Establishment Of Evaluation System For Nitrogen Use Efficiency (NUE) And Study On The Mechanisms Of High-NUE Development In Rice

Crop yields in the world have continuously increased partly because of the increasein fertilizer nutrient input, especially N fertilizer. However, nitrogen use efficiency (NUE)is relatively low in irrigated lowland rice because of rapid N losses from ammoniavolatilization, denitrification, surface runoff, and leaching in the soil-floodwater system.Irrigated rice in China accounts for nearly 37% of global nitrogen consumption in riceproduction. Average rate of N application for rice production in China is high andfertilizer-N use efficiency is low compared with other major rice growing countries. Themass N losses and low nitrogen use efficiency of rice production system has become athreat to the environment. Idea and practical approaches to improve nitrogen useefficiency and avoid nitrogen destruction of natural ecosystems are considered to use highNUE rice varieties and advanced fertilizer-N application technologies including newapplication methods and modified N sources, and optimizing the timing and rate of Napplication. Some efforts have been devoted to germplasm improvement in NUE. Inpresent study, pot experiments were carried out at the Huazhong Agricultural Universitygreenhouse in 2004 and 2005. Field experiments were conducted at the International RiceResearch Institute farm in dry and wet seasons in 2005. The objectives of this study were(1) to compare rice performance under fertilizer (+F) and non-fertilizer (-F) conditions, (2)to identify genotypic variation in NUE of different rice materials and among indicainbreds, indica hybrids, NPT lines and japonica inbreds, (3) to evaluate the characteristicsof NUE-related traits during N-efficiency formatting in rice, (4) to establish oneevaluation system for NUE in rice genotypic screening, (5) to understand thephysi-biochemistry mechanisms on genotypic variation in rice NUE. The main resultswere showed as follows:(1) Different performances were observed in rice under +F and -F conditions. Themean of plant height (PHP) at physiological maturity (PM) stage under +F condition washigher 11.6% to 22.1% than that under -F condition. Flag leaf of rice became shorterunder -F, depending on rice genotypes and growth seasons. TDW (total aboveground dryweight), FNUEb (Nitrogen biomass physiology efficiency), G_w/N_g (the ratio of Grain drymatter to Grain N accumulation), GNPP (Grain N content per panicle), and FNUEb-FL(Nitrogen biomass physiology efficiency at flowering stage) were positively associatedwith PHP. However, HI (Harvest index) and NTR (Grain nitrogen translocation ratio)showed negatively correlated with PHP.Genotypic differences in whole growth duration was observed under both -F and +EWhole growth duration was elongated by fertilizer-N application compared with thatunder-F. Increases in FNUEb-FL, FNUEb-PM (Nitrogen biomass physiology efficiencyat maturity stage), and PE (Nitrogen grain physiology efficiency) were found amonglonger duration genotypes. Nitrogen grain production efficiency (NUE_g) was higher ingenotypes with medium duration than that in genotypes with short duration. However, adecrease in NUE_g was observed among longer duration genotypes. Furthermore, RE(Recovery use efficiency) in longer duration genotypes was lower than that in shortduration genotypes at IRRI in 2005DS and WS.More tillers, more panicles, and lower productive tiller percentage were found under+F than those under -F. Leaf color represented by SPAD (Soil and plant analysis development section) reading was very susceptible to fertilizer-N application. Differencesin SPAD reading among rice genotypes and between two nitrogen levels were observeddue to the variation in nitrogen uptake.(2) The differences in dry matter among rice genotypes and groups at main growthstage, depending on both of the plant growth and development and environment.Increases in dry matter accumulation under +F than that under -F ranged from 3.31(Dular)to 5.66 (Starbonnet) folds in 2004HZAU (in 2004 at Huazhong Agricultural University),from 1.10 (Zheng dao 88) to 2.01(â…¡-32B) folds in 2005HZAU, from 0.61(IR72903-121-2-1-2) to 1.08 (IR71700-247-1-1-2) folds in 2005DS (in 2005 dry season)and from 0.13 (IR72) to 0.33 (IR75217H) folds in 2005WS (in 2005 wet season),respectively.(3) Total aboveground nitrogen accumulation increased with the growth anddevelopment in rice and approached to the maximum at PM stage. Genotypes differed intotal N uptake at all main growth stages. The differences in total N uptake amonggenotypes were consistent with dry matter at panicle initiation (PI) and flowering (FL)stages, respectively.(4) Genotypic differences in the responses of grain yield and its components tofertilizer-N application were observed. Increases in panicles per unit area were foundwhen fertilizer-N applied. In general, increases in panicles per unit area, spikelets perpanicle and grain weight, and decrease in grain filling occurred for almost all genotypeswhen fertilizer-N applied.