Crop Science Illumina
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 6 May 2005
Published in Crop Sci 45:1141-1150 (2005)
© 2005 Crop Science Society of America
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 Related articles in Crop Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (25)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barbottin, A.
Right arrow Articles by Jeuffroy, M.-H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Barbottin, A.
Right arrow Articles by Jeuffroy, M.-H.
Agricola
Right arrow Articles by Barbottin, A.
Right arrow Articles by Jeuffroy, M.-H.
Related Collections
Right arrow Nitrogen
Right arrow Wheat
Right arrow Plant and Environment Interactions

CROP PHYSIOLOGY & METABOLISM

Nitrogen Remobilization during Grain Filling in Wheat

Genotypic and Environmental Effects

Aude Barbottina, Christophe Lecomteb, Christine Bouchardc and Marie-Hélène Jeuffroyc,*

a Unité Mixte d‘Agronomie, INRA-INAPG, BP01, 78 850 Thiverval-Grignon, France
b Unité de Recherche en Génétique et Ecophysiologie des Légumineuses INRA, 17 rue de Sully, BP 86510, 21 065 Dijon, France
c INRA Unité Mixte d'Agronomie, INRA-INAPG, BP01, 78 850 Thiverval-Grignon, France

* Corresponding author (jeuffroy{at}grignon.inra.fr)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In wheat (Triticum aestivum L.), nitrogen remobilization from the vegetative organs of the crop to the grains has been shown to depend on environmental factors and genotype. We performed, for a set of 10 winter wheat genotypes, field experiments at six sites over a 2-yr period. By measuring nitrogen uptake at flowering (NUF from 32–284 kg ha–1), the amount of remobilized nitrogen (REMN from 24–228 kg ha–1) and nitrogen remobilization efficiency (NRE from 0.44–0.92) we were able to determine the effect of genotype and environment on the relationship between REMN and NUF. Environment and genotype had significant effects on nitrogen remobilization and nitrogen remobilization efficiency, which mainly depended on treatment (nitrogen and fungicide) and site. For environments without limiting factor during the grain-filling period, we found that REMN was not dependent on the genotype and could be estimated by a single two-parameter linear relationship (REMN = 4.13 + NUF x 0.76, r2 = 0.97). We analyzed the effect of drought stress before and after flowering, high temperature during these periods, nitrogen availability and disease pressure on REMN by comparing observed and estimated REMN. The effect of the environment on the relationship between nitrogen uptake at flowering and nitrogen remobilization depended on nitrogen uptake during grain-filling period and disease pressure and was also affected by genotype. Disease-resistant genotypes seemed to be able to keep remobilization efficiency stable in conditions of strong disease pressure, whereas nitrogen remobilization efficiency decreased strongly in susceptible genotypes under the same conditions.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
IN CEREALS, the main source of nitrogen for the grains is nitrogen remobilized from the vegetative parts (Simpson et al., 1983). This source accounts for 60 to 92% of the nitrogen accumulated in the grains at maturity (Austin et al., 1977; Cox et al., 1985b; Papakosta and Garianas, 1991). The amount of nitrogen remobilized depends on nitrogen remobilization efficiency and the amount of nitrogen available. The amount of nitrogen stored in the vegetative organs of the plant at flowering can be used to estimate the amount of nitrogen available for remobilization (Cox et al., 1986). The amount of stored nitrogen depends in turn on soil nitrogen supply, nitrogen fertilization level (Papakosta and Garianas, 1991), growing conditions during the preanthesis period and genotype (Cox et al., 1985b; Rodgers and Barneix, 1988). Some of the nitrogen stored in the plant is used for structural purposes (Lhuillier-Soundélé et al., 1999). Nitrogen remobilization efficiency (NRE) is therefore estimated as the fraction of stored nitrogen at flowering that is not recovered in the vegetative parts at maturity (Cox et al., 1985a, 1985b; Van Sanford and Mackown, 1987). NRE depends on growing conditions during the grain filling period and genotype.

