentration of photosynthetic enzymes in leaves，and the larger yields would dilute micronutrientcontents（Fig.7）.For Cu，the average content dropped from 2.87 mg kg-1at the 0 N level to 2.76 mg kg-1at the 240 Nlevel（-3.8%），and to 2.65 mg kg-1atthe 360 Nlevel（-7.7%）. Similar results were obtained for other micronutrients（Fig.7）. Comparisons between the cultivars within decades，showed thesame tendencies forallmicronutrients；thatis，the micronutrient contents substantially fromthe 1950s to 1980s，and then dropped from the 1980s to 2000s（Fig.7）.

Fig.6-Protein content in milled rice of cultivars released from the 1950s to 2000s.See caption to Fig.2 for other details.Data are average of two years±1 S.E.（vertical bars）.

3.6.Relationships among rice quality traits

As shown in Table 3，correlations between milling quality traits（i.e.BRP，MRP，HRP），appearance traits（i.e.CKP，CS，CK，LWR），cooking and eating quality traits（AC，GC，PV，HV，CV， BD，SB，CT，OP），and nutritive quality traits（i.e.PC，Cu，F(xiàn)e，Mn，Zn，Na，Ca，K，Mg，S，P，B）are very complex.In general，grains with chalkiness，indicative of loosely packed starch granules，were more likely to break and were negatively correlated with milling quality traits and cooking and eating quality traits. Higher protein content increased chalkiness，and was therefore negatively correlated with milling quality traits，appearance quality traits，and cooking and eating quality traits. Positive correlations between micronutrient contents wereobserved（Table 3），implying a coordinated response in root uptake abilities.

Fig.7-Cu，F(xiàn)e，Mn，Zn，Na concentrations（A-C），Ca，K，Mg，S，P concentrations（D-F），and B concentrations（G-I）of rice cultivars from the 1950s，1960s，1970s，1980s，1990s，and 2000s under 0 kg N ha-1（0 N），240 kg N ha-1（240 N），and 360 kg N ha-1（360 N）application levels.Data are average of two years with±1 S.E.（vertical bars）.

Table 3-Pearson correlation coefficients for rice grain quality traits when analyzed for each combination of years，N and cultivars.

4.Discussion

Rice production in China has tripled in the past six decades［10］due to high-yielding varieties and improvement in crop management such as fertilization and irrigation.Next to yield，grain quality is the most important factor in rice production as it is directly related to market value and thus influences farmer incomes.However，to date，little is known about the changes in grain quality over time and how fertilizer application influences grain quality.Some evidence indicates deterioration in grain quality in modern high yielding cultivars［2］.In the present work，we observed that milling quality，appearance quality，and cooking and eating quality were significantly improved with breeding progresses（Figs.2-6）.Cu，F(xiàn)e，Mn，Zn，Na，Ca，K，Mg，S，P，and B are important essential micronutrients in the human diet.Deficiency of these microelements can result serious diseases.During breeding micronutrient contents first increased and then decreased（Fig.7）.The increase in micronutrients from the 1950s to 1980s was partially due to improved micronutrient uptake［13，20］.However，the improvement in root uptake did not keep pace with the large increase in carbon assimilation capacity of super rice as reflected by the decreased of root activity during grain filling［21］，causing the dilution in protein content in the grain from the 1980s to 2000s.

Except for the slight decrease in protein content the above results suggest that simultaneous improvements in both grain yield and grain quality are possible through selective breeding. However，the improvements in grain quality were small with progression from old cultivars to modern cultivars（Figs.2-6）. Less genetic variation was also observed among modern cultivars（2000s）than in early cultivars（1950s-1970s）（Figs.2-C；3-A，C；4；5-E，F(xiàn)）.This may reflect a tendency in crop domestication and improvement as selection forspecific alleles controlling key morphological and agronomic traits［10，22］results in reduced genetic diversity relative to unselected genes［23］，and suggests that a greater effort should be made to improve grain quality in the future.Huang etal.［24］used a large and diverse sample of 950 worldwide rice varieties and found considerable variations in traits related to grain quality such as gelatinization temperature，amylose content，and grain protein content.Further progress could be achieved by introducing alleles conferring grain quality traits from exotic sources to broaden the genetic base of Chinese rice breeding.Alternatively，transgenic alteration ofthe expression patterns ofrelated genes can be tested for desired effects on relevant grain quality traits.