(5) Soil indigenous N supply and fertilizer-N input rate had a large influence onNUE. Significant genotypic variation in NUE was observed in rice by using differentNUE parameters, which represent different NUE components and can be divided intothree groups including uptake efficiency, utilization efficiency, and NUE-related traits. HI,NPI (Nitrogen production index), and NHI (Nitrogen harvest index) of japonica inbredsunder -F and +F conditions were lower than those of indica inbreds, while TDW andNUEb-PM were higher in japonica inbreds under -F and +F than those in indica inbreds.Indica hybrids had the highest HI and GNPP among 4 rice groups including indicainbreds, japonica inbreds, indica hybrids, and NPT (New plant type) lines. NPI, NTE andNHI/HI were higher in indica inbreds than those in NPT lines. Interestingly, indicahybrids had higher PE, RE, and AE than indica inbreds. PE and AE were higher in NPTlines than in indica inbreds. However, some NUE parameters were not significantlydifferent among rice genotypes.(6) Straw nitrogen concentration (Ns%) was negatively correlated with all ofnitrogen translocation efficiency (NTE), grain nitrogen per panicle, nitrogen productionindex (NPI), and nitrogen biomass physiology efficiency at flowering stage (FNUE_b-FL)at IRRI in 2005DS and at HZAU in 2005. Grain nitrogen concentration (Ng%) was alsonegatively correlated with total dry matter weight (TDW), straw nitrogen content,nitrogen dry matter production efficiency at flowering stage (NUE_b-FL), nitrogen drymatter production efficiency at PM stage (NUE_b-PM), and nitrogen biomass physiologyefficiency at physiological maturity (FNUE_b-PM) at IRRI in 2005DS and at HZAU in2005, respectively. However, negative correlations were only observed between Ns% andnitrogen grain production efficiency (NUE_g) and NPI, respectively at IRRI in 2005WS.Ng% was also negatively correlated with each of NUE_b-FL, NUE_b-PM, and FNUE_b-FL atIRRI in 2005WS. Genotypic variation in Ng% was caused largely by N supply andmanagement prastics.NHI was significantly and positively correlated with grain yield, total aboveground Ncontent (TN), and NUE_g, while it was negatively correlated with Ns% in 2005HZAU.NHI and grain nitrogen translocation ratio (NTR) were positively correlated with TN in 2005DS.(7) Various parameters were combined to evaluate NUE among rice genotypes,considering both high yield and high NUE. The rankings of genotypes were differences,depending on different N supply and growth season. Interestingly, GNPP provided wellgenotypic rankings that were consistent at all N levels and between different seasons. TheGNPP could be used as an important evaluation index for NUE.(8) Efforts were made to establish a simple and quick method for estimating NUEwith grain yield, total aboveground N content, NUEg, NUE_b-PM, and GNPP in this study.According to this rank resulted, sixteen genotypes were classified into four groupsincluding high NUE types, middle NUE type (above middle and below middle), and lowNUE types.(9) Grain yield and NUE differed significantly among genotypes and N supply inN-efficient and N-inferior rice groups. Total N uptake of all groups increased quickly withmore N supply. The increase amount in TN became small in N-inferior rice thanN-efficient at relatively high N supply level. Grain yield and total aboveground biomassalso increased consistently with more N applied. More N application resulted in more drymatter translocation to grains during filling period.(10) A slight decrease in NUE was observed in irrigated lowland rice at high N levelin comparison with that at low N supply. Excess of nitrogen during the late grain fillingperiod resulted in a significant accumulation of nitrogen in straw, which produced highstraw N concentration. Genotypes with lower percentage straw N (Ns%) at maturity hadhigher NUE_g NUE_b-PM and GNPP. NUE-related parameters were negatively correlatedwith grain and straw N concentration at maturity. NHI showed significantly and positivelycorrelations with NUE_g, NUE_b-PM, and grain N per panicle. Those results indicated thatlow grain or straw N concentrations may be indicators of higher NUE.(11) Genotypic differences in physio-biochemical response to fertilizer-N applicationwere significant. The soluble protein content in leaves was relatively lower inN-efficiency genotypes than N-inferior genotypes at panicle initiation stage, while theglutamine synthetase (GS) activity was higher significantly. Ribulose-1,5-bisphosphatecarboxylase/oxygenase content (Rubisco) in leaves appeared to be no difference amongvarious NUE genotypes. The net photosynthetic rate (P_n) in the N-inferior genotypes wassignificantly lower by 28.67% in comparison with that in N-efficiency genotypes at fullheading stage, regardless of no significant difference at panicle initiation stage.Interestingly, P_n per Chlorophyll unit (P_n/Ch1) was lower in N-inferior genotypes thanN-efficiency genotypes at both panicle initiation stage and full heading stage about by18.48% and 29.69%, respectively. P_n at full heading stage, P_n/Ch1 and GS activity weresignificantly and positively correlated with NUE-related traits including agronomyefficiency (AE), partial factor production of applied N (PFP), nitrogen grain physiologyefficiency (FNUEg-PM or PE), NUE_g, NUE_b-PM and NHI, respectively. However, PFP,AE, NUE_b-PM, NUE_g, NHI, and FNUE_b-PM were negatively correlated with solubleprotein in leaves, respectively. Also photosynthetic nitrogen use efficiency (PNUE) waspositively and significantly correlated with AE in six tropic rice genotypes. Theserelationships suggest that PNUE at reproductive growth stage of rice, P_n/Ch1, GS activityand soluble protein content in leaves might be candidate indicators for estimating NUE inrice.