Palta et al. (1994) showed that NRE was high in the Mediterranean-like conditions of western Australia, in which plants generally suffer from water stress during the grain-filling period. This increase in NRE may result indirectly from limits on nitrogen uptake during the grain-filling period, forcing the plant to make greater use of its stored nitrogen. Cox et al. (1986) showed that higher levels of nitrogen fertilization before flowering lead to a decrease in nitrogen remobilization efficiency as the resulting higher level of nitrogen availability just after flowering renders nitrogen remobilization unnecessary. Similar results were obtained by Moll et al. (1982) and Przulj and Momcilovic (2001) with barley (Hordeum vulgare L.). Halloran (1981) reported an increase in NRE in conditions unfavorable for nitrogen uptake before anthesis, linked to drought or high temperature. Foliar diseases, such as brown and yellow rusts (cause by Puccinia spp.), Septoria blotch (caused by Septoria tritici Roberge in Desmaz), and powdery mildew (caused by Erysiphe graminis DC. f. sp. tritici Em. Marchal) have also been shown to reduce nitrogen translocation from the vegetative parts of the plant to the grains during the grain-filling period (Dimmock and Gooding, 2002). Some studies have reported that NRE also varies according to genotype (Cox et al., 1986). Nevertheless, this genotype effect has been shown to depend on year (Przulj and Momcilovic, 2001) and fertilization level (Cox et al., 1986; Papakosta and Garianas, 1991).

Genotype and many environmental factors are known to affect nitrogen translocation. It is therefore important to analyze the effect of each factor (genotype and environment) on the amount of nitrogen remobilized and on nitrogen remobilization efficiency. In this study, we investigated the variability of nitrogen remobilization from vegetative organs to grains, in various environments and for various genotypes, to evaluate the effect of these factors on the relationship between remobilized nitrogen and nitrogen uptake at flowering.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Multisite Trials
Field trials were conducted over 2 yr (2001 and 2002) in France at six sites of the Institut National de la Recherche Agronomique (I.N.R.A) and one experimental station belonging to a plant-breeding company (SERASEM). The experimental sites were at Toulouse (T) (43°31' N, 01°28' E), Clermont-Ferrand (C) (45°46' N, 03°05' E), Dijon (D) (47°19' N, 5°01' E), Le Moulon (L) (48°48' N, 2°08' E), Mons (M) (49°56' N, 2°56' E), Lille (S) (50°39' N, 02°57' E) and Rennes (R) (48°06' N, 1°47 W). At each site, a complete randomized block design was used. Plot size varied among sites, from 5.9 m2 at Rennes to 7.8 m2 at Le Moulon. Wheat was sown at the optimal date for each region, varying from 15 October at Le Moulon to 10 November at Rennes (Table 1). At the end of winter, plant density exceeded 300 plants/m2 at each site. The experimental sites also differed in soil nitrogen supply at the end of winter. The amount of soil mineral nitrogen available for the crop at the beginning of active growth ranged from 30 kg ha–1 at Lille to 175 kg ha–1 at Toulouse in 2002. These differences in the amount of mineral nitrogen available at the end of winter resulted from differences, among sites and years, in the nature of the preceding crop. Climatic conditions throughout the crop cycle differed also among sites, in terms of mean temperature (from 9.71°C at Lille to 12.90°C at Toulouse), cumulative rainfall (from 320 mm at Clermont-Ferrand to 859 mm at Rennes) and cumulative solar radiation from 2702 MJ m–2 at Lille to 3607 MJ m–2 at Toulouse (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Characteristics of each experimental site; sowing date, climatic data during the crop cycle (mean temperature: Mean T, cumulative global radiation: CGR and cumulative rainfall: CR), soil nitrogen content at the end of winter (Nsoil) and nature of the preceding crop.

 
At each experimental site, two or three experimental treatments, differing in the amounts of nitrogen and fungicides applied (Table 2), were compared. The intensive treatment (in) correspond to the treatment for which nitrogen was applied to prevent nitrogen deficiency during the crop cycle. The total amount of nitrogen applied was calculated with the balance-sheet method (Machet et al., 1990, p. 21) on the basis of the potential yield specific to each site (from 9 Mg ha–1 at Clermont-Ferrand to 10 Mg ha–1 at Mons). Nitrogen was applied at two or three stages: at the end of winter, at the beginning of stem elongation and at heading. Pesticides were applied as necessary to prevent weeds and diseases from affecting crop growth. The reduced nitrogen treatment (nr) was obtained by applying a rate of 100 kg ha–1 of nitrogen lower than for the intensive treatment. The other techniques used were similar to those for the intensive treatment. The "no fungicide" treatment (nt) was defined as similar to the intensive treatment (all techniques were similar to those for the intensive treatment) except that no fungicide was applied to the crop during the crop cycle.


View this table:
[in this window]
[in a new window]
  		Table 2. Treatments applied at each experimental site in the multisite trial.