CT OP PC Cu Fe Mn Zn Na Ca K Mg S P -0.36 -0.01 -0.83** -0.36 0.53* -0.50* -0.48* 0.78** -0.64** 0.80** -0.43 0.80** -0.68** 0.90** 0.90** -0.64** 0.71** -0.44 0.74** 0.91** 0.88** -0.21 0.80** -0.79** 0.84** 0.83** 0.93** 0.69** -0.43 0.73** -0.50* 0.62** 0.63** 0.84** 0.71** 0.77** -0.48* 0.89** -0.70** 0.72** 0.84** 0.93** 0.83** 0.89** 0.92** -0.45 0.46 -0.36 0.90** 0.84** 0.84** 0.87** 0.67** 0.55* 0.64** -0.48* 0.42 -0.35 0.82** 0.85** 0.73** 0.83** 0.58* 0.33 0.53** 0.94** -0.22 0.83** -0.84** 0.77** 0.83** 0.87** 0.66** 0.93** 0.62** 0.83** 0.61** 0.59** -0.34 0.70** -0.62** 0.89** 0.81** 0.93** 0.77** 0.94** 0.76** 0.86** 0.76** 0.66** 0.84**

Another option to improve grain quality is through a better crop management，especially N management，as N fertilizer has a significant effect on grain quality（Table 2，F(xiàn)igs.3-7）.Nfertilizer significantly increased chalky kernel percentage，chalky area and chalkiness（Fig.3）.The high extentofchalkiness implies that rice grains grown under a higher N conditions have a lower density of starch granules and therefore become more prone to breakage during milling，as confirmed by negative correlation between chalky kernels percentage（r=-0.74，P＜0.01），chalky area（r=-0.67，P＜0.01），chalkiness（r=-0.76，P＜0.01）and head rice percentage when analyzed for each combination of years，N levels，and cultivars（Table 3）.Similar intrinsic relationships between chalkiness and milling quality are also reported in the literature［25，26］.In this study，a negative relationship between chalkiness and sensory properties was observed（Table 2），in agreement with other reports［5，27］.N fertilizer also significantly reduced the micronutrient content（Fig.8），probably due to the dilution effects of higher yields under high N regimes.Other studies also reported that N application promoted Fe，Mn，Cu，and Zn accumulation in grains of rice and wheat［28，29］.This discrepancy may be because the N application levels of 240 kg N ha-1and 360 kg N ha-1in our experiment were very high.Hao et al.（2007）also reported that N fertilizer decreased micronutrient contents when N applications were higher than 160 kg ha-1［30］.