 
Specific Experiment at Grignon
We performed a specific trial at the I.N.R.A. experimental site at Grignon (G) (48°N, 1.9°E) in 2001 and 2002, on a loamy soil with low levels of available nitrogen at the end of winter (18 kg ha–1 in 2001 and 20 kg ha–1 in 2002). We aimed to create a diversity of crop N nutrition conditions before flowering. Thus, various amounts of ammonium nitrate fertilizer were applied in 2001 and 2002, as described in Table 3. An irrigation system protected the crop against drought stress throughout the crop cycle and ensured that the N fertilizer applied rapidly became available. Frequent applications of pesticides during the crop cycle prevented weeds, pests and diseases.


View this table:
[in this window]
[in a new window]
  		Table 3. Nitrogen fertilizer application for the specific experiments at Grignon. Each environment created at the experimental site at Grignon is defined by a code indicating the site (G for Grignon), year (1: 2001, 2: 2002) and treatment. Treatments are defined as nn: no nitrogen supply treatment, lw: low nitrogen supply treatment, nr: reduce nitrogen supply treatment, in: intensive nitrogen supply treatment and hn: high nitrogen supply treatment. Reduce nitrogen (nr) and intensive treatment (in) are similar to those used in the multilocal trial.

 
Wheat Genotypes
Ten winter wheat genotypes differing in terms of earliness for heading and maturity, potential yield, and susceptibility to aerial diseases were used in this study (Table 4). They included high-yield genotypes (Arche, Isengrain, and Rumba) with low grain protein content, medium-yield cultivars with medium to high grain protein content (Camp-Rémy, DI9714, Récital, Soissons), a hybrid cultivar (Hynoprécia) and two multi-disease-resistant cultivars (Oratorio and Renan). Each genotype was present in each environment of the multisite and specific trials, except for Soissons, which was absent from the intensive treatment plots at Grignon in 2002.


View this table:
[in this window]
[in a new window]
  		Table 4. Genotype characteristics: year of registration, earliness at heading (from 1, very late, to 9, very early), susceptibility to the main foliar diseases (from 1, highly sensitive, to 9, highly resistant).

 
Plant Measurements
Measurements were made at flowering and just after grain maturity was reached, for each experimental site, treatment and genotype. Samples of 0.35 m2 (Grignon) and 0.40 m2 (other sites) from the center rows were cut at ground level for each of the three blocks. These samples were separated into vegetative (leaf and culm) and reproductive parts (ears). Each sample was oven-dried at 80°C for 72 h and weighed. At maturity, ears were separated into grains and chaff. Grains were oven-dried again and weighed. Chaff weight was calculated from the difference between ear dry matter and grain dry matter. After drying, all samples (ears and vegetative parts at flowering, grains, leaves plus culms at maturity) were ground in a mill to generate 1-mm particles. We analyzed the nitrogen content of all components, except chaff at maturity, with a Carlo-Erba NA 1500 CN Analyzer (Fisons Instrument-Thermoelectron) at Grignon's laboratory. Ten-milligram samples were analyzed by the Dumas (1831) method, which consisted of the combustion of the sample, separation of the different components (N2, H2O, CO2, O2), and quantification of the N2 content. Dumas method was used at all sites except Dijon where grain nitrogen content at maturity was determined by near infrared reflectance (NIRS 6500, Foss-France) on 120 g of grain. The nitrogen content values determined from NIR were checked by testing samples with the Kjeldahl procedure. The Dumas method involves the combustion of dehydrated and ground plant tissue at about 1800°C, the reduction of N oxides by reduced Cu at 600°C, and N2 analysis by catharometry. It can be used to estimate the total N content of the plant, including nitrate. The nitrogen content of the chaff at maturity was assumed to be equal to that of the leaves plus culms at maturity.

Crop biomass and the amount of nitrogen in the aerial parts of the crop were calculated at flowering and maturity, by summing values for individual organs. Grain yield and protein content were determined at harvest. Grain protein content at harvest was calculated as grain nitrogen concentration multiplied by a coefficient of 5.7.

The amount of remobilized nitrogen (REMN, kg ha–1) was estimated as grain nitrogen content at maturity (GNC, kg ha–1) not due to nitrogen uptake after flowering (NUAF, kg ha–1), as described by Cox et al. (1985a). Nitrogen uptake after flowering was estimated as the difference between total nitrogen uptake at maturity (grains + vegetative parts) and nitrogen uptake at flowering (ear + vegetative parts).

Nitrogen remobilization efficiency (NRE) was estimated as the fraction of nitrogen taken up at flowering (NUF, kg ha–1) that was remobilized (Cox et al., 1985b).


Characterization of the Environments
Environments were defined as combinations of experimental site, year, and treatment. In multienvironment trials, genotypes often differ in terms of earliness. It is therefore unlikely that all genotypes were subjected to the same limiting factors or to the same intensity of limiting factor. We characterized each environment for the main yield-limiting factors identified, with a diagnostic method and three probe genotypes, as described by Brancourt et al. (1999). These three genotypes were chosen on the basis of their known response to several environmental factors and to cover the range of earliness of the 10 genotypes used in this study. The probe genotypes used were Camp-Rémy, Récital, and Soissons. On a scale of earliness at heading from 1 (very late) to 9 (very early), Camp-Rémy scores 6, Soissons scores 7, and Récital scores 8. In each environment, the grain yields obtained were compared with potential yields for each probe genotype. Potential yields were determined as described by Brancourt et al. (1999), using a bootstrap procedure: 1000 samples of 200 points were taken from a large body of available experimental data (382 data for Camp-Rémy, 274 data for Récital, and 829 data for Soissons). Maximum yield was determined for each sample, and the mean and standard deviation of the maxima obtained for each sample were taken as the potential yield values (10.3 Mg ha–1 ± 0.5 for Camp-Rémy, 10.1 Mg ha–1 ± 0.4 for Récital, and 10.5 Mg ha–1 ± 0.6 for Soissons). The differences between observed and potential yields should result from limiting factors in the precise conditions in which the crop was grown.

The crop cycle was divided into several main periods, corresponding to the formation of the two yield components—grain number per square meter and mean weight per grain—and to the period of occurrence of the different limiting factors: sowing to heading (sw-h), heading to meiosis (h-m), meiosis to flowering (m-f), flowering to milky stage (f-ml), and milky stage to harvest (ml-hv). We used several variables as indicators of possible yield-limiting factors (high temperature, low radiation levels, water stress, nitrogen nutrition index, and diseases) to identify the limiting factors responsible for yield reduction at each site. Climatic indicators were deduced from the climatic data obtained from the weather station at each site (minimum and maximum temperature, solar radiation, rainfall and Penman's evapotranspiration).

Water deficit (WDT, mm) was determined as the sum of the daily difference between maximal evapotranspiration ETm, and actual evapotranspiration, ETa. ETa was deduced from potential evapotranspiration according to the relationship ETa = kc x ks x ETp, where kc is a coefficient that varies with the stage of the crop and ks is a coefficient that varies with root-zone water content (Gate, 1995). If root-zone water content is equal to the potential root-zone water content, ETa = ETm = kc x ETp (Gate, 1995). ETp was provided by the weather station at each site. Daily root-zone water content was estimated as the difference between the sum of water root-zone content (mm) and rainfall (mm), minus the evapotranspiration of the crop. Because of a lack of measurement, we considered the water content of the root-zone zero at each site on 1 August.

Heat stress (HTT, °Cd) was characterized in cumulative degree-days over 25°C (Gate, 1995; Stone and Nicolas, 1995). Low radiation levels were identified by calculating cumulative daily incident radiation from the beginning of stem elongation to flowering, or during the period around meiosis (± 3d) and during the flowering-maturity period. Nitrogen stress was characterized by estimating nitrogen nutrition index (NNI) at the beginning of stem elongation and at flowering. NNI was estimated as the ratio between the actual nitrogen concentration of aerial shoots and critical nitrogen concentration, calculated from the actual aerial biomass according to the reference curve described by Justes et al. (1994).

The disease scores obtained for each probe genotype and each treatment were used as indicators of disease intensity (1 = low to 9 = high disease level). Three diseases were observed during the flowering—milky stage (f-ml) or during the milky stage—harvest period: brown rust (Br), yellow rust (Yr), and powdery mildew (Pm).

Finally, agronomic diagnosis was performed by a stepwise linear regression with SAS software (SAS Institute, 1990), allowing the effect of each factor to be estimated. A factor was considered as limiting if its p value was significant (p < 0.15) in the regression.

On the basis of the results of Przulj and Momcilovic (2001) and Cox et al. (1986), showing that remobilization efficiency is decreased by late nitrogen absorption, mean nitrogen uptake after flowering (NUAF) in each environment was used as an indicator of nitrogen availability after flowering. If NUAF in an environment exceeded the mean NUAF estimated for all the environments (33 kg ha–1), this environment was classified as allowing a high level of N uptake during the grain-filling period.