We observed that the influence of N application on cooking and eating quality was very complex，as cooking and eating quality is described by a number of criteria.Starch，which comprises about 90%of the dry grain［31，32］，is generally believed to be the primary determinant of the cooking and eating quality.As a result，extensive research has been conducted on examining the effects of starch on cooking and eating quality with focus on amylose content［33，34］，solubility of amylose［35］，fine structure of amylopectin［36］，gelatinization and melting temperatures of amorphous and crystalline regions of amylopectin［37］，and the structure of starch granules after heating［38，39］.Endosperm contains two distinct groups of starch granules，A-and B-types.A-type granules（＞10 μm）represent the greatest proportion of endosperm starch by weight（50-90%），whereas B-type granules（＜10 μm）predominate numerically（as high as 99%）［40］. Generally A-type granules contain higher amylose content and exhibit lower gelatinization temperature and higher transition enthalpy than B-type granules［41］.It is reported that N fertilizer decreases the proportion of A-type granules［42］which could partly explain lower amylose content under a higher N levels in our study（Table 1；Fig.5）.Starch texture analyzers，e.g.，the Rapid Visco Analyzer［43］，and human sensory panels［1，25］have been employed to associate measures of starch characteristics in rice to cooking andeating quality.Although starch properties have received a high level of attention，there has been less emphasis on the influence of protein content on cooked rice texture and flavor［34，44］.Protein content in the grain affects the amount of water absorbed in early cooking，and the availability of water in early cooking determines the hydration of the protein and the concentration of the dispersed and viscous phases of the starch，which determines the texture of the cooked rice［45］.In our experiment，N fertilizer levels affected both protein and amylose contents.For all cultivars protein content was significantly higher（Table 2；Fig.6）whereas amylose content was significantly lower at higher N levels（Table 2；Fig.4）. Similar observations were reported in other studies［34，46，47］. Although the lower amylose content under higher N application contributed to improved cooking and eating qualities，the contrasting effects of higher protein content were larger than the amylose effects on grain quality with the increased N rates.Our results showed that palatability was significantly and negatively correlated with protein content（r=-0.83，P＜0.01）and there was no significant correlation between palatability and amylose content（r=-0.20，P＞0.05），implying that a higher N level may cause the deterioration of cooking and eating qualities.Moreover，over-use of N fertilizer often reduces grain yield because of excessive vegetative growth and increased lodging and pest damage.Thus，appropriate N management is essential for both high yield and high quality［44，48，49］.It is reported that，apart from N management，better water management，such as alternate wetting and moderate soil drying during grain filling period could increase both grain yield and grain quality［17］.

Climate trends over the past few decades have been significant.Atmospheric CO2during the industrial era has increased from 278 μmol mol-1to 381 μmol mol-1，and should reach 550 μmol mol-1by 2050［50］.Rising atmospheric CO2as the substrate for photosynthesis directly affects the yield and quality of rice.Increases in atmospheric CO2could lead to increased temperature.The available information in the literature indicates a clear tendency for quality deterioration with increased temperatures，especially during the grain filling period （see review of Wang et al.［26］）.In this experiment we studied the grain quality of japonica rice cultivars released from the 1950s to 2000s.Considering the climate change in the past 60 years，the grain quality of the tested cultivars under current climatic conditions might be not the same as those during the earlier decades.The improvement of grain quality during breeding selection from the 1950s to 2000s could be partly due to a response to climate change.Understanding the factors causing quality changes under conditions of temperature increase due to rising CO2and the related biological mechanisms might be crucial in achieving both high-quality and high-yield in rice production systems.

5.Conclusions

Grain yield progressively increased with the release of new cultivars.The response to N rates was greater for the cultivars bred in the early decades than for the modern super rice cultivars.The grain yield was MN （240 kg ha-1）＞HN（360 kg ha-1）＞0 N（N omission）for the cultivars bred in the 20th century，whereas it was HN＞MN＞0 N for super rice cultivars bred since the 20th century.Regardless of N rates，milling quality and appearance quality of japonica rice cultivars have progressively improved.Compared to early japonica rice cultivars，the taste quality of modern super rice varieties is much improved.Application of N fertilizer significantly increased protein content in the grain but did not improve other quality traits.As N rates changed from 0 to 360 kg ha-1，the appearance and taste qualities decrease.Further research is needed to simultaneously increase grain yield and quality and Nuse efficiency through the improvement in N management in rice production.

Acknowledgments

This work was supported by grants from the National Natural Science Foundation of China（31461143105，31271641，31471438），the National Key Technology R&D Program of China（2011BAD16B14，2012BAD04B08，2014AA10A605），and the Priority Academic Program Development of Jiangsu Higher Education Institutions（PAPD-2014-2）.