Statistical Analysis
We investigated the effects of genotype, year, experimental site, treatment, block, and genotype x environment interaction by means of an analysis of variance, with the SAS PROC GLM procedure (SAS Institute, 1990). The SAS PROC REG procedure (SAS Institute, 1990) was used to calculate linear regressions between the amounts of nitrogen remobilized (REMN) and nitrogen uptake at flowering (NUF) in environments with various levels of nitrogen availability before anthesis and with no other limiting factor. We fitted a two-parameter linear function to the data for the relationship between the amount of nitrogen remobilized and nitrogen uptake at flowering. Parameters were estimated with the least squares method. We investigated whether the inclusion of a genotypic coefficient improved the relationship by comparing the residues of the model adjusted for each genotype with the residues of the general regression model, using Fisher's test, as described by Borel et al. (1997):

{Sigma}SSi (sum of the residual sums of squares for individual fit to each genotype) was compared with SSc (residual sum of squares for common fit to all the genotypes) as follows:

which follows Fisher's law with (n – 1).k and (Ndata–k) degrees of freedom. Ndata is the total number of data points, n is the number of individual regressions and k is the number of fitted parameters for each regression (two in the case of a linear function).

The validity of the linear relationship between the amount of nitrogen remobilized and nitrogen uptake at flowering for situations characterized by various limiting factors, was assessed by the root mean squared error of prediction (RMSEP) (Wallach and Goffinet, 1987).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Characterization of the Environments
Environments differed in the nature and timing of the limiting factors identified. According to the nature of the limiting factors occurring during the nitrogen remobilization period, i.e. the grain filling period, environments were assigned into four groups. Considering the nature of the limiting factors, an environment could be assigned to one or more groups.

We identified environments with the limiting factors "heat stress and/or drought stress" during the grain-filling period (C1in, C2in, C2nr, D1in, D1nr, M1in, M1nr, R1in, R1nr, R2in, R2nr, T1in, T1nr, T2in, and T2nr). Most of these environments are located in the south and east of France, where high temperatures and drought frequently occur during the postanthesis period (Table 5). We also identified environments in which disease occurred during the grain-filling period, mostly in cases in which no fungicide was applied or the fungicide applied was inefficient (C1in, C1nr, D1in, L1nt, L2nt, M1nt, R1nt, R2nt, S2nt, T1in, and T1nr) (Table 5). The main diseases observed were brown rust and yellow rust. We identified environments for which no limiting factor occurred during the grain-filling period (G2in, G2nr, G2nn, G1nr, G1lw, G1nn, and S2in). In these environments, in which nitrogen availability varied during the preanthesis period from low (G1nn, G2nn) to high nitrogen (G2in, S2in), potential remobilization should have been maximal in the various genotypes (Table 5).


View this table:
[in this window]
[in a new window]
  		Table 5. Yield losses of the three probe genotypes and main features of the environment identified as limiting factors (NNI: nitrogen nutrition index, WDT: cumulative water deficit, HTT: cumulative high temperature, BR: brown rust and YR: yellow rust). The date of occurrence of each limiting factor was defined as: sw-st: sowing to stem elongation period, st–fl: stem elongation to flowering, st–me stem elongation to meiosis, me–fl: meiosis to flowering, h–fl: heading to flowering, f–ml: flowering to milky stage, ml–hv: milky stage to harvest.

 
Finally, we identified environments with levels of nitrogen uptake during the grain-filling period exceeding 33 kg ha–1 (C1nr, G1in, G1hn, G2hn, L1in, L1nr, L2in, and L2nr). Most of these environments were located in north-central France and in irrigated areas.

Grain Nitrogen Content, Crop Nitrogen Content at Flowering and Amount of Nitrogen Remobilized
The environment x genotype combinations studied here encompassed a wide range of grain protein content [7.6–17.3 g N (100 g)–1 dry wt.], yield (2.7–12.5 Mg ha–1), nitrogen uptake at flowering (32–284 kg ha–1), amount of nitrogen remobilized (24–228 kg ha–1) and nitrogen remobilization efficiency (0.44–0.92). The mean amount of nitrogen remobilized (121 kg ha–1) varied more among environments (28–196 kg ha–1) than among genotypes (104–133 kg ha–1). Variation in mean nitrogen remobilization efficiency (0.72) was also greater among environments (0.60–0.83) than among genotypes (0.70–0.76).