［1］M.A.Fitzgerald，S.R.McCouch，R.D.Hall，Not just a grain of rice:the quest for quality，Trends Plant Sci.14（2009）133-139.

［2］Q.Zhang，Strategies for developing green super rice，Proc. Natl.Acad.Sci.U.S.A.104（2007）16402-16409.

［3］Y.Han，M.Xu，X.Liu，C.Yan，S.S.Korban，X.Chen，M.Gu，Genes coding for starch branching enzymes are major contributors to starch viscosity characteristics in waxy rice（Oryza sativa L.），Plant Sci.166（2004）357-364.

［4］S.D.Koutroubas，F(xiàn).Mazzini，B.Pons，D.A.Ntanos，Grain quality variation and relationships with morphophysiological traits in rice（Oryza sativa L.）genetic resources in Europe，F(xiàn)ield Crop Res.86（2004）115-130.

［5］A.J.Lisle，M.Martin，M.A.Fitzgerald，Chalky and translucent rice grains differ in starch composition and structure and cooking properties，Cereal Chem.77（2000）627-632.

［6］B.O.Juliano，The rice caryopsis and its composition，in:D.F. Houston（Ed.），Rice:Chemistry and Technology，American Association Cereal Chemists，St.Paul，MN 1972，pp.16-74.

［7］C.M.Perez，B.O.Juliano，Modification of the simplified amylose test for milled rice，Starch-Starke 30（1978）424-426.

［8］J.S.Bao，M.Sun，H.Corke，Analysis of the genetic behavior of some starch properties in indica rice（Oryza sativa L.）:thermal properties，gel texture，swelling volume，Theor.Appl.Genet. 104（2002）408-413.

［9］S.H.Cheng，J.Y.Zhuang，Y.Y.Fan，J.H.Du，L.Y.Cao，Progress in research and development on hybrid rice:a super-domesticate in China，Ann.Bot.100（2007）959-966.

［10］S.Peng，Q.Tang，Y.Zou，Current status and challenges of rice production in China，Plant Prod.Sci.12（2009）3-8.

［11］FAOSTAT，F(xiàn)AO Statistical databases，F(xiàn)ood and Agriculture Organization（FAO）of the United Nations，Rome，http://www. fao.org/2008.

［12］S.Peng，R.J.Buresh，J.Huang，X.Zhong，Y.Zou，J.Yang，G. Wang，Y.Liu，R.Hu，Q.Tang，Improving nitrogen fertilization in rice by site-specific N management:a review，Agron. Sustain.Dev.30（2010）649-656.

［13］H.Zhang，T.Chen，L.Liu，Z.Wang，J.Yang，J.Zhang，Performance in grain yield and physiological traits in the Yangtze River Basin of China during the last 60 years，J. Integr.Agric.12（2013）57-66.

［14］Ministry of Agriculture P.R.China，Standard of Ministry of Agriculture，P.R.China（NY 83-88）-Methods for Rice Quality Measurement，Chinese Standard Press，Beijing，1988.（in Chinese）.

［15］International Organisation for Standardisation，ISO 6647: 1987E，Rice:Determination of Amylase Content1987.

［16］A.Borrell，A.Garside，S.Fukai，Improving efficiency of water use for irrigated rice in a semi-arid tropical environment，F(xiàn)ield Crop Res.52（1997）231-248.

［17］Z.Zhang，S.Zhang，J.Yang，J.Zhang，Yield，grain quality and water use efficiency of rice under non-flooded mulching cultivation，F(xiàn)ield Crop Res.108（2008）71-81.

［18］G.H.Ryu，P.E.Neumann，C.E.Walker，Pasting of wheat flour exudates containing conventional baking ingredients，J.Food Sci.58（1993）567-573.

［19］H.C.Lai，K.Motonobu，Z.D.Wang，M.Takashi，D.Q.Huang，H. Li，D.C.Lu，D.G.Zhou，S.C.Zhou，Cooking and eating quality of indica rice varieties from south China by using rice taste analyzer，Chin.J.Rice Sci.25（2011）435-438（in Chinese with English abstract）.