Analysis of variance for the amount of nitrogen remobilized (REMN), nitrogen remobilization efficiency (NRE), and nitrogen uptake at flowering (NUF) made it possible to identify the sources of variation (Table 6). The main source of variation in the amount of nitrogen remobilized, nitrogen uptake at flowering and nitrogen remobilization efficiency was the environment. We therefore divided the "environmental" effect into effects of site, treatment, and year. We found that variations in the amount of nitrogen remobilized, nitrogen remobilization efficiency and nitrogen uptake at flowering mainly depended on treatment, year, site, and site x year interaction. Genotype also had a highly significant effect on the amount of nitrogen remobilized and on nitrogen remobilization efficiency (p < 0.001). The effect of genotype depended on the environment, as shown by the sigificant genotype x environment interaction. Genotype x site and genotype x year interactions accounted for a small but highly significant (p < 0.001) proportion of the variation.


View this table:
[in this window]
[in a new window]
  		Table 6. Mean square data from the analysis of variance for nitrogen uptake at flowering (NUF), amount of nitrogen remobilized (REMN) and nitrogen remobilization efficiency (NRE).

 
The Student Newman and Keuls test (p ≤ 0.05) showed that the amount of nitrogen remobilized and nitrogen uptake at flowering increased with the level of nitrogen fertilization. The amount of nitrogen remobilized was larger in situations in which nitrogen was applied at the regional recommended rates (intensive treatment) and in the absence of fungicide application than in the case of "reduced nitrogen" treatments. Nevertheless, diseases had a significant effect on the amount of nitrogen remobilized, reducing the remobilization of the nitrogen stored at flowering, particularly if disease pressure appeared early in the cycle, with a high level of development during the grain-filling period. Situations favorable for nitrogen uptake before flowering, with no drought stress or nitrogen deficit during the pre-flowering period also resulted in the highest levels of nitrogen remobilization (from 199 kg ha–1 for G2in to 131 kg ha–1 for C1in). Conversely, the lowest levels of nitrogen remobilization were observed in situations with low levels of nitrogen supply (from 115 kg ha–1 at M1nr to 29 kg ha–1 for G1nn) because of low levels of nitrogen uptake at flowering. Nitrogen remobilization efficiency was also influenced by nitrogen fertilization level.

However, in contrast to what was observed for the amount of remobilized nitrogen, nitrogen remobilization efficiency was highest in situations of low nitrogen supply, in most cases without water stress. Nitrogen remobilization efficiency was highly variable among genotypes, varying between 0.84 (DI9714) and 0.82 (Récital) for situations with low levels of nitrogen fertilization and between 0.63 (Oratorio) and 0.56 (Récital) for situations with intensive disease pressure.

Relationship between Nitrogen Remobilization and Nitrogen Uptake at Flowering
We estimated the parameters of the relationship between the amount of nitrogen remobilized and nitrogen uptake at flowering for each of the 10 genotypes in the seven environments with no limiting factor during the grain-filling period. The regression model fitted well the observed points (r2 ≥ 0.92) and the intercepts of the linear regression curves were not significantly different from 0 (Table 7).


View this table:
[in this window]
[in a new window]
  		Table 7. Calculated slope, intercept and statistics for the linear regressions linking amount of nitrogen remobilized to nitrogen uptake at flowering. Relationships were estimated for each genotype and for all the genotypes (general model), for the group of environments without limiting factors during the grain-filling period.

 
The estimated values of the slope ranged from 0.69 ± 0.03 for Renan to 0.82 ± 0.05 for Récital, corresponding to the lowest and highest levels of remobilization, respectively. Arche was the only genotype for which the standard deviation of the slope was greater than 0.10 and the coefficient of variation was greater than 15%, indicating a high level of variation in N remobilization efficiency among the environments concerned. The value of the slope corresponding to all genotypes, in the general model, was estimated at 0.76 ± 0.02 and the general regression model obtained fitted well the observations (r2 = 0.97) (Fig. 1). Taking into account all the genotypes, there was no significant difference in the estimation of nitrogen remobilization between the general regression model and the individual genotypic models . Thus, for environments in which no limiting factor occurred during the grain-filling period, nitrogen remobilization for all the genotypes studied can be estimated with the same two-parameter linear regression equation: REMN (kg.ha–1) = NUF x 0.76 (± 0.02) + 4.13 (± 2.56) (r2 = 0.97, RMSE = 10.74 kg.ha–1).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 1. Relationship between the amount of nitrogen remobilized and nitrogen uptake at flowering estimated for all genotypes in the environments without limiting factors during the grain filling period (experiments in 2001 and 2002).