［20］L.Liu，H.Zhang，C.Ju，Y.Xiong，J.Bian，B.Zhao，J.Yang，Changes in grain yield and root morphology and physiology of mid-season rice in the Yangtze River Basin of China during the last 60 years，J.Agric.Sci.6（2014）1-15.

［21］H.Zhang，Y.Xue，Z.Wang，J.Yang，J.Zhang，Morphological and physiological traits of roots and their relationships with shoot growth in“super”rice，F(xiàn)ield Crop Res.113（2009）31-40.［22］S.Peng，G.S.Khush，P.Virk，Q.Tang，Y.Zou，Progress in ideotype breeding to increase rice yield potential，F(xiàn)ield Crop Res.108（2008）32-38.

［23］M.Yamasaki，A large-scale screen for artificial selection in maize identifies candidate agronomic loci for domestication and crop improvement，Plant Cell 17（2005）2859-2872.

［24］X.Huang，Y.Zhao，X.Wei，C.Li，A.Wang，Q.Zhao，W.Li，Y. Guo，L.Deng，C.Zhu，Genome-wide association study of flowering time and grain yield traits in a worldwide collection of rice germplasm，Nat.Genet.44（2012）32-39.

［25］L.Yang，Y.Wang，G.Dong，H.Gu，J.Huang，J.Zhu，H.Yang，G. Liu，Y.Han，The impact of free-air CO2enrichment（FACE）and nitrogen supply on grain quality of rice，F(xiàn)ield Crop Res.102（2007）128-140.

［26］Y.Wang，M.Frei，Q.Song，L.Yang，The impact of atmospheric CO2concentration enrichment on rice quality:a research review，Acta Ecol.Sin.31（2011）277-282.

［27］J.M.Sui，X.Li，S.Yan，C.J.Yan，R.Zhang，S.Z.Tang，J.F.Lu，Z.X. Chen，M.H.Gu，Studies on the rice RVA profile characteristics and its correlation with the quality，Sci.Agric.Sin.38（2005）657-663（in Chinese with English abstract）.

［28］S.Z.E，J.C.Yuan，Z.Y.Ding，F(xiàn).J.Yao，X.P.Yu，F(xiàn).X.Luo，Effect of N，P，K，fertilizer on Fe，Zn，Cu，Mn，Ca，and Mg contents and yields in rice，Chin.J.Rice Sci.19（2005）434-440（in Chinese with English abstract）.

［29］S.Z.Cheng，S.Y.Jie，Study of effect of N fertilizer on Zn，Mn，F(xiàn)e，Mg content of barley，Chin.J.Soil Sci.30（1995）71-73（in Chinese with English abstract）.

［30］H.L.Hao，Y.Z.Wei，X.E.Yang，Y.Feng，C.Y.Wu，Effects ofdifferent nitrogen fertilizer levels on Fe，Mn，Cu and Zn concentrations in shoot and grain quality in rice（Oryza sativa），Rice Sci.14（2007）289-294.

［31］S.Yoshida，Physiological aspects of grain yield，Annu.Rev. Plant Physiol.23（1972）437-464.

［32］J.Yang，J.Zhang，Grain-filling problem in‘super'rice，J.Exp. Bot.61（2010）1-5.

［33］B.O.Juliano，G.B.Cagampang，L.J.Cruz，R.G.Santiago，Some physicochemical properties of rice in Southeast Asia，Cereal Chem.20（1964）275-286.

［34］E.T.Champagne，K.L.Bett-Garber，J.L.Thomson，M.A. Fitzgerald，Unraveling the impact of nitrogen nutrition on cooked rice flavor and texture，Cereal Chem.86（2009）274-280.

［35］G.E.Vandeputte，V.Derycke，J.Geeroms，J.A.Delcour，rice starches:II.Structural aspects provide insight into swelling and pasting properties，J.Cereal Sci.38（2003）53-59.

［36］Z.A.Syahariza，S.Sar，J.Hasjim，M.J.Tizzotti，R.G.Gilbert，The importance of amylose and amylopectin fine structures for starch digestibility in cooked rice grains，F(xiàn)ood Chem.136（2013）742-749.

［37］D.L.E.Waters，R.J.Henry，R.F.Reinke，M.A.Fitzgerald，Gelatinization temperature of rice explained by polymorphisms in starch synthase，Plant Biotechnol.J.4（2006）115-122.

［38］S.Hizukuri，Relationship between the distribution of the chain length of amylopectin and the crystalline structure of starch granules，Carbohydr.Res.141（1985）295-306.

［39］C.Wei，F(xiàn).Qin，W.Zhou，B.Xu，C.Chen，Y.Chen，Y.Wang，M. Gu，Q.Liu，Comparison of the crystalline properties and structural changes of starches from high-amylose transgenic rice and its wild type during heating，F(xiàn)ood Chem.128（2011）645-652.

［40］S.V.Shinde，J.E.Nelson，K.C.Huber，Soft wheat starch pasting behavior in relation to A-and B-type granule content and composition，Cereal Chem.80（2003）91-98.

［41］P.Van Hung，N.Morita，Physicochemical properties of hydroxypropylated and cross-linked starches from A-type and B-type wheat starch granules，Carbohydr.Polym.59（2005）239-246.

［42］R.C.Kaufman，J.D.Wilson，S.R.Bean，D.R.Presley，H. Blanco-Canqui，M.Mikha，Effect of nitrogen fertilizer and cover cropping systems on sorghum grain characteristics，J. Agric.Food Chem.61（2013）5715-5719.

［43］E.T.Champagne，K.L.Bett-Garber，B.T.Vinyard，A.M. McClung，F(xiàn).E.Barton，K.Moldenhauer，S.Linscombe，K. McKenzie，Correlation between cooked rice texture and Rapid Visco Analyser measurements，Cereal Chem.76（1999）764-771.

［44］J.M.Alcantara，K.G.Cassman，M.P.Consuelo，O.J.Bienvenido，P.L.Samuel，Effects of late nitrogen fertilizer application on head rice yield，protein content，and grain quality of rice，Cereal Chem.73（1996）556-560.

［45］M.Martin，M.A.Fitzgerald，Proteins in rice grains influence cooking properties，J.Cereal Sci.36（2002）285-294.

［46］Y.S.Prakash，P.B.S.Bhadoria，A.Rakshit，Relative efficacy of organic manure in improving milling and cooking quality of rice，Int.Rice Res.Notes 27（2002）43-44.

［47］P.D.Larkin，A.M.McClung，N.M.Ayres，W.D.Park，The effect of the waxy locus（granule bound starch synthase）on pasting curve characteristics in specialty rices（Oryza sativa L.），Euphytica 131（2003）243-253.

［48］M.M.Wopereis-Pura，H.Watanabe，J.Moreira，M.C.S. Wopereis，Effect of late nitrogen application on rice yield，grain quality and profitability in the Senegal River valley，Eur. J.Agron.17（2002）191-198.

［49］M.Ghosh，B.K.Mandal，B.B.Mandal，S.B.Lodh，A.K.Dash，The effect of planting date and nitrogen management on yield and quality of aromatic rice（Oryza sativa），J.Agric.Sci.142（2004）183-191.

［50］IPCC，Climate change 2001:The Scientific basis，Intergovernmental Panel on Climate Change，Cambridge University Press，Cambridge，UK，2001.

*Corresponding author.

E-mail address:jcyang@yzu.edu.cn（J.Yang）.

Peer review under responsibility of Crop Science Society of China and Institute of Crop Science，CAAS.

http://dx.doi.org/10.1016/j.cj.2015.03.007

2214-5141/?2015 Crop Science Society of China and Institute of Crop Science，CAAS.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.