 
Variability of the Relationship in the Other Environments
For each group of environments characterized by particular limiting factors during the grain-filling period, we compared observed nitrogen remobilization with the values simulated by the single general model previously proposed. We analyzed the effect of genotype and environment on the accuracy of the model by root mean squared error of prediction (RMSEP) for the general model (Table 8).


View this table:
[in this window]
[in a new window]
  		Table 8. Root mean squared error of prediction and bias of the general model for each group of environments: (1) environments with heat stress and drought before anthesis; (2) environments with high postanthesis N uptake; (3) environments with weak and/or late disease pressure and; (4) environments with high disease pressure (see text for the classification of the environments in the various groups). Mean N remobilization efficiency (NRE) is also given for environments with strong disease pressure (4-4).

 
For the environments characterized by heat stress and drought (Group 1), high postanthesis N uptake (Group 2), and low disease intensity (Group 3), the mean RMSEP was close to the RMSE of the general model (10.74 kg ha–1), indicating that the accuracy of the model was not affected by these limiting factors (Table 8). Nevertheless, in environments with high postanthesis N uptake, estimates were good for some genotypes (Arche, Soissons, Renan, and Rumba) but not for others. This was probably because nitrogen remobilization efficiency is highly sensitive to the combined effect of irrigation and late nitrogen supply, as the error of prediction of the model concerned principally G1hn and G2hn (data not shown). In contrast, RMSEP was much higher in environments with strong disease pressure (mean of all genotypes = 24 kg ha–1), indicating that nitrogen remobilization was strongly affected in these conditions. Some simulated values were close to the observed values for these environments, but others were far from the 1:1 line (Fig. 2). The values far from the line corresponded to cases in which the model strongly overestimated the amount of nitrogen remobilized. However, some genotypes had a low RMSEP (Oratorio and Hynoprécia). These genotypes were resistant to the foliar diseases encountered during the trials: mainly brown rust and yellow rust (Table 4). The greatest overestimates obtained with the model concerned genotypes susceptible to these foliar diseases (Fig. 2). However, for some genotypes classified as resistant to the main foliar diseases (such as Renan), high errors of prediction were observed. These errors were due to the specific susceptibility of this variety to leaf blotch, which was observed in 2002. For genotypes well simulated by the model (Oratorio and Hynoprécia), the N remobilization efficiency estimated by the general model (0.76) was close to the mean N remobilization efficiency obtained for the environments considered (respectively 0.73 and 0.71). Even for genotypes for which overestimates were obtained, the calculated N remobilization efficiency was much lower than the estimates generated by the general model.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 2. Expected amount of remobilized N estimated from the general model (determined in Fig. 1) versus observed amount of remobilized N, in environments with high disease pressure. The line corresponds to the 1:1 bisecting line. Disease-resistant genotypes such as Oratorio (ORT) and Hynoprécia (HYP) and susceptible-genotypes such as Récital (REC) and Soissons (SOI) are identified.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
A wide range of nitrogen uptake conditions at flowering and during the grain-filling period for a set of 10 wheat genotypes was created in a multienvironment trial network. These different conditions resulted in differences in nitrogen remobilization efficiency among genotypes and environments. There were significant effects of genotype and environment on nitrogen remobilization efficiency. Environmental factors seemed to be the main sources of variation, with treatment and site having the greatest effect. Nitrogen remobilization from the vegetative parts of the plant was most efficient in situations of low nitrogen supply or low nitrogen availability during the preflowering period, consistent with the results of Papakosta and Garianas (1991) and Cox et al. (1985b). We analyzed the effect of genotype on nitrogen remobilization efficiency, assuming a linear relationship between nitrogen uptake at flowering and nitrogen remobilized to the grain at maturity. We showed that, for environments with no limiting factor during the grain-filling period, nitrogen remobilization could be estimated from a single linear relationship, regardless of genotype. In this kind of environment, the genotype effect on the amount of remobilized nitrogen is mostly accounted for by nitrogen uptake at flowering, as shown by Przulj and Momcilovic (2001) in barley. We also showed that for environments characterized by drought stress, heat stress, or low disease pressure, the relationship between nitrogen uptake at flowering and the amount of nitrogen remobilized was robust and did not seem to depend on genotype.

The relationship between nitrogen uptake at flowering and nitrogen remobilized to the grain is of the form A + B x N uptake before flowering, with the calculated coefficients A = 4.13 (±2.56) and B = 0.76 (±0.02). As we have shown that A was not significantly different from zero, B is likely to be close to the nitrogen remobilization efficiency value. Thus, the stable linear relationship observed in many environments indicates that nitrogen remobilization efficiency is stable across environments and genotypes.

Nevertheless, in cases of high disease pressure or high levels of nitrogen uptake during the grain-filling period, estimation was better for some genotypes than for others. In both cases, the estimation of remobilization retained a high level of accuracy because of genotypic adaptation to the factor considered. In the case of diseases, estimates were accurate for resistant genotypes, such as Oratorio, but overestimates were obtained for susceptible genotypes, such as Récital, demonstrating that the foliar diseases occurring during this 2-yr period decreased nitrogen remobilization efficiency for these kinds of genotypes. This result confirms the results of Dimmock and Gooding (2002) showing a decrease in N translocation because of aerial diseases in wheat and those of Garry et al. (1996) in pea (Pisum sativum L.). It also indicates that the genotype effects observed on NRE are accounted for mostly by a response of genotypes to specific environmental conditions.

In cases of high nitrogen availability after anthesis, the general linear model overestimated N remobilization for some genotypes. In our study, high N availability postanthesis occurred mostly if irrigation was applied after this stage and if high levels of nitrogen were applied at flowering, which is extremely rarely the case in agricultural conditions. This effect was weaker than that of high disease pressure. However, more detailed studies of the N remobilization of various genotypes in cases of high N availability postanthesis are required for the precise definition of the range of validity of the observed stable remobilization efficiency. This effect of the genotype on nitrogen remobilization efficiency could not be linked to a measurable genetic character, such as heading earliness, reflecting the lack of constancy of this effect within this kind of environment. The effect of genotype was highly dependent on the environment considered. Nevertheless, we tested only a small range of genotypes, and our results require confirmation with a larger range of genotypes, particularly genotypes that seem to be adapted to low levels of nitrogen, as described by Le Gouis et al. (2000).


   ACKNOWLEDGMENTS
 
We wish to thank P. Bérard (INRA Clermont-Ferrand), J. Troizier (INRA Grignon), S. Gilles and O. Gardet (INRA Le Moulon), M. Méosoone (SERASEM Lille), D. Béghin and D. Bouthor (INRA Mons), M. Trottet and J.Y. Morlais (INRA Rennes) and P. Bataillon (INRA Toulouse) and their collaborators for carrying out the experiments. We also thank F. Lafouge for sample analysis and B. Lefouillen, G. Grandeau, V. Tanneau, and C. Souin for technical assistance. We are thankful to Semences de France, SERASEM, Florimond-Desprez and Arvalis–Institut du végétal for financial support, and Alex Edelman & Associates for revision of the English version of this paper.

Received for publication August 4, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 


Related articles in Crop Science:

THIS ISSUE IN CROP SCIENCE

Crop Science 2005 45: xiii. [Full Text]  



This article has been cited by other articles:


Home page
 ANN BOT (LOND)Home page
P. Bancal
Decorrelating source and sink determinism of nitrogen remobilization during grain filling in wheat
Ann. Bot., June 1, 2009; 103(8): 1315 - 1324.
[Abstract] [Full Text] [PDF]

Home page
 ANN BOT (LOND)Home page
M.-O. Bancal, R. Roche, and P. Bancal
Late Foliar Diseases in Wheat Crops Decrease Nitrogen Yield Through N Uptake Rather than Through Variations in N Remobilization
Ann. Bot., October 1, 2008; 102(4): 579 - 590.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
S. Muurinen, J. Kleemola, and P. Peltonen-Sainio
Accumulation and Translocation of Nitrogen in Spring Cereal Cultivars Differing in Nitrogen Use Efficiency
Agron. J., February 6, 2007; 99(2): 441 - 449.
[Abstract] [Full Text] [PDF]

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 Related articles in Crop Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (25)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barbottin, A.
Right arrow Articles by Jeuffroy, M.-H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Barbottin, A.
Right arrow Articles by Jeuffroy, M.-H.
Agricola
Right arrow Articles by Barbottin, A.
Right arrow Articles by Jeuffroy, M.-H.
Related Collections
Right arrow Nitrogen
Right arrow Wheat
Right arrow Plant and Environment Interactions

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal 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